This is the unperfected transcript of Day 21 of the Horizon trial. Mr Patrick Green QC for the claimants presented his closing submission to the judge.
For every single word of what was said today, read on:
Monday, 1 July 2019
(10.35 am)
Closing submissions by MR GREEN
MR JUSTICE FRASER: Good morning.
MR GREEN: My Lord, I think it is me.
The focus of the claimants' submissions will be to
identify particular aspects of the way that Post Office
seeks to invite the court to determine the issues
against the background of what we hope is sort of
reasonably detailed treatment in our closing submissions
and to identify why we find ourselves in respectful
disagreement with the way Post Office seeks to do that.
And in particular, to identify to the court various
facets of the sort of overarching theme of the way
Post Office puts its case, which is, we respectfully
submit, a variant of the rather one-way approach that
was urged on the court in the Common Issues trial in
three particular respects.
First --
MR JUSTICE FRASER: Just pause. I'm afraid I have been
given a broken chair which does sometimes happen, but we
have a spare. Thank you.
MR GREEN: In three particular respects. Firstly,
Post Office effectively seeks to reframe the issues as
they have been agreed and ordered by the court. We
respectfully say that is the wrong approach. Secondly,
on analysis, the way Post Office has sought to adduce
its own evidence is effectively the same one-way
ratchet, for reasons that I will explain. And thirdly,
the way Post Office invites the court to treat the
evidence in the documents is also, we respectfully say,
oddly one-way.
And the short point, if I may, on point 2, which is
how they called their evidence, is simply this: that as
the court will have seen from our written closing
submissions and our written openings, effectively,
a common feature of much of Post Office's evidence is to
call Ms A to talk about what Mr B either knows or has
heard from Mr X, Ms Y and Ms Z, who are often not even
identified.
There has been six weeks of almost Where's Wally in
relation to Mr Gareth Jenkins who persistently pops up
in the source material for Post Office's evidence and
yet has not been called.
We respectfully say that the explanation for that is
unsatisfactory, which we will come to. But it is the
effect, it is the one-way effect of the way the evidence
has been presented to the court which is the focus of
this submission, and that one-way effect is essentially
this. If you call people who stand up and put their
name to an account of what generally the position is and
sign a statement of truth but they happen not to know
anything about the subject or not to know anything
reliable about it, the court on Post Office's approach
is presented with the following choice: (1) accept their
evidence; (2) if they are challenged on it, they get to
say "Well, I'm very sorry, I don't know about it". So
effectively Post Office seeks to be in a neutral
position if their evidence is rejected and a positive
position if their evidence is accepted.
We say that is the wrong approach. We also say that
the explanations for the way this is done are
underwhelming at best, and that goes to both the weight
to be attributed to Post Office's evidence throughout
and invites the court carefully to consider
Post Office's approach to adducing evidence in this
trial.
Briefly by way of introduction on point 3, which is
the way the Post Office invites the court to treat the
evidence and the documents, as the court will see as we
go through the material in a minute, there are some
frankly astonishing invitations to the court to construe
contemporaneous documents against what they say on their
face and to disavow witnesses where their evidence goes
badly. So Mr Godeseth, yes, we shouldn't have called
him and some of his evidence did require clarification
or correction, which then comes in correspondence after
he has given evidence. We say that is an unusual
approach at best.
Furthermore, a narrative of what happened in the
fact evidence that is very difficult to reconcile with
the view of someone who was actually present during the
trial. I appreciate that is a fairly strong pitch to
put it at, my Lord, and I propose to make that
submission good in a moment.
Can I just draw your Lordship's attention to
something else I'm going to come onto a bit later, but
just by way of example if your Lordship turns to
Post Office's closing submissions at page 134 at
paragraph 369.1 {A/6/134}, your Lordship will see:
"The Post Office documents were not drafted with the
benefit of the vast amount of work that has been carried
out by the experts for this trial. If the authors
considered that Horizon was not a good system, they were
wrong (although that is not even a fair summary of what
the documents say)."
Now, firstly there is the astonishing submission
that Post Office's own internal admissions as to the
shortcomings of the system by their own internal
employees and officers and directors should be ignored
and treated as wrong. Secondly, there's the premise in
that that those documents were not drafted with the
benefit of the vast amount of the work that has been
carried out by the experts for the trial.
Let's throw that back to the cross-examination of
Dr Worden who, when he was shown those documents about
Post Office's internal concerns, hadn't even read them.
So not only is that submission fatally flawed, but it
actually draws to the court's attention the documents
which Dr Worden did and did not take into account in
forming the views that he did. And the court will
remember his answers in cross-examination, "Do you
accept that that cannot be reconciled with the view that
you formed?", and him saying "Yes".
My Lord, I give that as just one example -- I was
going to come to it later -- of the approach that
Post Office invites the court to take in relation to
even Post Office's own documents, and we will see that
writ large.
Before I turn to this fact and evidence and
documents piece which is essentially those layers, 2 and
3, can I just invite the court to consider for a moment
the way in which the reframing of the issues has been
done by Post Office.
MR JUSTICE FRASER: This is your point 1, is it?
MR GREEN: Point 1.
MR JUSTICE FRASER: Yes.
MR GREEN: If I could just take the court back to the
Horizon Issues at {C1/1/1} and just take, by way of
totemic example, issue 1. And the issue itself says:
"To what extent was it possible or likely for bugs,
errors or defects of the nature alleged at [paras] 23
... [in the pleadings] ... to have the potential to ...
cause apparent or alleged discrepancies or shortfalls
relating to Subpostmasters’ branch accounts or
transactions, or (b) undermine the reliability of
Horizon accurately to process and to record transactions
..."
Now, pausing there, there are probably three layers
to the way in which certainly Dr Worden reframed them
and the Post Office off the back of that seeks to
reframe issue 1.
The first is that Dr Worden confined his analysis on
issue 1 by ignoring potential. Secondly, he added
a gloss to actual by saying that those should be
lasting, not transient, and when your Lordship asked him
about that, anything that was corrected ever, no matter
how long that took, is transient on his approach. We
respectfully submit it is a wrong approach.
The third wrong part of the approach adopted on
Post Office's behalf and reflected in their closing
submissions is the word "relating", three lines down,
about two-thirds of the way across:
"... discrepancies or shortfalls relating to
Subpostmasters' branch accounts or transactions ..."
Because what Post Office seeks to do is to say that
it has to be in the subpostmasters' branch account. So,
therefore, that discounts a situation if one follows
that to its logical conclusion where the initial figure
on Horizon is correct but the bug causes a different
figure in POLSAP or one of Post Office's other systems
from which a TC might subsequently be issued. We also
say that's the wrong approach.
We flagged this up in opening and we have followed
it up -- I think it is probably quicker to go to our
closings -- and we have identified at page {A/5/173} at
paragraph 500 -- in relation to the contrasting expert
approaches we submit that Mr Coyne correctly focused on
the issues as defined in his reports and was careful not
to draw conclusions about actual consequences where
there was no actual evidence about them but merely to
identify the potential for the effect that crystallised
where the evidence lay. Whereas, by contrast, we made
these points, my Lord, in relation to the lasting and
permanent matters already identified here.
Those are expanded and treated together in the
appendix at -- actually, while we are here can we go
forward to 562, which is at {A/5/198}.
This is the reframing by Dr Worden of Issue 2 which
he formulates as:
"Issue 2 appears to be asking - could Post Office
have given its Subpostmasters automated support in
Horizon, in the place of human support?"
To which he answers no. That isn't even
an available reading of Issue 2), let alone one which
ought to have been adopted by Dr Worden in that respect.
Issue 2 is perfectly clear on its face.
The gloss there of "in place of human support", for
example, speaks to a theme in Dr Worden's approach which
is then ultimately reflected to greater or lesser extent
in Post Office's submissions of changing the issues in
a way that favours Post Office.
When we go back to Issue 1, if we look at the
transcript at {Day18/156:2}, if we look at lines 2 to 6,
your Lordship will see:
"Question: And you describe section 8 as addressing
Horizon 1, yes?
"Answer: Yes.
"Question: "... the extent to which bugs in Horizon
may have affected the Claimants' branch accounts"?
"Answer: Yes, I've stuck in the word "claimants"
there."
So there is another feature which is not a feature
of the issue as the court defined it, and as had been
agreed by the parties for good reason, which Dr Worden
has feathered in. And your Lordship will also remember
how Mr Coyne was challenged about that and cautioned by
my learned friend about the suggestion that interest in
individual claimants had been rebuffed and we then saw
the RFI response to, which was to rebuff any enquiry
about individual claimants.
So there is a theme of reframing the issues in a way
that makes it harder for the claimants to succeed upon
them and comes close, the more one reframes them, to
a position where they are impossible for the claimants
to win. We see that with Dr Worden's assumptions in
relation to the issuing of transaction corrections,
which he says, he acknowledges, are outside the system,
he acknowledges that he doesn't know anything about
business processes and how they are done, and then he
draws inferences on two important premises.
The first premise is that even where documentary
evidence doesn't say that a transaction correction will
be issued and where there's no actual transaction
correction showing it was done, he infers that it would
be done. And he does that -- we will come to this in
passing in relation to the documents -- by saying where
a KEL or a PEAK refers to a TC or error notice needing
to be issued, he assumes that it is issued, and where it
doesn't, he assumes that's because it is written by
people who knew that they didn't need to mention that
because they knew it would happen.
So whatever the document says, it leads him to only
one conclusion and precludes the possibility of any
conclusion that might let in any justifiable complaint
about that by any claimant.
The second premise is that the TC procedure is done
so close to perfectly that the opportunity for error is
vanishingly small, and that's the second assumption that
he builds in. And then he builds all of that into his
version of the system as defined, which conflicts with
the definition of the system at the top of the page of
the Horizon Issues, and from that goes on to reach his
conclusions. And we respectfully say that is not the
right approach, and Mr Coyne took an approach which was
correct in his reports in relation to the issues as
defined.
I will deal with the issue of robustness in relation
to Issue 3 very briefly, reasonably briefly. There is
first a point of principle which is that the way --
robustness is the Post Office's favourite issue, that's
clear, something they insisted on and we agreed to and
was ordered. What it doesn't say is that the system is
robust? And then a second question: is it extremely
unlikely to be the cause of shortfalls in branches?
It was conjunctive in its formulation, it was not
split into subparagraphs like some of the others, and we
treated it as one issue. And on that basis and because
Post Office wants to conflate the word "robust", which
is in inverted commas in the issue as defined, with
extreme unlikelihood to be the cause of shortfalls in
branches, that is precisely the basis upon which the
claimants do not accept Post Office's analysis on that
issue, because they do not accept that it is robust in
the sense of being extremely unlikely to be the cause of
shortfalls in branches.
My Lord, there is an important point of drafting in
relation to Issue 3, which is Issue 3 does not say:
extremely unlikely to cause shortfalls in branches. So
it is not contemplating what proportion of 47 million
transactions a week would have errors caused in them.
What it actually contemplates is a situation where you
have an identified shortfall in a branch and the
question then is if you do have such a shortfall in
a branch, is it extremely unlikely that Horizon is the
cause of that shortfall in the branch?
We respectfully say no, it is not, or not
functionally so for the purposes of this dispute, and
particularly not so where the type of shortfall and
discrepancy that in reality is in issue in this case is
the unexplainable; the one where the subpostmaster
doesn't go "Oh yeah, that's probably Jeff remmed in too
many stamps there, yes, I remember he was a bit
distracted this morning" or "Oh I can see from the till
roll something has gone wrong". We are not talking
about that.
The focus of this litigation, as reflected also in
the Common Issues trial, is on what is the situation
where there is an unexplained shortfall that the
subpostmaster finds it difficult to get to the
bottom of.
So that is the background to those two facets of
Issue 3: robustness and extremely unlikely being part of
one concept; and secondly, the premise is the cause of
the shortfall in a branch.
MR JUSTICE FRASER: Where do I go to to find your benchmark
definition of robustness?
MR GREEN: My Lord --
MR JUSTICE FRASER: The claimants' benchmark definition.
MR GREEN: We found it difficult to define other than by the
extremely unlikelihood of causing shortfalls. So we
haven't got a definition that we advance as a meaningful
and satisfactory one for the purposes of this case. And
save insofar as it appears to be a correlation of
extremely unlikely to be the cause of shortfalls in
branches.
So that --
MR JUSTICE FRASER: So you are treating it as a summary term
that includes and consists of the second part of
Issue 3?
MR GREEN: That appears to be the only sensible objective
way of construing it, we say.
My Lord, it has never been the claimants' case that
there weren't numerous aspects in which Horizon is
relatively robust, and that is a point on which we
respectfully submit Post Office's closing rather
overshoots in the criticisms made of the position taken
by the claimants in the reply, which I would just like
to take your Lordship to, if I may, at paragraphs 36
and 37, which is {C3/4/21}.
So if we just start under "Operation of Horizon",
A.8 heading. 34:
"The Defendant admits that 'Horizon is not perfect'
... and that 'like all other IT systems, Horizon is not
a perfect system which has never had any errors or bugs'
..."
MR JUSTICE FRASER: Which document is this?
MR GREEN: So this is our reply, my Lord.
MR JUSTICE FRASER: Your reply?
MR GREEN: Yes.
35:
"It is also uncontroversial that the volume of
transactions effected through Horizon is very large."
36:
"Notwithstanding the various checks and controls
alleged (the existence and efficacy of which are not
admitted), the combination of Horizon's admitted
imperfections and the volume of transactions, is
entirely consistent with the levels of errors reflected
in the Claimants' case: the said imperfections are
likely to be the cause of discrepancies or apparent or
alleged shortfalls about which the Claimants complain."
37:
"It is therefore denied that Horizon 'is robust and
[...] is extremely unlikely to be the cause of losses in
branches' (paragraph 16). In fact, the relatively small
chance of errors admitted by the Defendant, would be
likely to produce the very picture reflected in the
Claimants' case."
Then there are observations in relation to the
existence of money in suspense accounts and so forth.
At paragraph 38 {C3/4/22}, the change of public
position was then noted. Then the clarification of the
defendants taking to make use of the pre-action
correspondence then follows at 39 onwards, which is
where the defendants effectively tried to bank the
absence of a systematic flaw across the whole of
Horizon.
We weren't in a position to allege that there was
a systematic flaw affecting everybody as meaning that
there was no claim that anyone had suffered
a Horizon-generated shortfall, which we just wanted to
make absolutely clear on the pleadings was not the case.
Then at paragraph 41 {C3/4/23}:
"As to the allegedly robust controls pleaded
variously ..."
41.1:
"the Defendant's case that ..."
Then it is set out, effectively saying:
" ... when supplemented by the various accounting
and cash controls applied in branches, make it very
unlikely indeed that an error in Horizon could affect a
Subpostmaster's financial position and go undetected'
..."
We say that:
"... implicitly accepts that Horizon is likely to be
the cause of discrepancies or apparent or alleged
shortfalls in branches, unless detected by Claimants."
Then they failed to plead to any material variations
because they only pleaded on the current version of
Horizon as it stood at the date of defence. So we
didn't know what the case was there.
Then, the defendant was put to strict proof to
matters which lay within their knowledge but not ours at
41.3. And at 41.4, {C3/4/24}:
" ... in respect of the alleged 'robust measures in
place in Horizon ... The Claimant denies that these
were effective, including on the basis that two of three
errors or bugs admitted by the Defendant (in Schedule 6
to the Letter of Response, referred to at paragraph 56
of the Defence) were not identified through Horizon's
own in-built checks and balances designed to identify
the same ..."
Then there is a passage about the position in
relation to the known error log, which we will come back
to. Just to give the court an indication of where we
were at this stage pleadings-wise, could I go, please,
to the generic defence at {C3/3/22}. This is
paragraph 50 from the previous page and it is primarily,
at this stage, addressing the known error log. The
second line of that paragraph says --
MR JUSTICE FRASER: Which paragraph?
MR GREEN: (4). I'm so sorry, I didn't say that. 50(4),
which is on {C3/3/22}, the second line of that
paragraph:
"To the best of Post Office's information and
belief, the Known Error Log is a knowledge base document
used by Fujitsu which explains how to deal with, or work
around, minor issues that can sometimes arise in Horizon
for which (often because of their triviality)
system-wide fixes have not been developed and
implemented. It is not a record of software coding
errors or bugs for which system-wide fixes have been
developed and implemented. To the best of Post Office's
knowledge and belief, there is no issue in the Known
Error Log that could affect the accuracy of a branch's
accounts or the secure transmission and storage of
transaction data."
So, my Lord, I just want your Lordship to have
a stake in the ground as to where we were pleadings-wise
at that point, in circumstances where Dr Worden's first
approach was to confine his analysis almost entirely but
not exclusively to KELs to identify where the bugs were
in Horizon that may have an impact on branch accounts.
So it is in that context that the reply is then pleaded
in that way in relation to robustness. And seeking to
make the distinction, ultimately reflected in the way
the Horizon system is defined for this trial, which is
slightly narrower, as the system itself. And then of
course Dr Worden is the one who says the subpostmasters
themselves form part of the countermeasures by manual
inspection of data MID and user error correction in his
analysis.
MR JUSTICE FRASER: So what's the stake in the ground? I'm
not --
MR GREEN: The stake in the ground is what Post Office were
saying about the existence or knowledge of information
about bugs at that stage by way of context to us.
So they are saying that the context in which we were
trying to plead was basically blindfolded and
Post Office was very keen for us to plead particulars of
things which a striking feature of this case is that the
subpostmasters did not know often about them at all and
were often not told what the underlying root cause of
things was even when, in Mrs Burke's case, for example,
it was not her fault at all and she gets a TC that says
"please take more care in the future". Totally a system
fault.
So at this stage we are pleading completely
blindfold to this, and Post Office are positively saying
that to the best of Post Office's knowledge and belief
there's no issue in the known error log that could
affect the accuracy of a branch's accounts or secure
transmission and storage of transaction data.
It is easy when one has been in a trial this long,
and it has been interrupted, to lose sight of the
utility of the listing of this trial given where we have
come from and where we are now in terms of the evidence
available for the court and indeed to the claimants.
I just wanted to put that stake in the ground to
give it at least a small indication of the distance of
travel. We have obviously mentioned the distance of
travel in relation to remote access in our closing
submissions, which is the position is completely
transformed. And it is also obviously apparent that
Post Office did not disclose in their pre-action
correspondence the existence of the Dalmellington bug at
all, notwithstanding that it affected 88 branches and
there were 112 known instances of it occurring over
a period of about five years.
That was left for Mr Coyne to stumble upon. There
was no disclosure of the APPSUP role, which the court
has now seen what a huge back door it was into the
system, and how the logs were not either kept or kept in
such a way as to allow there to be any sensible analysis
of how that privileged role had been used.
So it has transformed out of all recognition in
terms of the claimants' and the court's understanding of
remote access. None of that would have happened had the
court not ordered this trial to take place and
essentially the process of preparing for and undertaking
the trial actually happened as it has.
My Lord, just briefly in relation to robustness,
there is also the document at {C1/2/4}, which was our
provisional outline document, which the court will
remember Post Office were very keen for us to provide
but less keen to be forthcoming themselves about what
they knew to be the problems with the system.
So this provisional document was of course put
together when we still didn't know about the
Dalmellington bug:
"There's no single, authoritative definition of
'robust' in the context of a system such as Horizon,
which appears to be more commonly used in public
relations than as an objective performance standard.
Pending any clarification of its objective meaning, it
appears to relate to the ability of a system to perform
correctly in any scenario, including where invalid
inputs are introduced, namely, to have in place
effective error repellency."
{C1/2/5}:
"It is clear that, on a sensible construction of the
term 'robust', Horizon did not neat this standard
because:
"(a) it contained bugs, errors and defects as set
out in paragraph 1 above which created discrepancies in
the branch accounts of Subpostmasters;
"(b) it suffered failures of internal mechanisms
which were intended to ensure integrity of data."
There is a KEL in relation to recovery failures
there:
"(c) the system did not enable such discrepancies to
be detected, accurately identified and/or recorded
either reliably, consistently or at all;
"(d) the system did not reliably identify
'Mis-keying', which is inevitable in any system with
user input, and did not reliably have in place
functionality to restrict users from progressing a
mis-key;
"(e) ... numerous processes and workarounds ...
weaknesses and risks of errors ..."
And so forth.
That goes all the way down to paragraph 3.3 over the
page {C1/2/6}:
"However, as noted above, whether Horizon is
'robust' plainly depends upon the definition given to
that word. Even the small chance of errors, bugs and
defects admitted by the Defendant and/or supplemented by
those underlying the reports above, would be likely to
produce the result alleged by the Claimants. Therefore,
as noted above, even if the overall probability of a
bugs or errors affecting Branch accounts is small, the
sheer number of transactions undertaken by the Horizon
system is consistent with the level of discrepancies
arising from bugs and errors in issue in these
proceedings."
So, my Lord, that dovetails with those paragraphs in
the reply to which I have already taken the court.
So that is as far as we were able to go in relation
to robust. But we say for functional purposes it isn't
robust insofar as it does all of those things and those
things create a small but real chance of discrepancies
in relation to subpostmasters' branch accounts or their
transactions.
My Lord, for your note, just in relation to the
KELs, if we may. It is right that your Lordship should
know that at paragraph 29 in the Post Office's closing
submissions at {A/6/16}, Post Office suggests that it
has taken a co-operative approach to disclosure. For
reasons which will be already largely clear to the
court, we do not accept that is correct, and I will just
give the court two examples.
In our opening, claimants' written opening, which is
at {A/1/106}.
MR JUSTICE FRASER: This is your opening, is it?
MR GREEN: The claimants' opening.
MR JUSTICE FRASER: Yes.
MR GREEN: And we devoted an appendix, which is referred to
in the body, to looking at what the claimants had to go
through to get the KELs, given how large they loomed in
the trial. And at that appendix at page 106, just to
ask the court to turn to it for a moment in the opening,
the claimants first sought disclosure of the known error
logs on 28th April 2016 in the letter of claim, and they
said:
"We understand that Fujitsu maintained a 'Known
Error Log' for Horizon and that such reports will have
been provided to Post Office."
We thought the known error log was a proper problem
management system identifying all known errors in that
sense. It turns out that, in fact, the evidence of this
is much more atomised across the documentary landscape,
but we were expecting a log of all known errors.
Post Office's letter of response says in effect they
not only refused to provide the documents but denied
their relevance and cast doubt over whether they even
existed.
So Post Office's response said -- it is all
footnoted to the references; I won't take your Lordship
to the underlying documents --:
"In circumstances where you have not particularised
any factual basis on which horizon is defective,
disclosure of these documents (if they exist) is not
relevant reasonable or proportionate."
Leaving that matter at large for us.
We didn't accept that. No good reason to refuse
disclosure.
Then over the page at {A/1/107}, 317:
"Post Office's response on 13th October 2016 ..."
They say:
"The claims which you have particularised concern
the Core Audit Log. Following a review of the Known
Error Log, Fujitsu have confirmed that there have been
no logs in respect of Core Audit Log. The remainder of
the Known Error Log does not relate to the claim which
you have particularised and as such disclosure of this
document is not relevant."
My Lord, I'm not going to take you through it, but
your Lordship will see the history of this all the way
through, including paragraph 324 on page {A/1/109}, the
position of what the known error log was described as
before your Lordship at the CMC in October 2017:
"It contains things like there's a problem with
printers. There's a printer. You have to kick it on
the left-hand side to make the printer work. I mean
there's a vast range of hardware problems of that sort
and maybe some software problems (inaudible) but not the
kind of bugs, errors and defects that the claimants are
wishing to pursue in their particulars of claim so far
as Post Office is aware."
There was a sort of Nelsonian approach to the
obvious relevance of the known error log and a way of
presenting the known error log which we say is (a) far
from forthcoming, and (b) completely irreconcilable with
the claim that Post Office has been co-operative in
relation to disclosure.
My Lord, parenthetically they also queried the
relevance of OCPs and OCRs as well. I will give
your Lordship a reference to that later on. So for
those and other reasons that I will develop later on, we
do not accept that is the case; in fact, quite the
opposite. We ask the court to take note of Mr Coyne's
answer in re-examination that he had never encountered
a case like this where disclosure had come out in this
way in his entire career.
MR JUSTICE FRASER: Well, the Post Office's closing
submissions make a point in paragraph 1145 on page 377
that you haven't -- well, the suggestion is that
complaints about disclosure have emanated from the court
and that you haven't made any applications for specific
disclosure and you haven't advanced any complaint that
particular disclosure orders haven't been complied with.
MR GREEN: Well, my Lord, taking it in stages, that rather
sidesteps the issue that I have been making submissions
on. The issue that I have been making submissions on is
the manner in which Post Office -- we got clear examples
of documents that should have been disclosed; there is
a very recent one which they kept until May. They say
they intended to disclose it in March, for example. So
there are clear examples of things where they haven't.
MR JUSTICE FRASER: Did they say they intended to disclose
it in March or did they say they discovered it in March?
MR GREEN: There was one which was discovered in March which
they intended to disclose and erroneously forgot to.
MR JUSTICE FRASER: Where's that?
MR GREEN: I will get the reference to that in a minute, but
just by way of example. There is a whole pantheon of
disclosure issues in relation to which the claimants
have had to try to take a -- we simply could not have
issued applications on every possible complaint, not
least because the proportions approach, which the court
understandably took given what the court was being told
by Post Office, was to order focused disclosure and then
allow us to look at that, and then from there to then
ask sequentially for what we could then see would be
relevant.
On the basis of what the court was told, that was
plainly, on the face of it, the right way to approach
proportionality in disclosure. The point that I was
seeking to address here was the anterior point of when
we say the documents which were almost the sole source
for Dr Worden's first report, when we raised them in
2016, your Lordship has seen what was said about them
all the way through in the CMC as if they were of
absolutely marginal relevance.
If we had been given an honest answer, I say
"honest" carefully --
MR JUSTICE FRASER: I was about to pick you up on that.
MR GREEN: Yes. My Lord, the reason I say honest is this.
In a Nelsonian sense, to tell the court something where
you have made no enquiry to ensure that you are getting
a fair picture, is, we say, extremely unsatisfactory.
MR JUSTICE FRASER: Well, you have to be very careful about
making submissions that include effectively a criticism
of lack of honesty, don't you?
MR GREEN: My Lord, that is right. I'm trying to make it
clear what I'm saying and I will come back to that later
on. But there is a picture across the piece of --
I mean, perhaps I shall recede to "less than
forthcoming" for the moment.
MR JUSTICE FRASER: I think that would probably be best.
MR GREEN: Because it saves an argument about whether it
rises to that threshold or not. But there is a repeated
picture of the court being told things which turn out
not to be correct. And the explanation is, so we look
at Angela Van Den Bogerd's evidence which we will come
to in a minute, where she gives her evidence. She has
not looked at the underlying documents, which tends to
suggest it is not right, and she is specifically asked
by the court: did you look at those documents before you
made your witness statement or not? She says no.
So we have this constant picture of a situation in
which the implicit suggestion that one might expect to
be being made when a senior person gives evidence, for
example, that they have taken care to make sure what
they are saying is right, turns out not to be correct.
So my Lord, I'm happy to rest on "less than
forthcoming" for present purposes, and for present
purposes I'm going to withdraw the reference to had they
given an honest answer. What I shall say is had we been
given a correct answer which any reasonable enquiry
would have revealed, had we been given a correct answer
which any reasonable enquiry would have revealed in
2016, the proper thing for Post Office to do would have
been to provide disclosure at that stage of relevant
KELs that disclosed the existence of discrepancy of
bugs, defects and issues in Horizon which could give
rise to branch discrepancies at that stage.
That's what should have happened. So in terms of
where we were and what we were able to get later on, we
are about two years late getting things, compared to
where we should have been had Post Office given correct
information in responding to our enquiries, and from
where we would have been were we able to show our expert
those documents at a much earlier stage and not have the
situation which Mr Coyne found himself in in this case
with thousands of documents coming in a matter of days
before his second report, for example. So it has been
extremely unsatisfactory.
My Lord, in our closing at page {A/5/19}.
MR JUSTICE FRASER: Yes.
MR GREEN: -- which is {A/5/19} for Opus purposes,
paragraph 57. The PEAK at 1848.8.2 --
MR JUSTICE FRASER: Yes.
MR GREEN: -- was disclosed and was a document:
"... adverse to Post Office and would have been
directly relevant to the evidence which Mrs Van Den
Bogerd had given in her witness statement relating to
Drop & Go."
MR JUSTICE FRASER: Yes.
MR GREEN: "[It] also contained evidence of keystrokes which
were plainly available to the Post Office from Fujitsu
..."
Your Lordship will remember that document coming up
at the end:
" ... a point on which Post Office’s witnesses had
given very odd evidence (addressed further below). It
would therefore have been an obviously relevant document
for the Claimants to have referred to at that stage, had
it been disclosed to them. As it bore directly on
Issue 8, it is even more surprising that it was not
disclosed earlier..."
Because Issue 8 is about what information was
available to Post Office.
And we had the absolutely bizarre spectacle of
Mrs Van Den Bogerd accepting that keystrokes would be
provided in a Credence report, because Mr Patny, who was
accused of dishonesty, and your Lordship is very right
to point out the seriousness of alleging dishonesty,
Mr Aakash Patny was accused of dishonesty in
circumstances where he had repeatedly chased for
a Credence report because he wanted to find out what had
happened at a stage where Post Office internally
appeared to think that keystrokes would be provided with
that.
We have addressed that in relation to
Mr Aakash Patny and the assault on his honesty in our
closing submissions, but that is the context in relation
to which this keystrokes document would have been
extremely relevant. And then, if your Lordship looks at
paragraph 58:
"The explanation for the timing of the disclosure of
this PEAK given in the 18th witness statement of
Mr Parsons does little to improve the position or
assuage any concerns; it reveals that Post Office waited
2 months from discovering this PEAK on 29 March 2019 ...
before disclosing it to the Claimants, just before the
resumed hearing."
59:
"Further, despite the excuse given for this 2 month
delay being that 'For the next couple of months my firm
continued to investigate these bugs and a number of
further documents relating to these bugs came to light',
in fact, on 24 June 2019 at 17.42 ..."
In the evening:
"... [Wombles] provided yet further disclosure in
relation to this bug and email chain. In response to
a request for an explanation as to why this document had
not been disclosed before, [Wombles] said that the
document had been provided to it on 10 May 2019 was
intended to be disclosed on 31 May ..."
I think that should be March. May:
"... was intended to be disclosed on 31 May 2019,
but due to an oversight was not disclosed on that date
[in May]."
So there are examples in relation to that, but
there's also, my Lord, the other anterior feature of
Post Office's approach which is we were not able to
request documents which related to the Dalmellington bug
because the tenor of Post Office's letter of response
was such as to suggest that there had been three bugs,
which were the ones that Second Sight had identified,
and there obviously must be a possibility of other bugs.
In circumstances where, between the letter of claim
and the letter of response, there was the email saying
"stop investigating the Dalmellington bug", that is
a deliberate decision to stop investigating that bug at
that stage. And we respectfully submit that the only
available inference is it was a deliberate decision not
to mention the Dalmellington bug in the letter of
response.
We respectfully say that that is reaching the higher
echelons of being less than forthcoming. Similarly, the
death of a thousand cuts in relation to remote access.
Given that we know from the Paula Vennells email that
she specifically made an enquiry as to what the position
was, the way that evidence emerged is extremely
unsatisfactory, particularly because Mr Parker's first
round was to say Mr Roll's account was misleading;
positively to say that Mr Roll's account of remote
access was misleading, when (a) it wasn't, and (b) he
was right.
So, my Lord, with that general introduction can
I turn now to the approach in relation to fact evidence.
One of the points that the defendants/Post Office takes
against the claimants is to complain about the claimants
calling fact evidence from a handful of SPMs, and that
begins in their closings at {A/6/299} at paragraph 918.
We say it is a bad point to say the least. Briefly,
contrary to Post Office's approach before the Common
Issues trial, Post Office did not make any application
to strike out that witness evidence. Secondly, at the
pre-trial review my learned friend referred to his
anxiety on Post Office's part about that, which we can
see. The relevant part of the transcript is
{C8.14/3/13} at letter B. We see at letter B my learned
friend:
"I can tell your Lordship that a huge amount of
anxious thought went into that question. It was in the
aftermath of your Lordship’s strike-out judgment. The
reply that came from Freeths was, 'In the light of the
strike-out judgment we are not changing anything, we are
not changing course"
That's his suggestion:
"Given the anxiety, frankly, the fifth on this on
the part of Post Office --"
MR DE GARR ROBINSON: My recollection is that I said
"fearful".
MR JUSTICE FRASER: Fearful rather than what, sorry?
MR DE GARR ROBINSON: "... fifth on this on the part of
Post Office --"
MR JUSTICE FRASER: It should say:
"Given the anxiety, frankly, the --
MR GREEN: The fearfulness.
MR JUSTICE FRASER: Right. Thank you very much.
MR GREEN: Then your Lordship then explained to my learned
friend and the court at D:
"What I said ... in that judgment, I thought which
was obviously reserved and drafted, I applied the test
set down most usefully by Mann J about striking out
evidence of fact and what approach one takes to
relevance. You have 12 witnesses of fact for, let’s
call it, round two of Horizon. Mr Green has umpteen
witnesses of fact. Most of the witness statements
actually are relatively narrow in compass. There is
obviously more fact than one would necessarily expect
with the phrase 'limited evidence of fact', but I think
between you, you are agreed that it can be dealt with in
a reasonable number of days, and the only real issue
appears to be, given you each have an expert, how
Mr Henderson’s evidence is to be treated if it is to be
treated at all, that is really what it comes down to.
You say he is an expert by the back door."
MR JUSTICE FRASER: The focus at the pre-trial review
I think originally was about Professor McLachlan and
Mr Henderson.
MR GREEN: Originally.
MR JUSTICE FRASER: You said you weren't going to call
Professor McLachlan and I gave a ruling on
Mr Henderson's incorporation of the Second Sight report.
MR GREEN: Precisely. My Lord, the short point is that we
ended up with nine of the Post Office's witnesses of
fact giving evidence.
MR JUSTICE FRASER: Is that including Mr Membery?
MR GREEN: Excluding Mr Membery.
MR JUSTICE FRASER: It is really ten, then.
MR GREEN: It is really ten, but nine actually giving live
evidence, plus Mr Membery. And then Mr Roll and
Mr Henderson giving evidence, which was much narrower in
compass than he would have liked to have given, as must
have been apparent to the court. Plus five SPMs.
So Post Office sought to rely on ten witnesses of
fact and the claimants sought to rely on seven plus
Mr Singh, in respect of whom there was a notice to
admit, which I will just show your Lordship later. So
less.
MR JUSTICE FRASER: But the question about evidence is not
a question of counting up how many witnesses are on each
side.
MR GREEN: Your Lordship is quite right, but I just wanted
to clarify. It is not a counting exercise, but those
are the numbers. I will come to the substantive points.
In response to the complaint by Post Office about
this, just imagine what the trial would have looked like
if we had not had the facts of those individual SPM
cases to provide factual grist to the mill of
determining what the systems were and how they worked in
practice.
So the only way ARQ data came out was through the
individual cases of individual SPMs. Failed recoveries
was focused upon specifically, because of SPM evidence,
Ms Van Den Bogerd's response in her responsive witness
statement. It is important to identify that Post Office
hadn't addressed failed recoveries outside the context
of specific SPM evidence.
The claimants then had to prove their account to
contradict suggestions by Post Office of user error
being causative of discrepancies, which was demonstrated
to be wrong. There was a late amendment to Mrs Van Den
Bogerd's second witness statement in relation to
Angela Burke where she had directly sought to suggest
that Angela Burke was at fault, which was wrong. There
was no amendment in relation to Mr Tank's evidence in
Angela Van Den Bogerd's witness statement although it
was ultimately accepted that there wasn't user error
causative of the discrepancy in relation to the receipts
in Mr Tank's case.
Then the Post Office refused the claimants' notice
to admit specifically on the basis of causation of
discrepancies when correctly following the appropriate
course in relation to Mr Singh.
So if I can just take your Lordship very briefly to
that. {H/187/1} is the letter enclosing a notice to
admit, which is mentioned at the bottom. It is
explained at the bottom there:
"As you will see, we firstly invite an admission
that it is possible that a phantom transaction similar
to that described by Mr Setpal Singh at paragraph 6 of
his witness statement dated 28 September 2018 occurred
at the Reddish Post Office in or around the summer
of 2003."
It is a very limited admission because Ms Van Den
Bogerd's submission, as we see over the page, was that
there was insufficient documentary evidence available to
determine whether or not that was the case {H/187/2}.
And that was one admission that was sought on the basis
of what she herself had said in her witness statement.
Then at the bottom of the page, the second admission
that was sought was to invite Post Office to agree, to
admit that it was:
"... aware of occasions where Horizon froze or would
stop working or screens would go black as a result of a
system crash, and discrepancies in accounts did
sometimes arise after SPMs followed, or attempted to
follow the recovery process."
There are two alternatives there. Your Lordship
just briefly will see {H/188/1}. That is the actual
notice to admit, and paragraph 2 deals with that.
Then if we go to {H/200/1}:
"The second alleged fact is drafted in a tendentious
manner and, as drafted, is not true (or even supported
by the evidence that you cite in your letter).
Specifically, there is no evidence cited that
discrepancies arose after system crashes where the SPM
did in fact follow the recovery processes properly, as
opposed to merely attempting to follow those processes."
So they refused to admit something that the court
can now see is true, and Angela Burke's case is
a perfectly good example of exactly that.
See also Mr Tank. So seeing in reality what event
logs actually looked like for specific transactions,
what transaction logs looked like for specific
transactions was necessary, and there was -- how else
were we going to do it?
Again, Post Office would have had a trial in which
it was able to say what would happen or could happen and
what it would normally do and where the claimants would
be unable to effectively analyse or challenge or unpick
those general assertions which were almost exclusively
if not wholly exclusively self-serving.
Another feature of the one-way street approach to
determining the outcome of this trial, my Lord, is in
relation to so-called unchallenged evidence, and this is
quite a big point in the closing submissions and quite
heavy reliance is placed on it, and it has quite a long
history. And your Lordship has repeatedly made clear
that a point of that type would be given short shrift.
My learned friend raised it at the PTR and it was
made perfectly clear what the position was. It was
raised at the PTR, I do not think we need to go it but
I will give the court the references at {C8.14/3/14} at
D to E and {C8.14/3/15} at E to H.
That led to the court's observations at
{C8.14/3/17} --
MR JUSTICE FRASER: Maybe we should go there.
MR GREEN: Shall we go to that part. {C8.14/3/17}, top of
the page, 17, your Lordship to me:
"Mr Green, I am going to make certain observations
to you, which really are for the purposes of both of
you, and that you might find most useful now that you
have stood up. In a way, it is possibly, I suppose, a
misunderstanding as to the way time-limited trials work.
I know this is a QB case but it is being tried in this
building using many of the trial management procedures
that we use in this building, not only in the TCC but
the other specialist courts. In any time-limited trial
any party would get pretty short shrift if they tried to
make a formal point if a particular subject had not been
challenged within the time available. Therefore, it
occurs to me that at least part of the Post Office's
concern arises out of a misapprehension of your
technical ability to do that. Are you aware of your
technical inability to manage such an objective?"
It might originally have been an objection. So it
was absolutely clear, and yet what we then find in the
defendants'/Post Office's closings at {A/6/78} -- I'm
only going to give your Lordship examples, but just to
explore the point generally. We have at paragraph 176:
"Nine PEAKs, limited to two categories, across
17 years does not suggest that the final response codes
were regularly applied incorrectly."
MR JUSTICE FRASER: Where are you looking?
MR GREEN: At paragraph 176, my Lord.
MR JUSTICE FRASER: Of the Post Office's closings?
MR GREEN: Indeed, at page 78.
MR JUSTICE FRASER: Yes.
MR GREEN: So that is the point in relation to the nine
PEAKs.
Then at paragraph 177, the very next paragraph
{A/6/79}, it says this was:
"... limited to only two codes used across the PEAKs
and categorised by Mr Parker, namely ... 70 and 68."
It is actually not correct. It also included 62,
but let's leave that by the by for a moment.
Not only is it generally wrong to take that
approach, but there was, we respectfully say,
a divergence of approach between the claimants and
Post Office during the trial.
The claimants sought, where they could, not just to
give an example of a point but to try to approach points
in a way that the court could fairly regard as
representative. And that was the basis upon which bugs
1 to 10 were cross-examined on rather than cherrypicking
favourite bugs. It is the basis upon which later today
I'm going to invite your Lordship to look at 11 to 15.
So I'm not cherrypicking 17 or 21.
Similarly, in relation to the PEAKs that we dealt
with with Mr Parker and the closure codes, those PEAKs
were PEAKs that were already in play in the trial bundle
for the most part. I'm not aware of any that we
specifically uploaded to demonstrate the absurdness of
the closure codes. And we saw them because we were
looking at PEAKs that were already in play and they were
good examples. But not only that, my Lord, that
I specifically said at {Day12/53:23} in the transcript
that I was not going to be able to go through all of the
examples and specifically explained that I was going to
look at the big categories where there were large
numbers of PEAKs for which that code had been adopted in
the table that Mr Parker was referring to.
So he showed in the table there were thousands of
PEAKs in some categories and we chose to look at the
closure codes where there were the largest number of
closures being affected and where he had not included
those closure codes as having any content in software
errors that Mr Roll might have been working on. That
piece ended at {Day12/86:13} and line 18 with
your Lordship saying:
"I don't think you are going to have time to do any
more of these:
"MR GREEN: I'm not going to do any more of those, my
Lord, I'm cutting it there because -- not least because
the witness accepts that he didn't go through them
before he gave his witness statement."
So there were two points there: (a) I went as far as
I could to try and give a representative impression of
how this had been done; and (b) because it became
absolutely clear that Mr Parker had not gone through
them himself before and also the extent to which he had
used the definitions in the underlying document which
had been put to Mr Roll was seriously in doubt, and he
positively said at one point in his evidence, which we
addressed in our closings, that he actually deliberately
adopted his own definition in one of the categories.
So for those reasons we didn't take that any further
than necessary.
MR JUSTICE FRASER: But Mr de Garr Robinson is entitled to
submit, isn't he, that it wasn't a representative
impression? The fact it is a time limited trial doesn't
stop him making a submission.
MR GREEN: Your Lordship is absolutely right that the fact
that -- he can say it is not representative and put
forward submissions to make that point good. That's
always available. Your Lordship is absolutely right.
MR JUSTICE FRASER: But isn't that what he and his team are
doing in paragraph 177? {A/6/79}
The time limited trial point is that the party is
not permitted to submit that a point is accepted or has
not been challenged, and hence certain conclusions can
be drawn.
MR GREEN: Indeed.
MR JUSTICE FRASER: But that's rather different, isn't it,
from saying it is not a representative exercise?
MR GREEN: Well, my Lord, your Lordship is absolutely right.
But because no counter exercise is advanced to make that
point good, and because of the way it plays out through
the rest of the submissions, it appears to us that the
substance of the point there --
MR JUSTICE FRASER: I understand.
MR GREEN: -- is that. And if we look at {A/6/88}, the
entire heading, your Lordship will see, is evidence that
was not challenged.
When we go over the page {A/6/89} to the conclusion
at paragraph 208, it is just a bare assertion that large
parts of Mr Parker's evidence was not challenged.
Having read all that, we revisited what the substantive
point seemed to be in other places.
MR JUSTICE FRASER: Understood.
MR GREEN: And that's what we wished to draw the court's
attention to.
MR JUSTICE FRASER: We are probably going to need a break at
some point.
MR GREEN: My Lord, I have two small points to finish this
off and give the transcript writers a break.
MR JUSTICE FRASER: Yes.
MR GREEN: The evidence said not to have been challenged
includes his evidence about the APPSUP role at 205.9
{A/6/89}:
"This includes his evidence that he cannot recall
any instance in which the APPSUP role has been used to
change transaction data, although he cannot state
unequivocally that it has not happened.
We didn't cross-examine further on that because his
evidence carries so little weight in relation to that
that it is effectively worthless.
His paragraph 15 of his witness statement in which
the belated discovery of the APPSUP role is reflected,
which in itself is bizarre and speaks volumes, this is
at {E2/13/4}, reads as follows:
"APPSUP is used by SSC for updates to and
maintenance of the BRDB that would not involve changing
transaction data. I have not examined the privileged
user logs, but based on my experience my expectation is
that these uses of APPSUP or at least the vast majority,
are for support work that does not involve changes to
transaction data. I cannot recall any cases in which it
has been used to change transaction data, but I cannot
state unequivocally that there are no circumstances in
which it has ever happened."
Given the fact that he didn't know it had happened
at all, it was absolutely pointless to challenge
a witness of that sort in a time limited trial on the
point.
So it's not just the bare not challenged point, it's
ignoring the obvious context in which particular
evidence was not challenged.
My Lord, would that be a convenient moment?
MR JUSTICE FRASER: I think so, and we will have ten minutes
I think for the shorthand writers. Thank you very much.
(11.56 am)
(A short break)
(12.05 pm)
MR GREEN: My Lord, can I now deal briefly with the approach
to some of the fact witnesses that Post Office called.
In relation to Mr Godeseth at {A/6/63}, which is page 63
of Post Office's written closings.
MR JUSTICE FRASER: Go on.
MR GREEN: Post Office's paragraph deals with Mr Godeseth's
evidence there, and what is said there is, in line 3,
some of his evidence was unsatisfactory:
"With the benefit of hindsight, Post Office would
not have asked Mr Godeseth to cover several matters that
were addressed in his first two witness statements -
although it is right to point out that if Post Office
had only called first-hand evidence, the trial would
have been wholly unworkable."
The obvious answer to that is to call Mr Jenkins,
who was relied on repeatedly by a number of witnesses.
But let's leave that aside for a moment.
Just analysing what is said by Post Office there.
The contention that with the benefit of hindsight they
would not have asked him to deal with several matters
that were addressed is amplified at paragraph 146 on
page {A/6/67}:
"In Mr Godeseth's cross-examination, some of the
points he made on the basis of information provided by
others were shown to require correction or at least
clarification. This took Post Office by surprise. With
the benefit of hindsight, Post Office accepts that there
are points on which, if it wished to adduce any evidence
at all, it should have ensured that witness statements
were prepared for the individuals who were the sources
of the relevant information."
It is not clear why that is accepted in particular
in relation to Mr Godeseth rather than across the piece
on every occasion when that problem arose, but let's
just focus on this for the moment.
The Post Office didn't need hindsight to realise
that calling witnesses who don't actually know the
subject matter is wide open to error and will cause
difficulties for the experts, the other party and the
court, because the assumption when someone gives
a statement of truth that they believe something to be
true is at least likely to be that there is some
foundation for that belief. The difficulty for the
claimants was flagged at the pre-trial review in the
context of responding to complaints about Mr Henderson's
evidence. I won't take your Lordship to it but it is at
{C8.14/3/20}, which is me making the submission that
I effectively have to challenge someone who is not even
there in relation to Mr Jenkins who appeared to be the
source of much of the evidence.
So the difficulty of a number of Post Office's
witnesses giving evidence about things they didn't know
about, which they got from others, was flagged up at the
PTR at {C8.14/3/20}.
Mr Coyne had himself identified difficulties with
Mr Godeseth's evidence in Coyne 2, which is at
{D2/4.1/98} at paragraphs 4.3(a) and (b). If we can get
that up, that will be helpful.
It is 4.3(a) and (b), specifically referring to
speaking with Mr Jenkins and so forth and not referring
to supportive documents. Post Office should not have
been surprised about the errors revealed in
cross-examination because those errors were errors that
the claimants were able to identify in relation to his
approach to the bugs and treatment of them, which were
apparent on the face of such documents as were
disclosed.
The number of errors were identified in Mr Coyne's
second report, but not subsequently corrected. You can
refer to Mr Coyne's second report; we don't need to
bring it up, but paragraph 3.30, {D2/4.1/21}.
Particularly with regard to the number of branches
affected in the receipts and payments mismatch which we
find on page 24 of the report and the number of
occurrences at Callendar Square.
Your Lordship will remember that even in
January 2017 the Post Office was asserting that the
Callendar Square bug affected only one branch. We now
know that not to be the case.
The suggestion that Post Office were surprised is
resurrected further at paragraph 144.3, which is at
{A/6/66}:
" ... so far as Post Office was aware, the relevant
parts of Godeseth 2 were most unlikely to be
controversial. For example, the Misra trial was a
matter of public record, the four bugs were covered by
contemporaneous documentation and Post Office had no
reason to doubt Fujitsu's account of the documents it
held."
It is extremely surprising that it is said there
that his evidence of the bugs was most likely to be
uncontroversial. The impact and affect of bugs was
always going to be hotly contested and, indeed,
I actually made that submission expressly about
Mr Godeseth's second witness statement at the PTR at
{C8.14/3/18} between pages 18 and 19.
The second line on the right-hand side of the foot
of page 18:
"If your Lordship then looks at what then follows,
there is a treatment of the individual known bugs that
Mr Coyne is reporting, so Callendar Square, these are
actually front and centre main --"
MR JUSTICE FRASER: Where are you? At the very bottom.
MR GREEN: Three lines up from the bottom. Go over the page
{C8.14/3/19}. Something has gone wrong there. I will
just read it out:
"... main fighting ground for the Horizon trial, so
Callendar Square is immediately below, and you then get
some of the bugs, later on, payment mismatches ...
dealing in this section ..."
There's absolutely no doubt that the evidence about
bugs is going to be central to the testing of the
evidence in this trial.
Mr Coyne in his second report raised multiple issues
with Mr Godeseth's second witness statement and whether
it captured the full extent of the bugs. And that was
paragraph 4.3(d) on page {C8.14/98} and the further
points that follow at 4.4 over the page. And as to the
language that Fujitsu's account of the documents it
held, which is Post Office's language in its closing
submissions.
MR JUSTICE FRASER: Which language are you talking about?
MR GREEN: In Post Office's explanation of having no reason
to doubt Fujitsu's account of documents which it held.
Many of the documents were effectively created by
Fujitsu for Post Office, if they weren't Post Office
documents themselves. So, for example, the receipts and
PEAK payments mismatch documents is a note of a meeting
attended by Mr Andy Wynn of Post Office.
So it is not simply a matter of Fujitsu's account of
documents that helped. And in any event the documents
when taking a witness statement from anyone, the proper
approach is to say: what's the source of that? Is there
a document that supports that and, if so, should it be
exhibited? And that didn't happen, and that's part of
what went wrong, although not all of what went wrong.
So there are two things in combination. There is
calling witnesses who aren't properly placed and it
hasn't been made sure that they are properly placed to
give the evidence that they are giving. And secondly,
the fact that very frequently no documentary support was
cited for general propositions of what would or could or
should have been happening is why (a) it was difficult
for the claimants to unpick it, and we had to go through
the documents ourselves and find them, but (b) it must
have contributed to any misapprehension, if there was
such, by Post Office about whether Mr Godeseth was
properly placed to give the evidence at all.
Had the witness statements been done in the usual
way by seeking to identify what the basis for things
that are being said is and exhibiting the relevant
documents, it would have become apparent much sooner,
even if it wasn't fully apparent from all the
observations made by Mr Coyne and others, that
Mr Godeseth's evidence was very unsatisfactory in that
respect.
As to Mr Jenkins, the explanation for him not being
called is provided at paragraph 138, and that's at
{A/6/64}, page 64 of the Post Office's closing.
Paragraph 138 says:
"Post Office wanted to provide a simple and
uncontroversial overview of Horizon and its relevant
features."
My Lord, just pausing there. That is exactly the
problem we faced in the Common Issues trial, that there
was this wish to present a general overview that seemed
extremely rosy and which, when unpicked, fell apart.
That was also the difference reflected in the
approach of the two experts. Dr Worden provided this
top down overview-based look at the Horizon system, in
part largely relying on Mr Godeseth, as we will come to
in a minute, and Mr Parker's evidence and others.
Whereas Mr Coyne was trying from a very early stage to
actually understand how any errors were recorded, were
they recorded, obtain the documents and analyse them for
the purposes of answering the Horizon questions. In
that respect -- we will come to it in a bit more
detail -- we say the criticisms made of Mr Coyne are
extremely unfair.
Just focusing on the explanation from Mr Jenkins not
being called, which we find at paragraph 138, it says
Post Office says there that:
"It recognised that it was not possible for one
person to have a complete understanding of all the
corners of the Horizon system ..."
Pausing there, Mr Jenkins appears to have had nearly
a complete knowledge from what we have heard:
" ... but, on the basis that there would not be room
in the timetable for multiple witnesses, it took the
view that this overview should be provided by one
person."
Just pausing there. That isn't even what happened.
It wasn't only Mr Godeseth who referred to Mr Jenkins.
And Mr Jenkins could certainly have collected
information from other team members who other witnesses
collected information from. But they then go on to say:
"Two possible candidates were Torstein Godeseth and
Gareth Jenkins. Taking into account the involvement
that Mr Jenkins had had in a number of criminal
prosecutions that are currently being looked at by the
criminal Cases Review Commission (eg the Misra case),
Post Office asked Mr Godeseth to do so."
Well, we respectfully say it is a slightly
surprising approach because we see it further amplified
at 144.1 on page {A/6/66}. It says:
"... Post Office was concerned that the Horizon
Issues trial could become an investigation of
[Mr Jenkins'] role in this and other criminal cases."
My Lord, two points arise on that. Firstly, those
cases are stayed in the GLO, the criminal cases.
MR JUSTICE FRASER: This court is not concerned with and has
no jurisdiction in respect of any of the criminal cases.
MR GREEN: Precisely. There is absolutely no prospect of
either of us seeking to do that or of your Lordship
allowing it. So we don't accept that is a good reason
at all. Mr Jenkins clearly had the firsthand knowledge
and the court wouldn't have allowed the sort of
investigation of which Post Office was fearful in this
respect. But it does suggest a defensiveness which is
not completely isolated in these proceedings.
MR JUSTICE FRASER: I don't have the reference immediately
to hand but I seem to recall one of the witnesses said
that Mr Jenkins had retired, hadn't he, or am I wrong
about that?
MR GREEN: He has retired, but he had been consulted. He
provided comments --
MR JUSTICE FRASER: I know the extent to which witnesses
have relied on him etc, but I just wanted to check
I wasn't imagining that.
MR GREEN: Your Lordship is absolutely right. Of course,
that's not actually the reason relied on. And then
there has then been correspondence about his
availability, which I will very briefly touch on.
{H/184/1}, {H/201/1} and {H/203/1} are effectively the
three letters.
{H/184/1}, if we start with that. A very short
letter:
"We note that Mr Jenkins is the author of many
important contemporaneous documents, and that both of
Mr Godeseth's witness statements give hearsay evidence
on the basis of conversations he has apparently had with
Mr Jenkins.
"Is Mr Jenkins available during the trial period?"
{H/201/1} is the Wombles response of
12th February 2019, and the second line of that says:
"The information that you have sought regarding
Mr Jenkins is clearly privileged."
Then they say it would have been open to us to
approach Mr Jenkins or apply to call him, which is
correct.
Then they add there if we were going to make
an application out of time, we should bear in mind that:
"... he acted as our clients' [Post Office's] expert
witness in relation to a number of criminal prosecutions
... being looked at by the [CCRC]," as a suggestion to
us that we should not seek to contact him, it appeared
to be.
Then at {H/203/1} the letter from Freeths:
"We note your response to our letter regarding the
reason for not calling Mr Jenkins and why he is not
available to give evidence during the trial: you have
declined to give any or even to tell us whether he is
available during the trial.
"You have cited that the explanation for him not
himself giving evidence is privileged."
So our just simple question, "Is he available?", was
not: we don't know, or anything of that sort, it is that
it is privileged. So we assumed that he must be part of
the shadow expert team.
We have now been informed apparently he is not.
MR JUSTICE FRASER: He is not what?
MR GREEN: Part of the shadow experts team that the
Post Office is entitled to use on the case and has been
using and included in their budget.
So we enquired whether he was in fact part of that
team and the answer is no. We hadn't appreciated that.
MR JUSTICE FRASER: I don't think he -- I think the item of
cost was originally included in the budget and was then
withdrawn. I do not think it is correct to say it is
included in the budget.
MR GREEN: No, my Lord, it was originally included and then
disallowed.
MR JUSTICE FRASER: Well, no, it was not disallowed. It was
withdrawn because it was --
MR GREEN: We objected to it and it was not included in the
approved budget. That's absolutely correct.
My Lord, can I move on to the issue taken with
cross-examining witnesses by reference to documents,
because this is something in respect of which
Post Office complains that it is unfair to cross-examine
witnesses by taking them to documents.
The criticism is levelled at us for doing this at
paragraph 850, which is {A/6/277}. Paragraph 850 makes
the point three lines up from the bottom on the
right-hand side:
"She was, however, taken in cross-examination to
many documents which she had not seen before -
a recurring theme of Cs' cross-examination. It is
submit that such an exercise is of very limited
utility."
Pausing there. We say it is not an unusual way to
cross-examine to put contemporaneous documents that have
been disclosed as relevant to a witness whose evidence
appears to suggest that they are aware of at least the
general area in which those documents arise. We say
that's actually common. Questions like that have been
put up and down the country every second of the working
day across all courts and tribunals. So it is not
an unusual approach to take witnesses to contemporaneous
documents.
But the complaint that's made is that Ms Van Den
Bogerd was taken to many documents which she had not
seen before. Now, that is a prisoner to how many
documents she has chosen to look at, because if
a witness has not looked at, on the face of it, relevant
contemporaneous documents at all then every document
will fall into the category of document to which
Post Office appears to complain that she was taken. And
if we take an example, Mrs Van Den Bogerd's witness
statement speculated about the potential cause of
a spike in declared losses on the introduction of
Horizon Online. Paragraph 183 at {E2/5/42}.
This was effectively SPMs tidying up their accounts.
That speculation was challenged in cross-examination on
the footing that the claimants would expect a senior
witness giving evidence on an issue of that type to
review the available documentary record before
speculating in that way and providing the court with
a witness statement signed and verified with a statement
of truth.
Mrs Van Den Bogerd specifically said she had not
done this in answer to one of your Lordship's questions
which we find at {Day5/167:22} onwards. At 167, line 22
your Lordship says:
"MR JUSTICE FRASER: Understood. And then the
second point is when you were preparing your witness
statement and in particular the paragraphs at 180 to
183 --
"Answer: Yes.
"MR JUSTICE FRASER: -- did you do any investigation
in respect of what might have been happening that you
didn't know at the time in 2000 --
"Answer: Not back to 2000, no.
"MR JUSTICE FRASER: -- or in 2010 when the change
was from Legacy Horizon to Horizon Online?
"Answer: So in 2010 I was in a different role and
had broader responsibility and I knew what -- what we
did, again we replicated a similar approach to make sure
we supported branches at the time, but as for any detail
of Information, I didn't research into that, no."
So that was an example of one of the matters on
which Mrs Van Den Bogerd was challenged by reference to
contemporaneous documents. She hadn't seen them, but
that is not a proper or legitimate criticism of the
claimants' side, we respectfully submit, and might
fairly be thought to go the other way.
In moving on to rewriting the fact evidence in the
closing submissions, which is the next theme, some of
these aspects are quite surprising. One example is the
problem management procedure.
The defendants' closing addressed the problem
management procedure albeit in the context of
Mr Godeseth's evidence being unsatisfactory, and used
that as an excuse to try and rewrite the evidence which
the court has actually heard.
We see that at page 68 of the defendants' closing
{A/6/68} at paragraph 147.4, where it says:
"In paragraph 63, he [Mr Godeseth] appeared to be
saying that Fujitsu’s Post Office Account Customer
Service Problem Management Procedure document 223 was
not implemented following Mr Salawu's departure as
Horizon Head Lead Service Delivery manager, when in fact
it was merely section 1.4 of that document that was not
implemented."
Well, that came as news to the claimants, my Lord,
not least because of the way the issue arose and because
of how it was cross-examined on and not re-examined on.
MR JUSTICE FRASER: Does that last part of the sentence come
from evidence or is that --
MR GREEN: No, that's new.
MR JUSTICE FRASER: All right.
MR GREEN: So it arose because Mr Coyne was proceeding on
the basis that there was a proper problem management
system in place, and therefore there should be documents
available that would collate bugs and errors and show
lots of information, so we could get disclosure. We
thought that would be very helpful, and we can see that
from Mr Coyne's first report at {D2/1/97} at
paragraph 5.158 and 5.159:
" ... is the POL monitor that tracks the number of
records arising directly as a result of managed change
activities."
You can see the problem management procedure
footnoted at footnote 150 at the bottom of the page:
"No disclosed logs have been found in respect of
these problem records that are listed as being reported
monthly."
5.159:
"Requests have been made in relation to making such
Management Information reports. At the time of writing,
these have not been made available for analysis."
So Mr Coyne's assumption, which is actually shared
by Dr Worden, was that the problem management procedure
document was in force and therefore there should be
a repository of documents, which we have not been aware
of at that stage, which would be very helpful because
they would show what was going on.
Then in response to that, with the looming
possibility of disclosure, we get Mr Godeseth's second
witness statement served saying in fact it was not
implemented, and we find that at paragraph 63,
{E2/7/16}. In paragraph 63 we get the conversation with
Mr Bansal.
And then four lines up from the bottom:
"I understand from Steve that Saheed Salawu's
replacement did not wish to implement the changes and
therefore the records referred to by Mr Coyne ... of his
report do not exist, as we continued to follow the
previous existing reporting methodology."
That is referring back up to the situation before
the new procedure, which was to achieve the same,
problem management seeks to establish the root cause of
incidents and then start actions to improve or correct
the situation.
So your Lordship may remember the cross-examination
obviously that followed what we heralded in our written
openings at paragraph 6(4) at {A/1/6} where we made
clear our understanding was that this had not been
brought in and that Mr Godeseth suggested this followed
the departure of the Horizon lead service delivery
manager.
MR JUSTICE FRASER: Where are you looking now?
MR GREEN: This is paragraph 6(4).
MR JUSTICE FRASER: Yes.
MR GREEN: The bottom two lines, the departure, this was not
brought in. Three lines up from the right-hand side.
Mr Godeseth suggests and refers to those paragraphs.
No whiff of disagreement from Post Office at this
stage, in fact until closing submissions. Evidence at
trial was Mr Godeseth was cross-examined on
paragraph 63. I will give your Lordship the reference
without going there. {Day7/157:4} to line 21. We also
put to Mr Parker the problem management system wasn't
brought in. We did that at {Day12/53:4}.
MR JUSTICE FRASER: What did Mr Parker say?
MR GREEN: He didn't suggest it had been brought in either.
MR JUSTICE FRASER: Let's go to --
MR GREEN: Let's go to {Day12/53:4}, if we may.
We can see it says "proper management", but we can
see later on it is clearly problem management:
"... just as an aside -- the proper", it says proper
on the transcript, "wasn't brought in either. We
covered that with Mr Godeseth, you were here for that?
"Answer: Yes.
"Question: So there was no problem management
system brought in notwithstanding it was internally
recommended, and so all you're left with is this system
of looking at the codes and seeing how they have been
categorised on closure."
Absolutely no demur at all. It has been there in
Mr Godeseth's evidence, not a scintilla of disagreement
from anyone, no re-examination at that point of
Mr Godeseth or Mr Parker.
We also put it to Dr Worden on {Day19/166:15} to
line 16 on that footing. About halfway down the page.
I'm not sure that's quite right. Can we just check that
reference. I will come back to you on that.
MR JUSTICE FRASER: That doesn't appear to be the right
reference.
MR GREEN: It is not the right reference.
MR JUSTICE FRASER: You say you put it to Dr Worden?
MR GREEN: Put it to Dr Worden as well. Again, no
re-examination on that.
So we have an unusual situation, we respectfully
submit, to say the least that we filed our written
closing submissions on the basis of the evidence that
was actually heard and was tested because with
Mr Godeseth we said "Well, how come there are all these
extra versions of this if it wasn't introduced?"
Because it looked weird to us. We challenged it,
apparently it was not introduced. And we filed our
written closing submissions, and we learn then for the
first time from Post Office that actually it was brought
in. It was only an individual paragraph that wasn't.
We respectfully say obviously that's completely
unsatisfactory, but also if it is true it is again less
than forthcoming and showing a striking lack of candour
about what documents might be available.
My Lord, just to give another example of a sort of
rewrite which sort of bleeds into the attacks on the
documents where they are unfavourable to Post Office.
So we have this strange part, which I have already shown
your Lordship, where they say if internal Post Office
documents say Horizon was problematic, the authors have
got it wrong. And that's part of this theme.
If we look at the document recording the visit to
Mr Bates' branch, it is at {F/99.1/1}. It was
a document put to both Mrs Van Den Bogerd and Dr Worden.
If I can just give the references {Day5/168:14} and
{Day18/123:12}.
If we look at {F/99.1/1}, the point about that
document was that the visit being made, the
Post Office -- it should be, I think, F/99 --
MR JUSTICE FRASER: You said 99.1.
MR GREEN: That may have been my error, my Lord.
MR JUSTICE FRASER: Is it {F/99/1}?
MR GREEN: I think it may be {F/99/1}. I will just make the
point and we will check the reference on that.
It is {F/99.1/1}, I think. The short point,
my Lord, is this is the one where Mr Bates is visited,
and the officer visiting says "I couldn't get a correct
read on the cash account because Horizon intermittently
adds the cash from the previous day".
MR JUSTICE FRASER: This is the audit report, isn't it?
MR GREEN: Exactly. So the Post Office auditor -- two of my
Opus screens have gone down -- but at {F/99.1/1}. We
seem to be able to see it on our screens.
MR JUSTICE FRASER: Which page are you going to?
MR GREEN: It is {F/99.1/4}.
MR JUSTICE FRASER: Page 4. Yes. I have got it on my own
separate screen.
MR GREEN: I'm most grateful. The short point is that it is
striking that someone attending to do an audit of the
correct cash totals -- thank you very much; if we could
slide to page 4 of that, I think -- is unable -- if we
go towards the bottom, "Comments", just above "National
Savings":
"A correct assessment of cash holdings could not be
made because the Horizon system intermittently adds the
previous days cash holdings to the daily declaration."
There was no attempt to re-examine Mrs Van Den
Bogerd or Dr Worden about that, but what is now said is
that this is actually a designed function in Horizon in
the closing submissions that Post Office has filed at
{A/6/279}, paragraph 856 about carrying forward balances
from a previous day. And we respectfully submit (a)
that's completely new, and (b) it just doesn't chime
with the reality of an auditor who well knows how
Horizon should operate encountering that difficulty in
Mr Bates's branch when he goes to get a cash reading.
The word "intermittently adding" is extremely
difficult to reconcile with simply carrying a value
forward from the day before.
My Lord, just to give your Lordship the correct
reference to the problem management being put to
Dr Worden, that was at {Day19/176:13} to line 17.
I apologise for that mistake.
MR JUSTICE FRASER: Yes.
MR GREEN: Then moving forward to --
MR JUSTICE FRASER: Just before you move on though, isn't
the point it was not re-examined upon, doesn't that fall
within the same basket of it is a time-limited trial,
one doesn't have time to re-examine on everything?
MR GREEN: My Lord, absolutely right, but if there's going
to be a --
MR JUSTICE FRASER: Positive explanation.
MR GREEN: Positive case suddenly introduced for the first
time, it is quite nice to have a passing mention in
re-examination.
MR JUSTICE FRASER: All right.
MR GREEN: Take, for example, keystrokes where the position
is absolutely bizarre because you get Mrs Van Den Bogerd
accepting initially that there were keystrokes
available. Then you get the sheepish re-examination
that maybe they are not, and then you get the reference
to Mrs Mather whose witness statement expressly said
that keystrokes were available and then in chief is
invited to say what she meant by the word "keystrokes".
She means or transactions and sales data.
Then the hammer drops later on when we get
disclosure of a whole load of keystrokes which were
available from Fujitsu. I mean, it is against that
background that there is some concern about introducing
completely new explanations without any foundation in
the evidence.
My Lord, can I just deal with a couple of
documentary examples. First of all, the phantom
transaction PEAK and the Romec engineer.
That was quite surprising to say the least. What
effectively was said, as your Lordship may remember, is
that the Romec engineer's view of having actually
observed phantom transactions happening was not
reliable. That was how it was put to Mr Coyne. And
that PEAK is {F/97/1}, and the reference to the Romec
engineer advising that he has witnessed further phantom
transactions whilst on site is on page {F/97/5} of that
PEAK.
It is right your Lordship should be reminded, if
I may, that it wasn't only that Romec had seen it,
because we know that Mr Carroll had also seen it. So
when we look at page {F/97/7} of that PEAK, in the
bottom yellow box:
"I now have pressing evidence to suggest that
unwanted peripheral input is occurring, the likely
source being the screen. This has been seen at Old
Iselworth (OI) and Wawne (W) with OI being the best
site; when the PM has been asked to leave the screen on
overnight I have observed system activity corresponding
to screen presses happening with no corresponding
evidence of either routine system activity or human
interference, the way forward now is to correlate this
with the Microtouch supplied monitoring software and to
this end Wendy is arranging for installation of Kit at
OI on Friday, we can then, provided the PM agrees, leave
screens on over the weekend and record what happens.
Once these results have been analysed I feel sure that
we will be in a position to move forwards at OI. all
other cases should be considered on their individual
merits but you must appreciate that this is a fairly
intensive analytical activity and I cannot hope to
provide answers on all cases in the short term."
So it was not just Romec who had reached that view,
but what's relied on by Post Office when they say a host
of possible explanations at page 475 of their closing
submissions, paragraphs 10 to 11, is Mr Carroll's view,
effectively, that we find on page 9 of the PEAK {F/97/9}
where, in closing this down, Mr Carroll says:
"Phantom [transactions] have not been proven in
certifications which preclude user error. In all cases
where these have occurred a user error related cause can
be attributed to the phenomenon."
Three things going on. First, rather overlooking
what Pat Carroll had previously said about it. Secondly
seeking to elevate the slightly bizarre way of
determining how this should be treated, as user error or
not, to a finding that it was in fact user error, and
Mr Coyne being challenged on the footing that he agreed
he was he was not in a position to say that Mr Carroll
was wrong, which Mr Coyne very fairly accepted: well,
I can't say he is wrong that it couldn't preclude user
error.
But we respectfully say that is a pretty astonishing
way of dealing with the PEAK which is pretty
unsatisfactory, to say that actually Mr Carroll and the
Romec engineer plainly got it wrong.
The Helen Rose report goes a step further, if that's
possible. The treatment of that is, we say, absolutely
extraordinary. The Helen Rose report is at page 1082
and includes --
MR JUSTICE FRASER: Of?
MR GREEN: Sorry, {F/1082/1}, and on page {F/1082/2} of that
document under "Reviewing the data", it says:
"On looking at the credence data, it clearly
indicates that the reversal was completed by JAR001
(postmaster) at 10:37 ... and was reversal indicator 1
(existing reversal) and settled to cash. An existing
reversal is where the session number/Automated Payment
number has to be entered to reverse the item."
So the point about this was the initial view was
an assumption that there had been a reversal by the SPM.
And the Helen Rose report then, as your Lordship will
remember from dealing with this document on several
occasions, goes through to identify that that actually
had not been the case.
Mr Coyne flagged this up at {D2/4.1/113} in his
second report. He says:
"Credence data, most commonly used by Post Office
for their investigations, is either wrong or does not
provide sufficient information to complete the full
picture ..."
This is an example of one of his concerns. On what
was presented in the Helen Rose report, that seems to be
correct. But Mr Coyne's account of the Helen Rose
report was made subject to very extensive
cross-examination, pressing him to accept that
Helen Rose had in fact misinterpreted the underlying
data and put 2 and 2 together to make 5. That's what
was put. In fact, my learned friend said put 2 and 2
together to make 4 and then changed it to 5, but the
point is it was being suggested that she had got it
wrong, not right. That's at {Day15/38:15} to
{Day15/39:24}.
What's effectively being put to Mr Coyne there is
that Helen Rose has got it wrong. And just pausing
there, Mr Coyne's interpretation of what appears on the
face of the report was totally reasonable and, we say,
right, and Post Office not only runs the, we say,
extraordinary misinterpretation point in their closing
submissions at paragraph 529 on page 194, but also used
that as a basis to subject Mr Coyne to very heavy
criticism for his evidence. There are pages and pages
of criticism on that footing on an extremely strained
interpretation of a document which appears to say the
opposite to what Helen Rose actually said.
Bear in mind, my Lord, this was a document about
which, for example, Angela Van Den Bogerd's evidence was
initially completely wrong. So it is actually the
Post Office's treatment of the Helen Rose report which
is properly the subject of criticism, and it is bizarre
that it is Post Office's case that it is Helen Rose's
fault that she misinterpreted the data, given that she
was a security fraud analyst who was evidently involved
in prosecutions.
My Lord, I have touched on the internet documents
critical of the Post Office and we identified a number
in opening. We put them to Dr Worden and addressed them
again in closing by reference to what the documents
actually say; for example, paragraph 614 at {A/5/213}.
MR JUSTICE FRASER: This is your closing, yes?
MR GREEN: This is our closing. It is to that that
Post Office give their response. It is a section in our
closing that begins at paragraph 614 on page 213.
MR JUSTICE FRASER: Yes.
MR GREEN: It is that section to which Post Office are
responding with reference to which I took your Lordship
out of order to at the beginning, where they say
{A/6/134}:
"The Post Office documents were not drafted with the
benefit of the vast amount of work that has been carried
out by the experts for this trial. If the authors
considered that Horizon was not a good system, they were
wrong (although that is not even a fair summary of what
the documents say)."
We say that is an absolutely astonishing submission,
particularly where Dr Worden hadn't read any of them and
the contention is that he is in a better position, for
example, than Mr Rob Houghton, who is Post Office's
chief information officer who presented to the board on
Post Office's IT strategy in January 2017, which is one
of the documents which we see at {F/1611/87}.
Then, just to make the one-way street point good,
that is to be contrasted with the, as it turned out,
wholly misplaced reliance on the ISAE service audits in
relation to change management and remote access, which
we have addressed separately why those audits (a)
weren't addressed to the same point as the 2011 audit,
and secondly, why they plainly didn't pick up what we've
seen about remote access in the APPSUP role in the
APPSUP PEAK.
But what they say about that, and this is part of
Post Office's cross-examination of Mr Coyne, was to say
to Mr Coyne -- we can look at it at {Day16/174:11} to
line 15. The tee-up is that unhelpful documents to
Post Office, even if drafted by Post Office, even if
drafted by their own IT chief information officer and
presented to the board, they are wrong and the experts,
including Dr Worden, who had never read any of them, are
right. Leave aside the fact that Dr Worden accepted
they were irreconcilable with his views.
Then when something is thought, albeit wrongly, to
be in Post Office's favour, what you get is the
cross-examination at page 174, lines 11 to 15:
"Question: Mr Coyne, would you agree with me that
in principle the best people to judge whether action is
being taken to address recommendations made by auditors
is the auditors themselves rather than you, would you
agree with that?
"Answer: Yes.
"Question: Let's look to see what the auditors say
in later years."
So effectively where the documents are in
Post Office's favour you take them on their face and the
authors are right, even if they are the wrong documents
and so forth, contradicted by the PEAK at the time, well
known to Fujitsu. But when the documents are against
Post Office, the authors, even if they are Post Office's
own employees charged with those specific roles,
internal experts, they are apparently wrong. And we
respectfully say even by the standards of Post Office
conduct in this litigation, that is a bizarre approach
to urge upon this court to deal with the relevant
contemporaneous documents.
My Lord, that's a natural break. Would it be
convenient to rise a couple of minutes early?
MR JUSTICE FRASER: I think it would. Thank you very much.
2 o'clock.
(1.00 pm)
(The short adjournment)
(2.00 pm)
MR JUSTICE FRASER: Mr Green.
MR GREEN: My Lord, can I just pick up one brief observation
in relation to disclosure.
The Post Office made various criticisms in their
closing submissions relating to claimants' disclosure,
and the short point is that the claimants were directed
to disclose the documents they relied on and any known
adverse documents, which is what they in fact did. And
that order is from the fourth CMC order, paragraph 5,
which is at {C7/18/2}.
We needn't go to it. There is correspondence where
Post Office has sought to press the claimants in
relation to those issues in the light of evidence given
at trial. And the relevant -- I'm not going to take
your Lordship to the correspondence, but just in case,
so your Lordship has the picture in case something is
made of it, they are letters of 15th March, 17th March,
14th May and 29th May of this year, and they are to be
found respectively at {H/242.6/1}, 15th March; {H/255/1}
is 27th March --
MR JUSTICE FRASER: You said 27th March?
MR GREEN: 27th March.
MR JUSTICE FRASER: I thought you said 17th March.
MR GREEN: If I did I misspoke, I'm sorry. 27th March.
{H/280/1} is 14th May and {H/303/1} is 29th May.
MR JUSTICE FRASER: Those are letters in both directions,
are they?
MR GREEN: Toing and froing in relation to that. It is of
course open to a party to make submissions in relation
to how document have been disclosed, as, indeed,
Post Office seeks to do going back that. A bit more
resistant to the stream going the other way, as
your Lordship has already seen.
Just very briefly in relation to that, the adoption
of model C disclosure by the court, including in
particular the obligation to disclose known adverse
documents, does not make a party's explanation for the
emergence at a late stage of documents irrelevant or
immune to scrutiny from the court, nor is an approach by
the claimants to seek at successive hearings to improve
on the position in relation to documents, some of which
they have been seeking since 2016, an answer by way of
saying, well, you never sought specific disclosure.
Neither of those are answers in relation to a fair
appraisal being made of how Post Office documents have
come to be disclosed as late as they have.
That's a very brief response in relation to that
point, my Lord. We just don't accept that for a moment,
particularly when the evidence, for example the
mis-keying, flies in the face of the most extraordinary
changes of evidence before the court. Perfectly
legitimate for the court to want to know how that
document came to be disclosed when it did.
So with that brief footnote in relation to
disclosure I just wanted to make a very few brief
submissions in relation to the treatment of Mr Coyne,
who your Lordship has obviously appreciated has been
heavily criticised, we respectfully say very unfairly.
His approach was correctly to try and see if there
was evidence of any relevant bugs, to try and identify
what remote access evidence there was, to identify
relevant lines of enquiry, which we see in his early
requests as long ago as April 2018, as we flagged up in
our closing submissions, and to seek documents in the
RFIs that they did in the summer of 2018.
His attempt to be careful and try and work out what
has happened and what there was and was not evidence of,
including what there was not strong evidence of, which
is the wording of paragraph 1.15 on which he was
extensively cross-examined, that was not only the right
approach, but a helpful approach and one that was framed
by reference to the Horizon Issues as they were actually
ordered by the court.
There is a criticism of him in relation to the
Helen Rose report to which I have already averted, but
we respectfully invite the court to contrast that with
the defendants'/Post Office's treatment of the evidence
of Mrs Van Den Bogerd in relation to Helen Rose, which
turned out to be wrong, which Post Office, at
paragraph 852 {A/6/278} of their closings explained that
her evidence was to give early notice of Post Office's
case but now effectively overtaken by more detailed
consideration in the expert evidence.
That is an attempt (a) to gloss over the fact that
the evidence was wrong, but (b) it stands in stark
contrast to the pages, ten pages of criticism,
critiquing Mr Coyne's account of the Helen Rose report
that we find between paragraph 521 at {A/6/192} and 552
at {A/6/202}.
The reality is that Mrs Van Den Bogerd
misrepresented, whether unwittingly or by not taking
care, what the Helen Rose report said and what her
evidence said was wrong, as she accepted.
We see that evidence at {Day5/90:1}. So, again, we
have got a sort of one-way street where Mr Coyne
actually gets the Helen Rose report right and is
criticised; Mrs Van Den Bogerd gets it wrong and accepts
it is wrong, and that's just clarification and overtaken
by more detailed consideration, all in circumstances
where the premise of the criticism of Mr Coyne is that
Helen Rose, Post Office's own investigator, as
I submitted before lunch, completely misinterpreted the
data before her.
That is one example, my Lord, of a fairly wholesale
attempt to discredit Mr Coyne by suggesting, we say
unfairly, that his essential endeavour was to throw mud
at Horizon in the expectation that at least some of it
would stick. And those words are taken from
paragraph 267 at {A/6/103}, bottom four lines:
"His essential endeavour was to throw mud at Horizon
in the expectation that at least some of it would stick,
and not to worry too much or at all about giving the
other side of the picture or even presenting a fair view
of the documents to which he referred."
That's both a surprising submission and an unfair
one. And where, for example, Mr Coyne said that the
position on transaction corrections might make the
position better or worse, that was put to Dr Worden. He
is being scrupulously fair there and Dr Worden accepted
that he was. So it flies in the face of concessions
made by Dr Worden and it also flies in the face of the
way in which Mr Coyne gave his evidence.
That's obviously a matter for the court, but we
respectfully submit that Mr Coyne was conscientiously
trying to answer questions that he was asked, not always
in circumstances where he was very comfortable with the
premise that he was being boxed into for those
questions. But he sought to try and answer them fairly,
and a fair reading of his evidence would have regard to
the joint statements, the fact that 1.15 was necessarily
a compromise between two experts who differed on the
definition of lasting, amongst other things, who agreed,
as we see at joint 3, paragraph 4.1, which is {D1/4/7}
that Issue 4 fed back into Issue 1, and when the joint
statement is read as a whole rather than saying
an isolated paragraph and inviting an expert who has
taken a completely different approach to grind very
finely those aspects of his analysis which are
sufficient to show strong evidence of the type of bugs
that was being considered.
I mention that feature of 1.15 because it was
particularly telling for Mr Coyne who was very careful
in his answers to stop at the point at which there was
no evidence going further.
So when the question was put to him, where he felt
he couldn't go further because he didn't have the
evidence one way or the other, he would stop at that
point. He would not go on to draw what Dr Worden in
some places called weak inferences about various things,
or other inferences, or proceed on assumptions.
MR JUSTICE FRASER: You said 1.15 a couple of times, I
think.
MR GREEN: Yes, 1.15, my Lord, which was the bit that my
learned friend cross-examined him about in joint 2.
MR JUSTICE FRASER: In joint 2. I thought you said joint 4.
MR GREEN: I'm sorry. The Issue 4 point is paragraph 4.1 of
joint 3.
MR JUSTICE FRASER: Of joint 3.
MR GREEN: I should have made clear that the 1.15 point is
the one that my learned friend cross-examined on fairly
extensively, which is in --
MR JUSTICE FRASER: Joint 2.
MR GREEN: Joint 2. That's at page 29.
MR JUSTICE FRASER: Yes.
MR GREEN: Your Lordship will remember that the 1.9 on the
previous page is the experts' differing view on branch
impact.
MR JUSTICE FRASER: Yes.
MR GREEN: So we respectfully say it is a little artificial
and unfair where the experts try to reach a bracket
where there is more than one variable in their
difference of view and reach that compromise at 1.15, to
subject it to the sort of drafting scrutiny that
Parliamentary statute might bear, and that the fair way
to approach it is in the light of the joint statement
read as a whole in the context of the approaches of the
two experts.
Parenthetically, my Lord, when my learned friend
revisited the 40 bugs analysis with Mr Coyne on
{Day15/165:17}, our submission about the layered
assumptions and confinements on the basis of which he
was cross-examined on 1.15 become clear when you look at
his answer at 165.17. He makes clear that he still
regards the position he described in his report as
correct.
Turning to Dr Worden. Your Lordship has our
submissions on the statistical analysis and so forth.
And we are obviously not going to repeat those, but
there are a couple of points we would like briefly to
pick up on.
The Post Office submits at {A/6/99} at paragraph 249
of their closing that Dr Worden did not confine his
efforts to looking only for problems or only for
evidence of Horizon working well.
We respectfully say that's not correct. Not only
was there no consideration of documents pointing in the
other direction to any significant extent, but if we
look at joint 1 at {D1/1/10}, Dr Worden, in the top box,
penultimate paragraph, explains what he is going to do.
He says:
"In my report I shall survey the evidence I have
found that Fujitsu paid sufficient attention to the
dimensions of robustness, and that they did so
successfully. I shall also address evidence from
Mr Coyne implying that Horizon fell short of its
robustness objectives."
What's missing from that summary is an enquiry by
him in the same vein as Mr Coyne. So not only was that
his approach, but he actually telegraphed it in that
description of what he proposed to do. And that's
vividly found in the approach in his report.
Secondly, at paragraph 251 the Post Office goes on
to say he did not inappropriately prefer the evidence of
the Post Office witnesses. {A/6/99}
That's just not right. He did so. It was put to
him, and we have got {Day18/56:5} to line 7. He was
cross-examined specifically about where he had said in
his report he had established that something had
happened based on the evidence of Post Office witnesses,
and he accepted at that place in the transcript,
{Day18/56:5} to 7 -- I think that piece begins at
page 53, line 3 -- that he shouldn't have done that.
So this is not even an available submission, let
alone one which, in our respectful submission, should be
referred by the court. The fact that his report is
replete with references to on the basis that the
defendants' evidence is right, underscores that that was
a running theme. And the fact that when Mr Parker was
saying Mr Roll was wrong, Dr Worden was happy to rely on
that, but when Mr Parker in his second witness statement
then agreed with Mr Roll, Dr Worden said he was confused
and that's why he couldn't agree with it but took
absolutely no steps whatsoever to seek any clarification
that would have alleviated any genuine confusion.
So we respectfully say that that is not correct to
say he did not inappropriately prefer the evidence of
Post Office witnesses. He did and he admitted doing so.
Finally, in relation to Dr Worden, at paragraph 316
of the Post Office's closings, which is {A/6/120}, the
Post Office says:
"Dr Worden stated in cross-examination that he had
been 'told' to send his report direct to the Court ..."
Pausing there. It was actually in a question from
your Lordship that he explained that. Then Post Office
says:
"... but that word could give the wrong impression."
Pausing there. Post Office's closing submissions
are replete with footnotes everywhere, but that word
"told" is not footnoted. So it is worth your Lordship
having a reference to the relevant part of the
transcript, which is {Day20/189:12}.
If we look at line 12, your Lordship asked:
"Have you ever served one of your expert reports
directly on the court before?
"Answer: I have never done that myself before.
I had it done to me."
Your Lordship says:
"But you have never done it before?"
Then Dr Worden then volunteers:
"Answer: A kind of late report, no. I mean the
issue of serving direct on the court rather than through
lawyers, I don't recall how that happened in the past.
I suspect it was all done through lawyers."
Then your Lordship at line 21:
"But in this case everyone knows you sent an email
to my clerk?"
Then line 23 Dr Worden then says:
"Answer: I did, yes."
Then he volunteers:
"That's what I was advised to do, that was how I was
advised to do it."
Your Lordship:
"You were advised to do it?"
"Answer: By Post Office lawyers, yes."
So the word that is quoted, "told", is not the word
that was used. The word that was used was "advised" and
it was volunteered by Dr Worden. Your Lordship didn't
ask who told him to do it; he volunteered that answer.
And at the end of that exchange your Lordship turned to
both counsel and said: are there any follow-up questions
from that exchange?" And there were none from either me
or my learned friend.
Now, that account is not identical to the accounts
given to the court by Post Office lawyers, especially in
Mr Parsons' 17th witness statement at {C11/22/1}, which
made no mention of any advice given to Dr Worden and at
least --
MR JUSTICE FRASER: Where are we going now?
MR GREEN: {C11/22/1}.
MR JUSTICE FRASER: Page?
MR GREEN: If we go forward to page {C11/22/6} and if we
look at paragraph 22:
"His covering email to the court provided that:
"The further work ... was done at his own
instigation and not prompted by Post Office or its
lawyers."
22.2:
"In Dr Worden's opinion, this work led to a material
change in his opinions and that he believed he was
obliged to inform the Court ..."
22.3:
"A draft version of the report was provided to
Mr Coyne ..."
And so forth.
Then it deals with Dr Worden's email to the court.
That does not make clear what we find in the transcript
at 190, lines 9 to 10, which is where he tells
your Lordship that he had been given advice on the
content of the email that was going to be sent as
a covering email.
So we are now faced with a situation in which
Post Office's closing submissions at 316 say that the
word "told" could give the wrong impression {A/6/120}.
And in our submission, if Post Office are saying that
Dr Worden's evidence is misleading, they should disclose
all communications with Dr Worden on that issue, which,
in the circumstances, either are not privileged, or if
they are privileged, that privilege has been waived by
saying that it could give a misleading impression.
And if Post Office has any resistance to that
course, they should be put to their election to withdraw
the suggestion in that paragraph that Dr Worden's
evidence gives a misleading impression, or waive any
privilege there might be in relation to their dealings
with Dr Worden on those two points.
My Lord, I'm now turning to a new topic which is
bugs, and I mentioned this morning that we obviously
covered the ground with Dr Worden on bugs 1 to 10 in
cross-examination to try and give a sufficient number of
bugs and a grouping of bugs that was not arbitrary or
cherry-picked to try and give the court a reasonably
representative sample by which to judge the reliability
of the analysis urged on the court by Dr Worden.
We explained those in our closing submissions at
paragraph 535 onwards. However, in Post Office's
closing submissions, notwithstanding that there are
really only three bugs touched on in cross-examination
where they had four days, there is now an analysis,
detailed submissions on every one of the 29 bugs in the
bugs table, and that's in their closing submissions at
appendix 2 and it runs from page 400 to page 537.
{A/6/400}
Given the time since receiving that document and the
time available today, I hope your Lordship will forgive
me for taking the next five bugs, 11 to 15, and
identifying whether or not Post Office submissions in
this new insight are reliable at all.
I respectfully invite your Lordship to note as we go
through two features. The submission that is made where
there is a document I will show your Lordship frequently
is simply not borne out, particularly where a gloss is
put onto things to suggest user error when, as we saw
with the Romec PEAK before Pat Carroll's final sign-off,
cannot preclude user error. It is very different from
reaching a conclusion that it was.
We will show your Lordship why that's also
a particularly unreliable conclusion for other reasons.
So what use they make when there are documents, and the
second point is where we just have assertions not
footnoted and not tethered to evidence the court
actually has before it. And those are the two
particular features.
Now, in Post Office's treatment of the bugs, they
are sorted out into different categories, and I will
just get your Lordship's eye on it, if I may. {A/6/400}
is the beginning of this 137-page appendix of the 29
bugs.
Where they break down is that the submission now is
that eight are not bugs at all. That's paragraph 3.1 on
page 400. So eight are said to be not bugs at all.
Three had no branch impact. Nine had, or potentially
had only transient impact, and nine caused or had the
potential to cause lasting impact but were resolved by
Post Office and Fujitsu.
They then list: paragraph 4 are the ones that are
not bugs at all; paragraph 5, no impact; paragraph 6,
transient impact; paragraph 7, lasting. {A/6/401}
My Lord, can I just identify even at that stage that
Post Office has listed nine bugs having a transient
impact.
MR JUSTICE FRASER: They are the ones in paragraph 6,
I think.
MR GREEN: Exactly. And nine bugs in paragraph 7.
So pausing there, on a correct construction of
Issue 1 there are 18 bugs meeting the definition of
Issue 1 even on the Post Office analysis in its closing
submissions here.
I'm going to take your Lordship, if I may, just
briefly through bugs 11 to 15. I have to deal with bugs
11 to 12. Bugs 13 and 14 were accepted by Dr Worden,
but bug 13 Post Office has changed its mind on. So I'm
going to have to deal briefly with 13.
MR JUSTICE FRASER: Which ones are you dealing with?
MR GREEN: So 11 and 12 weren't agreed by Dr Worden anyway.
MR JUSTICE FRASER: Yes.
MR GREEN: 13 I still have to deal with because although it
was agreed by Dr Worden, it is not anymore by
Post Office.
MR JUSTICE FRASER: 13?
MR GREEN: Then 14 was accepted by Dr Worden and is still
accepted, so I won't trouble your Lordship with that,
but 15 is still in dispute. So it is 11, 12, 13,
because of the change of position, and 15.
So on number 11, so this is one said to have no
branch impact, we see that at {A/6/401}, paragraph 6.
And Post Office describe the relevant PEAK --
MR JUSTICE FRASER: This is Girobank, yes?
MR GREEN: Sorry?
MR JUSTICE FRASER: Girobank?
MR GREEN: My Lord, yes. Girobank discrepancies, exactly.
MR JUSTICE FRASER: Okay.
MR GREEN: At page 452, paragraph 9, {A/6/452} your Lordship
will see the Issue 1 heading under which Post Office
analyses what happens as follows.
Paragraph 10 identifies the main PEAK for Girobank
noticing there was a £505.72 discrepancy, and says at
(1):
"This was a known issue dealt with by KEL
MWright531p. This KEL is now deleted and irretrievable,
but details about it can be gleaned from its associated
Peaks."
Then 10(2):
"The issue arose when a giro transaction was entered
and then reversed, with the reversal being entered after
the report cut-off time."
So reversal was not included in the following day's
report.
Paragraph 10(3), rather importantly:
"This led to an error notice being issued on the
mistaken basis that the branch had a discrepancy."
So pausing there, Post Office actually themselves
set this out that the way this arose was because
an error notice was issued on a mistaken basis in
relation to the original 505.72 because of the fact that
the post-cut-off time problem meant the following day's
report didn't capture the reversal.
Paragraph 10(4) says:
"The fact that the reversal, performed after the
daily cut-off, did not show on that day's report
reflects the intended operation of Horizon.
Subpostmasters were instructed that if a reversal is
carried out to giro transactions after cut-off, a manual
summary will need to be produced for Girobank. Issue 1
is therefore not a 'bug'."
Now, obviously how well they were instructed to do
that and all that sort of thing, whether they were at
all and in what terms, is at large. But the suggestion
there is the way that the system worked in this respect,
creating this mistaken footing for an error notice, is
not a bug, a defect or error in the data because that's
the full definition in Issue 1. And we respectfully
disagree with that. But that's what's said. So not
a bug, conflating all three into one.
Then paragraph 10(5):
"Rather, Issue 1 relates to reporting. The
underlying data is correct and the branch's accounts
would have been correct at the end of the trading
period, once the reversal had been recognised (at the
time of this Peak, the trading period was weekly).
Mr Coyne appeared to accept this in cross-examination.
In this particular case, the only possible impact would
be if the branch had accepted the error notice received
because of the reporting issue."
Now, if we just move forward if we may now to
Issue 2, which is at paragraph 17 on {A/6/455}. This is
what's described as the secondary problem initially by
Mr Coyne.
At paragraph 17 your Lordship will see:
"Issue 2 was that an £81 giro deposit was included
on two consecutive daily reports. This is because the
transaction was entered onto Horizon in a precise (and
very small) window of time between two system calls
being undertaken, resulting in a duplication. The
overall branch position would still have been correct,
but the daily reports to Girobank may have been wrong.
If they were (ie if the same transaction was included on
two consecutive daily reports), it is expected that this
would have been spotted and a TC would not have been
issued to the branch."
So pausing there, that is a pretty astonishing
example of the one-way street.
So we have got the fact of an error notice being
mistakenly issued as the premise upon which the £81
problem is discovered. You have got the assertion that
this whole thing is not a bug when that plainly is
a bug. In fact, we say both are bugs or defects or
errors in the data, and obviously so. Post Office says
neither of them are and does it effectively on opposite
footings for 1 and 2.
We say that is strikingly a one-way approach to
resolving that issue. It is then helpful to look at the
underlying PEAK, which is at {F/25/1}.
If we begin on page 1. In the top light green box
there is a reference:
"04/05/00, 13:44 system error - giro bank said there
is a discrepancy on the giro figures."
Come down a bit, username "1ha001":
"Girobank have been in touch to say that there is
software problem as the figures are not correct. Daily
figures when totalled are more than the cash account
giro figures."
Then if we go down to the words just below the
"daily figures 85990.88."
MR JUSTICE FRASER: Sorry, where are you now?
MR GREEN: Halfway down the box, my Lord.
MR JUSTICE FRASER: The same box?
MR GREEN: Yes.
MR JUSTICE FRASER: "Daily figures 85990.88."
MR GREEN: Exactly, and if we come down below that:
"The pm has checked all dockets and all reversals
that may have been done and cannot find anything.
Therefore he would like this investigated further as an
error notice has now been provided and he does not want
this to happen again."
So this is the postmaster or postmistress
challenging the error notice having tried to check
everything and not being able to find any justification
for the error notice.
Then if we go to the bottom box, just underneath "F)
Response":
"This difference (£505.72) between the Cash Account
and the Daily reports is explained [in the MWright KEL].
There was a giro for this amount that was entered on the
13th Apr then reversed AFTER cutoff then re-entered
again and reversed again. The Daily report would have
shown the original £505.72 but the daily reports never
show reversals.
"It would be nice to close the call as known error,
however while investigating the message store I have
identified another problem ... there is a Giro Deposit
for £81 (1-17240) that is being calculated in TWO
consecutive cutoffs (18th AND 19th April). I have
attached the full message store as evidence, however the
error happens in message ..."
Then there is a typed number.
If we go over the page, please, {F/25/2}, in the top
box, in the second paragraph, three lines down on the
right:
"There are two separate calls --"
MR JUSTICE FRASER: Where are you looking now?
MR GREEN: The top box, my Lord. In the second paragraph in
the top box which begins "after further investigation".
MR JUSTICE FRASER: Yes, I have that.
MR GREEN: On the right-hand side, three lines down.
MR JUSTICE FRASER: I see, yes.
MR GREEN: "There are two separate calls to find the latest
messages, and this gives a very small window of
opportunity for another transaction to have been
registered (The £81 giro was entered at EXACTLY the same
time as the TideMark was generated). The chance of
having a transaction entered at the same time... "
MR JUSTICE FRASER: No, not entered at the same time,
entered at the time.
MR GREEN: I'm sorry:
"... entered at the time between the two calls AND
causes the SEQ number to be greater than the tidemark is
very small but real."
We then see in the next box it is passed to EPOS FP
to correct the problem caused.
MR JUSTICE FRASER: By the difference between the mark and
sequence attribute.
MR GREEN: Exactly, and there is a fix that should address
all cut-off reporting, not just Girobank reports.
So we can see there that at times it requires a code
fix.
It is not limited to Girobank, the problem, and it
is a small but real problem which has been uncovered
effectively by accident.
If one looks back at Post Office's closing at
paragraph 17 on page {A/6/455}, it effectively
acknowledges that there is a discrepancy caused. It
says it is expected that it be spotted before TC is
issued. But there's actually no evidential foundation
for this, particularly in circumstances where what we do
learn from the PEAK is that an error notice had been
issued by mistake in relation to the sum of over £500
through which the £81 mistake was discovered.
So it doesn't say it, and what evidence it does
provide tends to be, if anything, pointing in the
opposite direction to the assumption the court is
invited to make about a process which falls outside the
system as defined in the Horizon Issues.
Post Office's submissions in relation to this at
paragraph 39 {A/6/460} in their conclusion on page 460
effectively involve a criticism of Mr Coyne in this
respect.
They say:
"That analysis is incorrect. None of the Peaks
referred to by Mr Coyne demonstrate a direct financial
impact on branches; in most cases this is because the
issue affects reporting whilst the underlying data
remains unaffected."
And so what they are doing is they are confining
their data for the purpose of Issue 1 to what's actually
in the branch count on the face of Horizon without
really grappling with the word discrepancy of course, or
shortfall, which implies a comparison between one thing
and something else.
So it is wrong as a matter of principle and, we say,
and characterisation and fact.
My Lord, can I turn now to bug 12, please. That
starts at page {A/6/461}. It is the counter replacement
causing one-sided transaction and this is said to be
a bug with transient but not lasting impact.
Your Lordship has our submissions that that's more
than sufficient for Issue 1 and that Dr Worden's
approach to permanence is wrong. Post Office's closing
submissions at paragraph 3 under the nature of the issue
refer to the PEAK PC0058528, which is a replacement of
a counter's hard drive, and that replacement caused two
messages relating to an OBCS transaction to be
overwritten, resulting in a receipts and payments
mismatch. A transaction with a value of 167.12 was not
added to the cash account. So there is a discrepancy of
£167.12 on Post Office's own description.
Then paragraphs 4 to 10 set out technical detail
which is not referred to {A/6/462}. There is no
reference to where that's found in the evidence. And
paragraph 11 confirms the discrepancy of £167 and says
it would have been flagged to the SPM. So this is
rather like the would have evidence we had in Common
Issues. It invites an assumption of a uniform
favourable practice in Post Office's favour performed
faultlessly.
Then at paragraph 13 {F/6/463}:
"Information of the overwritten messages was passed
to MSU who created a BIMS report for Post Office and an
error notice would have been issued to hold the branch
harmless thereafter."
Against that background it is worth looking at the
PEAK at {F/77.1/1}, Mr Miletic rightly reminds me.
That's the reference to the discrepancy of £167.12 we
see on page 1. If we go over to page {F/77.1/2},
please, in the third box down underneath "F} Response",
halfway down in the third box:
"This is a single counter outlet,n and the counter
was replaed on 22nd and two messages were overwritten.
KEL JBallantyne5328R. The messages retrieved from the
mirror disc show that a transaction for product number
184 value £167.12 was overwritten. I have attached the
two sets of messages as evidence."
Then if we go over the page {F/77.1/3} we can see at
the top of that KEL, top of the page, the PEAK:
"Like the other cases this is a single counter
office which had its hard drive replaced due to problems
with it."
So we can tell there are other cases of this type.
Then if we go to the KEL that's mentioned there, which
is the Ballantyne 5328R KEL, which is at {F/421/1}, what
we actually find in the KEL is underneath "Solution -
ATOS" towards the bottom:
"To find the overwritten transactions for
reconciliation we need to look at the Ripostemirror
messagestore."
And we can see the number attribute prior to the
rogue data in the third line. So that is being done
there, as we can see on the right-hand side. Halfway
down is to tidy up the counter and analyse the content
of the messagestore:
"For a multi-counter outlet (MCO) need to retrieve
the messagestore from another counter, as well as the
affected counter ... transaction numbers for the
RiposteVersionString messages should reveal the original
transactions. When you have identified any missing
transactions attach the details to the PinICL and route
to MSU."
So what is not there is a reference to the BIMS
report there or error notices, or TC or an instruction
to do that in the KEL itself.
MR JUSTICE FRASER: And where are you comparing that with?
MR GREEN: What is suggested at the closing submissions is
paragraph 13 where they say:
"An error notice would have been issued to hold the
branch harmless thereafter."
MR JUSTICE FRASER: So your point is that hasn't come from
the KEL?
MR GREEN: No. If it is somewhere that our attention has
not been directed to by the way it has been presented,
there are plenty of other examples of points without
foundations in the underlying documents.
MR JUSTICE FRASER: For example, in paragraph 14, the final
line:
"A further change was made to stop Riposte writing
messages as it came online."
MR GREEN: I'm not quite sure where that is. We haven't had
a chance to search randomly. It may be in one of the
documents, but given the pressure of responding to
545 pages.
MR JUSTICE FRASER: All right.
MR GREEN: In that case, my Lord, it is right that the PEAK
says a final BIMS should be issued, but a KEL which is
passed across says pass to MSU.
MR DE GARR ROBINSON: Sorry, is my learned friend suggesting
that those things two are inconsistent with each other?
MR GREEN: Well, they are different and whether or not the
transaction corrections are actually issued is at large.
What is suggested to Mr Coyne on {Day17/98:9}, if we
have a look there -- if we look at line 9 on page 98.
MR JUSTICE FRASER: Where?
MR GREEN: You will see:
"Question: Here's what interests me, Mr Coyne.
What you are saying is -- let me do it this way.
I would suggest to you that on any fair and reasonable
reading what this PEAK demonstrates is, first of all,
that Fujitsu spotted that there was a failed recovery
situation?
"Answer: Yes.
"Question: Very reliably. One can reliably assume
that's going to happen, yes?
"Answer: Yes.
"Question: Looked into the underlying circumstances
at the branch at the time of the recovery. Again one
can reliably assume that's going to happen?
"Answer: Yes.
"Question: Then formed the view it was necessary to
work out what had happened on the ground in order to
know whether any discrepancy had been created or not,
yes?
"Answer: Yes.
"Question: Then sent through a BIMS to Post Office
to tell Post Office to reach out to the postmaster and
ask what actually happened on the ground?
"Answer: Yes.
"Question: And I further suggest to you, Mr Coyne,
that the reason why Fujitsu sent that BIMS and the
reason why Post Office received that BIMS, they don't
receive these documents in order to put them in a pile
in some warehouse and never look at them, they receive
them so that they can be acted upon?
"Answer: Yes.
"Question: And on any fair reading of the evidence,
it would be extraordinary in this case to assume that
having received that BIMS, Post Office would not have
reached out to the postmaster, ascertain what had
happened and sent a TC or not depending on the
postmaster's answer."
So that is the footing on which it is put to
Mr Coyne.
His answer at line 17:
"Answer: Yes, but this is quite clear, when you
read the heading "Recovery Failures", that it is seeking
to address Horizon Issue 4: to what extent has there
been the potential for errors in the data recorded in
Horizon?"
So the point Mr Coyne is also making in relation to
that is the need for a transaction correction creating
the discrepancy in the first place.
So where we end up in relation to this is the
financial impact is admitted by Post Office and so it is
sufficient in any event. But it is clear the court can
have no confidence in this not being a lasting
discrepancy because although there is a reference to
a BIMS being sent through, there is no evidence it was,
and there is no evidence about what action Post Office
took in relation to it. And that all falls outside the
system as defined in Horizon Issues, which is the reason
that there wasn't disclosure in relation to the TCs
beyond the overall numbers and so forth which we had.
So, my Lord, I would now like to move to bug 13
which is the one that is now said not to be a bug at
all. This is addressed at Post Office's closing
submissions at {A/6/466}.
Post Office dispute this is relevant to Issue 1
because, as they set out at paragraph 2, it is not a bug
at all, and it is curious for two reasons. Firstly, it
is in Dr Worden's list of 12 bugs which had a financial
impact in joint 2 at paragraph 112, the reference for
which -- we needn't go to it -- is {D1/2/27}.
The second reason becomes apparent when we look at
it more closely. Post Office's submissions at
paragraphs 528 on page 467 describe the initial issue in
the PEAK, and essentially what happens is Post Office
withdraws stock, namely a £5 saving stamp. The SPM
returns the stamp as required but does not rem them out.
This leads to a £685 discrepancy which paragraph 6
describes -- and this is rather important -- as "pure
user error" {A/6/467}.
It is said at paragraph 8 that:
"The [SPM] elected to make good the shortfall and
a credit transaction correction was subsequently issued
for.£685 to rectify the issue."
Then at paragraph 9:
"However, a bug in Horizon ..."
Bearing in mind that this is not a bug apparently:
" ... caused the £685 of stamps to be subsequently
reintroduced into the branch’s accounts on two
occasions. By this point, Horizon was showing that the
branch was holding £1,370 of the stamps."
So there's, at the very lowest, something of a
tension in Post Office's analysis of whether this is a
bug, error or defect as defined in Horizon Issue 1.
They say not. And that's the basis upon which they
departed from Dr Worden's approach.
Paragraph 10 is also important because there's
another unrelated trading issue that was said to have
likely caused the SPM not to notice the first instance
of the withdrawn stock being introduced.
That's obviously unhelpful for the countermeasures
that Dr Worden relies on if there are any difficulties
in accounts not being able to immediately identify, as
Post Office acknowledges you may not be able to, the
cause of Horizon system-generated problems.
If one then looks at the actual PEAK itself, which
is at {F/765/1}, it doesn't actually support the
submission that is made that this was pure user error.
And we see that at -- we start on page 1 of 765. We can
see under "user names" in that top box, go down just
over halfway down:
"User names - SK1001 and PCA001."
It says under that:
"This office physically held 137 £5 PO saving stamps
..."
And this is really important:
"... and did not rem them out before the date the
rem out icon disappeared. The office physically
returned the stamps to Transaction Processing as advised
and the office then did a Trading Period balance ..."
So pausing there. We do not get from the PEAK
itself the fact that the SPM failed to do what they were
advised to do in terms of failing to rem them out. What
we do get is that they returned the stamps to
transaction processing as advised.
We can see towards the bottom of that box, there are
two two-line paragraphs:
"I have spoken to Phil Herrett in Transaction
Processing who has confirmed he is aware of about
8 offices with similar issues with the stamps still
showing on the stock."
Gives an example:
"Can this be investigated to see why the stock is
still showing in the office, and if this keeps giving
the office a loss of £685 every time they do a Trading
Period balance."
Then what we then find is over the page {F/765/2}:
"This is an example of the problem described in KEL
PothapragadaC4913L."
Underneath that:
"The change to prevent the withdrawn product being
put back into stock has been rolled out across the
estate so there should be no further new occurrences of
this problem."
We can see in the rest of the PEAK, if we go over to
page {F/765/4}, just over halfway down in the yellow
box:
"I have spoken to Gareth Jenkins ref this, he is
going to find some time to go through the issue with me
this week."
Then Mr Charlton, the penultimate green box, halfway
down that:
"... also checking to see how many other offices are
affected by this issue as there may be some who have not
reported the problem."
Which we say is realistic.
Then there is a workaround over the page at the top
{F/765/5}.
MR JUSTICE FRASER: We are still in the PEAK, are we?
MR GREEN: We are still in the PEAK, my Lord, yes. In the
top yellow box:
"To continue investigation into the root cause of
this issue I need to request some information from the
audit team ... continue monitoring the progress ..."
If we go to the end, {F/765/6}, three yellow boxes
down, the audit data that Mr Charlton had been supplied
with was blank. So he is trying to get another disk
with the required data on it, and the closure,
notwithstanding that had there is a fix rolled out to
cover that, is:
"Development [confirms] a refdata fix will be
delivered to prevent any further occurrences of this
problem.
"As NBSC have a circumvention for the issue, a KEL
covers the scenario and a fix is pending I am closing
this call.
"... Category 95 ... Advice after investigation."
Which does not capture the reference data error
problem that was underlined in the system which was
corrected by the fix.
MR JUSTICE FRASER: But that KEL is the PothapragadaC4913L,
is it?
MR GREEN: That is the KEL being referred to, yes.
MR JUSTICE FRASER: Do you have a reference for that KEL?
MR GREEN: My Lord, I don't off the top of my head.
MR JUSTICE FRASER: I don't expect you --
MR DE GARR ROBINSON: It may be {F/678/1}. I'm just going
from the appendix. If I have got that wrong let me
apologise in advance.
MR JUSTICE FRASER: I'm looking at the appendix, but in
paragraph 12 it mentions that KEL but it doesn't have a
footnote, so ...
MR DE GARR ROBINSON: I'm looking at paragraph 121.2 on
page 466 and there is a footnote to the
PothapragadaC4913L.
MR JUSTICE FRASER: F/678, thank you very much. Yes, thank
you very much, Mr de Garr Robinson.
MR GREEN: My Lord, could I take your Lordship back, with
that background, to what's said about this in the
closing submissions at 466 and 467.
MR JUSTICE FRASER: Yes.
MR GREEN: It is {A/6/466}. So we have got the contention
that it is not a bug at all at paragraph 2, and over the
page, {A/6/467} it says at paragraph 5:
"Subpostmasters would have been instructed to rem
out any excess stock ..."
MR JUSTICE FRASER: Yes.
MR GREEN: That appears to be the foundation for the
assertion that this was pure user error in paragraph 6.
Then paragraph 9 openly recites that a bug in
Horizon caused £685 of stamps to be subsequently
reintroduced.
MR JUSTICE FRASER: Where are you looking now?
MR GREEN: Paragraph 9, my Lord.
MR JUSTICE FRASER: Yes.
MR GREEN: A bug in Horizon caused £685 to be subsequently
reintroduced, which we respectfully say is just
hopelessly irreconcilable with saying "this isn't
a bug".
MR JUSTICE FRASER: I think I have got that point.
MR GREEN: I'm grateful.
MR JUSTICE FRASER: I think you did make that point a little
earlier.
MR GREEN: I'm grateful. Two layers --
MR JUSTICE FRASER: I do generally get points first time
round.
MR GREEN: Noted.
Bug 14 we don't need to deal with because that's
agreed by Dr Worden and Post Office haven't changed
their mind about it. Bug 15 is phantom transactions
which is picked up at 473.
MR JUSTICE FRASER: Do you want to have a break for the
shorthand writers before you go on to bug 15?
MR GREEN: My Lord, I think I was asked to go a bit further
last week before stopping.
MR JUSTICE FRASER: That's fine.
MR GREEN: If that's all right.
We pick it up at 473 {A/6/473}, which is phantom
transactions. At paragraph 2:
"Post Office submits that this is not a bug at all.
Manifestations of this alleged bug are either design
features of Horizon or user error."
The footing for that, the key aspects of the
Post Office analysis are paragraph 8 on page {A/6/474}
where they criticise Mr Coyne for failing to refer to
the other two PEAKs, both of which state the issues at
Old Isleworth were attributable to user error.
As your Lordship has already seen in relation to
that PEAK that is the Romec one and Pat Carroll.
MR JUSTICE FRASER: Yes.
MR GREEN: That's a bold gloss to put on that PEAK and
a pretty bold departure point for criticising Mr Coyne.
Paragraph 11:
"It is not possible from the PEAK to know what the
Romec engineer saw. Indeed there are a host of possible
explanations ..."
Then Post Office's closing submissions:
"Mr Carroll ultimately determined that there was not
a fault in Horizon." {A/6/476}
Which is the not so as to preclude user error point.
Then 477, paragraph 16 contends that:
"Mr Coyne gives a misleading impression of what the
Peak actually says and ignores the later content of Peak
PC0062561 and indeed the conclusions of the Peak.
Mr Coyne fails to mention at all the conclusion to Peak
PC0065021 that all reported cases are attributable to
user error."
So that is a premise for criticising Mr Coyne for
not acknowledging that conclusion: that all reported
cases are attributable to user error.
I'm not going to take your Lordship back to the
Romec/Pat Carroll PEAK, so I think we can take that
reasonably hopefully as read, but that's {F/97/1}.
That is where Pat Carroll on page 7 obviously has
himself observed the problems overnight and Romec have
separately observed the problems when visiting, and then
it is closed in a way which we say is thoroughly
unsatisfactory.
Mrs Van Den Bogerd was cross-examined on this
{Day5/40:3} to line 14. She didn't seek to suggest in
her evidence that the Romec engineer was likely to have
been in error in some way, notwithstanding her
experience of dealing with postmasters herself,
subpostmasters herself in her previous roles.
The Post Office says the user error was the
conclusion in the other PEAKs, and so we need to look at
{F/100.1/2} which is the conclusion which is PEAK
0068327.
On page 2 there, it says:
"Following a significant amount of monitoring we
have been unable to definitively link any
equipment/environmental issues to any particular event.
There have been incidents which showed a possible
correlation between system activity and phantom
[transactions], these pointed to a touch screen problem
and as a result the screen was replaced with a resistive
model. As this produced no measurable improvement it
has to be assumed that the problems were user related."
So there's no definitive determination, and the
basis that Pat Carroll says it has to be assumed
problems were user related is the screen has been
changed.
Let's go to the other PEAK, which is at {F/88.2/2}
and then perhaps take a break.
If we look at page 2 of that PEAK we can see the
bottom or penultimate green box, the large green box,
17th August 2001, 9.13:
"This outlet has reported continual phantom
transaction problems causing us to exhaust every
possible course of action in trying to solve them."
We have, and then there is a list of what they have
done. Then below "holiday":
"After all this the PM is still experiencing Phantom
transactions but they are mainly on counter positon 1
and this is always used by Robert Parker (PM). I have
asked Robert if I can spend some time at the outlet with
him so I can be present when the phantoms occur but he
is not keen for this to take place as he feels the
outlet is too small and gets too heated as it is.
"I spoke with HSH this morning and she advises that
since power help was last archived, Mr Parker has logged
34 calls to the helpdesk and a vast amount are advice
and guidance. My personal feeling is that Mr Parker
could do with some further training and I feel that this
should be our next course of action. The only other
option we have open to us is to change the ISDN line
which is the old style, but myself and HSH feel that
this is an expensive option to go down when it may be
user error at fault."
Then over the page {F/88.2/3}, the yellow box, third
line:
"From RNM," the regional network manager:
"I spoke to training and Dev this afternoon and
arranged 2 days training for next week, when I rang
Mr Parker he told me that he did not need the extra
training so I have now cancelled it. He also told me
that the phantom transactions have stopped.
"PON to RNM: 'There seems to be no issues at this
outlet if you are happy with the postmasters response.
"Is there anything else that needs investigating at
the outlet proven to be directly liked with phantom
[transactions] ... as there are none recorded? If not I
would like your agreement to close down this problem as
now resolved. I would like to make you aware though
that the postmaster does seem to be making quite a few
calls still to the HSH helpdesk, mainly around simple
things such as reversals.
"RNM to PON: Thanks for making aware about the
number of calls your still receiving, I don't think we
will ever stop him from making these. I see no reason
why this call cannot not be closed. As I said the
Postmaster said he is no longer getting these
transactions. Calls have actually reduced in September,
there are ... only 4. I have agreed with PON that there
is little else which can be done. The PM is not making
errors with his work and the call volume has improved.
I have agreed to close the problem."
So what we get is it may be user error at fault and
the problems stopped, and that's the point at which they
decide to close it down.
If we then go forward to the KEL that relates to
these, which I can do in about one minute:
It is {F/174/1}. It is a one-page document:
"Symptoms.
"There have been several calls over the last few
months where postmasters have reported phantom sales.
Items appear by themselves for which the PM has not
pressed an icon. These may be individual items or
several of the same item. Sometimes when no one has
been near the screen items may appear."
Then under the "Problem", the last line of the
problem:
"A more recent case revealed that the cable between
the screen and the base unit was the root cause."
That rather undermines the reliability of
a conclusion that Pat Carroll thought was the only
available assumption, that once you change the screen
and the problems continue, you have to assume it is user
error. Because it turned out, as we can see from the
KEL, that in one of the cases it was the cable that was
the problem.
My Lord, that's the end of the piece in relation to
that. We respectfully say Post Office is putting
a massive gloss on the underlying documents in asserting
what they are asserting, and it is a thoroughly
unreliable account of that bug.
MR JUSTICE FRASER: Thank you very much. We will have
10 minutes.
(3.21 pm)
(A short break)
(3.31 pm)
MR GREEN: My Lord, I now propose to take a particular
example issue and just trace it through, if I may.
I would like to start, please, with {F/908/1} which
is an internal Fujitsu/Post Office document of
22nd March 2012, which is a review of various computer
weekly articles. And this shows that it appears I think
to have been authored by Fujitsu and this goes to the
identification of issues certainly by this date which
are still live in these proceedings.
Your Lordship will see at the top, first paragraph:
"A number of articles have been written by Computer
Weekly relating to the Horizon System and the issues
postmasters have had with deficits. The main article
was published in May 2009 and can be found in
Appendix 1."
There is a summary, it highlights seven case studies
where they claim faults.
If we go forward to page {F/908/3}, we can see that
the second case study is that of Jo Hamilton, and we can
see in the key points box on the right, second bullet
point:
"Postmasters claim faults with the technology are
generating unexplained losses."
Third bullet:
"Post Office denies IT fault could cause accounting
system to show incorrect balances."
So this is the Computer Weekly article of
11th May 2009, as we can see from the top.
If we go over the page to {F/908/4}, we see a brief
reference to Jo Hamilton's case being ultimately:
"... signing her accounts even when she knew they
were wrong, because, she says, calls to the Horizon
helpline didn't stop the deficits occurring and she felt
backed into a corner. She was convicted of false
accounting, but was spared a prison sentence after local
villagers organised a collection to pay the debt."
If we go forward to page {F/908/6}, the relevant
case study is at the foot of page 6. Her name appears,
Jo Hamilton, at the bottom of page 6 and over the page
to page 7.
The specific point that I just want to trace through
in the evidence is the idea of problems with sums
doubling, because it was one of the things that was
certainly known to Post Office to be an issue.
As we see in Jo Hamilton's case {F/908/7}:
"'One time it said I was down £2,000, so I rang the
Horizon helpdesk. The supervisor told me to do various
things, and three minutes later I was £4,000 down.
Whatever I did after that, I couldn't get it to come up
any different,' she says."
So that is the Computer Weekly report of what
Jo Hamilton had said there. So we know that doubling
had been identified certainly as an issue, although it
sounds very implausible when you first hear about it
perhaps.
If we then go forward to {F/930/1}. I should
mention that I think the correct date on that
document --
MR JUSTICE FRASER: By that document, you mean 930?
MR GREEN: The first one is 2012 at 9.08. It is in Opus
I think as 2010, but it bears clearly a date of 2012.
So if we go back to 17th May 2010, this is a pack
that was prepared in advance of a meeting with
James Arburthnot and Oliver Letwin, both MPs who had
affected constituents, and it is a plan for what was
going to be said. And we can see the agenda on page
{F/930/2}.
There is a reference to "Review Jo Hamilton case" at
6a, and over the page on page {F/930/3} "Introductions",
third bullet point:
"We understand you have raised some concerns, and
are representing the concerns of subpostmasters in your
areas.
"We are open to feedback and we will provide you the
information we have available, our aim is to be open and
transparent."
Then over the page {F/930/4} at item 3 "Horizon -
background", second bullet point:
"Although we recognise that Horizon is not perfect,
no computer systems is, it has been audited by internal
and external teams, it has also been tested in the
courts and no evidence of problems found (of the nature
suggested by the JFSA) ..."
Third bullet point second line:
"Both versions of Horizon were built on the same
principles of reliability and integrity."
Then if we come down to 6a {F/930/5} on the next
page, there is a summary of the "Review Jo Hamilton
case":
"Cash holdings ...
"Audit findings.
"She was in personal financial difficulties.
"She was provided an opportunity for an explanation.
"She did plead guilty to fraud."
It is described there.
If we go over the page, it says {F/930/6}:
"What are your thoughts on the meeting? Do you have
any areas of concern?
"We are considering commissioning an independent
audit as an assurance measure, but in light that there
is no evidence that there is a problem, we need to
determine if this is a good use of public money."
Then if we go forward to page {F/930/8} at
paragraph 2:
"What is our view of Computer Weekly."
They are obviously respected.
Third paragraph:
"As we have external and internal experts available
we don't believe Computer Weekly can assist us in this
specific case. Although there is no evidence of
problems with Horizon ..."
This is in 2010:
" ... Horizon, as an assurance exercise we are
considering an audit of our processes, data, and IT
systems. If we do proceed with this audit, it is likely
that we will use a professional audit organisation that
are tried and tested in this area as we require
organisation who have the reputation and the experience
of defending audits against external scrutiny."
Then in paragraph 3, the second paragraph:
"In cases where an auditor has found evidence of
fraud, the previous trial balance (which the
sub-postmaster has approved) will be the baseline
record."
Then in paragraph 5 {F/930/9}:
"Why are we considering Deloittes to perform the
audit?"
What's said there is:
"KPMG are excluded as they are Fujitsu's auditor.
"Ernst & Young are excluded as they are
Post Office's auditor."
So that was the view taken at the time, which gives
further background to the Post Office reliance now on
the ISAE service audits which conflict with the APPSUP
PEAK.
The audit at paragraph 6 is:
"The audit envisioned is a thorough end-to-end
review of processes, systems and data which not only
could reveal potential improvements but could be used as
an assurance for court future cases."
Now, against that background there is an explanation
in a little bit more detail in relation to Horizon on
page {F/930/12}. There is really one key reference
there which is the third bullet point down in the
"Summary", which is:
"Each transaction is audited and protected to
prevent change or tampering."
Now, your Lordship may remember that Mr Dunks was
cross-examined in relation to that point. We find that
cross-examination by Mr Miletic at {Day7/32:18}.
Line 18 is the beginning of the section. He recites
what's said at paragraph 4 of Mr Dunks' witness
statement, then the question at line 24:
"I just want to be precise about the language there.
"When you say 'audited transaction records', you are
talking about transaction records generated from the
audit store; what you're not talking about -- it's not
the case that those records are actually audited prior
to going into the audit archive, correct?"
"Answer: No, it is just the extraction data that
I'm taking out."
"Question: Exactly. And so ..."
It goes on to page 33.
We referred to that in closing at paragraph 219,
my Lord, for your Lordship's note, page 92. I needn't
take you there.
We then get on page {F/930/16}, back in the
Post Office document, Jo Hamilton's case and the
timeline. And if we go on to the second page of that
{F/930/17} we have the entry on 5th May, and the third
bullet point:
"SPM provides a pre-prepared written statement. The
statement states that the SPM did not receive adequate
training and that the operation manuals provided were
out of date. Statement also makes reference to an error
for £1,500 which is alleged to have doubled to £3,000
when attempts were made to correct it."
So whatever the precise amount, Post Office knew
from -- it looks like 2006 but recognised and recorded
it in 2010 in this document, that the issue of
an apparent disputed discrepancy, having doubled when
help was sought, was one that was being complained about
on any view.
Now, it may initially appear that that seems to be,
without knowing about the case, somewhat implausible or
surprising if Horizon is working as the explanation in
the pack explained it should.
If we go, please, now to {F/1333/1}, which is the
Second Sight report. Second report. If we go, please,
to page {F/1333/26} of that report, my Lord, not for the
purposes of establishing the facts that are set out in
it but for identifying that it is an issue that has been
raised, we can see that under the helpline section at
paragraph 12, which makes various complaints, people
having difficulty contacting, script-based responses,
instructions later countermanded and then being told
don't worry, it will sort itself out, people not knowing
how long they should wait for that to happen, how they
are supposed to balance the books in an intervening
period, and so forth.
We can see at 12.4:
"Many of the shortfalls suffered by Applicants to
the Scheme have, on the balance of probabilities, been
attributed to 'errors made at the counter' but that does
not, in our view always mean that more extensive initial
training would have eliminated all of those errors
although it would obviously have helped."
12.5:
"What we have observed is that, in many instances,
the biggest shortages seem to have arisen as a result of
'errors made while trying to correct earlier errors'.
We attribute this less to inadequate initial training
than to inadequate subsequent support when branch staff,
when they were attempting to correct errors that they
had previously made, just made matters worse."
Then if we go over the page, 12.6, {F/1333/27} we
can see:
"There have been numerous references to shortages
doubling, trebling or even quadrupling as branch staff
tried to correct, under instruction from the Helpline,
errors that they had previously made."
Then there is a reference to improved error
repellency at the end of 12.7 and the extent to which
transcripts are available, if at all, at 12.8, and so
forth.
Now, against that background it is interesting to
identify the trouble that SPMs had with doubles. If we
start, if we may, at page {F/24/1}, which is PEAK
0043811. My Lord, I should say this is a slimmed down
selection of the available doubling problems.
If we look page {F/24/6} of that PEAK, if we may.
The second green box up from the bottom, 16th May, 16.11
exactly, the three lines up from the bottom halfway
across:
"The RNM," the regional network manager, "put
£6,343.07 into the suspense account. The discrepancy
has now doubled and is showing as a £12,686.14 surplus."
Over the page on page {F/24/8} there is a suggestion
that this is another instance of a different PEAK where
the data server trees have failed to build and now have
been fixed hopefully.
If we go forward now to {F/149/1}, we are now in
2003.
MR JUSTICE FRASER: What date was that PEAK we were just
looking at?
MR GREEN: I'm so sorry, I should have mentioned it,
my Lord.
MR JUSTICE FRASER: 2000?
MR GREEN: That's 2000.
MR JUSTICE FRASER: Where are we going now?
MR GREEN: 25th April 2003 at {F/149/1}. On the first page
of that, the first substantive paragraph there, 24/04/03
at 15.44:
"Darren from NBSC states that the BM ..."
I think it should be PM.
MR JUSTICE FRASER: I see, 24th April 2003. 15.44.
MR GREEN: That's it. The target date on the PEAK at the
top is 28th April 2003.
MR JUSTICE FRASER: That's rather what threw me. Yes.
Darren from NBSC.
MR GREEN: "... states that the PM," it should be, "is
trying to reverse a Rem, but when this has been reversed
it is doubling up on a balance snapshot."
If we go over the page to page {F/149/2} and we come
down to the word "contacted", about two-thirds of the
way down the left-hand margin just below 25/04/03,
09.15:
"Contacted: Pm confirmed all previous information.
Summarise. PM was in SU 'Y' he remmed in £13910 and
continued to trade in that SU in error, he should have
been in 'I' SU, so he did reversals of all transactions
in 'Y' including the REM IN and expected to see a zero,
but his REM IN had DOUBLED to £27,820.
"He went through all CA checks with NBSC and Horizon
and a reversal was attempted but a message that a
'reversal CANNOT be reversed' came up which indicated
that a reversal WAS CORRECTLY done."
So the system was reacting on the basis that you
can't reverse something you have already reversed, but
the effect of it was to double the figure to £27,820
rather than put it back to zero.
If we go forward to page {F/149/4}, the second last
yellow box, 1st May 2003, 06.34, Walter Wright.:
"Notes for testers:
"As well as testing that the fix addresses the
problem described in this PinICL, it should be verified
that the fix for the original bug ... Has not been
undone. Other desktop modes should also be checked to
ensure that no regression has taken place."
It appears that there was a fix developed for the
original version of this bug, but nonetheless this has
happened again and there is a further fix being dealt
with. We can see that appears to be right because on
page {F/149/5} we can see in the top green box, four
lines up from the bottom:
"... call type L ... Category 46 - Product Error
Fixed."
MR JUSTICE FRASER: That's also on 1st May, I think,
isn't it?
MR GREEN: 1st May 2003.
MR JUSTICE FRASER: They are both on 1st May, aren't they?
MR GREEN: Exactly.
MR JUSTICE FRASER: Okay.
MR GREEN: Then we move forward to January 2004 at
{F/184/1}.
MR JUSTICE FRASER: Where are we going now?
MR GREEN: {F/184/1}, which is PEAK 0098230.
This PEAK is opened on 13th January 2004 at
15:54:26 18 15:48:19. If we go over the page {F/184/2}, in the top
box six lines down, just under 13/01/2004 at 15.43:
"Information: The RLM has been through the cash
account with the PM and tried to adjust the figures but
they keep doubling up."
So this is not the postmaster or postmistress on
their own.
MR JUSTICE FRASER: Does that say RLM?
MR GREEN: I think it is meant to be RNM, unless it was
regional line manager at the time. I can't remember
when it changed.
MR JUSTICE FRASER: Yes.
MR GREEN: If we go down to the penultimate green box on the
same page, 14th January 2004, 17.55.56, underneath
halfway down:
"This results in a discrepancy between the system
cheque figure and the declared figure."
We get:
"Something has changed in the counter code recently
(I think at COUNTER_EPOSS 20_3; released end Nov) which
causes the discrepancy to be recorded wrongly; so the
cheque discrepancy; instead of being cleared; is
doubled; and the cash is also wrongly adjusted."
When we go forward to page {F/184/3},
15 January 2004 at 14.32.05, which is the large green
box beginning just before halfway down the page, we can
see there is a response. Then just underneath where the
second paragraph starts, just underneath "3rd December",
does your Lordship have:
"This is a nasty problem ..."?
MR JUSTICE FRASER: Yes.
MR GREEN: "This is a nasty problem as; if cheques continue
to be declared each week on the Declare Stock screen;
the numbers keep doubling. However there is a simple
circumvention and; as they should not be declaring
cheques in this way; it shouldn't have much impact. But
I think it should be looked at and resolved (any chance
for S60??) - might be instances where it really does
cause problems."
Then if we go to {F/207/1}, this is a different bit
of double trouble. It is the Horizon KEL GC Simpson
1049L. We are in April 2004 to May 2004.
Symptoms:
"The PM balanced on Wednesday ... and noticed that
all currencies on hand had doubled up."
MR JUSTICE FRASER: That is an 04 KEL, isn't it?
MR GREEN: This is an 04 KEL, my Lord, yes. It is the 29th.
It is raised on 29th April 2004 and updated on
5th May 2004 by Mr Simpson.
MR JUSTICE FRASER: It is just for some reason it has
"March 19" on the top right-hand corner. That's just
why I asked. All right. Yes.
MR GREEN: If we go to page {F/366/1} and we come down on
{F/366/1} to the penultimate yellow box. Cheryl Card.
24th November 2006, 14.56.15, penultimate yellow box:
"Problem appears to be related to smartpost products
... by 1pm Bulk/PrePaidBulk. Credit and Debit figures
appear to have doubled - message for product 7967,
SaleValue is 7.40 but Credit is 1480; message for
product 8058, SaleValue is -7.40 but Debit is again
1480.
"Problem has occurred 3 times - see also," some
related PEAKs.
If we go over the page to {F/366/2}. If we come
down to the third green box, 29th November 2006.
MR JUSTICE FRASER: At 10.26?
MR GREEN: 10.26, my Lord, yes:
"In the 3 cases seen so far each riposte message has
had the <Credit:> attribute written immediately before
after the <Mode:SC>> with the value doubled."
I'm not quite sure what that means:
"Normally the credit attribute is seen at the end of
the message. We have not as yet been able to reproduce
the fault, but will continue to try and reproduce the
fault."
So what they can see is what has been in fact
written and it is not as it should be written but they
can't reproduce the fault.
Then we see the PEAK appears to be a duplicate.
This is three yellow boxes up from the bottom. It
appears to be a duplicate of another PEAK. If we go on
to page {F/366/3}, that PEAK is then being used to
progress the fault. So there is an administrative
response, closure code for that.
Then if we go to {F/590/1} we are now in
7th March 2010. It is raised on 4th March 2010 with
a target date of the 7th. This is PEAK 0195561. If we
go to page {F/590/2} and we look at the penultimate
yellow box, 5th March 2010 at 16.58.10. Cheryl Card:
"On 02/03/10 on counter 2 at 15.04, the clerk
attempted a Transfer Out of 4,000.00 from stock unit BB
to MS. Due to a system problem the Transfer Out doubled
up, so when the Transfer In was done on counter 1 ... it
was for 8,000.00. The branch now has a loss of
4,000.00.
"I phoned the PM and explained that the problem was
under investigation. The PM would like to have it
sorted out before she rolls into the next TP, which is
due on Wed 17th March."
Then {F/590/3}, the third yellow box down in the
middle of the page:
"Advised PM to print a balance snapshot. Ran the
Transaction Correction tool."
MR JUSTICE FRASER: Sorry, where are you?
MR GREEN: Halfway down the page, my Lord. The yellow box,
11th March 2010, 15.22.14:
"Advised PM to print a balance snapshot. Ran the
Transaction Correction tool."
So this is the use of the transaction correction
tool and it is then successfully repaired, as we can see
at the bottom of page 3, penultimate box.
Then there's a little bit of insight into the
context in which those at SSC are working, which we find
on page {F/590/5} onwards.
18th March 2010, the penultimate green box. Second
paragraph:
"As the fix is already released, I would like to
request the priority of the issue to be downgraded to C
as we are trying to investigate the root cause only.
Please let me know your thoughts on this."
Then two boxes below:
"What is missing from this PEAK is the explanation
of the events in terms of the requests, how they were
ordered and when any was committed. Only then can we
qualify the priority. The assumption is that we have a
fix. The facts are -
"1: A settlement request to timed.
"2: A retry of request timeout occurred.
"3: According to the DB entries both later
succeeded.
"Now unlike other reconciliation Peaks this stands
out because only one of the requests are specific to
settlement. They should have worked because there would
have been a unique constraint violation on the journal
entry of one of them and if we are not getting this then
this is still an issue!"
So he says he can't reduce the priority {F/590/6}.
There is further investigation. The request is failing
in the large -- over the page, on page 6. So there is
still an issue is at the top. Then:
"We can't reduce the priority ..."
Then the big yellow box says:
"Yes, the second request is failing and the first
request is committing ..."
MR JUSTICE FRASER: Where?
MR GREEN: In the large yellow box, 18th March, 14.26,
my Lord.
Then over the page {F/590/7}, if we look at the
third yellow box down:
"Timeouts were the underlying cause of the issue and
that there were long delays waiting on the DB to process
the 4 requests. In this case two of the requests were
committed and two correctly detected that the
transaction had already succeeded. There is an issue
with the 2 commits because this shouldn't have
happened."
Then a bit further down:
"We would like to find the root cause of the issue
as to how the duplicate entry was committed in the db."
There is further investigation, checking the
relevant database tables on page {F/590/8}. Then we go
to page {F/590/9}. That's the large yellow box at the
bottom of the page, 22nd April 2010, 15.35.54:
"I have gone through the counter logs, OSR logs and
the DB dumps provided in the peak. Lets analyze this
from the scratch.
"Peak has been raised when a clerk attempted a
Transfer Out of 4000.00 from stock unit BB to MS. Due to
a system problem, the Transfer out doubled up, so when
the Transfer In was done on counter 1 at 16:15, it was
for 8,000.00. The branch now has a loss of 4,000.00.
"I checked the counter logs and analysed the
request ..."
Second line:
"But note that in the BAL/OSR side this request was
ignored by the time out monitor and continued to execute
and hence updated the table."
That seems to be the problem.
About halfway down that box there is a two-line
paragraph:
"I have requested for the journal table dump, to
check where duplicate JSN entries exists in the table.
But from the DB dump, I couldn't find any duplicates."
Then over the page {F/590/10}, top box, halfway
down:
"So, we can see that the insert time stamp is
different for these 2 records and hence it might have
entered from 2 different requests."
MR JUSTICE FRASER: Sorry, where are you looking?
MR GREEN: Halfway down the top box, the yellow box at the
top. Halfway down:
"So, we can see that the insert time stamp is
different for these 2 records and hence it might have
entered from 2 different requests.
"I have no doubt that one of the records was
inserted by the request id ..."
Then the number. It was an original request:
"But I am not sure how the second report was
inserted. But I have doubt on the retried request ...
which didn't fail at the journal filter stage.
"So, I would request you suggest on this, since
there wasn't any evidence which shows that 2
[transactions] has happened and updated the tables."
So at the bottom of the page on page 10, Cheryl Card
is not happy with the classification of duplicate call
which appears to be because there was another similar
PEAK which was not immediately spotted to be the same
problem, and she says:
"I'm sending this call back with Response Rejected.
"Closing a call as 'Duplicate Call' results in
a black mark against me. It basically means that I
should not have sent the call over since the same
problem has already been sent over in a previous call.
"... (duplicate transfer of 4,000 cash) may have
been caused by the same underlying fault as PC0194893
(banking reconciliation), however I could not have been
reasonably expected to link the 2 calls and take the
decision that it was not necessary to send PC0195561
over for further investigation."
So we can see from there that there was also another
PEAK, the link to which was not initially appreciated,
and there was some pressure on Cheryl Card not to
identify as a distinct problem something that might be
linked to something else.
Then we look at, if we can move forward to
23rd March 2010, which is at {F/596/1}.
MR JUSTICE FRASER: We are going into a different PEAK?
MR GREEN: A new PEAK which is PEAK 0196154.
Summary:
"... remmed in some currencies and the figures got
doubled up."
Then halfway down under the second lot of equals
lines:
"PM states he remmed in some currencies and the
figures got doubled up on the system - previous call
regarding this was closed. POL state this is a system
issue."
Then over the page {F/596/2} we can see the problem
with the sterling value of dollars. I think it should
be $2,000 with a value of £1,320 at the top, rather than
200.
MR JUSTICE FRASER: Yes.
MR GREEN: Then 2,104 matches the pouch value and then there
is the third paragraph. It says:
"Then got another slip printed showing a different
session number ...
"When she went to do her report to check the bureau
stock ... her system showed as having a value of
£4,208.84 in the bureau stock which is exactly double
the original amount she scanned ..."
Then there's a further analysis suggesting that the
pouch, if we look at the penultimate green box, third
paragraph, suggesting that it was accepted at the
branch, accepted twice at the branch:
"... due to a system problem when the clerk used the
Prev button several times during the Delivery acceptance
... POL may have to issue a TC."
There is another PEAK currently with development in
relation to that problem. So that is doubling in
relation to currencies.
Then at {F/630/1}, new PEAK, 019877, which is the
Hucclecote fix PEAK that your Lordship may remember from
the trial. We see this at {F/630/2}. This is where
there is an additional office, which is the Hucclecote
one which has had a problem in the yellow box at the
both bottom:
"NBSC has just advised that another office had
a similar problem, although the discrepancy has now been
sorted out. Details of the site and problem are below
for information ...
"Office - Hucclecote SPSO ...
"Office was dealing with the discrepancy in the
office following the TP rollover, and selected settle
centrally. The office reports that nothing happened and
they ended up doing this a further 2 times before they
could proceed. This has resulted in the office settling
the loss centrally 3 times. This showed as such as the
total on the final balance. The Trading Statement and
suspense account seemed to be correct though. On
Monday, 19th April the office reported they showed a
cash gain of double the original loss and after further
investigation a suspense account was produced that
showed 2 clear loss from local suspense entries. We
have now cleared this by clearing gain from local
suspense, which should clear the gain in the office."
Then it says {F/630/3} immediately under that, the
first green box at the top:
"The solution we thought we had for Hucclecote ...
has not resolved the problem, but has actually doubled
the discrepancy. The original figures in suspense were
clear loss from local suspense ... of £998.81 which was
the original loss in for the branch and this shows
twice. We have entered a clear gain from local suspense
but has doubled the discrepancy that was showing on AA
su from £1,997.62 to £3,995.24."
So that's one where they think they have fixed and
what in fact happens is it has not been fixed at all.
In fact, it has doubled the discrepancy.
Then if we go forward to May 2017.
MR JUSTICE FRASER: Different PEAK now.
MR GREEN: My Lord, it is a different document entirely. It
is {F/1794.1/1}, which is the agenda for the Post Office
operations board of 23rd May 2018. The point 1 refers
to the fact that 1794 was the redacted version and the
version we are going to look at is the version with less
redactions on.
MR JUSTICE FRASER: Fewer redactions.
MR GREEN: Fewer.
If we look at page {F/1794.1/34} of that document.
It says:
"There has been 3 reported instances."
This is SAP, GUI Bureau Value Issues.
2018.
"There has been 3 reported instances this week where
the sterling value of remittances of bureau to branches
has doubled when the delivery has been booked into the
branch, For example, Aylesbury GPO were sent a bureau
rem with a total value of £6.219,81, however when they
booked it into branch it populated as £12,439.62"
That passage was previously redacted but we can see
it in 1794.1 finally, if we go to {F/1857/1} which is
hot off the press. This is a document very properly
recently disclosed because it is recent and it is
referring to the date of an incident on 21 June, 2019.
It is surpassing the doubling problem with
a tripling problem, namely the triplication of
transaction acknowledgements into branches for Lottery
transactions in the order of 2 and a half million
pounds.
"TAs for lottery TAs (Transaction Acknowledgements)
for Lottery transactions taken on 20/6/2019 have been
triplicated in branch causing discrepancies.
"Accenture loaded the file 2 times in error, once
correctly."
That appears to be, my Lord, three times in total
because we get a triplication of the sums involved.
MR JUSTICE FRASER: Two times an error, once correctly.
MR GREEN: Exactly. It is believe this is due to the
introduction of new RPOS system into pilot.
MR JUSTICE FRASER: What's RPOS?
MR GREEN: I think it is a new bit of script but I am not
sure.
MR JUSTICE FRASER: All right.
MR GREEN: "All Lottery branches are unable to reconcile
their stock and cash. GL accounts in the Finance systems
have triplicated data. If a technical fix is not in
place then FSC will need to issue Transaction
Corrections (TCs) to enable branches to balance.
"If a technical fix is not in place and FSC do not
issue TCs, then emergency suspense accounting will cause
significant issues operationally for FSC to manage.
"Communication sent to branches for the issues and
timeframe to resolve:
"ATOS/FUJITSU/ACCENTURE is engaged in finding the
root cause and developing a fix. Conference call
arranged to discuss and if a fix is not in place TCs
will be issued from 25/06/19."
Then there is a note which we have asked in
correspondence for an explanation of because we only got
this very recently. Which says:
"Please note:
"Materiality - the threshold would be anything with
a financial impact of over £50,000 and/or a major
adverse reputational or regulatory reaction; the 'Daily
Mail test'. At this stage, it is more helpful to
overreport than to miss the opportunity."
We have enquired where that materiality filter is
applied, whether it is prior to this report being made
or whether it is who that gets sent onto. But the short
point, my Lord, is that there we have a picture of what
Post Office explained it knew in both 2010 and 2012,
where doubling was in issue in a very serious case,
Jo Hamilton's case --
MR DE GARR ROBINSON: My Lord, is my learned friend seeking
to -- inviting your Lordship to make findings about
a criminal case that's before the CCRC?.
MR GREEN: I'm not.
MR DE GARR ROBINSON: Then why does my learned friend keep
talking about Jo Hamilton? I do find it striking that
sometimes he appears to be addressing a jury rather than
a judge.
MR GREEN: My Lord, it may be uncomfortable for the
Post Office to be faced with documents that show they
well knew that there was an issue about doubling, but it
doesn't matter whose cases they arise in --
MR JUSTICE FRASER: What you have been doing in the last
however many minutes, and I didn't stop you with the
references to Jo Hamilton at the beginning, you used
that as a springboard to take me chronologically through
the sequence.
MR GREEN: Indeed.
MR JUSTICE FRASER: I have already made clear I think on at
least three or four occasions this court's position on
the criminal cases.
MR GREEN: Of course my Lord. My Lord, I'm absolutely not
trying to invite your Lordship to make any finding
whatsoever about merits one way or the other. The only
point I'm making is that this is a matter which
Second Sight had identified had been raised by multiple
SPOs in the Second Sight report and on any view was
known to be an issue by Post Office from the documents
we have seen. That's the only point and it is not right
for my learned friend to try to seek to distract from
the stinging nature of the underlying documents by
objecting.
MR DE GARR ROBINSON: My Lord, there are rules about the
conduct of commercial litigation.
MR JUSTICE FRASER: There are.
MR DE GARR ROBINSON: One of those rules is that one doesn't
say things incautiously that might have an impact on
evaluations being done in another place in relation to
different proceedings.
Another one of those rules is not to conflate a
whole series of issues, all of which are very different,
bureau rems in, we have got other kinds of rems, we have
got transfers between stock units, all sorts of
different issues which my learned friend picks up
randomly over a period of 20 years and seeks to promote
them to your Lordship as a single problem which
Post Office has known about for all that time.
Your Lordship will have seen from the PEAKs that
that's not the case. There were individual instances of
individual problems, as far as I can tell, some of these
documents I have not seen before of course, which itself
is extraordinary given the nature of this trial and that
we are at the end of it.
But my Lord what's happened is my learned friend is
seeking to jumble things up and then create
an impression. If I may say so, he has given the game
away in this particular exercise over the last 45
minutes and I would invite him to stop.
MR GREEN: My Lord, doubtless my learned friend can amplify
tomorrow. What I was about to say was specifically to
identify the fact that, on any view, Second Sight had
known and we have seen the other documents. This was
an issue and what we see in the PEAKs is not only the
situation where there is an attempted correction, where
there's doubling, but we see a variety of different
causes other. I was not going to conflate them at all.
I wanted the court to see we have found a wide range
of strands of causes of doubling issues in the accounts
of SPMs. We don't want to conflate them. We want to
identify that, not only are there cases where a SPM --
for example, the regional line manager puts it into
suspense and it doubles, Hucclecote -- your Lordship has
got the point.
MR JUSTICE FRASER: I have been following your analysis of
the PEAKs very carefully.
MR GREEN: I'm most grateful, my Lord.
MR JUSTICE FRASER: I think Mr de Garr Robinson's point is
that you are going rather wider than the confines of
this trial by referencing, firstly, criminal cases over
which I have made it clear I have no coverage and
jurisdiction and they are not part of the Horizon trial.
MR GREEN: My Lord, I agree.
MR JUSTICE FRASER: And he has also suggested that you are
giving a misleading impression by conflating different
technical issues all into one single issue as though it
started in 2000 and has run uninterrupted or unbroken
through to 2019.
MR GREEN: Yes and I'm not seeking to do that. What I'm
seeking to do is identify by way of one problem which
might be faced by SPMs without the insight that you get
from the PEAKs, as to: they have a problem where
a figure doubles in their accounts and they don't
understand why.
Not only is it correct that one of the problems
Second Sight reported about the number of people
complaining, which is when you seek help it doubles, not
only do we find that, we also find a whole load of other
reasons why figures might be doubling in their accounts.
MR JUSTICE FRASER: Understood.
MR GREEN: I don't want to conflate them at all. The short
point is, stepping well back from this, what we see as
an overall picture in relation to the documents the
court has was fairly reflected by Mr Coyne when he
identified that of course the documents we have got are
the ones that successfully got through, made it through
to the third layer, a PEAK was created and is then
ultimately followed through and so we have more
documentation where more investigation was done, less in
documentation where less investigation was done and we
can also see examples in the documents where
subpostmasters, even in the cases that do ultimately
reach a PEAK, subpostmasters either at their or at
intervention logs can be seen to be being bounced back
to NBSC.
My Lord, against that background we have answered
the Horizon Issues as we have sought to do in a matter
of a few words in our closing submissions. We have done
so, we respectfully submit, on a correct formulation of
those issues without any glosses which would distract
from the issues as they were properly formulated by the
court and the way in which those issues fall to be
answered by this court are totally unrecognisable from
the positions that Post Office has taken in relation to
them in a number of material respects, including into
the course of this very trial.
My Lord unless I have anything on which I can assist
you further, those are my submissions.
MR JUSTICE FRASER: Well, I have some questions. Just give
me a second.
Some of them are just minor sweep up points. There
have been various references to a figure for the
claimants' accounting losses generally as about
£18 million.
MR GREEN: My Lord, yes.
MR JUSTICE FRASER: And that is mentioned in one of the
expert statements. I don't know what the origin of that
figure is.
MR GREEN: My Lord it is totted up from the SOCIs, the
schedules of claimant information, that were identified.
MR JUSTICE FRASER: All right. If it is the arithmetic
total of that then I don't need any references.
You mentioned Mr Singh.
MR GREEN: Yes.
MR JUSTICE FRASER: What's your position on Mr Singh?
MR GREEN: The position that we didn't call Mr Singh --
MR JUSTICE FRASER: I am aware of that that is rather why
I'm asking you.
MR GREEN: The position on the evidence is that he said he
had this bizarre huge sum on a phantom transaction.
Mrs Van Den Bogerd says this could have happened,
I can't say it is possible or not possible.
MR JUSTICE FRASER: No, but you haven't called him as
a witness?
MR GREEN: We have not called him as a witness.
MR JUSTICE FRASER: His evidence is not admitted?
MR GREEN: His evidence is not admitted?
MR JUSTICE FRASER: And you showed me the notice to admit.
MR GREEN: And we showed the notice to submit and the basis
upon which it was refused.
MR JUSTICE FRASER: That was my understanding that you did
not call him.
MR GREEN: We did not call him and his evidence is not
before your Lordship.
MR JUSTICE FRASER: That's rather why I was asking the
question.
At the end of the oral evidence when I think the
question of disclosure was being discussed, you stood up
and from memory referred to a solicitor's letter which
you said you thought was 5th August of last year. But
you weren't sure. Now often when counsel says that they
know jolly well it is 5th August and they are
demonstrating their memory recall.
MR GREEN: I'm not sure I was on that occasion.
MR JUSTICE FRASER: But when I go to the H bundle and look
for a letter of 5th August I can't find it. I'm not
saying you have to give me a reference for it, but if
there is a reference for the document that you were
referring to, I would just like to know what it is
because it will save me some time. The authorities
bundle is an agreed bundle I think, yes?
MR GREEN: To which we haven't contributed.
MR JUSTICE FRASER: It is an agreed bundle of authorities?
MR GREEN: It is an agreed bundle.
MR JUSTICE FRASER: You have made various submissions about
Mr Jenkins.
MR GREEN: Indeed.
MR JUSTICE FRASER: And you have repeated them today or
expanded on them today.
MR GREEN: Indeed.
MR JUSTICE FRASER: I'm not inviting you to do this, I'm
just drawing your attention, there is a lot of law or
there's some law on how the court approaches matters if
it is being invited to draw an adverse inference from
someone's absence. I'm proceeding on the basis I'm not
invited to draw an adverse inference because none of the
law relating to it is in the agreed authorities bundle.
MR GREEN: My Lord, the position in relations to that is
reasonably well known but the only -- the way we have
put it is that it is unhelpful not to have called
Mr Jenkins and the reason for it doesn't seem
satisfactory.
MR JUSTICE FRASER: I understand that.
MR GREEN: We have said throughout to Post Office and
I think also to the court, that therefore the weight to
be attached to what's been passed down is very low
indeed.
MR JUSTICE FRASER: I understand that too. That is a rather
different point.
MR GREEN: We also respectfully submit that it feeds into
the impression of Post Office being less than
forthcoming about the true position in relation to --
MR JUSTICE FRASER: I understand that too.
MR GREEN: Beyond that we don't go further.
MR JUSTICE FRASER: That's also a different point.
The next two points really are practical, pragmatic
points, or the next one certainly is, for me which I am
going to raise now because they also affect
Mr de Garr Robinson's and his team's closing.
I happen to take the view that appendix 2 is very,
very useful.
MR DE GARR ROBINSON: Of?
MR JUSTICE FRASER: Of your closing submissions. It
collates a large amount of information in one place. It
is effectively a very useful narrative. You have,
however, made certain submissions this afternoon that
some passages of it, you say, aren't reflected in the
evidence.
MR GREEN: Indeed.
MR JUSTICE FRASER: Now, that may or may not be right and
I have no position on it one way or the other. But in
view of the number of bugs, the length of the appendix
and the quantity of the documents, it is a nettle I have
to grasp now about how it is going to be dealt with.
MR GREEN: Indeed.
MR JUSTICE FRASER: Just to put Mr de Garr Robinson's mind
at rest, I'm not going to ask for any sort of exercise
that has to be done before tomorrow and I'm not going to
ask for any exercise at all other than a purely
referencing one; but it seems to me that for my purposes
in writing the judgment I have to know which passages in
there you say are effectively submission rather than
evidence, so that the Post Office has an opportunity to
direct my attention to wherever it is in the evidence if
it is in the evidence or it might be a submission or it
might be a point of basic computing knowledge which one
is expected to have.
So I would like the two of you, please, just to
address your minds as to how you are going to go about
that. I envisage notification of paragraph numbers from
you to the Post Office where you say that, for example,
paragraph X on page Y, which isn't footnoted, is not
anywhere in the evidence and that then gives the
Post Office the opportunity simply to say the reference
is here or it is a submission.
MR GREEN: My Lord, our approach, given that where there is
a document it appears to be footnoted, where insertions
are made for which no reference is given, we have not
been able quickly to identify what the foundation for
that is.
MR JUSTICE FRASER: Mr Green, I'm not requesting or ordering
anything to be done particularly quickly.
MR GREEN: I'm grateful.
MR JUSTICE FRASER: It is simply -- and I want to make two
points absolutely crystal clear; this judgment is not
going to be out at any time in the next couple of weeks
for obvious reasons and we are going to address
housekeeping tomorrow.
MR GREEN: Indeed.
MR JUSTICE FRASER: When I say housekeeping I mean in terms
of the shape of the rest of 2019; not to conduct a CMC,
just to discuss the point I mentioned to both of you two
weeks ago, what are we going to do about the CMC in July
and the one in September, is it sensible to have either
and/or one of them, etc?
But, as you were taking me through the bugs which
you chose to address this afternoon, there were a number
of paragraphs which you said in relation to those issues
they are not in the evidence or we don't know where that
information is from.
If that, in respect of any particular paragraph, is
the case, then it is very unsatisfactory for me to be
left in a grey area where I don't know has it come from
the evidence or is it a submission? I just need to
know. It is very simple. It is a mechanical task. It
won't be very complicated and it doesn't need doing any
time in the next 7 days, but it does need doing.
MR GREEN: My Lord, yes. I mean, I understand. The basis
on which we have approached --
MR JUSTICE FRASER: I know you have explained that.
MR GREEN: So we should identify --
MR JUSTICE FRASER: I think as a matter of fairness to both
of you, you have to identify a paragraph that you say is
not taken from the evidence if you say it is not taken
froth evidence and then the Post Office has to have the
opportunity simply to give me a reference or say it is
a submission. It is not massively complicated.
MR GREEN: If they have got sources in mind for things --
MR JUSTICE FRASER: Mr Green, I'm not saying --
MR GREEN: But anyway I appreciate your Lordship says it is
a task that has to be done.
MR JUSTICE FRASER: It is a task that has to be done, it is
not designed to impose an enormously expensive or
onerous burden on the legal advisers and it doesn't
necessarily have to be done, if I can put it this way,
to the nth degree for every separate sentence in
a paragraph that might have ten sentences. But it does
have to be done. Because the only other way I'm going
to be in a position to know: is that paragraph based on
any evidence? is by going through the whole of the
evidence myself, which I will be doing but I won't be
doing separately in respect of each paragraph in
appendix 2.
MR GREEN: I see.
MR JUSTICE FRASER: It is fairly straightforward.
MR GREEN: I understand.
MR JUSTICE FRASER: I don't imagine -- when you have had
time to look at it, it might be more obvious that either
there are a list of however long it is or a much shorter
list, but it is not going to be -- and if the answer
from the Post Office is: well, yes you need to look at
this paragraph of Dr Worden 2 or you need to look at
this witness statement of Mr X or Ms Y, that's the only
response I need.
I'm not asking for extra submissions. I would just
like to know because at the moment you are saying it is
not based on the evidence. So just have a talk between
yourselves about the mechanics of that.
MR GREEN: I'm grateful.
MR JUSTICE FRASER: I think that's everything save for one
point. Mr de Garr Robinson, at the end of the afternoon
when there was the exchange between counsel about what
Mr Green was and wasn't doing in mentioning the criminal
cases, you said that one of the rules in commercial
litigation is that one doesn't say things incautiously
that might have an impact on evaluations being done in
proceedings elsewhere.
By that do I take it to mean in relation to
proceedings before the CCRC or are you talking about
something else?
MR DE GARR ROBINSON: I had the CCRC in mind my Lord but
I made the point advisedly as a wider application but my
objection that I made to your Lordship was with respect
to the CCRC.
MR JUSTICE FRASER: Because there are two issues and it is
something that I just need to address now with you
because of the way in which it has arisen in the last
half an hour.
Firstly, if there are any criminal proceedings
currently underway as of today, then, obviously, there
are potential jury issues in terms of any sort of
publication of today's proceedings. So that is a point
which seems to me does have some potential concern
because those sorts of proceedings could be influenced
by anything that was reported about this case.
I'm not sure the CCRC could because that's not
a jury environment and I am sure they would be
sufficiently trained to exclude from their
considerations anything that was reported about this
case rather than looking at the actual judgment itself.
But because of the fact the point has been raised
I wanted to raise it full square with you now.
MR DE GARR ROBINSON: I'm very much obliged to
your Lordship, could I take instructions very briefly?
MR JUSTICE FRASER: I think so.
(Pause).
MR DE GARR ROBINSON: I'm obliged to your Lordship for
giving me that moment. My Lord, there are no
prosecutions being brought by Post Office as we speak.
Post Office understands that there are
investigations being made by the police and possibly
other authorities which may lead to criminal
prosecutions but my instructing solicitors and indeed
a legal officer from the Post Office is not aware that
those investigations have actually materialised into
a prosecution yet. That's not to say that they are
certain there isn't one, but they are not aware of one.
MR JUSTICE FRASER: That is entirely understood. My
potential concern was, obviously, if there were
a criminal trial underway this week, then we may well
have to or would have needed to address whether or not
it was necessary to take any sort of measures to prevent
those Crown Court proceedings being potentially derailed
and having to be started again as a result of anything
that's happened in the last 40 minutes. But on the
basis of what you have said, I'm perfectly content and
unless either of you have anything to say about that?
No? Thank you very much.
Final point. When I say we are going to address the
shape of 2019 at the end of tomorrow, that's not
intended to be an enormous exercise. It is just a very
straightforward pragmatic analysis of how long this
judgment might take and the sequence of the litigation
following on from that and it may be that it is a short
point and it is decided we don't need the CMC in July,
we just have the one in September or something like that
but we can address that tomorrow when you have finished.
MR DE GARR ROBINSON: My Lord, yes.
MR JUSTICE FRASER: Anything else?
MR GREEN: My Lord, no.
MR JUSTICE FRASER: Thank you very much. 10.30 tomorrow.
(4.42 pm)
(The court adjourned until 10.30 am on Tuesday,
2nd July 2019)
(10.35 am)
Closing submissions by MR GREEN
MR JUSTICE FRASER: Good morning.
MR GREEN: My Lord, I think it is me.
The focus of the claimants' submissions will be to
identify particular aspects of the way that Post Office
seeks to invite the court to determine the issues
against the background of what we hope is sort of
reasonably detailed treatment in our closing submissions
and to identify why we find ourselves in respectful
disagreement with the way Post Office seeks to do that.
And in particular, to identify to the court various
facets of the sort of overarching theme of the way
Post Office puts its case, which is, we respectfully
submit, a variant of the rather one-way approach that
was urged on the court in the Common Issues trial in
three particular respects.
First --
MR JUSTICE FRASER: Just pause. I'm afraid I have been
given a broken chair which does sometimes happen, but we
have a spare. Thank you.
MR GREEN: In three particular respects. Firstly,
Post Office effectively seeks to reframe the issues as
they have been agreed and ordered by the court. We
respectfully say that is the wrong approach. Secondly,
on analysis, the way Post Office has sought to adduce
its own evidence is effectively the same one-way
ratchet, for reasons that I will explain. And thirdly,
the way Post Office invites the court to treat the
evidence in the documents is also, we respectfully say,
oddly one-way.
And the short point, if I may, on point 2, which is
how they called their evidence, is simply this: that as
the court will have seen from our written closing
submissions and our written openings, effectively,
a common feature of much of Post Office's evidence is to
call Ms A to talk about what Mr B either knows or has
heard from Mr X, Ms Y and Ms Z, who are often not even
identified.
There has been six weeks of almost Where's Wally in
relation to Mr Gareth Jenkins who persistently pops up
in the source material for Post Office's evidence and
yet has not been called.
We respectfully say that the explanation for that is
unsatisfactory, which we will come to. But it is the
effect, it is the one-way effect of the way the evidence
has been presented to the court which is the focus of
this submission, and that one-way effect is essentially
this. If you call people who stand up and put their
name to an account of what generally the position is and
sign a statement of truth but they happen not to know
anything about the subject or not to know anything
reliable about it, the court on Post Office's approach
is presented with the following choice: (1) accept their
evidence; (2) if they are challenged on it, they get to
say "Well, I'm very sorry, I don't know about it". So
effectively Post Office seeks to be in a neutral
position if their evidence is rejected and a positive
position if their evidence is accepted.
We say that is the wrong approach. We also say that
the explanations for the way this is done are
underwhelming at best, and that goes to both the weight
to be attributed to Post Office's evidence throughout
and invites the court carefully to consider
Post Office's approach to adducing evidence in this
trial.
Briefly by way of introduction on point 3, which is
the way the Post Office invites the court to treat the
evidence and the documents, as the court will see as we
go through the material in a minute, there are some
frankly astonishing invitations to the court to construe
contemporaneous documents against what they say on their
face and to disavow witnesses where their evidence goes
badly. So Mr Godeseth, yes, we shouldn't have called
him and some of his evidence did require clarification
or correction, which then comes in correspondence after
he has given evidence. We say that is an unusual
approach at best.
Furthermore, a narrative of what happened in the
fact evidence that is very difficult to reconcile with
the view of someone who was actually present during the
trial. I appreciate that is a fairly strong pitch to
put it at, my Lord, and I propose to make that
submission good in a moment.
Can I just draw your Lordship's attention to
something else I'm going to come onto a bit later, but
just by way of example if your Lordship turns to
Post Office's closing submissions at page 134 at
paragraph 369.1 {A/6/134}, your Lordship will see:
"The Post Office documents were not drafted with the
benefit of the vast amount of work that has been carried
out by the experts for this trial. If the authors
considered that Horizon was not a good system, they were
wrong (although that is not even a fair summary of what
the documents say)."
Now, firstly there is the astonishing submission
that Post Office's own internal admissions as to the
shortcomings of the system by their own internal
employees and officers and directors should be ignored
and treated as wrong. Secondly, there's the premise in
that that those documents were not drafted with the
benefit of the vast amount of the work that has been
carried out by the experts for the trial.
Let's throw that back to the cross-examination of
Dr Worden who, when he was shown those documents about
Post Office's internal concerns, hadn't even read them.
So not only is that submission fatally flawed, but it
actually draws to the court's attention the documents
which Dr Worden did and did not take into account in
forming the views that he did. And the court will
remember his answers in cross-examination, "Do you
accept that that cannot be reconciled with the view that
you formed?", and him saying "Yes".
My Lord, I give that as just one example -- I was
going to come to it later -- of the approach that
Post Office invites the court to take in relation to
even Post Office's own documents, and we will see that
writ large.
Before I turn to this fact and evidence and
documents piece which is essentially those layers, 2 and
3, can I just invite the court to consider for a moment
the way in which the reframing of the issues has been
done by Post Office.
MR JUSTICE FRASER: This is your point 1, is it?
MR GREEN: Point 1.
MR JUSTICE FRASER: Yes.
MR GREEN: If I could just take the court back to the
Horizon Issues at {C1/1/1} and just take, by way of
totemic example, issue 1. And the issue itself says:
"To what extent was it possible or likely for bugs,
errors or defects of the nature alleged at [paras] 23
... [in the pleadings] ... to have the potential to ...
cause apparent or alleged discrepancies or shortfalls
relating to Subpostmasters’ branch accounts or
transactions, or (b) undermine the reliability of
Horizon accurately to process and to record transactions
..."
Now, pausing there, there are probably three layers
to the way in which certainly Dr Worden reframed them
and the Post Office off the back of that seeks to
reframe issue 1.
The first is that Dr Worden confined his analysis on
issue 1 by ignoring potential. Secondly, he added
a gloss to actual by saying that those should be
lasting, not transient, and when your Lordship asked him
about that, anything that was corrected ever, no matter
how long that took, is transient on his approach. We
respectfully submit it is a wrong approach.
The third wrong part of the approach adopted on
Post Office's behalf and reflected in their closing
submissions is the word "relating", three lines down,
about two-thirds of the way across:
"... discrepancies or shortfalls relating to
Subpostmasters' branch accounts or transactions ..."
Because what Post Office seeks to do is to say that
it has to be in the subpostmasters' branch account. So,
therefore, that discounts a situation if one follows
that to its logical conclusion where the initial figure
on Horizon is correct but the bug causes a different
figure in POLSAP or one of Post Office's other systems
from which a TC might subsequently be issued. We also
say that's the wrong approach.
We flagged this up in opening and we have followed
it up -- I think it is probably quicker to go to our
closings -- and we have identified at page {A/5/173} at
paragraph 500 -- in relation to the contrasting expert
approaches we submit that Mr Coyne correctly focused on
the issues as defined in his reports and was careful not
to draw conclusions about actual consequences where
there was no actual evidence about them but merely to
identify the potential for the effect that crystallised
where the evidence lay. Whereas, by contrast, we made
these points, my Lord, in relation to the lasting and
permanent matters already identified here.
Those are expanded and treated together in the
appendix at -- actually, while we are here can we go
forward to 562, which is at {A/5/198}.
This is the reframing by Dr Worden of Issue 2 which
he formulates as:
"Issue 2 appears to be asking - could Post Office
have given its Subpostmasters automated support in
Horizon, in the place of human support?"
To which he answers no. That isn't even
an available reading of Issue 2), let alone one which
ought to have been adopted by Dr Worden in that respect.
Issue 2 is perfectly clear on its face.
The gloss there of "in place of human support", for
example, speaks to a theme in Dr Worden's approach which
is then ultimately reflected to greater or lesser extent
in Post Office's submissions of changing the issues in
a way that favours Post Office.
When we go back to Issue 1, if we look at the
transcript at {Day18/156:2}, if we look at lines 2 to 6,
your Lordship will see:
"Question: And you describe section 8 as addressing
Horizon 1, yes?
"Answer: Yes.
"Question: "... the extent to which bugs in Horizon
may have affected the Claimants' branch accounts"?
"Answer: Yes, I've stuck in the word "claimants"
there."
So there is another feature which is not a feature
of the issue as the court defined it, and as had been
agreed by the parties for good reason, which Dr Worden
has feathered in. And your Lordship will also remember
how Mr Coyne was challenged about that and cautioned by
my learned friend about the suggestion that interest in
individual claimants had been rebuffed and we then saw
the RFI response to, which was to rebuff any enquiry
about individual claimants.
So there is a theme of reframing the issues in a way
that makes it harder for the claimants to succeed upon
them and comes close, the more one reframes them, to
a position where they are impossible for the claimants
to win. We see that with Dr Worden's assumptions in
relation to the issuing of transaction corrections,
which he says, he acknowledges, are outside the system,
he acknowledges that he doesn't know anything about
business processes and how they are done, and then he
draws inferences on two important premises.
The first premise is that even where documentary
evidence doesn't say that a transaction correction will
be issued and where there's no actual transaction
correction showing it was done, he infers that it would
be done. And he does that -- we will come to this in
passing in relation to the documents -- by saying where
a KEL or a PEAK refers to a TC or error notice needing
to be issued, he assumes that it is issued, and where it
doesn't, he assumes that's because it is written by
people who knew that they didn't need to mention that
because they knew it would happen.
So whatever the document says, it leads him to only
one conclusion and precludes the possibility of any
conclusion that might let in any justifiable complaint
about that by any claimant.
The second premise is that the TC procedure is done
so close to perfectly that the opportunity for error is
vanishingly small, and that's the second assumption that
he builds in. And then he builds all of that into his
version of the system as defined, which conflicts with
the definition of the system at the top of the page of
the Horizon Issues, and from that goes on to reach his
conclusions. And we respectfully say that is not the
right approach, and Mr Coyne took an approach which was
correct in his reports in relation to the issues as
defined.
I will deal with the issue of robustness in relation
to Issue 3 very briefly, reasonably briefly. There is
first a point of principle which is that the way --
robustness is the Post Office's favourite issue, that's
clear, something they insisted on and we agreed to and
was ordered. What it doesn't say is that the system is
robust? And then a second question: is it extremely
unlikely to be the cause of shortfalls in branches?
It was conjunctive in its formulation, it was not
split into subparagraphs like some of the others, and we
treated it as one issue. And on that basis and because
Post Office wants to conflate the word "robust", which
is in inverted commas in the issue as defined, with
extreme unlikelihood to be the cause of shortfalls in
branches, that is precisely the basis upon which the
claimants do not accept Post Office's analysis on that
issue, because they do not accept that it is robust in
the sense of being extremely unlikely to be the cause of
shortfalls in branches.
My Lord, there is an important point of drafting in
relation to Issue 3, which is Issue 3 does not say:
extremely unlikely to cause shortfalls in branches. So
it is not contemplating what proportion of 47 million
transactions a week would have errors caused in them.
What it actually contemplates is a situation where you
have an identified shortfall in a branch and the
question then is if you do have such a shortfall in
a branch, is it extremely unlikely that Horizon is the
cause of that shortfall in the branch?
We respectfully say no, it is not, or not
functionally so for the purposes of this dispute, and
particularly not so where the type of shortfall and
discrepancy that in reality is in issue in this case is
the unexplainable; the one where the subpostmaster
doesn't go "Oh yeah, that's probably Jeff remmed in too
many stamps there, yes, I remember he was a bit
distracted this morning" or "Oh I can see from the till
roll something has gone wrong". We are not talking
about that.
The focus of this litigation, as reflected also in
the Common Issues trial, is on what is the situation
where there is an unexplained shortfall that the
subpostmaster finds it difficult to get to the
bottom of.
So that is the background to those two facets of
Issue 3: robustness and extremely unlikely being part of
one concept; and secondly, the premise is the cause of
the shortfall in a branch.
MR JUSTICE FRASER: Where do I go to to find your benchmark
definition of robustness?
MR GREEN: My Lord --
MR JUSTICE FRASER: The claimants' benchmark definition.
MR GREEN: We found it difficult to define other than by the
extremely unlikelihood of causing shortfalls. So we
haven't got a definition that we advance as a meaningful
and satisfactory one for the purposes of this case. And
save insofar as it appears to be a correlation of
extremely unlikely to be the cause of shortfalls in
branches.
So that --
MR JUSTICE FRASER: So you are treating it as a summary term
that includes and consists of the second part of
Issue 3?
MR GREEN: That appears to be the only sensible objective
way of construing it, we say.
My Lord, it has never been the claimants' case that
there weren't numerous aspects in which Horizon is
relatively robust, and that is a point on which we
respectfully submit Post Office's closing rather
overshoots in the criticisms made of the position taken
by the claimants in the reply, which I would just like
to take your Lordship to, if I may, at paragraphs 36
and 37, which is {C3/4/21}.
So if we just start under "Operation of Horizon",
A.8 heading. 34:
"The Defendant admits that 'Horizon is not perfect'
... and that 'like all other IT systems, Horizon is not
a perfect system which has never had any errors or bugs'
..."
MR JUSTICE FRASER: Which document is this?
MR GREEN: So this is our reply, my Lord.
MR JUSTICE FRASER: Your reply?
MR GREEN: Yes.
35:
"It is also uncontroversial that the volume of
transactions effected through Horizon is very large."
36:
"Notwithstanding the various checks and controls
alleged (the existence and efficacy of which are not
admitted), the combination of Horizon's admitted
imperfections and the volume of transactions, is
entirely consistent with the levels of errors reflected
in the Claimants' case: the said imperfections are
likely to be the cause of discrepancies or apparent or
alleged shortfalls about which the Claimants complain."
37:
"It is therefore denied that Horizon 'is robust and
[...] is extremely unlikely to be the cause of losses in
branches' (paragraph 16). In fact, the relatively small
chance of errors admitted by the Defendant, would be
likely to produce the very picture reflected in the
Claimants' case."
Then there are observations in relation to the
existence of money in suspense accounts and so forth.
At paragraph 38 {C3/4/22}, the change of public
position was then noted. Then the clarification of the
defendants taking to make use of the pre-action
correspondence then follows at 39 onwards, which is
where the defendants effectively tried to bank the
absence of a systematic flaw across the whole of
Horizon.
We weren't in a position to allege that there was
a systematic flaw affecting everybody as meaning that
there was no claim that anyone had suffered
a Horizon-generated shortfall, which we just wanted to
make absolutely clear on the pleadings was not the case.
Then at paragraph 41 {C3/4/23}:
"As to the allegedly robust controls pleaded
variously ..."
41.1:
"the Defendant's case that ..."
Then it is set out, effectively saying:
" ... when supplemented by the various accounting
and cash controls applied in branches, make it very
unlikely indeed that an error in Horizon could affect a
Subpostmaster's financial position and go undetected'
..."
We say that:
"... implicitly accepts that Horizon is likely to be
the cause of discrepancies or apparent or alleged
shortfalls in branches, unless detected by Claimants."
Then they failed to plead to any material variations
because they only pleaded on the current version of
Horizon as it stood at the date of defence. So we
didn't know what the case was there.
Then, the defendant was put to strict proof to
matters which lay within their knowledge but not ours at
41.3. And at 41.4, {C3/4/24}:
" ... in respect of the alleged 'robust measures in
place in Horizon ... The Claimant denies that these
were effective, including on the basis that two of three
errors or bugs admitted by the Defendant (in Schedule 6
to the Letter of Response, referred to at paragraph 56
of the Defence) were not identified through Horizon's
own in-built checks and balances designed to identify
the same ..."
Then there is a passage about the position in
relation to the known error log, which we will come back
to. Just to give the court an indication of where we
were at this stage pleadings-wise, could I go, please,
to the generic defence at {C3/3/22}. This is
paragraph 50 from the previous page and it is primarily,
at this stage, addressing the known error log. The
second line of that paragraph says --
MR JUSTICE FRASER: Which paragraph?
MR GREEN: (4). I'm so sorry, I didn't say that. 50(4),
which is on {C3/3/22}, the second line of that
paragraph:
"To the best of Post Office's information and
belief, the Known Error Log is a knowledge base document
used by Fujitsu which explains how to deal with, or work
around, minor issues that can sometimes arise in Horizon
for which (often because of their triviality)
system-wide fixes have not been developed and
implemented. It is not a record of software coding
errors or bugs for which system-wide fixes have been
developed and implemented. To the best of Post Office's
knowledge and belief, there is no issue in the Known
Error Log that could affect the accuracy of a branch's
accounts or the secure transmission and storage of
transaction data."
So, my Lord, I just want your Lordship to have
a stake in the ground as to where we were pleadings-wise
at that point, in circumstances where Dr Worden's first
approach was to confine his analysis almost entirely but
not exclusively to KELs to identify where the bugs were
in Horizon that may have an impact on branch accounts.
So it is in that context that the reply is then pleaded
in that way in relation to robustness. And seeking to
make the distinction, ultimately reflected in the way
the Horizon system is defined for this trial, which is
slightly narrower, as the system itself. And then of
course Dr Worden is the one who says the subpostmasters
themselves form part of the countermeasures by manual
inspection of data MID and user error correction in his
analysis.
MR JUSTICE FRASER: So what's the stake in the ground? I'm
not --
MR GREEN: The stake in the ground is what Post Office were
saying about the existence or knowledge of information
about bugs at that stage by way of context to us.
So they are saying that the context in which we were
trying to plead was basically blindfolded and
Post Office was very keen for us to plead particulars of
things which a striking feature of this case is that the
subpostmasters did not know often about them at all and
were often not told what the underlying root cause of
things was even when, in Mrs Burke's case, for example,
it was not her fault at all and she gets a TC that says
"please take more care in the future". Totally a system
fault.
So at this stage we are pleading completely
blindfold to this, and Post Office are positively saying
that to the best of Post Office's knowledge and belief
there's no issue in the known error log that could
affect the accuracy of a branch's accounts or secure
transmission and storage of transaction data.
It is easy when one has been in a trial this long,
and it has been interrupted, to lose sight of the
utility of the listing of this trial given where we have
come from and where we are now in terms of the evidence
available for the court and indeed to the claimants.
I just wanted to put that stake in the ground to
give it at least a small indication of the distance of
travel. We have obviously mentioned the distance of
travel in relation to remote access in our closing
submissions, which is the position is completely
transformed. And it is also obviously apparent that
Post Office did not disclose in their pre-action
correspondence the existence of the Dalmellington bug at
all, notwithstanding that it affected 88 branches and
there were 112 known instances of it occurring over
a period of about five years.
That was left for Mr Coyne to stumble upon. There
was no disclosure of the APPSUP role, which the court
has now seen what a huge back door it was into the
system, and how the logs were not either kept or kept in
such a way as to allow there to be any sensible analysis
of how that privileged role had been used.
So it has transformed out of all recognition in
terms of the claimants' and the court's understanding of
remote access. None of that would have happened had the
court not ordered this trial to take place and
essentially the process of preparing for and undertaking
the trial actually happened as it has.
My Lord, just briefly in relation to robustness,
there is also the document at {C1/2/4}, which was our
provisional outline document, which the court will
remember Post Office were very keen for us to provide
but less keen to be forthcoming themselves about what
they knew to be the problems with the system.
So this provisional document was of course put
together when we still didn't know about the
Dalmellington bug:
"There's no single, authoritative definition of
'robust' in the context of a system such as Horizon,
which appears to be more commonly used in public
relations than as an objective performance standard.
Pending any clarification of its objective meaning, it
appears to relate to the ability of a system to perform
correctly in any scenario, including where invalid
inputs are introduced, namely, to have in place
effective error repellency."
{C1/2/5}:
"It is clear that, on a sensible construction of the
term 'robust', Horizon did not neat this standard
because:
"(a) it contained bugs, errors and defects as set
out in paragraph 1 above which created discrepancies in
the branch accounts of Subpostmasters;
"(b) it suffered failures of internal mechanisms
which were intended to ensure integrity of data."
There is a KEL in relation to recovery failures
there:
"(c) the system did not enable such discrepancies to
be detected, accurately identified and/or recorded
either reliably, consistently or at all;
"(d) the system did not reliably identify
'Mis-keying', which is inevitable in any system with
user input, and did not reliably have in place
functionality to restrict users from progressing a
mis-key;
"(e) ... numerous processes and workarounds ...
weaknesses and risks of errors ..."
And so forth.
That goes all the way down to paragraph 3.3 over the
page {C1/2/6}:
"However, as noted above, whether Horizon is
'robust' plainly depends upon the definition given to
that word. Even the small chance of errors, bugs and
defects admitted by the Defendant and/or supplemented by
those underlying the reports above, would be likely to
produce the result alleged by the Claimants. Therefore,
as noted above, even if the overall probability of a
bugs or errors affecting Branch accounts is small, the
sheer number of transactions undertaken by the Horizon
system is consistent with the level of discrepancies
arising from bugs and errors in issue in these
proceedings."
So, my Lord, that dovetails with those paragraphs in
the reply to which I have already taken the court.
So that is as far as we were able to go in relation
to robust. But we say for functional purposes it isn't
robust insofar as it does all of those things and those
things create a small but real chance of discrepancies
in relation to subpostmasters' branch accounts or their
transactions.
My Lord, for your note, just in relation to the
KELs, if we may. It is right that your Lordship should
know that at paragraph 29 in the Post Office's closing
submissions at {A/6/16}, Post Office suggests that it
has taken a co-operative approach to disclosure. For
reasons which will be already largely clear to the
court, we do not accept that is correct, and I will just
give the court two examples.
In our opening, claimants' written opening, which is
at {A/1/106}.
MR JUSTICE FRASER: This is your opening, is it?
MR GREEN: The claimants' opening.
MR JUSTICE FRASER: Yes.
MR GREEN: And we devoted an appendix, which is referred to
in the body, to looking at what the claimants had to go
through to get the KELs, given how large they loomed in
the trial. And at that appendix at page 106, just to
ask the court to turn to it for a moment in the opening,
the claimants first sought disclosure of the known error
logs on 28th April 2016 in the letter of claim, and they
said:
"We understand that Fujitsu maintained a 'Known
Error Log' for Horizon and that such reports will have
been provided to Post Office."
We thought the known error log was a proper problem
management system identifying all known errors in that
sense. It turns out that, in fact, the evidence of this
is much more atomised across the documentary landscape,
but we were expecting a log of all known errors.
Post Office's letter of response says in effect they
not only refused to provide the documents but denied
their relevance and cast doubt over whether they even
existed.
So Post Office's response said -- it is all
footnoted to the references; I won't take your Lordship
to the underlying documents --:
"In circumstances where you have not particularised
any factual basis on which horizon is defective,
disclosure of these documents (if they exist) is not
relevant reasonable or proportionate."
Leaving that matter at large for us.
We didn't accept that. No good reason to refuse
disclosure.
Then over the page at {A/1/107}, 317:
"Post Office's response on 13th October 2016 ..."
They say:
"The claims which you have particularised concern
the Core Audit Log. Following a review of the Known
Error Log, Fujitsu have confirmed that there have been
no logs in respect of Core Audit Log. The remainder of
the Known Error Log does not relate to the claim which
you have particularised and as such disclosure of this
document is not relevant."
My Lord, I'm not going to take you through it, but
your Lordship will see the history of this all the way
through, including paragraph 324 on page {A/1/109}, the
position of what the known error log was described as
before your Lordship at the CMC in October 2017:
"It contains things like there's a problem with
printers. There's a printer. You have to kick it on
the left-hand side to make the printer work. I mean
there's a vast range of hardware problems of that sort
and maybe some software problems (inaudible) but not the
kind of bugs, errors and defects that the claimants are
wishing to pursue in their particulars of claim so far
as Post Office is aware."
There was a sort of Nelsonian approach to the
obvious relevance of the known error log and a way of
presenting the known error log which we say is (a) far
from forthcoming, and (b) completely irreconcilable with
the claim that Post Office has been co-operative in
relation to disclosure.
My Lord, parenthetically they also queried the
relevance of OCPs and OCRs as well. I will give
your Lordship a reference to that later on. So for
those and other reasons that I will develop later on, we
do not accept that is the case; in fact, quite the
opposite. We ask the court to take note of Mr Coyne's
answer in re-examination that he had never encountered
a case like this where disclosure had come out in this
way in his entire career.
MR JUSTICE FRASER: Well, the Post Office's closing
submissions make a point in paragraph 1145 on page 377
that you haven't -- well, the suggestion is that
complaints about disclosure have emanated from the court
and that you haven't made any applications for specific
disclosure and you haven't advanced any complaint that
particular disclosure orders haven't been complied with.
MR GREEN: Well, my Lord, taking it in stages, that rather
sidesteps the issue that I have been making submissions
on. The issue that I have been making submissions on is
the manner in which Post Office -- we got clear examples
of documents that should have been disclosed; there is
a very recent one which they kept until May. They say
they intended to disclose it in March, for example. So
there are clear examples of things where they haven't.
MR JUSTICE FRASER: Did they say they intended to disclose
it in March or did they say they discovered it in March?
MR GREEN: There was one which was discovered in March which
they intended to disclose and erroneously forgot to.
MR JUSTICE FRASER: Where's that?
MR GREEN: I will get the reference to that in a minute, but
just by way of example. There is a whole pantheon of
disclosure issues in relation to which the claimants
have had to try to take a -- we simply could not have
issued applications on every possible complaint, not
least because the proportions approach, which the court
understandably took given what the court was being told
by Post Office, was to order focused disclosure and then
allow us to look at that, and then from there to then
ask sequentially for what we could then see would be
relevant.
On the basis of what the court was told, that was
plainly, on the face of it, the right way to approach
proportionality in disclosure. The point that I was
seeking to address here was the anterior point of when
we say the documents which were almost the sole source
for Dr Worden's first report, when we raised them in
2016, your Lordship has seen what was said about them
all the way through in the CMC as if they were of
absolutely marginal relevance.
If we had been given an honest answer, I say
"honest" carefully --
MR JUSTICE FRASER: I was about to pick you up on that.
MR GREEN: Yes. My Lord, the reason I say honest is this.
In a Nelsonian sense, to tell the court something where
you have made no enquiry to ensure that you are getting
a fair picture, is, we say, extremely unsatisfactory.
MR JUSTICE FRASER: Well, you have to be very careful about
making submissions that include effectively a criticism
of lack of honesty, don't you?
MR GREEN: My Lord, that is right. I'm trying to make it
clear what I'm saying and I will come back to that later
on. But there is a picture across the piece of --
I mean, perhaps I shall recede to "less than
forthcoming" for the moment.
MR JUSTICE FRASER: I think that would probably be best.
MR GREEN: Because it saves an argument about whether it
rises to that threshold or not. But there is a repeated
picture of the court being told things which turn out
not to be correct. And the explanation is, so we look
at Angela Van Den Bogerd's evidence which we will come
to in a minute, where she gives her evidence. She has
not looked at the underlying documents, which tends to
suggest it is not right, and she is specifically asked
by the court: did you look at those documents before you
made your witness statement or not? She says no.
So we have this constant picture of a situation in
which the implicit suggestion that one might expect to
be being made when a senior person gives evidence, for
example, that they have taken care to make sure what
they are saying is right, turns out not to be correct.
So my Lord, I'm happy to rest on "less than
forthcoming" for present purposes, and for present
purposes I'm going to withdraw the reference to had they
given an honest answer. What I shall say is had we been
given a correct answer which any reasonable enquiry
would have revealed, had we been given a correct answer
which any reasonable enquiry would have revealed in
2016, the proper thing for Post Office to do would have
been to provide disclosure at that stage of relevant
KELs that disclosed the existence of discrepancy of
bugs, defects and issues in Horizon which could give
rise to branch discrepancies at that stage.
That's what should have happened. So in terms of
where we were and what we were able to get later on, we
are about two years late getting things, compared to
where we should have been had Post Office given correct
information in responding to our enquiries, and from
where we would have been were we able to show our expert
those documents at a much earlier stage and not have the
situation which Mr Coyne found himself in in this case
with thousands of documents coming in a matter of days
before his second report, for example. So it has been
extremely unsatisfactory.
My Lord, in our closing at page {A/5/19}.
MR JUSTICE FRASER: Yes.
MR GREEN: -- which is {A/5/19} for Opus purposes,
paragraph 57. The PEAK at 1848.8.2 --
MR JUSTICE FRASER: Yes.
MR GREEN: -- was disclosed and was a document:
"... adverse to Post Office and would have been
directly relevant to the evidence which Mrs Van Den
Bogerd had given in her witness statement relating to
Drop & Go."
MR JUSTICE FRASER: Yes.
MR GREEN: "[It] also contained evidence of keystrokes which
were plainly available to the Post Office from Fujitsu
..."
Your Lordship will remember that document coming up
at the end:
" ... a point on which Post Office’s witnesses had
given very odd evidence (addressed further below). It
would therefore have been an obviously relevant document
for the Claimants to have referred to at that stage, had
it been disclosed to them. As it bore directly on
Issue 8, it is even more surprising that it was not
disclosed earlier..."
Because Issue 8 is about what information was
available to Post Office.
And we had the absolutely bizarre spectacle of
Mrs Van Den Bogerd accepting that keystrokes would be
provided in a Credence report, because Mr Patny, who was
accused of dishonesty, and your Lordship is very right
to point out the seriousness of alleging dishonesty,
Mr Aakash Patny was accused of dishonesty in
circumstances where he had repeatedly chased for
a Credence report because he wanted to find out what had
happened at a stage where Post Office internally
appeared to think that keystrokes would be provided with
that.
We have addressed that in relation to
Mr Aakash Patny and the assault on his honesty in our
closing submissions, but that is the context in relation
to which this keystrokes document would have been
extremely relevant. And then, if your Lordship looks at
paragraph 58:
"The explanation for the timing of the disclosure of
this PEAK given in the 18th witness statement of
Mr Parsons does little to improve the position or
assuage any concerns; it reveals that Post Office waited
2 months from discovering this PEAK on 29 March 2019 ...
before disclosing it to the Claimants, just before the
resumed hearing."
59:
"Further, despite the excuse given for this 2 month
delay being that 'For the next couple of months my firm
continued to investigate these bugs and a number of
further documents relating to these bugs came to light',
in fact, on 24 June 2019 at 17.42 ..."
In the evening:
"... [Wombles] provided yet further disclosure in
relation to this bug and email chain. In response to
a request for an explanation as to why this document had
not been disclosed before, [Wombles] said that the
document had been provided to it on 10 May 2019 was
intended to be disclosed on 31 May ..."
I think that should be March. May:
"... was intended to be disclosed on 31 May 2019,
but due to an oversight was not disclosed on that date
[in May]."
So there are examples in relation to that, but
there's also, my Lord, the other anterior feature of
Post Office's approach which is we were not able to
request documents which related to the Dalmellington bug
because the tenor of Post Office's letter of response
was such as to suggest that there had been three bugs,
which were the ones that Second Sight had identified,
and there obviously must be a possibility of other bugs.
In circumstances where, between the letter of claim
and the letter of response, there was the email saying
"stop investigating the Dalmellington bug", that is
a deliberate decision to stop investigating that bug at
that stage. And we respectfully submit that the only
available inference is it was a deliberate decision not
to mention the Dalmellington bug in the letter of
response.
We respectfully say that that is reaching the higher
echelons of being less than forthcoming. Similarly, the
death of a thousand cuts in relation to remote access.
Given that we know from the Paula Vennells email that
she specifically made an enquiry as to what the position
was, the way that evidence emerged is extremely
unsatisfactory, particularly because Mr Parker's first
round was to say Mr Roll's account was misleading;
positively to say that Mr Roll's account of remote
access was misleading, when (a) it wasn't, and (b) he
was right.
So, my Lord, with that general introduction can
I turn now to the approach in relation to fact evidence.
One of the points that the defendants/Post Office takes
against the claimants is to complain about the claimants
calling fact evidence from a handful of SPMs, and that
begins in their closings at {A/6/299} at paragraph 918.
We say it is a bad point to say the least. Briefly,
contrary to Post Office's approach before the Common
Issues trial, Post Office did not make any application
to strike out that witness evidence. Secondly, at the
pre-trial review my learned friend referred to his
anxiety on Post Office's part about that, which we can
see. The relevant part of the transcript is
{C8.14/3/13} at letter B. We see at letter B my learned
friend:
"I can tell your Lordship that a huge amount of
anxious thought went into that question. It was in the
aftermath of your Lordship’s strike-out judgment. The
reply that came from Freeths was, 'In the light of the
strike-out judgment we are not changing anything, we are
not changing course"
That's his suggestion:
"Given the anxiety, frankly, the fifth on this on
the part of Post Office --"
MR DE GARR ROBINSON: My recollection is that I said
"fearful".
MR JUSTICE FRASER: Fearful rather than what, sorry?
MR DE GARR ROBINSON: "... fifth on this on the part of
Post Office --"
MR JUSTICE FRASER: It should say:
"Given the anxiety, frankly, the --
MR GREEN: The fearfulness.
MR JUSTICE FRASER: Right. Thank you very much.
MR GREEN: Then your Lordship then explained to my learned
friend and the court at D:
"What I said ... in that judgment, I thought which
was obviously reserved and drafted, I applied the test
set down most usefully by Mann J about striking out
evidence of fact and what approach one takes to
relevance. You have 12 witnesses of fact for, let’s
call it, round two of Horizon. Mr Green has umpteen
witnesses of fact. Most of the witness statements
actually are relatively narrow in compass. There is
obviously more fact than one would necessarily expect
with the phrase 'limited evidence of fact', but I think
between you, you are agreed that it can be dealt with in
a reasonable number of days, and the only real issue
appears to be, given you each have an expert, how
Mr Henderson’s evidence is to be treated if it is to be
treated at all, that is really what it comes down to.
You say he is an expert by the back door."
MR JUSTICE FRASER: The focus at the pre-trial review
I think originally was about Professor McLachlan and
Mr Henderson.
MR GREEN: Originally.
MR JUSTICE FRASER: You said you weren't going to call
Professor McLachlan and I gave a ruling on
Mr Henderson's incorporation of the Second Sight report.
MR GREEN: Precisely. My Lord, the short point is that we
ended up with nine of the Post Office's witnesses of
fact giving evidence.
MR JUSTICE FRASER: Is that including Mr Membery?
MR GREEN: Excluding Mr Membery.
MR JUSTICE FRASER: It is really ten, then.
MR GREEN: It is really ten, but nine actually giving live
evidence, plus Mr Membery. And then Mr Roll and
Mr Henderson giving evidence, which was much narrower in
compass than he would have liked to have given, as must
have been apparent to the court. Plus five SPMs.
So Post Office sought to rely on ten witnesses of
fact and the claimants sought to rely on seven plus
Mr Singh, in respect of whom there was a notice to
admit, which I will just show your Lordship later. So
less.
MR JUSTICE FRASER: But the question about evidence is not
a question of counting up how many witnesses are on each
side.
MR GREEN: Your Lordship is quite right, but I just wanted
to clarify. It is not a counting exercise, but those
are the numbers. I will come to the substantive points.
In response to the complaint by Post Office about
this, just imagine what the trial would have looked like
if we had not had the facts of those individual SPM
cases to provide factual grist to the mill of
determining what the systems were and how they worked in
practice.
So the only way ARQ data came out was through the
individual cases of individual SPMs. Failed recoveries
was focused upon specifically, because of SPM evidence,
Ms Van Den Bogerd's response in her responsive witness
statement. It is important to identify that Post Office
hadn't addressed failed recoveries outside the context
of specific SPM evidence.
The claimants then had to prove their account to
contradict suggestions by Post Office of user error
being causative of discrepancies, which was demonstrated
to be wrong. There was a late amendment to Mrs Van Den
Bogerd's second witness statement in relation to
Angela Burke where she had directly sought to suggest
that Angela Burke was at fault, which was wrong. There
was no amendment in relation to Mr Tank's evidence in
Angela Van Den Bogerd's witness statement although it
was ultimately accepted that there wasn't user error
causative of the discrepancy in relation to the receipts
in Mr Tank's case.
Then the Post Office refused the claimants' notice
to admit specifically on the basis of causation of
discrepancies when correctly following the appropriate
course in relation to Mr Singh.
So if I can just take your Lordship very briefly to
that. {H/187/1} is the letter enclosing a notice to
admit, which is mentioned at the bottom. It is
explained at the bottom there:
"As you will see, we firstly invite an admission
that it is possible that a phantom transaction similar
to that described by Mr Setpal Singh at paragraph 6 of
his witness statement dated 28 September 2018 occurred
at the Reddish Post Office in or around the summer
of 2003."
It is a very limited admission because Ms Van Den
Bogerd's submission, as we see over the page, was that
there was insufficient documentary evidence available to
determine whether or not that was the case {H/187/2}.
And that was one admission that was sought on the basis
of what she herself had said in her witness statement.
Then at the bottom of the page, the second admission
that was sought was to invite Post Office to agree, to
admit that it was:
"... aware of occasions where Horizon froze or would
stop working or screens would go black as a result of a
system crash, and discrepancies in accounts did
sometimes arise after SPMs followed, or attempted to
follow the recovery process."
There are two alternatives there. Your Lordship
just briefly will see {H/188/1}. That is the actual
notice to admit, and paragraph 2 deals with that.
Then if we go to {H/200/1}:
"The second alleged fact is drafted in a tendentious
manner and, as drafted, is not true (or even supported
by the evidence that you cite in your letter).
Specifically, there is no evidence cited that
discrepancies arose after system crashes where the SPM
did in fact follow the recovery processes properly, as
opposed to merely attempting to follow those processes."
So they refused to admit something that the court
can now see is true, and Angela Burke's case is
a perfectly good example of exactly that.
See also Mr Tank. So seeing in reality what event
logs actually looked like for specific transactions,
what transaction logs looked like for specific
transactions was necessary, and there was -- how else
were we going to do it?
Again, Post Office would have had a trial in which
it was able to say what would happen or could happen and
what it would normally do and where the claimants would
be unable to effectively analyse or challenge or unpick
those general assertions which were almost exclusively
if not wholly exclusively self-serving.
Another feature of the one-way street approach to
determining the outcome of this trial, my Lord, is in
relation to so-called unchallenged evidence, and this is
quite a big point in the closing submissions and quite
heavy reliance is placed on it, and it has quite a long
history. And your Lordship has repeatedly made clear
that a point of that type would be given short shrift.
My learned friend raised it at the PTR and it was
made perfectly clear what the position was. It was
raised at the PTR, I do not think we need to go it but
I will give the court the references at {C8.14/3/14} at
D to E and {C8.14/3/15} at E to H.
That led to the court's observations at
{C8.14/3/17} --
MR JUSTICE FRASER: Maybe we should go there.
MR GREEN: Shall we go to that part. {C8.14/3/17}, top of
the page, 17, your Lordship to me:
"Mr Green, I am going to make certain observations
to you, which really are for the purposes of both of
you, and that you might find most useful now that you
have stood up. In a way, it is possibly, I suppose, a
misunderstanding as to the way time-limited trials work.
I know this is a QB case but it is being tried in this
building using many of the trial management procedures
that we use in this building, not only in the TCC but
the other specialist courts. In any time-limited trial
any party would get pretty short shrift if they tried to
make a formal point if a particular subject had not been
challenged within the time available. Therefore, it
occurs to me that at least part of the Post Office's
concern arises out of a misapprehension of your
technical ability to do that. Are you aware of your
technical inability to manage such an objective?"
It might originally have been an objection. So it
was absolutely clear, and yet what we then find in the
defendants'/Post Office's closings at {A/6/78} -- I'm
only going to give your Lordship examples, but just to
explore the point generally. We have at paragraph 176:
"Nine PEAKs, limited to two categories, across
17 years does not suggest that the final response codes
were regularly applied incorrectly."
MR JUSTICE FRASER: Where are you looking?
MR GREEN: At paragraph 176, my Lord.
MR JUSTICE FRASER: Of the Post Office's closings?
MR GREEN: Indeed, at page 78.
MR JUSTICE FRASER: Yes.
MR GREEN: So that is the point in relation to the nine
PEAKs.
Then at paragraph 177, the very next paragraph
{A/6/79}, it says this was:
"... limited to only two codes used across the PEAKs
and categorised by Mr Parker, namely ... 70 and 68."
It is actually not correct. It also included 62,
but let's leave that by the by for a moment.
Not only is it generally wrong to take that
approach, but there was, we respectfully say,
a divergence of approach between the claimants and
Post Office during the trial.
The claimants sought, where they could, not just to
give an example of a point but to try to approach points
in a way that the court could fairly regard as
representative. And that was the basis upon which bugs
1 to 10 were cross-examined on rather than cherrypicking
favourite bugs. It is the basis upon which later today
I'm going to invite your Lordship to look at 11 to 15.
So I'm not cherrypicking 17 or 21.
Similarly, in relation to the PEAKs that we dealt
with with Mr Parker and the closure codes, those PEAKs
were PEAKs that were already in play in the trial bundle
for the most part. I'm not aware of any that we
specifically uploaded to demonstrate the absurdness of
the closure codes. And we saw them because we were
looking at PEAKs that were already in play and they were
good examples. But not only that, my Lord, that
I specifically said at {Day12/53:23} in the transcript
that I was not going to be able to go through all of the
examples and specifically explained that I was going to
look at the big categories where there were large
numbers of PEAKs for which that code had been adopted in
the table that Mr Parker was referring to.
So he showed in the table there were thousands of
PEAKs in some categories and we chose to look at the
closure codes where there were the largest number of
closures being affected and where he had not included
those closure codes as having any content in software
errors that Mr Roll might have been working on. That
piece ended at {Day12/86:13} and line 18 with
your Lordship saying:
"I don't think you are going to have time to do any
more of these:
"MR GREEN: I'm not going to do any more of those, my
Lord, I'm cutting it there because -- not least because
the witness accepts that he didn't go through them
before he gave his witness statement."
So there were two points there: (a) I went as far as
I could to try and give a representative impression of
how this had been done; and (b) because it became
absolutely clear that Mr Parker had not gone through
them himself before and also the extent to which he had
used the definitions in the underlying document which
had been put to Mr Roll was seriously in doubt, and he
positively said at one point in his evidence, which we
addressed in our closings, that he actually deliberately
adopted his own definition in one of the categories.
So for those reasons we didn't take that any further
than necessary.
MR JUSTICE FRASER: But Mr de Garr Robinson is entitled to
submit, isn't he, that it wasn't a representative
impression? The fact it is a time limited trial doesn't
stop him making a submission.
MR GREEN: Your Lordship is absolutely right that the fact
that -- he can say it is not representative and put
forward submissions to make that point good. That's
always available. Your Lordship is absolutely right.
MR JUSTICE FRASER: But isn't that what he and his team are
doing in paragraph 177? {A/6/79}
The time limited trial point is that the party is
not permitted to submit that a point is accepted or has
not been challenged, and hence certain conclusions can
be drawn.
MR GREEN: Indeed.
MR JUSTICE FRASER: But that's rather different, isn't it,
from saying it is not a representative exercise?
MR GREEN: Well, my Lord, your Lordship is absolutely right.
But because no counter exercise is advanced to make that
point good, and because of the way it plays out through
the rest of the submissions, it appears to us that the
substance of the point there --
MR JUSTICE FRASER: I understand.
MR GREEN: -- is that. And if we look at {A/6/88}, the
entire heading, your Lordship will see, is evidence that
was not challenged.
When we go over the page {A/6/89} to the conclusion
at paragraph 208, it is just a bare assertion that large
parts of Mr Parker's evidence was not challenged.
Having read all that, we revisited what the substantive
point seemed to be in other places.
MR JUSTICE FRASER: Understood.
MR GREEN: And that's what we wished to draw the court's
attention to.
MR JUSTICE FRASER: We are probably going to need a break at
some point.
MR GREEN: My Lord, I have two small points to finish this
off and give the transcript writers a break.
MR JUSTICE FRASER: Yes.
MR GREEN: The evidence said not to have been challenged
includes his evidence about the APPSUP role at 205.9
{A/6/89}:
"This includes his evidence that he cannot recall
any instance in which the APPSUP role has been used to
change transaction data, although he cannot state
unequivocally that it has not happened.
We didn't cross-examine further on that because his
evidence carries so little weight in relation to that
that it is effectively worthless.
His paragraph 15 of his witness statement in which
the belated discovery of the APPSUP role is reflected,
which in itself is bizarre and speaks volumes, this is
at {E2/13/4}, reads as follows:
"APPSUP is used by SSC for updates to and
maintenance of the BRDB that would not involve changing
transaction data. I have not examined the privileged
user logs, but based on my experience my expectation is
that these uses of APPSUP or at least the vast majority,
are for support work that does not involve changes to
transaction data. I cannot recall any cases in which it
has been used to change transaction data, but I cannot
state unequivocally that there are no circumstances in
which it has ever happened."
Given the fact that he didn't know it had happened
at all, it was absolutely pointless to challenge
a witness of that sort in a time limited trial on the
point.
So it's not just the bare not challenged point, it's
ignoring the obvious context in which particular
evidence was not challenged.
My Lord, would that be a convenient moment?
MR JUSTICE FRASER: I think so, and we will have ten minutes
I think for the shorthand writers. Thank you very much.
(11.56 am)
(A short break)
(12.05 pm)
MR GREEN: My Lord, can I now deal briefly with the approach
to some of the fact witnesses that Post Office called.
In relation to Mr Godeseth at {A/6/63}, which is page 63
of Post Office's written closings.
MR JUSTICE FRASER: Go on.
MR GREEN: Post Office's paragraph deals with Mr Godeseth's
evidence there, and what is said there is, in line 3,
some of his evidence was unsatisfactory:
"With the benefit of hindsight, Post Office would
not have asked Mr Godeseth to cover several matters that
were addressed in his first two witness statements -
although it is right to point out that if Post Office
had only called first-hand evidence, the trial would
have been wholly unworkable."
The obvious answer to that is to call Mr Jenkins,
who was relied on repeatedly by a number of witnesses.
But let's leave that aside for a moment.
Just analysing what is said by Post Office there.
The contention that with the benefit of hindsight they
would not have asked him to deal with several matters
that were addressed is amplified at paragraph 146 on
page {A/6/67}:
"In Mr Godeseth's cross-examination, some of the
points he made on the basis of information provided by
others were shown to require correction or at least
clarification. This took Post Office by surprise. With
the benefit of hindsight, Post Office accepts that there
are points on which, if it wished to adduce any evidence
at all, it should have ensured that witness statements
were prepared for the individuals who were the sources
of the relevant information."
It is not clear why that is accepted in particular
in relation to Mr Godeseth rather than across the piece
on every occasion when that problem arose, but let's
just focus on this for the moment.
The Post Office didn't need hindsight to realise
that calling witnesses who don't actually know the
subject matter is wide open to error and will cause
difficulties for the experts, the other party and the
court, because the assumption when someone gives
a statement of truth that they believe something to be
true is at least likely to be that there is some
foundation for that belief. The difficulty for the
claimants was flagged at the pre-trial review in the
context of responding to complaints about Mr Henderson's
evidence. I won't take your Lordship to it but it is at
{C8.14/3/20}, which is me making the submission that
I effectively have to challenge someone who is not even
there in relation to Mr Jenkins who appeared to be the
source of much of the evidence.
So the difficulty of a number of Post Office's
witnesses giving evidence about things they didn't know
about, which they got from others, was flagged up at the
PTR at {C8.14/3/20}.
Mr Coyne had himself identified difficulties with
Mr Godeseth's evidence in Coyne 2, which is at
{D2/4.1/98} at paragraphs 4.3(a) and (b). If we can get
that up, that will be helpful.
It is 4.3(a) and (b), specifically referring to
speaking with Mr Jenkins and so forth and not referring
to supportive documents. Post Office should not have
been surprised about the errors revealed in
cross-examination because those errors were errors that
the claimants were able to identify in relation to his
approach to the bugs and treatment of them, which were
apparent on the face of such documents as were
disclosed.
The number of errors were identified in Mr Coyne's
second report, but not subsequently corrected. You can
refer to Mr Coyne's second report; we don't need to
bring it up, but paragraph 3.30, {D2/4.1/21}.
Particularly with regard to the number of branches
affected in the receipts and payments mismatch which we
find on page 24 of the report and the number of
occurrences at Callendar Square.
Your Lordship will remember that even in
January 2017 the Post Office was asserting that the
Callendar Square bug affected only one branch. We now
know that not to be the case.
The suggestion that Post Office were surprised is
resurrected further at paragraph 144.3, which is at
{A/6/66}:
" ... so far as Post Office was aware, the relevant
parts of Godeseth 2 were most unlikely to be
controversial. For example, the Misra trial was a
matter of public record, the four bugs were covered by
contemporaneous documentation and Post Office had no
reason to doubt Fujitsu's account of the documents it
held."
It is extremely surprising that it is said there
that his evidence of the bugs was most likely to be
uncontroversial. The impact and affect of bugs was
always going to be hotly contested and, indeed,
I actually made that submission expressly about
Mr Godeseth's second witness statement at the PTR at
{C8.14/3/18} between pages 18 and 19.
The second line on the right-hand side of the foot
of page 18:
"If your Lordship then looks at what then follows,
there is a treatment of the individual known bugs that
Mr Coyne is reporting, so Callendar Square, these are
actually front and centre main --"
MR JUSTICE FRASER: Where are you? At the very bottom.
MR GREEN: Three lines up from the bottom. Go over the page
{C8.14/3/19}. Something has gone wrong there. I will
just read it out:
"... main fighting ground for the Horizon trial, so
Callendar Square is immediately below, and you then get
some of the bugs, later on, payment mismatches ...
dealing in this section ..."
There's absolutely no doubt that the evidence about
bugs is going to be central to the testing of the
evidence in this trial.
Mr Coyne in his second report raised multiple issues
with Mr Godeseth's second witness statement and whether
it captured the full extent of the bugs. And that was
paragraph 4.3(d) on page {C8.14/98} and the further
points that follow at 4.4 over the page. And as to the
language that Fujitsu's account of the documents it
held, which is Post Office's language in its closing
submissions.
MR JUSTICE FRASER: Which language are you talking about?
MR GREEN: In Post Office's explanation of having no reason
to doubt Fujitsu's account of documents which it held.
Many of the documents were effectively created by
Fujitsu for Post Office, if they weren't Post Office
documents themselves. So, for example, the receipts and
PEAK payments mismatch documents is a note of a meeting
attended by Mr Andy Wynn of Post Office.
So it is not simply a matter of Fujitsu's account of
documents that helped. And in any event the documents
when taking a witness statement from anyone, the proper
approach is to say: what's the source of that? Is there
a document that supports that and, if so, should it be
exhibited? And that didn't happen, and that's part of
what went wrong, although not all of what went wrong.
So there are two things in combination. There is
calling witnesses who aren't properly placed and it
hasn't been made sure that they are properly placed to
give the evidence that they are giving. And secondly,
the fact that very frequently no documentary support was
cited for general propositions of what would or could or
should have been happening is why (a) it was difficult
for the claimants to unpick it, and we had to go through
the documents ourselves and find them, but (b) it must
have contributed to any misapprehension, if there was
such, by Post Office about whether Mr Godeseth was
properly placed to give the evidence at all.
Had the witness statements been done in the usual
way by seeking to identify what the basis for things
that are being said is and exhibiting the relevant
documents, it would have become apparent much sooner,
even if it wasn't fully apparent from all the
observations made by Mr Coyne and others, that
Mr Godeseth's evidence was very unsatisfactory in that
respect.
As to Mr Jenkins, the explanation for him not being
called is provided at paragraph 138, and that's at
{A/6/64}, page 64 of the Post Office's closing.
Paragraph 138 says:
"Post Office wanted to provide a simple and
uncontroversial overview of Horizon and its relevant
features."
My Lord, just pausing there. That is exactly the
problem we faced in the Common Issues trial, that there
was this wish to present a general overview that seemed
extremely rosy and which, when unpicked, fell apart.
That was also the difference reflected in the
approach of the two experts. Dr Worden provided this
top down overview-based look at the Horizon system, in
part largely relying on Mr Godeseth, as we will come to
in a minute, and Mr Parker's evidence and others.
Whereas Mr Coyne was trying from a very early stage to
actually understand how any errors were recorded, were
they recorded, obtain the documents and analyse them for
the purposes of answering the Horizon questions. In
that respect -- we will come to it in a bit more
detail -- we say the criticisms made of Mr Coyne are
extremely unfair.
Just focusing on the explanation from Mr Jenkins not
being called, which we find at paragraph 138, it says
Post Office says there that:
"It recognised that it was not possible for one
person to have a complete understanding of all the
corners of the Horizon system ..."
Pausing there, Mr Jenkins appears to have had nearly
a complete knowledge from what we have heard:
" ... but, on the basis that there would not be room
in the timetable for multiple witnesses, it took the
view that this overview should be provided by one
person."
Just pausing there. That isn't even what happened.
It wasn't only Mr Godeseth who referred to Mr Jenkins.
And Mr Jenkins could certainly have collected
information from other team members who other witnesses
collected information from. But they then go on to say:
"Two possible candidates were Torstein Godeseth and
Gareth Jenkins. Taking into account the involvement
that Mr Jenkins had had in a number of criminal
prosecutions that are currently being looked at by the
criminal Cases Review Commission (eg the Misra case),
Post Office asked Mr Godeseth to do so."
Well, we respectfully say it is a slightly
surprising approach because we see it further amplified
at 144.1 on page {A/6/66}. It says:
"... Post Office was concerned that the Horizon
Issues trial could become an investigation of
[Mr Jenkins'] role in this and other criminal cases."
My Lord, two points arise on that. Firstly, those
cases are stayed in the GLO, the criminal cases.
MR JUSTICE FRASER: This court is not concerned with and has
no jurisdiction in respect of any of the criminal cases.
MR GREEN: Precisely. There is absolutely no prospect of
either of us seeking to do that or of your Lordship
allowing it. So we don't accept that is a good reason
at all. Mr Jenkins clearly had the firsthand knowledge
and the court wouldn't have allowed the sort of
investigation of which Post Office was fearful in this
respect. But it does suggest a defensiveness which is
not completely isolated in these proceedings.
MR JUSTICE FRASER: I don't have the reference immediately
to hand but I seem to recall one of the witnesses said
that Mr Jenkins had retired, hadn't he, or am I wrong
about that?
MR GREEN: He has retired, but he had been consulted. He
provided comments --
MR JUSTICE FRASER: I know the extent to which witnesses
have relied on him etc, but I just wanted to check
I wasn't imagining that.
MR GREEN: Your Lordship is absolutely right. Of course,
that's not actually the reason relied on. And then
there has then been correspondence about his
availability, which I will very briefly touch on.
{H/184/1}, {H/201/1} and {H/203/1} are effectively the
three letters.
{H/184/1}, if we start with that. A very short
letter:
"We note that Mr Jenkins is the author of many
important contemporaneous documents, and that both of
Mr Godeseth's witness statements give hearsay evidence
on the basis of conversations he has apparently had with
Mr Jenkins.
"Is Mr Jenkins available during the trial period?"
{H/201/1} is the Wombles response of
12th February 2019, and the second line of that says:
"The information that you have sought regarding
Mr Jenkins is clearly privileged."
Then they say it would have been open to us to
approach Mr Jenkins or apply to call him, which is
correct.
Then they add there if we were going to make
an application out of time, we should bear in mind that:
"... he acted as our clients' [Post Office's] expert
witness in relation to a number of criminal prosecutions
... being looked at by the [CCRC]," as a suggestion to
us that we should not seek to contact him, it appeared
to be.
Then at {H/203/1} the letter from Freeths:
"We note your response to our letter regarding the
reason for not calling Mr Jenkins and why he is not
available to give evidence during the trial: you have
declined to give any or even to tell us whether he is
available during the trial.
"You have cited that the explanation for him not
himself giving evidence is privileged."
So our just simple question, "Is he available?", was
not: we don't know, or anything of that sort, it is that
it is privileged. So we assumed that he must be part of
the shadow expert team.
We have now been informed apparently he is not.
MR JUSTICE FRASER: He is not what?
MR GREEN: Part of the shadow experts team that the
Post Office is entitled to use on the case and has been
using and included in their budget.
So we enquired whether he was in fact part of that
team and the answer is no. We hadn't appreciated that.
MR JUSTICE FRASER: I don't think he -- I think the item of
cost was originally included in the budget and was then
withdrawn. I do not think it is correct to say it is
included in the budget.
MR GREEN: No, my Lord, it was originally included and then
disallowed.
MR JUSTICE FRASER: Well, no, it was not disallowed. It was
withdrawn because it was --
MR GREEN: We objected to it and it was not included in the
approved budget. That's absolutely correct.
My Lord, can I move on to the issue taken with
cross-examining witnesses by reference to documents,
because this is something in respect of which
Post Office complains that it is unfair to cross-examine
witnesses by taking them to documents.
The criticism is levelled at us for doing this at
paragraph 850, which is {A/6/277}. Paragraph 850 makes
the point three lines up from the bottom on the
right-hand side:
"She was, however, taken in cross-examination to
many documents which she had not seen before -
a recurring theme of Cs' cross-examination. It is
submit that such an exercise is of very limited
utility."
Pausing there. We say it is not an unusual way to
cross-examine to put contemporaneous documents that have
been disclosed as relevant to a witness whose evidence
appears to suggest that they are aware of at least the
general area in which those documents arise. We say
that's actually common. Questions like that have been
put up and down the country every second of the working
day across all courts and tribunals. So it is not
an unusual approach to take witnesses to contemporaneous
documents.
But the complaint that's made is that Ms Van Den
Bogerd was taken to many documents which she had not
seen before. Now, that is a prisoner to how many
documents she has chosen to look at, because if
a witness has not looked at, on the face of it, relevant
contemporaneous documents at all then every document
will fall into the category of document to which
Post Office appears to complain that she was taken. And
if we take an example, Mrs Van Den Bogerd's witness
statement speculated about the potential cause of
a spike in declared losses on the introduction of
Horizon Online. Paragraph 183 at {E2/5/42}.
This was effectively SPMs tidying up their accounts.
That speculation was challenged in cross-examination on
the footing that the claimants would expect a senior
witness giving evidence on an issue of that type to
review the available documentary record before
speculating in that way and providing the court with
a witness statement signed and verified with a statement
of truth.
Mrs Van Den Bogerd specifically said she had not
done this in answer to one of your Lordship's questions
which we find at {Day5/167:22} onwards. At 167, line 22
your Lordship says:
"MR JUSTICE FRASER: Understood. And then the
second point is when you were preparing your witness
statement and in particular the paragraphs at 180 to
183 --
"Answer: Yes.
"MR JUSTICE FRASER: -- did you do any investigation
in respect of what might have been happening that you
didn't know at the time in 2000 --
"Answer: Not back to 2000, no.
"MR JUSTICE FRASER: -- or in 2010 when the change
was from Legacy Horizon to Horizon Online?
"Answer: So in 2010 I was in a different role and
had broader responsibility and I knew what -- what we
did, again we replicated a similar approach to make sure
we supported branches at the time, but as for any detail
of Information, I didn't research into that, no."
So that was an example of one of the matters on
which Mrs Van Den Bogerd was challenged by reference to
contemporaneous documents. She hadn't seen them, but
that is not a proper or legitimate criticism of the
claimants' side, we respectfully submit, and might
fairly be thought to go the other way.
In moving on to rewriting the fact evidence in the
closing submissions, which is the next theme, some of
these aspects are quite surprising. One example is the
problem management procedure.
The defendants' closing addressed the problem
management procedure albeit in the context of
Mr Godeseth's evidence being unsatisfactory, and used
that as an excuse to try and rewrite the evidence which
the court has actually heard.
We see that at page 68 of the defendants' closing
{A/6/68} at paragraph 147.4, where it says:
"In paragraph 63, he [Mr Godeseth] appeared to be
saying that Fujitsu’s Post Office Account Customer
Service Problem Management Procedure document 223 was
not implemented following Mr Salawu's departure as
Horizon Head Lead Service Delivery manager, when in fact
it was merely section 1.4 of that document that was not
implemented."
Well, that came as news to the claimants, my Lord,
not least because of the way the issue arose and because
of how it was cross-examined on and not re-examined on.
MR JUSTICE FRASER: Does that last part of the sentence come
from evidence or is that --
MR GREEN: No, that's new.
MR JUSTICE FRASER: All right.
MR GREEN: So it arose because Mr Coyne was proceeding on
the basis that there was a proper problem management
system in place, and therefore there should be documents
available that would collate bugs and errors and show
lots of information, so we could get disclosure. We
thought that would be very helpful, and we can see that
from Mr Coyne's first report at {D2/1/97} at
paragraph 5.158 and 5.159:
" ... is the POL monitor that tracks the number of
records arising directly as a result of managed change
activities."
You can see the problem management procedure
footnoted at footnote 150 at the bottom of the page:
"No disclosed logs have been found in respect of
these problem records that are listed as being reported
monthly."
5.159:
"Requests have been made in relation to making such
Management Information reports. At the time of writing,
these have not been made available for analysis."
So Mr Coyne's assumption, which is actually shared
by Dr Worden, was that the problem management procedure
document was in force and therefore there should be
a repository of documents, which we have not been aware
of at that stage, which would be very helpful because
they would show what was going on.
Then in response to that, with the looming
possibility of disclosure, we get Mr Godeseth's second
witness statement served saying in fact it was not
implemented, and we find that at paragraph 63,
{E2/7/16}. In paragraph 63 we get the conversation with
Mr Bansal.
And then four lines up from the bottom:
"I understand from Steve that Saheed Salawu's
replacement did not wish to implement the changes and
therefore the records referred to by Mr Coyne ... of his
report do not exist, as we continued to follow the
previous existing reporting methodology."
That is referring back up to the situation before
the new procedure, which was to achieve the same,
problem management seeks to establish the root cause of
incidents and then start actions to improve or correct
the situation.
So your Lordship may remember the cross-examination
obviously that followed what we heralded in our written
openings at paragraph 6(4) at {A/1/6} where we made
clear our understanding was that this had not been
brought in and that Mr Godeseth suggested this followed
the departure of the Horizon lead service delivery
manager.
MR JUSTICE FRASER: Where are you looking now?
MR GREEN: This is paragraph 6(4).
MR JUSTICE FRASER: Yes.
MR GREEN: The bottom two lines, the departure, this was not
brought in. Three lines up from the right-hand side.
Mr Godeseth suggests and refers to those paragraphs.
No whiff of disagreement from Post Office at this
stage, in fact until closing submissions. Evidence at
trial was Mr Godeseth was cross-examined on
paragraph 63. I will give your Lordship the reference
without going there. {Day7/157:4} to line 21. We also
put to Mr Parker the problem management system wasn't
brought in. We did that at {Day12/53:4}.
MR JUSTICE FRASER: What did Mr Parker say?
MR GREEN: He didn't suggest it had been brought in either.
MR JUSTICE FRASER: Let's go to --
MR GREEN: Let's go to {Day12/53:4}, if we may.
We can see it says "proper management", but we can
see later on it is clearly problem management:
"... just as an aside -- the proper", it says proper
on the transcript, "wasn't brought in either. We
covered that with Mr Godeseth, you were here for that?
"Answer: Yes.
"Question: So there was no problem management
system brought in notwithstanding it was internally
recommended, and so all you're left with is this system
of looking at the codes and seeing how they have been
categorised on closure."
Absolutely no demur at all. It has been there in
Mr Godeseth's evidence, not a scintilla of disagreement
from anyone, no re-examination at that point of
Mr Godeseth or Mr Parker.
We also put it to Dr Worden on {Day19/166:15} to
line 16 on that footing. About halfway down the page.
I'm not sure that's quite right. Can we just check that
reference. I will come back to you on that.
MR JUSTICE FRASER: That doesn't appear to be the right
reference.
MR GREEN: It is not the right reference.
MR JUSTICE FRASER: You say you put it to Dr Worden?
MR GREEN: Put it to Dr Worden as well. Again, no
re-examination on that.
So we have an unusual situation, we respectfully
submit, to say the least that we filed our written
closing submissions on the basis of the evidence that
was actually heard and was tested because with
Mr Godeseth we said "Well, how come there are all these
extra versions of this if it wasn't introduced?"
Because it looked weird to us. We challenged it,
apparently it was not introduced. And we filed our
written closing submissions, and we learn then for the
first time from Post Office that actually it was brought
in. It was only an individual paragraph that wasn't.
We respectfully say obviously that's completely
unsatisfactory, but also if it is true it is again less
than forthcoming and showing a striking lack of candour
about what documents might be available.
My Lord, just to give another example of a sort of
rewrite which sort of bleeds into the attacks on the
documents where they are unfavourable to Post Office.
So we have this strange part, which I have already shown
your Lordship, where they say if internal Post Office
documents say Horizon was problematic, the authors have
got it wrong. And that's part of this theme.
If we look at the document recording the visit to
Mr Bates' branch, it is at {F/99.1/1}. It was
a document put to both Mrs Van Den Bogerd and Dr Worden.
If I can just give the references {Day5/168:14} and
{Day18/123:12}.
If we look at {F/99.1/1}, the point about that
document was that the visit being made, the
Post Office -- it should be, I think, F/99 --
MR JUSTICE FRASER: You said 99.1.
MR GREEN: That may have been my error, my Lord.
MR JUSTICE FRASER: Is it {F/99/1}?
MR GREEN: I think it may be {F/99/1}. I will just make the
point and we will check the reference on that.
It is {F/99.1/1}, I think. The short point,
my Lord, is this is the one where Mr Bates is visited,
and the officer visiting says "I couldn't get a correct
read on the cash account because Horizon intermittently
adds the cash from the previous day".
MR JUSTICE FRASER: This is the audit report, isn't it?
MR GREEN: Exactly. So the Post Office auditor -- two of my
Opus screens have gone down -- but at {F/99.1/1}. We
seem to be able to see it on our screens.
MR JUSTICE FRASER: Which page are you going to?
MR GREEN: It is {F/99.1/4}.
MR JUSTICE FRASER: Page 4. Yes. I have got it on my own
separate screen.
MR GREEN: I'm most grateful. The short point is that it is
striking that someone attending to do an audit of the
correct cash totals -- thank you very much; if we could
slide to page 4 of that, I think -- is unable -- if we
go towards the bottom, "Comments", just above "National
Savings":
"A correct assessment of cash holdings could not be
made because the Horizon system intermittently adds the
previous days cash holdings to the daily declaration."
There was no attempt to re-examine Mrs Van Den
Bogerd or Dr Worden about that, but what is now said is
that this is actually a designed function in Horizon in
the closing submissions that Post Office has filed at
{A/6/279}, paragraph 856 about carrying forward balances
from a previous day. And we respectfully submit (a)
that's completely new, and (b) it just doesn't chime
with the reality of an auditor who well knows how
Horizon should operate encountering that difficulty in
Mr Bates's branch when he goes to get a cash reading.
The word "intermittently adding" is extremely
difficult to reconcile with simply carrying a value
forward from the day before.
My Lord, just to give your Lordship the correct
reference to the problem management being put to
Dr Worden, that was at {Day19/176:13} to line 17.
I apologise for that mistake.
MR JUSTICE FRASER: Yes.
MR GREEN: Then moving forward to --
MR JUSTICE FRASER: Just before you move on though, isn't
the point it was not re-examined upon, doesn't that fall
within the same basket of it is a time-limited trial,
one doesn't have time to re-examine on everything?
MR GREEN: My Lord, absolutely right, but if there's going
to be a --
MR JUSTICE FRASER: Positive explanation.
MR GREEN: Positive case suddenly introduced for the first
time, it is quite nice to have a passing mention in
re-examination.
MR JUSTICE FRASER: All right.
MR GREEN: Take, for example, keystrokes where the position
is absolutely bizarre because you get Mrs Van Den Bogerd
accepting initially that there were keystrokes
available. Then you get the sheepish re-examination
that maybe they are not, and then you get the reference
to Mrs Mather whose witness statement expressly said
that keystrokes were available and then in chief is
invited to say what she meant by the word "keystrokes".
She means or transactions and sales data.
Then the hammer drops later on when we get
disclosure of a whole load of keystrokes which were
available from Fujitsu. I mean, it is against that
background that there is some concern about introducing
completely new explanations without any foundation in
the evidence.
My Lord, can I just deal with a couple of
documentary examples. First of all, the phantom
transaction PEAK and the Romec engineer.
That was quite surprising to say the least. What
effectively was said, as your Lordship may remember, is
that the Romec engineer's view of having actually
observed phantom transactions happening was not
reliable. That was how it was put to Mr Coyne. And
that PEAK is {F/97/1}, and the reference to the Romec
engineer advising that he has witnessed further phantom
transactions whilst on site is on page {F/97/5} of that
PEAK.
It is right your Lordship should be reminded, if
I may, that it wasn't only that Romec had seen it,
because we know that Mr Carroll had also seen it. So
when we look at page {F/97/7} of that PEAK, in the
bottom yellow box:
"I now have pressing evidence to suggest that
unwanted peripheral input is occurring, the likely
source being the screen. This has been seen at Old
Iselworth (OI) and Wawne (W) with OI being the best
site; when the PM has been asked to leave the screen on
overnight I have observed system activity corresponding
to screen presses happening with no corresponding
evidence of either routine system activity or human
interference, the way forward now is to correlate this
with the Microtouch supplied monitoring software and to
this end Wendy is arranging for installation of Kit at
OI on Friday, we can then, provided the PM agrees, leave
screens on over the weekend and record what happens.
Once these results have been analysed I feel sure that
we will be in a position to move forwards at OI. all
other cases should be considered on their individual
merits but you must appreciate that this is a fairly
intensive analytical activity and I cannot hope to
provide answers on all cases in the short term."
So it was not just Romec who had reached that view,
but what's relied on by Post Office when they say a host
of possible explanations at page 475 of their closing
submissions, paragraphs 10 to 11, is Mr Carroll's view,
effectively, that we find on page 9 of the PEAK {F/97/9}
where, in closing this down, Mr Carroll says:
"Phantom [transactions] have not been proven in
certifications which preclude user error. In all cases
where these have occurred a user error related cause can
be attributed to the phenomenon."
Three things going on. First, rather overlooking
what Pat Carroll had previously said about it. Secondly
seeking to elevate the slightly bizarre way of
determining how this should be treated, as user error or
not, to a finding that it was in fact user error, and
Mr Coyne being challenged on the footing that he agreed
he was he was not in a position to say that Mr Carroll
was wrong, which Mr Coyne very fairly accepted: well,
I can't say he is wrong that it couldn't preclude user
error.
But we respectfully say that is a pretty astonishing
way of dealing with the PEAK which is pretty
unsatisfactory, to say that actually Mr Carroll and the
Romec engineer plainly got it wrong.
The Helen Rose report goes a step further, if that's
possible. The treatment of that is, we say, absolutely
extraordinary. The Helen Rose report is at page 1082
and includes --
MR JUSTICE FRASER: Of?
MR GREEN: Sorry, {F/1082/1}, and on page {F/1082/2} of that
document under "Reviewing the data", it says:
"On looking at the credence data, it clearly
indicates that the reversal was completed by JAR001
(postmaster) at 10:37 ... and was reversal indicator 1
(existing reversal) and settled to cash. An existing
reversal is where the session number/Automated Payment
number has to be entered to reverse the item."
So the point about this was the initial view was
an assumption that there had been a reversal by the SPM.
And the Helen Rose report then, as your Lordship will
remember from dealing with this document on several
occasions, goes through to identify that that actually
had not been the case.
Mr Coyne flagged this up at {D2/4.1/113} in his
second report. He says:
"Credence data, most commonly used by Post Office
for their investigations, is either wrong or does not
provide sufficient information to complete the full
picture ..."
This is an example of one of his concerns. On what
was presented in the Helen Rose report, that seems to be
correct. But Mr Coyne's account of the Helen Rose
report was made subject to very extensive
cross-examination, pressing him to accept that
Helen Rose had in fact misinterpreted the underlying
data and put 2 and 2 together to make 5. That's what
was put. In fact, my learned friend said put 2 and 2
together to make 4 and then changed it to 5, but the
point is it was being suggested that she had got it
wrong, not right. That's at {Day15/38:15} to
{Day15/39:24}.
What's effectively being put to Mr Coyne there is
that Helen Rose has got it wrong. And just pausing
there, Mr Coyne's interpretation of what appears on the
face of the report was totally reasonable and, we say,
right, and Post Office not only runs the, we say,
extraordinary misinterpretation point in their closing
submissions at paragraph 529 on page 194, but also used
that as a basis to subject Mr Coyne to very heavy
criticism for his evidence. There are pages and pages
of criticism on that footing on an extremely strained
interpretation of a document which appears to say the
opposite to what Helen Rose actually said.
Bear in mind, my Lord, this was a document about
which, for example, Angela Van Den Bogerd's evidence was
initially completely wrong. So it is actually the
Post Office's treatment of the Helen Rose report which
is properly the subject of criticism, and it is bizarre
that it is Post Office's case that it is Helen Rose's
fault that she misinterpreted the data, given that she
was a security fraud analyst who was evidently involved
in prosecutions.
My Lord, I have touched on the internet documents
critical of the Post Office and we identified a number
in opening. We put them to Dr Worden and addressed them
again in closing by reference to what the documents
actually say; for example, paragraph 614 at {A/5/213}.
MR JUSTICE FRASER: This is your closing, yes?
MR GREEN: This is our closing. It is to that that
Post Office give their response. It is a section in our
closing that begins at paragraph 614 on page 213.
MR JUSTICE FRASER: Yes.
MR GREEN: It is that section to which Post Office are
responding with reference to which I took your Lordship
out of order to at the beginning, where they say
{A/6/134}:
"The Post Office documents were not drafted with the
benefit of the vast amount of work that has been carried
out by the experts for this trial. If the authors
considered that Horizon was not a good system, they were
wrong (although that is not even a fair summary of what
the documents say)."
We say that is an absolutely astonishing submission,
particularly where Dr Worden hadn't read any of them and
the contention is that he is in a better position, for
example, than Mr Rob Houghton, who is Post Office's
chief information officer who presented to the board on
Post Office's IT strategy in January 2017, which is one
of the documents which we see at {F/1611/87}.
Then, just to make the one-way street point good,
that is to be contrasted with the, as it turned out,
wholly misplaced reliance on the ISAE service audits in
relation to change management and remote access, which
we have addressed separately why those audits (a)
weren't addressed to the same point as the 2011 audit,
and secondly, why they plainly didn't pick up what we've
seen about remote access in the APPSUP role in the
APPSUP PEAK.
But what they say about that, and this is part of
Post Office's cross-examination of Mr Coyne, was to say
to Mr Coyne -- we can look at it at {Day16/174:11} to
line 15. The tee-up is that unhelpful documents to
Post Office, even if drafted by Post Office, even if
drafted by their own IT chief information officer and
presented to the board, they are wrong and the experts,
including Dr Worden, who had never read any of them, are
right. Leave aside the fact that Dr Worden accepted
they were irreconcilable with his views.
Then when something is thought, albeit wrongly, to
be in Post Office's favour, what you get is the
cross-examination at page 174, lines 11 to 15:
"Question: Mr Coyne, would you agree with me that
in principle the best people to judge whether action is
being taken to address recommendations made by auditors
is the auditors themselves rather than you, would you
agree with that?
"Answer: Yes.
"Question: Let's look to see what the auditors say
in later years."
So effectively where the documents are in
Post Office's favour you take them on their face and the
authors are right, even if they are the wrong documents
and so forth, contradicted by the PEAK at the time, well
known to Fujitsu. But when the documents are against
Post Office, the authors, even if they are Post Office's
own employees charged with those specific roles,
internal experts, they are apparently wrong. And we
respectfully say even by the standards of Post Office
conduct in this litigation, that is a bizarre approach
to urge upon this court to deal with the relevant
contemporaneous documents.
My Lord, that's a natural break. Would it be
convenient to rise a couple of minutes early?
MR JUSTICE FRASER: I think it would. Thank you very much.
2 o'clock.
(1.00 pm)
(The short adjournment)
(2.00 pm)
MR JUSTICE FRASER: Mr Green.
MR GREEN: My Lord, can I just pick up one brief observation
in relation to disclosure.
The Post Office made various criticisms in their
closing submissions relating to claimants' disclosure,
and the short point is that the claimants were directed
to disclose the documents they relied on and any known
adverse documents, which is what they in fact did. And
that order is from the fourth CMC order, paragraph 5,
which is at {C7/18/2}.
We needn't go to it. There is correspondence where
Post Office has sought to press the claimants in
relation to those issues in the light of evidence given
at trial. And the relevant -- I'm not going to take
your Lordship to the correspondence, but just in case,
so your Lordship has the picture in case something is
made of it, they are letters of 15th March, 17th March,
14th May and 29th May of this year, and they are to be
found respectively at {H/242.6/1}, 15th March; {H/255/1}
is 27th March --
MR JUSTICE FRASER: You said 27th March?
MR GREEN: 27th March.
MR JUSTICE FRASER: I thought you said 17th March.
MR GREEN: If I did I misspoke, I'm sorry. 27th March.
{H/280/1} is 14th May and {H/303/1} is 29th May.
MR JUSTICE FRASER: Those are letters in both directions,
are they?
MR GREEN: Toing and froing in relation to that. It is of
course open to a party to make submissions in relation
to how document have been disclosed, as, indeed,
Post Office seeks to do going back that. A bit more
resistant to the stream going the other way, as
your Lordship has already seen.
Just very briefly in relation to that, the adoption
of model C disclosure by the court, including in
particular the obligation to disclose known adverse
documents, does not make a party's explanation for the
emergence at a late stage of documents irrelevant or
immune to scrutiny from the court, nor is an approach by
the claimants to seek at successive hearings to improve
on the position in relation to documents, some of which
they have been seeking since 2016, an answer by way of
saying, well, you never sought specific disclosure.
Neither of those are answers in relation to a fair
appraisal being made of how Post Office documents have
come to be disclosed as late as they have.
That's a very brief response in relation to that
point, my Lord. We just don't accept that for a moment,
particularly when the evidence, for example the
mis-keying, flies in the face of the most extraordinary
changes of evidence before the court. Perfectly
legitimate for the court to want to know how that
document came to be disclosed when it did.
So with that brief footnote in relation to
disclosure I just wanted to make a very few brief
submissions in relation to the treatment of Mr Coyne,
who your Lordship has obviously appreciated has been
heavily criticised, we respectfully say very unfairly.
His approach was correctly to try and see if there
was evidence of any relevant bugs, to try and identify
what remote access evidence there was, to identify
relevant lines of enquiry, which we see in his early
requests as long ago as April 2018, as we flagged up in
our closing submissions, and to seek documents in the
RFIs that they did in the summer of 2018.
His attempt to be careful and try and work out what
has happened and what there was and was not evidence of,
including what there was not strong evidence of, which
is the wording of paragraph 1.15 on which he was
extensively cross-examined, that was not only the right
approach, but a helpful approach and one that was framed
by reference to the Horizon Issues as they were actually
ordered by the court.
There is a criticism of him in relation to the
Helen Rose report to which I have already averted, but
we respectfully invite the court to contrast that with
the defendants'/Post Office's treatment of the evidence
of Mrs Van Den Bogerd in relation to Helen Rose, which
turned out to be wrong, which Post Office, at
paragraph 852 {A/6/278} of their closings explained that
her evidence was to give early notice of Post Office's
case but now effectively overtaken by more detailed
consideration in the expert evidence.
That is an attempt (a) to gloss over the fact that
the evidence was wrong, but (b) it stands in stark
contrast to the pages, ten pages of criticism,
critiquing Mr Coyne's account of the Helen Rose report
that we find between paragraph 521 at {A/6/192} and 552
at {A/6/202}.
The reality is that Mrs Van Den Bogerd
misrepresented, whether unwittingly or by not taking
care, what the Helen Rose report said and what her
evidence said was wrong, as she accepted.
We see that evidence at {Day5/90:1}. So, again, we
have got a sort of one-way street where Mr Coyne
actually gets the Helen Rose report right and is
criticised; Mrs Van Den Bogerd gets it wrong and accepts
it is wrong, and that's just clarification and overtaken
by more detailed consideration, all in circumstances
where the premise of the criticism of Mr Coyne is that
Helen Rose, Post Office's own investigator, as
I submitted before lunch, completely misinterpreted the
data before her.
That is one example, my Lord, of a fairly wholesale
attempt to discredit Mr Coyne by suggesting, we say
unfairly, that his essential endeavour was to throw mud
at Horizon in the expectation that at least some of it
would stick. And those words are taken from
paragraph 267 at {A/6/103}, bottom four lines:
"His essential endeavour was to throw mud at Horizon
in the expectation that at least some of it would stick,
and not to worry too much or at all about giving the
other side of the picture or even presenting a fair view
of the documents to which he referred."
That's both a surprising submission and an unfair
one. And where, for example, Mr Coyne said that the
position on transaction corrections might make the
position better or worse, that was put to Dr Worden. He
is being scrupulously fair there and Dr Worden accepted
that he was. So it flies in the face of concessions
made by Dr Worden and it also flies in the face of the
way in which Mr Coyne gave his evidence.
That's obviously a matter for the court, but we
respectfully submit that Mr Coyne was conscientiously
trying to answer questions that he was asked, not always
in circumstances where he was very comfortable with the
premise that he was being boxed into for those
questions. But he sought to try and answer them fairly,
and a fair reading of his evidence would have regard to
the joint statements, the fact that 1.15 was necessarily
a compromise between two experts who differed on the
definition of lasting, amongst other things, who agreed,
as we see at joint 3, paragraph 4.1, which is {D1/4/7}
that Issue 4 fed back into Issue 1, and when the joint
statement is read as a whole rather than saying
an isolated paragraph and inviting an expert who has
taken a completely different approach to grind very
finely those aspects of his analysis which are
sufficient to show strong evidence of the type of bugs
that was being considered.
I mention that feature of 1.15 because it was
particularly telling for Mr Coyne who was very careful
in his answers to stop at the point at which there was
no evidence going further.
So when the question was put to him, where he felt
he couldn't go further because he didn't have the
evidence one way or the other, he would stop at that
point. He would not go on to draw what Dr Worden in
some places called weak inferences about various things,
or other inferences, or proceed on assumptions.
MR JUSTICE FRASER: You said 1.15 a couple of times, I
think.
MR GREEN: Yes, 1.15, my Lord, which was the bit that my
learned friend cross-examined him about in joint 2.
MR JUSTICE FRASER: In joint 2. I thought you said joint 4.
MR GREEN: I'm sorry. The Issue 4 point is paragraph 4.1 of
joint 3.
MR JUSTICE FRASER: Of joint 3.
MR GREEN: I should have made clear that the 1.15 point is
the one that my learned friend cross-examined on fairly
extensively, which is in --
MR JUSTICE FRASER: Joint 2.
MR GREEN: Joint 2. That's at page 29.
MR JUSTICE FRASER: Yes.
MR GREEN: Your Lordship will remember that the 1.9 on the
previous page is the experts' differing view on branch
impact.
MR JUSTICE FRASER: Yes.
MR GREEN: So we respectfully say it is a little artificial
and unfair where the experts try to reach a bracket
where there is more than one variable in their
difference of view and reach that compromise at 1.15, to
subject it to the sort of drafting scrutiny that
Parliamentary statute might bear, and that the fair way
to approach it is in the light of the joint statement
read as a whole in the context of the approaches of the
two experts.
Parenthetically, my Lord, when my learned friend
revisited the 40 bugs analysis with Mr Coyne on
{Day15/165:17}, our submission about the layered
assumptions and confinements on the basis of which he
was cross-examined on 1.15 become clear when you look at
his answer at 165.17. He makes clear that he still
regards the position he described in his report as
correct.
Turning to Dr Worden. Your Lordship has our
submissions on the statistical analysis and so forth.
And we are obviously not going to repeat those, but
there are a couple of points we would like briefly to
pick up on.
The Post Office submits at {A/6/99} at paragraph 249
of their closing that Dr Worden did not confine his
efforts to looking only for problems or only for
evidence of Horizon working well.
We respectfully say that's not correct. Not only
was there no consideration of documents pointing in the
other direction to any significant extent, but if we
look at joint 1 at {D1/1/10}, Dr Worden, in the top box,
penultimate paragraph, explains what he is going to do.
He says:
"In my report I shall survey the evidence I have
found that Fujitsu paid sufficient attention to the
dimensions of robustness, and that they did so
successfully. I shall also address evidence from
Mr Coyne implying that Horizon fell short of its
robustness objectives."
What's missing from that summary is an enquiry by
him in the same vein as Mr Coyne. So not only was that
his approach, but he actually telegraphed it in that
description of what he proposed to do. And that's
vividly found in the approach in his report.
Secondly, at paragraph 251 the Post Office goes on
to say he did not inappropriately prefer the evidence of
the Post Office witnesses. {A/6/99}
That's just not right. He did so. It was put to
him, and we have got {Day18/56:5} to line 7. He was
cross-examined specifically about where he had said in
his report he had established that something had
happened based on the evidence of Post Office witnesses,
and he accepted at that place in the transcript,
{Day18/56:5} to 7 -- I think that piece begins at
page 53, line 3 -- that he shouldn't have done that.
So this is not even an available submission, let
alone one which, in our respectful submission, should be
referred by the court. The fact that his report is
replete with references to on the basis that the
defendants' evidence is right, underscores that that was
a running theme. And the fact that when Mr Parker was
saying Mr Roll was wrong, Dr Worden was happy to rely on
that, but when Mr Parker in his second witness statement
then agreed with Mr Roll, Dr Worden said he was confused
and that's why he couldn't agree with it but took
absolutely no steps whatsoever to seek any clarification
that would have alleviated any genuine confusion.
So we respectfully say that that is not correct to
say he did not inappropriately prefer the evidence of
Post Office witnesses. He did and he admitted doing so.
Finally, in relation to Dr Worden, at paragraph 316
of the Post Office's closings, which is {A/6/120}, the
Post Office says:
"Dr Worden stated in cross-examination that he had
been 'told' to send his report direct to the Court ..."
Pausing there. It was actually in a question from
your Lordship that he explained that. Then Post Office
says:
"... but that word could give the wrong impression."
Pausing there. Post Office's closing submissions
are replete with footnotes everywhere, but that word
"told" is not footnoted. So it is worth your Lordship
having a reference to the relevant part of the
transcript, which is {Day20/189:12}.
If we look at line 12, your Lordship asked:
"Have you ever served one of your expert reports
directly on the court before?
"Answer: I have never done that myself before.
I had it done to me."
Your Lordship says:
"But you have never done it before?"
Then Dr Worden then volunteers:
"Answer: A kind of late report, no. I mean the
issue of serving direct on the court rather than through
lawyers, I don't recall how that happened in the past.
I suspect it was all done through lawyers."
Then your Lordship at line 21:
"But in this case everyone knows you sent an email
to my clerk?"
Then line 23 Dr Worden then says:
"Answer: I did, yes."
Then he volunteers:
"That's what I was advised to do, that was how I was
advised to do it."
Your Lordship:
"You were advised to do it?"
"Answer: By Post Office lawyers, yes."
So the word that is quoted, "told", is not the word
that was used. The word that was used was "advised" and
it was volunteered by Dr Worden. Your Lordship didn't
ask who told him to do it; he volunteered that answer.
And at the end of that exchange your Lordship turned to
both counsel and said: are there any follow-up questions
from that exchange?" And there were none from either me
or my learned friend.
Now, that account is not identical to the accounts
given to the court by Post Office lawyers, especially in
Mr Parsons' 17th witness statement at {C11/22/1}, which
made no mention of any advice given to Dr Worden and at
least --
MR JUSTICE FRASER: Where are we going now?
MR GREEN: {C11/22/1}.
MR JUSTICE FRASER: Page?
MR GREEN: If we go forward to page {C11/22/6} and if we
look at paragraph 22:
"His covering email to the court provided that:
"The further work ... was done at his own
instigation and not prompted by Post Office or its
lawyers."
22.2:
"In Dr Worden's opinion, this work led to a material
change in his opinions and that he believed he was
obliged to inform the Court ..."
22.3:
"A draft version of the report was provided to
Mr Coyne ..."
And so forth.
Then it deals with Dr Worden's email to the court.
That does not make clear what we find in the transcript
at 190, lines 9 to 10, which is where he tells
your Lordship that he had been given advice on the
content of the email that was going to be sent as
a covering email.
So we are now faced with a situation in which
Post Office's closing submissions at 316 say that the
word "told" could give the wrong impression {A/6/120}.
And in our submission, if Post Office are saying that
Dr Worden's evidence is misleading, they should disclose
all communications with Dr Worden on that issue, which,
in the circumstances, either are not privileged, or if
they are privileged, that privilege has been waived by
saying that it could give a misleading impression.
And if Post Office has any resistance to that
course, they should be put to their election to withdraw
the suggestion in that paragraph that Dr Worden's
evidence gives a misleading impression, or waive any
privilege there might be in relation to their dealings
with Dr Worden on those two points.
My Lord, I'm now turning to a new topic which is
bugs, and I mentioned this morning that we obviously
covered the ground with Dr Worden on bugs 1 to 10 in
cross-examination to try and give a sufficient number of
bugs and a grouping of bugs that was not arbitrary or
cherry-picked to try and give the court a reasonably
representative sample by which to judge the reliability
of the analysis urged on the court by Dr Worden.
We explained those in our closing submissions at
paragraph 535 onwards. However, in Post Office's
closing submissions, notwithstanding that there are
really only three bugs touched on in cross-examination
where they had four days, there is now an analysis,
detailed submissions on every one of the 29 bugs in the
bugs table, and that's in their closing submissions at
appendix 2 and it runs from page 400 to page 537.
{A/6/400}
Given the time since receiving that document and the
time available today, I hope your Lordship will forgive
me for taking the next five bugs, 11 to 15, and
identifying whether or not Post Office submissions in
this new insight are reliable at all.
I respectfully invite your Lordship to note as we go
through two features. The submission that is made where
there is a document I will show your Lordship frequently
is simply not borne out, particularly where a gloss is
put onto things to suggest user error when, as we saw
with the Romec PEAK before Pat Carroll's final sign-off,
cannot preclude user error. It is very different from
reaching a conclusion that it was.
We will show your Lordship why that's also
a particularly unreliable conclusion for other reasons.
So what use they make when there are documents, and the
second point is where we just have assertions not
footnoted and not tethered to evidence the court
actually has before it. And those are the two
particular features.
Now, in Post Office's treatment of the bugs, they
are sorted out into different categories, and I will
just get your Lordship's eye on it, if I may. {A/6/400}
is the beginning of this 137-page appendix of the 29
bugs.
Where they break down is that the submission now is
that eight are not bugs at all. That's paragraph 3.1 on
page 400. So eight are said to be not bugs at all.
Three had no branch impact. Nine had, or potentially
had only transient impact, and nine caused or had the
potential to cause lasting impact but were resolved by
Post Office and Fujitsu.
They then list: paragraph 4 are the ones that are
not bugs at all; paragraph 5, no impact; paragraph 6,
transient impact; paragraph 7, lasting. {A/6/401}
My Lord, can I just identify even at that stage that
Post Office has listed nine bugs having a transient
impact.
MR JUSTICE FRASER: They are the ones in paragraph 6,
I think.
MR GREEN: Exactly. And nine bugs in paragraph 7.
So pausing there, on a correct construction of
Issue 1 there are 18 bugs meeting the definition of
Issue 1 even on the Post Office analysis in its closing
submissions here.
I'm going to take your Lordship, if I may, just
briefly through bugs 11 to 15. I have to deal with bugs
11 to 12. Bugs 13 and 14 were accepted by Dr Worden,
but bug 13 Post Office has changed its mind on. So I'm
going to have to deal briefly with 13.
MR JUSTICE FRASER: Which ones are you dealing with?
MR GREEN: So 11 and 12 weren't agreed by Dr Worden anyway.
MR JUSTICE FRASER: Yes.
MR GREEN: 13 I still have to deal with because although it
was agreed by Dr Worden, it is not anymore by
Post Office.
MR JUSTICE FRASER: 13?
MR GREEN: Then 14 was accepted by Dr Worden and is still
accepted, so I won't trouble your Lordship with that,
but 15 is still in dispute. So it is 11, 12, 13,
because of the change of position, and 15.
So on number 11, so this is one said to have no
branch impact, we see that at {A/6/401}, paragraph 6.
And Post Office describe the relevant PEAK --
MR JUSTICE FRASER: This is Girobank, yes?
MR GREEN: Sorry?
MR JUSTICE FRASER: Girobank?
MR GREEN: My Lord, yes. Girobank discrepancies, exactly.
MR JUSTICE FRASER: Okay.
MR GREEN: At page 452, paragraph 9, {A/6/452} your Lordship
will see the Issue 1 heading under which Post Office
analyses what happens as follows.
Paragraph 10 identifies the main PEAK for Girobank
noticing there was a £505.72 discrepancy, and says at
(1):
"This was a known issue dealt with by KEL
MWright531p. This KEL is now deleted and irretrievable,
but details about it can be gleaned from its associated
Peaks."
Then 10(2):
"The issue arose when a giro transaction was entered
and then reversed, with the reversal being entered after
the report cut-off time."
So reversal was not included in the following day's
report.
Paragraph 10(3), rather importantly:
"This led to an error notice being issued on the
mistaken basis that the branch had a discrepancy."
So pausing there, Post Office actually themselves
set this out that the way this arose was because
an error notice was issued on a mistaken basis in
relation to the original 505.72 because of the fact that
the post-cut-off time problem meant the following day's
report didn't capture the reversal.
Paragraph 10(4) says:
"The fact that the reversal, performed after the
daily cut-off, did not show on that day's report
reflects the intended operation of Horizon.
Subpostmasters were instructed that if a reversal is
carried out to giro transactions after cut-off, a manual
summary will need to be produced for Girobank. Issue 1
is therefore not a 'bug'."
Now, obviously how well they were instructed to do
that and all that sort of thing, whether they were at
all and in what terms, is at large. But the suggestion
there is the way that the system worked in this respect,
creating this mistaken footing for an error notice, is
not a bug, a defect or error in the data because that's
the full definition in Issue 1. And we respectfully
disagree with that. But that's what's said. So not
a bug, conflating all three into one.
Then paragraph 10(5):
"Rather, Issue 1 relates to reporting. The
underlying data is correct and the branch's accounts
would have been correct at the end of the trading
period, once the reversal had been recognised (at the
time of this Peak, the trading period was weekly).
Mr Coyne appeared to accept this in cross-examination.
In this particular case, the only possible impact would
be if the branch had accepted the error notice received
because of the reporting issue."
Now, if we just move forward if we may now to
Issue 2, which is at paragraph 17 on {A/6/455}. This is
what's described as the secondary problem initially by
Mr Coyne.
At paragraph 17 your Lordship will see:
"Issue 2 was that an £81 giro deposit was included
on two consecutive daily reports. This is because the
transaction was entered onto Horizon in a precise (and
very small) window of time between two system calls
being undertaken, resulting in a duplication. The
overall branch position would still have been correct,
but the daily reports to Girobank may have been wrong.
If they were (ie if the same transaction was included on
two consecutive daily reports), it is expected that this
would have been spotted and a TC would not have been
issued to the branch."
So pausing there, that is a pretty astonishing
example of the one-way street.
So we have got the fact of an error notice being
mistakenly issued as the premise upon which the £81
problem is discovered. You have got the assertion that
this whole thing is not a bug when that plainly is
a bug. In fact, we say both are bugs or defects or
errors in the data, and obviously so. Post Office says
neither of them are and does it effectively on opposite
footings for 1 and 2.
We say that is strikingly a one-way approach to
resolving that issue. It is then helpful to look at the
underlying PEAK, which is at {F/25/1}.
If we begin on page 1. In the top light green box
there is a reference:
"04/05/00, 13:44 system error - giro bank said there
is a discrepancy on the giro figures."
Come down a bit, username "1ha001":
"Girobank have been in touch to say that there is
software problem as the figures are not correct. Daily
figures when totalled are more than the cash account
giro figures."
Then if we go down to the words just below the
"daily figures 85990.88."
MR JUSTICE FRASER: Sorry, where are you now?
MR GREEN: Halfway down the box, my Lord.
MR JUSTICE FRASER: The same box?
MR GREEN: Yes.
MR JUSTICE FRASER: "Daily figures 85990.88."
MR GREEN: Exactly, and if we come down below that:
"The pm has checked all dockets and all reversals
that may have been done and cannot find anything.
Therefore he would like this investigated further as an
error notice has now been provided and he does not want
this to happen again."
So this is the postmaster or postmistress
challenging the error notice having tried to check
everything and not being able to find any justification
for the error notice.
Then if we go to the bottom box, just underneath "F)
Response":
"This difference (£505.72) between the Cash Account
and the Daily reports is explained [in the MWright KEL].
There was a giro for this amount that was entered on the
13th Apr then reversed AFTER cutoff then re-entered
again and reversed again. The Daily report would have
shown the original £505.72 but the daily reports never
show reversals.
"It would be nice to close the call as known error,
however while investigating the message store I have
identified another problem ... there is a Giro Deposit
for £81 (1-17240) that is being calculated in TWO
consecutive cutoffs (18th AND 19th April). I have
attached the full message store as evidence, however the
error happens in message ..."
Then there is a typed number.
If we go over the page, please, {F/25/2}, in the top
box, in the second paragraph, three lines down on the
right:
"There are two separate calls --"
MR JUSTICE FRASER: Where are you looking now?
MR GREEN: The top box, my Lord. In the second paragraph in
the top box which begins "after further investigation".
MR JUSTICE FRASER: Yes, I have that.
MR GREEN: On the right-hand side, three lines down.
MR JUSTICE FRASER: I see, yes.
MR GREEN: "There are two separate calls to find the latest
messages, and this gives a very small window of
opportunity for another transaction to have been
registered (The £81 giro was entered at EXACTLY the same
time as the TideMark was generated). The chance of
having a transaction entered at the same time... "
MR JUSTICE FRASER: No, not entered at the same time,
entered at the time.
MR GREEN: I'm sorry:
"... entered at the time between the two calls AND
causes the SEQ number to be greater than the tidemark is
very small but real."
We then see in the next box it is passed to EPOS FP
to correct the problem caused.
MR JUSTICE FRASER: By the difference between the mark and
sequence attribute.
MR GREEN: Exactly, and there is a fix that should address
all cut-off reporting, not just Girobank reports.
So we can see there that at times it requires a code
fix.
It is not limited to Girobank, the problem, and it
is a small but real problem which has been uncovered
effectively by accident.
If one looks back at Post Office's closing at
paragraph 17 on page {A/6/455}, it effectively
acknowledges that there is a discrepancy caused. It
says it is expected that it be spotted before TC is
issued. But there's actually no evidential foundation
for this, particularly in circumstances where what we do
learn from the PEAK is that an error notice had been
issued by mistake in relation to the sum of over £500
through which the £81 mistake was discovered.
So it doesn't say it, and what evidence it does
provide tends to be, if anything, pointing in the
opposite direction to the assumption the court is
invited to make about a process which falls outside the
system as defined in the Horizon Issues.
Post Office's submissions in relation to this at
paragraph 39 {A/6/460} in their conclusion on page 460
effectively involve a criticism of Mr Coyne in this
respect.
They say:
"That analysis is incorrect. None of the Peaks
referred to by Mr Coyne demonstrate a direct financial
impact on branches; in most cases this is because the
issue affects reporting whilst the underlying data
remains unaffected."
And so what they are doing is they are confining
their data for the purpose of Issue 1 to what's actually
in the branch count on the face of Horizon without
really grappling with the word discrepancy of course, or
shortfall, which implies a comparison between one thing
and something else.
So it is wrong as a matter of principle and, we say,
and characterisation and fact.
My Lord, can I turn now to bug 12, please. That
starts at page {A/6/461}. It is the counter replacement
causing one-sided transaction and this is said to be
a bug with transient but not lasting impact.
Your Lordship has our submissions that that's more
than sufficient for Issue 1 and that Dr Worden's
approach to permanence is wrong. Post Office's closing
submissions at paragraph 3 under the nature of the issue
refer to the PEAK PC0058528, which is a replacement of
a counter's hard drive, and that replacement caused two
messages relating to an OBCS transaction to be
overwritten, resulting in a receipts and payments
mismatch. A transaction with a value of 167.12 was not
added to the cash account. So there is a discrepancy of
£167.12 on Post Office's own description.
Then paragraphs 4 to 10 set out technical detail
which is not referred to {A/6/462}. There is no
reference to where that's found in the evidence. And
paragraph 11 confirms the discrepancy of £167 and says
it would have been flagged to the SPM. So this is
rather like the would have evidence we had in Common
Issues. It invites an assumption of a uniform
favourable practice in Post Office's favour performed
faultlessly.
Then at paragraph 13 {F/6/463}:
"Information of the overwritten messages was passed
to MSU who created a BIMS report for Post Office and an
error notice would have been issued to hold the branch
harmless thereafter."
Against that background it is worth looking at the
PEAK at {F/77.1/1}, Mr Miletic rightly reminds me.
That's the reference to the discrepancy of £167.12 we
see on page 1. If we go over to page {F/77.1/2},
please, in the third box down underneath "F} Response",
halfway down in the third box:
"This is a single counter outlet,n and the counter
was replaed on 22nd and two messages were overwritten.
KEL JBallantyne5328R. The messages retrieved from the
mirror disc show that a transaction for product number
184 value £167.12 was overwritten. I have attached the
two sets of messages as evidence."
Then if we go over the page {F/77.1/3} we can see at
the top of that KEL, top of the page, the PEAK:
"Like the other cases this is a single counter
office which had its hard drive replaced due to problems
with it."
So we can tell there are other cases of this type.
Then if we go to the KEL that's mentioned there, which
is the Ballantyne 5328R KEL, which is at {F/421/1}, what
we actually find in the KEL is underneath "Solution -
ATOS" towards the bottom:
"To find the overwritten transactions for
reconciliation we need to look at the Ripostemirror
messagestore."
And we can see the number attribute prior to the
rogue data in the third line. So that is being done
there, as we can see on the right-hand side. Halfway
down is to tidy up the counter and analyse the content
of the messagestore:
"For a multi-counter outlet (MCO) need to retrieve
the messagestore from another counter, as well as the
affected counter ... transaction numbers for the
RiposteVersionString messages should reveal the original
transactions. When you have identified any missing
transactions attach the details to the PinICL and route
to MSU."
So what is not there is a reference to the BIMS
report there or error notices, or TC or an instruction
to do that in the KEL itself.
MR JUSTICE FRASER: And where are you comparing that with?
MR GREEN: What is suggested at the closing submissions is
paragraph 13 where they say:
"An error notice would have been issued to hold the
branch harmless thereafter."
MR JUSTICE FRASER: So your point is that hasn't come from
the KEL?
MR GREEN: No. If it is somewhere that our attention has
not been directed to by the way it has been presented,
there are plenty of other examples of points without
foundations in the underlying documents.
MR JUSTICE FRASER: For example, in paragraph 14, the final
line:
"A further change was made to stop Riposte writing
messages as it came online."
MR GREEN: I'm not quite sure where that is. We haven't had
a chance to search randomly. It may be in one of the
documents, but given the pressure of responding to
545 pages.
MR JUSTICE FRASER: All right.
MR GREEN: In that case, my Lord, it is right that the PEAK
says a final BIMS should be issued, but a KEL which is
passed across says pass to MSU.
MR DE GARR ROBINSON: Sorry, is my learned friend suggesting
that those things two are inconsistent with each other?
MR GREEN: Well, they are different and whether or not the
transaction corrections are actually issued is at large.
What is suggested to Mr Coyne on {Day17/98:9}, if we
have a look there -- if we look at line 9 on page 98.
MR JUSTICE FRASER: Where?
MR GREEN: You will see:
"Question: Here's what interests me, Mr Coyne.
What you are saying is -- let me do it this way.
I would suggest to you that on any fair and reasonable
reading what this PEAK demonstrates is, first of all,
that Fujitsu spotted that there was a failed recovery
situation?
"Answer: Yes.
"Question: Very reliably. One can reliably assume
that's going to happen, yes?
"Answer: Yes.
"Question: Looked into the underlying circumstances
at the branch at the time of the recovery. Again one
can reliably assume that's going to happen?
"Answer: Yes.
"Question: Then formed the view it was necessary to
work out what had happened on the ground in order to
know whether any discrepancy had been created or not,
yes?
"Answer: Yes.
"Question: Then sent through a BIMS to Post Office
to tell Post Office to reach out to the postmaster and
ask what actually happened on the ground?
"Answer: Yes.
"Question: And I further suggest to you, Mr Coyne,
that the reason why Fujitsu sent that BIMS and the
reason why Post Office received that BIMS, they don't
receive these documents in order to put them in a pile
in some warehouse and never look at them, they receive
them so that they can be acted upon?
"Answer: Yes.
"Question: And on any fair reading of the evidence,
it would be extraordinary in this case to assume that
having received that BIMS, Post Office would not have
reached out to the postmaster, ascertain what had
happened and sent a TC or not depending on the
postmaster's answer."
So that is the footing on which it is put to
Mr Coyne.
His answer at line 17:
"Answer: Yes, but this is quite clear, when you
read the heading "Recovery Failures", that it is seeking
to address Horizon Issue 4: to what extent has there
been the potential for errors in the data recorded in
Horizon?"
So the point Mr Coyne is also making in relation to
that is the need for a transaction correction creating
the discrepancy in the first place.
So where we end up in relation to this is the
financial impact is admitted by Post Office and so it is
sufficient in any event. But it is clear the court can
have no confidence in this not being a lasting
discrepancy because although there is a reference to
a BIMS being sent through, there is no evidence it was,
and there is no evidence about what action Post Office
took in relation to it. And that all falls outside the
system as defined in Horizon Issues, which is the reason
that there wasn't disclosure in relation to the TCs
beyond the overall numbers and so forth which we had.
So, my Lord, I would now like to move to bug 13
which is the one that is now said not to be a bug at
all. This is addressed at Post Office's closing
submissions at {A/6/466}.
Post Office dispute this is relevant to Issue 1
because, as they set out at paragraph 2, it is not a bug
at all, and it is curious for two reasons. Firstly, it
is in Dr Worden's list of 12 bugs which had a financial
impact in joint 2 at paragraph 112, the reference for
which -- we needn't go to it -- is {D1/2/27}.
The second reason becomes apparent when we look at
it more closely. Post Office's submissions at
paragraphs 528 on page 467 describe the initial issue in
the PEAK, and essentially what happens is Post Office
withdraws stock, namely a £5 saving stamp. The SPM
returns the stamp as required but does not rem them out.
This leads to a £685 discrepancy which paragraph 6
describes -- and this is rather important -- as "pure
user error" {A/6/467}.
It is said at paragraph 8 that:
"The [SPM] elected to make good the shortfall and
a credit transaction correction was subsequently issued
for.£685 to rectify the issue."
Then at paragraph 9:
"However, a bug in Horizon ..."
Bearing in mind that this is not a bug apparently:
" ... caused the £685 of stamps to be subsequently
reintroduced into the branch’s accounts on two
occasions. By this point, Horizon was showing that the
branch was holding £1,370 of the stamps."
So there's, at the very lowest, something of a
tension in Post Office's analysis of whether this is a
bug, error or defect as defined in Horizon Issue 1.
They say not. And that's the basis upon which they
departed from Dr Worden's approach.
Paragraph 10 is also important because there's
another unrelated trading issue that was said to have
likely caused the SPM not to notice the first instance
of the withdrawn stock being introduced.
That's obviously unhelpful for the countermeasures
that Dr Worden relies on if there are any difficulties
in accounts not being able to immediately identify, as
Post Office acknowledges you may not be able to, the
cause of Horizon system-generated problems.
If one then looks at the actual PEAK itself, which
is at {F/765/1}, it doesn't actually support the
submission that is made that this was pure user error.
And we see that at -- we start on page 1 of 765. We can
see under "user names" in that top box, go down just
over halfway down:
"User names - SK1001 and PCA001."
It says under that:
"This office physically held 137 £5 PO saving stamps
..."
And this is really important:
"... and did not rem them out before the date the
rem out icon disappeared. The office physically
returned the stamps to Transaction Processing as advised
and the office then did a Trading Period balance ..."
So pausing there. We do not get from the PEAK
itself the fact that the SPM failed to do what they were
advised to do in terms of failing to rem them out. What
we do get is that they returned the stamps to
transaction processing as advised.
We can see towards the bottom of that box, there are
two two-line paragraphs:
"I have spoken to Phil Herrett in Transaction
Processing who has confirmed he is aware of about
8 offices with similar issues with the stamps still
showing on the stock."
Gives an example:
"Can this be investigated to see why the stock is
still showing in the office, and if this keeps giving
the office a loss of £685 every time they do a Trading
Period balance."
Then what we then find is over the page {F/765/2}:
"This is an example of the problem described in KEL
PothapragadaC4913L."
Underneath that:
"The change to prevent the withdrawn product being
put back into stock has been rolled out across the
estate so there should be no further new occurrences of
this problem."
We can see in the rest of the PEAK, if we go over to
page {F/765/4}, just over halfway down in the yellow
box:
"I have spoken to Gareth Jenkins ref this, he is
going to find some time to go through the issue with me
this week."
Then Mr Charlton, the penultimate green box, halfway
down that:
"... also checking to see how many other offices are
affected by this issue as there may be some who have not
reported the problem."
Which we say is realistic.
Then there is a workaround over the page at the top
{F/765/5}.
MR JUSTICE FRASER: We are still in the PEAK, are we?
MR GREEN: We are still in the PEAK, my Lord, yes. In the
top yellow box:
"To continue investigation into the root cause of
this issue I need to request some information from the
audit team ... continue monitoring the progress ..."
If we go to the end, {F/765/6}, three yellow boxes
down, the audit data that Mr Charlton had been supplied
with was blank. So he is trying to get another disk
with the required data on it, and the closure,
notwithstanding that had there is a fix rolled out to
cover that, is:
"Development [confirms] a refdata fix will be
delivered to prevent any further occurrences of this
problem.
"As NBSC have a circumvention for the issue, a KEL
covers the scenario and a fix is pending I am closing
this call.
"... Category 95 ... Advice after investigation."
Which does not capture the reference data error
problem that was underlined in the system which was
corrected by the fix.
MR JUSTICE FRASER: But that KEL is the PothapragadaC4913L,
is it?
MR GREEN: That is the KEL being referred to, yes.
MR JUSTICE FRASER: Do you have a reference for that KEL?
MR GREEN: My Lord, I don't off the top of my head.
MR JUSTICE FRASER: I don't expect you --
MR DE GARR ROBINSON: It may be {F/678/1}. I'm just going
from the appendix. If I have got that wrong let me
apologise in advance.
MR JUSTICE FRASER: I'm looking at the appendix, but in
paragraph 12 it mentions that KEL but it doesn't have a
footnote, so ...
MR DE GARR ROBINSON: I'm looking at paragraph 121.2 on
page 466 and there is a footnote to the
PothapragadaC4913L.
MR JUSTICE FRASER: F/678, thank you very much. Yes, thank
you very much, Mr de Garr Robinson.
MR GREEN: My Lord, could I take your Lordship back, with
that background, to what's said about this in the
closing submissions at 466 and 467.
MR JUSTICE FRASER: Yes.
MR GREEN: It is {A/6/466}. So we have got the contention
that it is not a bug at all at paragraph 2, and over the
page, {A/6/467} it says at paragraph 5:
"Subpostmasters would have been instructed to rem
out any excess stock ..."
MR JUSTICE FRASER: Yes.
MR GREEN: That appears to be the foundation for the
assertion that this was pure user error in paragraph 6.
Then paragraph 9 openly recites that a bug in
Horizon caused £685 of stamps to be subsequently
reintroduced.
MR JUSTICE FRASER: Where are you looking now?
MR GREEN: Paragraph 9, my Lord.
MR JUSTICE FRASER: Yes.
MR GREEN: A bug in Horizon caused £685 to be subsequently
reintroduced, which we respectfully say is just
hopelessly irreconcilable with saying "this isn't
a bug".
MR JUSTICE FRASER: I think I have got that point.
MR GREEN: I'm grateful.
MR JUSTICE FRASER: I think you did make that point a little
earlier.
MR GREEN: I'm grateful. Two layers --
MR JUSTICE FRASER: I do generally get points first time
round.
MR GREEN: Noted.
Bug 14 we don't need to deal with because that's
agreed by Dr Worden and Post Office haven't changed
their mind about it. Bug 15 is phantom transactions
which is picked up at 473.
MR JUSTICE FRASER: Do you want to have a break for the
shorthand writers before you go on to bug 15?
MR GREEN: My Lord, I think I was asked to go a bit further
last week before stopping.
MR JUSTICE FRASER: That's fine.
MR GREEN: If that's all right.
We pick it up at 473 {A/6/473}, which is phantom
transactions. At paragraph 2:
"Post Office submits that this is not a bug at all.
Manifestations of this alleged bug are either design
features of Horizon or user error."
The footing for that, the key aspects of the
Post Office analysis are paragraph 8 on page {A/6/474}
where they criticise Mr Coyne for failing to refer to
the other two PEAKs, both of which state the issues at
Old Isleworth were attributable to user error.
As your Lordship has already seen in relation to
that PEAK that is the Romec one and Pat Carroll.
MR JUSTICE FRASER: Yes.
MR GREEN: That's a bold gloss to put on that PEAK and
a pretty bold departure point for criticising Mr Coyne.
Paragraph 11:
"It is not possible from the PEAK to know what the
Romec engineer saw. Indeed there are a host of possible
explanations ..."
Then Post Office's closing submissions:
"Mr Carroll ultimately determined that there was not
a fault in Horizon." {A/6/476}
Which is the not so as to preclude user error point.
Then 477, paragraph 16 contends that:
"Mr Coyne gives a misleading impression of what the
Peak actually says and ignores the later content of Peak
PC0062561 and indeed the conclusions of the Peak.
Mr Coyne fails to mention at all the conclusion to Peak
PC0065021 that all reported cases are attributable to
user error."
So that is a premise for criticising Mr Coyne for
not acknowledging that conclusion: that all reported
cases are attributable to user error.
I'm not going to take your Lordship back to the
Romec/Pat Carroll PEAK, so I think we can take that
reasonably hopefully as read, but that's {F/97/1}.
That is where Pat Carroll on page 7 obviously has
himself observed the problems overnight and Romec have
separately observed the problems when visiting, and then
it is closed in a way which we say is thoroughly
unsatisfactory.
Mrs Van Den Bogerd was cross-examined on this
{Day5/40:3} to line 14. She didn't seek to suggest in
her evidence that the Romec engineer was likely to have
been in error in some way, notwithstanding her
experience of dealing with postmasters herself,
subpostmasters herself in her previous roles.
The Post Office says the user error was the
conclusion in the other PEAKs, and so we need to look at
{F/100.1/2} which is the conclusion which is PEAK
0068327.
On page 2 there, it says:
"Following a significant amount of monitoring we
have been unable to definitively link any
equipment/environmental issues to any particular event.
There have been incidents which showed a possible
correlation between system activity and phantom
[transactions], these pointed to a touch screen problem
and as a result the screen was replaced with a resistive
model. As this produced no measurable improvement it
has to be assumed that the problems were user related."
So there's no definitive determination, and the
basis that Pat Carroll says it has to be assumed
problems were user related is the screen has been
changed.
Let's go to the other PEAK, which is at {F/88.2/2}
and then perhaps take a break.
If we look at page 2 of that PEAK we can see the
bottom or penultimate green box, the large green box,
17th August 2001, 9.13:
"This outlet has reported continual phantom
transaction problems causing us to exhaust every
possible course of action in trying to solve them."
We have, and then there is a list of what they have
done. Then below "holiday":
"After all this the PM is still experiencing Phantom
transactions but they are mainly on counter positon 1
and this is always used by Robert Parker (PM). I have
asked Robert if I can spend some time at the outlet with
him so I can be present when the phantoms occur but he
is not keen for this to take place as he feels the
outlet is too small and gets too heated as it is.
"I spoke with HSH this morning and she advises that
since power help was last archived, Mr Parker has logged
34 calls to the helpdesk and a vast amount are advice
and guidance. My personal feeling is that Mr Parker
could do with some further training and I feel that this
should be our next course of action. The only other
option we have open to us is to change the ISDN line
which is the old style, but myself and HSH feel that
this is an expensive option to go down when it may be
user error at fault."
Then over the page {F/88.2/3}, the yellow box, third
line:
"From RNM," the regional network manager:
"I spoke to training and Dev this afternoon and
arranged 2 days training for next week, when I rang
Mr Parker he told me that he did not need the extra
training so I have now cancelled it. He also told me
that the phantom transactions have stopped.
"PON to RNM: 'There seems to be no issues at this
outlet if you are happy with the postmasters response.
"Is there anything else that needs investigating at
the outlet proven to be directly liked with phantom
[transactions] ... as there are none recorded? If not I
would like your agreement to close down this problem as
now resolved. I would like to make you aware though
that the postmaster does seem to be making quite a few
calls still to the HSH helpdesk, mainly around simple
things such as reversals.
"RNM to PON: Thanks for making aware about the
number of calls your still receiving, I don't think we
will ever stop him from making these. I see no reason
why this call cannot not be closed. As I said the
Postmaster said he is no longer getting these
transactions. Calls have actually reduced in September,
there are ... only 4. I have agreed with PON that there
is little else which can be done. The PM is not making
errors with his work and the call volume has improved.
I have agreed to close the problem."
So what we get is it may be user error at fault and
the problems stopped, and that's the point at which they
decide to close it down.
If we then go forward to the KEL that relates to
these, which I can do in about one minute:
It is {F/174/1}. It is a one-page document:
"Symptoms.
"There have been several calls over the last few
months where postmasters have reported phantom sales.
Items appear by themselves for which the PM has not
pressed an icon. These may be individual items or
several of the same item. Sometimes when no one has
been near the screen items may appear."
Then under the "Problem", the last line of the
problem:
"A more recent case revealed that the cable between
the screen and the base unit was the root cause."
That rather undermines the reliability of
a conclusion that Pat Carroll thought was the only
available assumption, that once you change the screen
and the problems continue, you have to assume it is user
error. Because it turned out, as we can see from the
KEL, that in one of the cases it was the cable that was
the problem.
My Lord, that's the end of the piece in relation to
that. We respectfully say Post Office is putting
a massive gloss on the underlying documents in asserting
what they are asserting, and it is a thoroughly
unreliable account of that bug.
MR JUSTICE FRASER: Thank you very much. We will have
10 minutes.
(3.21 pm)
(A short break)
(3.31 pm)
MR GREEN: My Lord, I now propose to take a particular
example issue and just trace it through, if I may.
I would like to start, please, with {F/908/1} which
is an internal Fujitsu/Post Office document of
22nd March 2012, which is a review of various computer
weekly articles. And this shows that it appears I think
to have been authored by Fujitsu and this goes to the
identification of issues certainly by this date which
are still live in these proceedings.
Your Lordship will see at the top, first paragraph:
"A number of articles have been written by Computer
Weekly relating to the Horizon System and the issues
postmasters have had with deficits. The main article
was published in May 2009 and can be found in
Appendix 1."
There is a summary, it highlights seven case studies
where they claim faults.
If we go forward to page {F/908/3}, we can see that
the second case study is that of Jo Hamilton, and we can
see in the key points box on the right, second bullet
point:
"Postmasters claim faults with the technology are
generating unexplained losses."
Third bullet:
"Post Office denies IT fault could cause accounting
system to show incorrect balances."
So this is the Computer Weekly article of
11th May 2009, as we can see from the top.
If we go over the page to {F/908/4}, we see a brief
reference to Jo Hamilton's case being ultimately:
"... signing her accounts even when she knew they
were wrong, because, she says, calls to the Horizon
helpline didn't stop the deficits occurring and she felt
backed into a corner. She was convicted of false
accounting, but was spared a prison sentence after local
villagers organised a collection to pay the debt."
If we go forward to page {F/908/6}, the relevant
case study is at the foot of page 6. Her name appears,
Jo Hamilton, at the bottom of page 6 and over the page
to page 7.
The specific point that I just want to trace through
in the evidence is the idea of problems with sums
doubling, because it was one of the things that was
certainly known to Post Office to be an issue.
As we see in Jo Hamilton's case {F/908/7}:
"'One time it said I was down £2,000, so I rang the
Horizon helpdesk. The supervisor told me to do various
things, and three minutes later I was £4,000 down.
Whatever I did after that, I couldn't get it to come up
any different,' she says."
So that is the Computer Weekly report of what
Jo Hamilton had said there. So we know that doubling
had been identified certainly as an issue, although it
sounds very implausible when you first hear about it
perhaps.
If we then go forward to {F/930/1}. I should
mention that I think the correct date on that
document --
MR JUSTICE FRASER: By that document, you mean 930?
MR GREEN: The first one is 2012 at 9.08. It is in Opus
I think as 2010, but it bears clearly a date of 2012.
So if we go back to 17th May 2010, this is a pack
that was prepared in advance of a meeting with
James Arburthnot and Oliver Letwin, both MPs who had
affected constituents, and it is a plan for what was
going to be said. And we can see the agenda on page
{F/930/2}.
There is a reference to "Review Jo Hamilton case" at
6a, and over the page on page {F/930/3} "Introductions",
third bullet point:
"We understand you have raised some concerns, and
are representing the concerns of subpostmasters in your
areas.
"We are open to feedback and we will provide you the
information we have available, our aim is to be open and
transparent."
Then over the page {F/930/4} at item 3 "Horizon -
background", second bullet point:
"Although we recognise that Horizon is not perfect,
no computer systems is, it has been audited by internal
and external teams, it has also been tested in the
courts and no evidence of problems found (of the nature
suggested by the JFSA) ..."
Third bullet point second line:
"Both versions of Horizon were built on the same
principles of reliability and integrity."
Then if we come down to 6a {F/930/5} on the next
page, there is a summary of the "Review Jo Hamilton
case":
"Cash holdings ...
"Audit findings.
"She was in personal financial difficulties.
"She was provided an opportunity for an explanation.
"She did plead guilty to fraud."
It is described there.
If we go over the page, it says {F/930/6}:
"What are your thoughts on the meeting? Do you have
any areas of concern?
"We are considering commissioning an independent
audit as an assurance measure, but in light that there
is no evidence that there is a problem, we need to
determine if this is a good use of public money."
Then if we go forward to page {F/930/8} at
paragraph 2:
"What is our view of Computer Weekly."
They are obviously respected.
Third paragraph:
"As we have external and internal experts available
we don't believe Computer Weekly can assist us in this
specific case. Although there is no evidence of
problems with Horizon ..."
This is in 2010:
" ... Horizon, as an assurance exercise we are
considering an audit of our processes, data, and IT
systems. If we do proceed with this audit, it is likely
that we will use a professional audit organisation that
are tried and tested in this area as we require
organisation who have the reputation and the experience
of defending audits against external scrutiny."
Then in paragraph 3, the second paragraph:
"In cases where an auditor has found evidence of
fraud, the previous trial balance (which the
sub-postmaster has approved) will be the baseline
record."
Then in paragraph 5 {F/930/9}:
"Why are we considering Deloittes to perform the
audit?"
What's said there is:
"KPMG are excluded as they are Fujitsu's auditor.
"Ernst & Young are excluded as they are
Post Office's auditor."
So that was the view taken at the time, which gives
further background to the Post Office reliance now on
the ISAE service audits which conflict with the APPSUP
PEAK.
The audit at paragraph 6 is:
"The audit envisioned is a thorough end-to-end
review of processes, systems and data which not only
could reveal potential improvements but could be used as
an assurance for court future cases."
Now, against that background there is an explanation
in a little bit more detail in relation to Horizon on
page {F/930/12}. There is really one key reference
there which is the third bullet point down in the
"Summary", which is:
"Each transaction is audited and protected to
prevent change or tampering."
Now, your Lordship may remember that Mr Dunks was
cross-examined in relation to that point. We find that
cross-examination by Mr Miletic at {Day7/32:18}.
Line 18 is the beginning of the section. He recites
what's said at paragraph 4 of Mr Dunks' witness
statement, then the question at line 24:
"I just want to be precise about the language there.
"When you say 'audited transaction records', you are
talking about transaction records generated from the
audit store; what you're not talking about -- it's not
the case that those records are actually audited prior
to going into the audit archive, correct?"
"Answer: No, it is just the extraction data that
I'm taking out."
"Question: Exactly. And so ..."
It goes on to page 33.
We referred to that in closing at paragraph 219,
my Lord, for your Lordship's note, page 92. I needn't
take you there.
We then get on page {F/930/16}, back in the
Post Office document, Jo Hamilton's case and the
timeline. And if we go on to the second page of that
{F/930/17} we have the entry on 5th May, and the third
bullet point:
"SPM provides a pre-prepared written statement. The
statement states that the SPM did not receive adequate
training and that the operation manuals provided were
out of date. Statement also makes reference to an error
for £1,500 which is alleged to have doubled to £3,000
when attempts were made to correct it."
So whatever the precise amount, Post Office knew
from -- it looks like 2006 but recognised and recorded
it in 2010 in this document, that the issue of
an apparent disputed discrepancy, having doubled when
help was sought, was one that was being complained about
on any view.
Now, it may initially appear that that seems to be,
without knowing about the case, somewhat implausible or
surprising if Horizon is working as the explanation in
the pack explained it should.
If we go, please, now to {F/1333/1}, which is the
Second Sight report. Second report. If we go, please,
to page {F/1333/26} of that report, my Lord, not for the
purposes of establishing the facts that are set out in
it but for identifying that it is an issue that has been
raised, we can see that under the helpline section at
paragraph 12, which makes various complaints, people
having difficulty contacting, script-based responses,
instructions later countermanded and then being told
don't worry, it will sort itself out, people not knowing
how long they should wait for that to happen, how they
are supposed to balance the books in an intervening
period, and so forth.
We can see at 12.4:
"Many of the shortfalls suffered by Applicants to
the Scheme have, on the balance of probabilities, been
attributed to 'errors made at the counter' but that does
not, in our view always mean that more extensive initial
training would have eliminated all of those errors
although it would obviously have helped."
12.5:
"What we have observed is that, in many instances,
the biggest shortages seem to have arisen as a result of
'errors made while trying to correct earlier errors'.
We attribute this less to inadequate initial training
than to inadequate subsequent support when branch staff,
when they were attempting to correct errors that they
had previously made, just made matters worse."
Then if we go over the page, 12.6, {F/1333/27} we
can see:
"There have been numerous references to shortages
doubling, trebling or even quadrupling as branch staff
tried to correct, under instruction from the Helpline,
errors that they had previously made."
Then there is a reference to improved error
repellency at the end of 12.7 and the extent to which
transcripts are available, if at all, at 12.8, and so
forth.
Now, against that background it is interesting to
identify the trouble that SPMs had with doubles. If we
start, if we may, at page {F/24/1}, which is PEAK
0043811. My Lord, I should say this is a slimmed down
selection of the available doubling problems.
If we look page {F/24/6} of that PEAK, if we may.
The second green box up from the bottom, 16th May, 16.11
exactly, the three lines up from the bottom halfway
across:
"The RNM," the regional network manager, "put
£6,343.07 into the suspense account. The discrepancy
has now doubled and is showing as a £12,686.14 surplus."
Over the page on page {F/24/8} there is a suggestion
that this is another instance of a different PEAK where
the data server trees have failed to build and now have
been fixed hopefully.
If we go forward now to {F/149/1}, we are now in
2003.
MR JUSTICE FRASER: What date was that PEAK we were just
looking at?
MR GREEN: I'm so sorry, I should have mentioned it,
my Lord.
MR JUSTICE FRASER: 2000?
MR GREEN: That's 2000.
MR JUSTICE FRASER: Where are we going now?
MR GREEN: 25th April 2003 at {F/149/1}. On the first page
of that, the first substantive paragraph there, 24/04/03
at 15.44:
"Darren from NBSC states that the BM ..."
I think it should be PM.
MR JUSTICE FRASER: I see, 24th April 2003. 15.44.
MR GREEN: That's it. The target date on the PEAK at the
top is 28th April 2003.
MR JUSTICE FRASER: That's rather what threw me. Yes.
Darren from NBSC.
MR GREEN: "... states that the PM," it should be, "is
trying to reverse a Rem, but when this has been reversed
it is doubling up on a balance snapshot."
If we go over the page to page {F/149/2} and we come
down to the word "contacted", about two-thirds of the
way down the left-hand margin just below 25/04/03,
09.15:
"Contacted: Pm confirmed all previous information.
Summarise. PM was in SU 'Y' he remmed in £13910 and
continued to trade in that SU in error, he should have
been in 'I' SU, so he did reversals of all transactions
in 'Y' including the REM IN and expected to see a zero,
but his REM IN had DOUBLED to £27,820.
"He went through all CA checks with NBSC and Horizon
and a reversal was attempted but a message that a
'reversal CANNOT be reversed' came up which indicated
that a reversal WAS CORRECTLY done."
So the system was reacting on the basis that you
can't reverse something you have already reversed, but
the effect of it was to double the figure to £27,820
rather than put it back to zero.
If we go forward to page {F/149/4}, the second last
yellow box, 1st May 2003, 06.34, Walter Wright.:
"Notes for testers:
"As well as testing that the fix addresses the
problem described in this PinICL, it should be verified
that the fix for the original bug ... Has not been
undone. Other desktop modes should also be checked to
ensure that no regression has taken place."
It appears that there was a fix developed for the
original version of this bug, but nonetheless this has
happened again and there is a further fix being dealt
with. We can see that appears to be right because on
page {F/149/5} we can see in the top green box, four
lines up from the bottom:
"... call type L ... Category 46 - Product Error
Fixed."
MR JUSTICE FRASER: That's also on 1st May, I think,
isn't it?
MR GREEN: 1st May 2003.
MR JUSTICE FRASER: They are both on 1st May, aren't they?
MR GREEN: Exactly.
MR JUSTICE FRASER: Okay.
MR GREEN: Then we move forward to January 2004 at
{F/184/1}.
MR JUSTICE FRASER: Where are we going now?
MR GREEN: {F/184/1}, which is PEAK 0098230.
This PEAK is opened on 13th January 2004 at
15:54:26 18 15:48:19. If we go over the page {F/184/2}, in the top
box six lines down, just under 13/01/2004 at 15.43:
"Information: The RLM has been through the cash
account with the PM and tried to adjust the figures but
they keep doubling up."
So this is not the postmaster or postmistress on
their own.
MR JUSTICE FRASER: Does that say RLM?
MR GREEN: I think it is meant to be RNM, unless it was
regional line manager at the time. I can't remember
when it changed.
MR JUSTICE FRASER: Yes.
MR GREEN: If we go down to the penultimate green box on the
same page, 14th January 2004, 17.55.56, underneath
halfway down:
"This results in a discrepancy between the system
cheque figure and the declared figure."
We get:
"Something has changed in the counter code recently
(I think at COUNTER_EPOSS 20_3; released end Nov) which
causes the discrepancy to be recorded wrongly; so the
cheque discrepancy; instead of being cleared; is
doubled; and the cash is also wrongly adjusted."
When we go forward to page {F/184/3},
15 January 2004 at 14.32.05, which is the large green
box beginning just before halfway down the page, we can
see there is a response. Then just underneath where the
second paragraph starts, just underneath "3rd December",
does your Lordship have:
"This is a nasty problem ..."?
MR JUSTICE FRASER: Yes.
MR GREEN: "This is a nasty problem as; if cheques continue
to be declared each week on the Declare Stock screen;
the numbers keep doubling. However there is a simple
circumvention and; as they should not be declaring
cheques in this way; it shouldn't have much impact. But
I think it should be looked at and resolved (any chance
for S60??) - might be instances where it really does
cause problems."
Then if we go to {F/207/1}, this is a different bit
of double trouble. It is the Horizon KEL GC Simpson
1049L. We are in April 2004 to May 2004.
Symptoms:
"The PM balanced on Wednesday ... and noticed that
all currencies on hand had doubled up."
MR JUSTICE FRASER: That is an 04 KEL, isn't it?
MR GREEN: This is an 04 KEL, my Lord, yes. It is the 29th.
It is raised on 29th April 2004 and updated on
5th May 2004 by Mr Simpson.
MR JUSTICE FRASER: It is just for some reason it has
"March 19" on the top right-hand corner. That's just
why I asked. All right. Yes.
MR GREEN: If we go to page {F/366/1} and we come down on
{F/366/1} to the penultimate yellow box. Cheryl Card.
24th November 2006, 14.56.15, penultimate yellow box:
"Problem appears to be related to smartpost products
... by 1pm Bulk/PrePaidBulk. Credit and Debit figures
appear to have doubled - message for product 7967,
SaleValue is 7.40 but Credit is 1480; message for
product 8058, SaleValue is -7.40 but Debit is again
1480.
"Problem has occurred 3 times - see also," some
related PEAKs.
If we go over the page to {F/366/2}. If we come
down to the third green box, 29th November 2006.
MR JUSTICE FRASER: At 10.26?
MR GREEN: 10.26, my Lord, yes:
"In the 3 cases seen so far each riposte message has
had the <Credit:> attribute written immediately before
after the <Mode:SC>> with the value doubled."
I'm not quite sure what that means:
"Normally the credit attribute is seen at the end of
the message. We have not as yet been able to reproduce
the fault, but will continue to try and reproduce the
fault."
So what they can see is what has been in fact
written and it is not as it should be written but they
can't reproduce the fault.
Then we see the PEAK appears to be a duplicate.
This is three yellow boxes up from the bottom. It
appears to be a duplicate of another PEAK. If we go on
to page {F/366/3}, that PEAK is then being used to
progress the fault. So there is an administrative
response, closure code for that.
Then if we go to {F/590/1} we are now in
7th March 2010. It is raised on 4th March 2010 with
a target date of the 7th. This is PEAK 0195561. If we
go to page {F/590/2} and we look at the penultimate
yellow box, 5th March 2010 at 16.58.10. Cheryl Card:
"On 02/03/10 on counter 2 at 15.04, the clerk
attempted a Transfer Out of 4,000.00 from stock unit BB
to MS. Due to a system problem the Transfer Out doubled
up, so when the Transfer In was done on counter 1 ... it
was for 8,000.00. The branch now has a loss of
4,000.00.
"I phoned the PM and explained that the problem was
under investigation. The PM would like to have it
sorted out before she rolls into the next TP, which is
due on Wed 17th March."
Then {F/590/3}, the third yellow box down in the
middle of the page:
"Advised PM to print a balance snapshot. Ran the
Transaction Correction tool."
MR JUSTICE FRASER: Sorry, where are you?
MR GREEN: Halfway down the page, my Lord. The yellow box,
11th March 2010, 15.22.14:
"Advised PM to print a balance snapshot. Ran the
Transaction Correction tool."
So this is the use of the transaction correction
tool and it is then successfully repaired, as we can see
at the bottom of page 3, penultimate box.
Then there's a little bit of insight into the
context in which those at SSC are working, which we find
on page {F/590/5} onwards.
18th March 2010, the penultimate green box. Second
paragraph:
"As the fix is already released, I would like to
request the priority of the issue to be downgraded to C
as we are trying to investigate the root cause only.
Please let me know your thoughts on this."
Then two boxes below:
"What is missing from this PEAK is the explanation
of the events in terms of the requests, how they were
ordered and when any was committed. Only then can we
qualify the priority. The assumption is that we have a
fix. The facts are -
"1: A settlement request to timed.
"2: A retry of request timeout occurred.
"3: According to the DB entries both later
succeeded.
"Now unlike other reconciliation Peaks this stands
out because only one of the requests are specific to
settlement. They should have worked because there would
have been a unique constraint violation on the journal
entry of one of them and if we are not getting this then
this is still an issue!"
So he says he can't reduce the priority {F/590/6}.
There is further investigation. The request is failing
in the large -- over the page, on page 6. So there is
still an issue is at the top. Then:
"We can't reduce the priority ..."
Then the big yellow box says:
"Yes, the second request is failing and the first
request is committing ..."
MR JUSTICE FRASER: Where?
MR GREEN: In the large yellow box, 18th March, 14.26,
my Lord.
Then over the page {F/590/7}, if we look at the
third yellow box down:
"Timeouts were the underlying cause of the issue and
that there were long delays waiting on the DB to process
the 4 requests. In this case two of the requests were
committed and two correctly detected that the
transaction had already succeeded. There is an issue
with the 2 commits because this shouldn't have
happened."
Then a bit further down:
"We would like to find the root cause of the issue
as to how the duplicate entry was committed in the db."
There is further investigation, checking the
relevant database tables on page {F/590/8}. Then we go
to page {F/590/9}. That's the large yellow box at the
bottom of the page, 22nd April 2010, 15.35.54:
"I have gone through the counter logs, OSR logs and
the DB dumps provided in the peak. Lets analyze this
from the scratch.
"Peak has been raised when a clerk attempted a
Transfer Out of 4000.00 from stock unit BB to MS. Due to
a system problem, the Transfer out doubled up, so when
the Transfer In was done on counter 1 at 16:15, it was
for 8,000.00. The branch now has a loss of 4,000.00.
"I checked the counter logs and analysed the
request ..."
Second line:
"But note that in the BAL/OSR side this request was
ignored by the time out monitor and continued to execute
and hence updated the table."
That seems to be the problem.
About halfway down that box there is a two-line
paragraph:
"I have requested for the journal table dump, to
check where duplicate JSN entries exists in the table.
But from the DB dump, I couldn't find any duplicates."
Then over the page {F/590/10}, top box, halfway
down:
"So, we can see that the insert time stamp is
different for these 2 records and hence it might have
entered from 2 different requests."
MR JUSTICE FRASER: Sorry, where are you looking?
MR GREEN: Halfway down the top box, the yellow box at the
top. Halfway down:
"So, we can see that the insert time stamp is
different for these 2 records and hence it might have
entered from 2 different requests.
"I have no doubt that one of the records was
inserted by the request id ..."
Then the number. It was an original request:
"But I am not sure how the second report was
inserted. But I have doubt on the retried request ...
which didn't fail at the journal filter stage.
"So, I would request you suggest on this, since
there wasn't any evidence which shows that 2
[transactions] has happened and updated the tables."
So at the bottom of the page on page 10, Cheryl Card
is not happy with the classification of duplicate call
which appears to be because there was another similar
PEAK which was not immediately spotted to be the same
problem, and she says:
"I'm sending this call back with Response Rejected.
"Closing a call as 'Duplicate Call' results in
a black mark against me. It basically means that I
should not have sent the call over since the same
problem has already been sent over in a previous call.
"... (duplicate transfer of 4,000 cash) may have
been caused by the same underlying fault as PC0194893
(banking reconciliation), however I could not have been
reasonably expected to link the 2 calls and take the
decision that it was not necessary to send PC0195561
over for further investigation."
So we can see from there that there was also another
PEAK, the link to which was not initially appreciated,
and there was some pressure on Cheryl Card not to
identify as a distinct problem something that might be
linked to something else.
Then we look at, if we can move forward to
23rd March 2010, which is at {F/596/1}.
MR JUSTICE FRASER: We are going into a different PEAK?
MR GREEN: A new PEAK which is PEAK 0196154.
Summary:
"... remmed in some currencies and the figures got
doubled up."
Then halfway down under the second lot of equals
lines:
"PM states he remmed in some currencies and the
figures got doubled up on the system - previous call
regarding this was closed. POL state this is a system
issue."
Then over the page {F/596/2} we can see the problem
with the sterling value of dollars. I think it should
be $2,000 with a value of £1,320 at the top, rather than
200.
MR JUSTICE FRASER: Yes.
MR GREEN: Then 2,104 matches the pouch value and then there
is the third paragraph. It says:
"Then got another slip printed showing a different
session number ...
"When she went to do her report to check the bureau
stock ... her system showed as having a value of
£4,208.84 in the bureau stock which is exactly double
the original amount she scanned ..."
Then there's a further analysis suggesting that the
pouch, if we look at the penultimate green box, third
paragraph, suggesting that it was accepted at the
branch, accepted twice at the branch:
"... due to a system problem when the clerk used the
Prev button several times during the Delivery acceptance
... POL may have to issue a TC."
There is another PEAK currently with development in
relation to that problem. So that is doubling in
relation to currencies.
Then at {F/630/1}, new PEAK, 019877, which is the
Hucclecote fix PEAK that your Lordship may remember from
the trial. We see this at {F/630/2}. This is where
there is an additional office, which is the Hucclecote
one which has had a problem in the yellow box at the
both bottom:
"NBSC has just advised that another office had
a similar problem, although the discrepancy has now been
sorted out. Details of the site and problem are below
for information ...
"Office - Hucclecote SPSO ...
"Office was dealing with the discrepancy in the
office following the TP rollover, and selected settle
centrally. The office reports that nothing happened and
they ended up doing this a further 2 times before they
could proceed. This has resulted in the office settling
the loss centrally 3 times. This showed as such as the
total on the final balance. The Trading Statement and
suspense account seemed to be correct though. On
Monday, 19th April the office reported they showed a
cash gain of double the original loss and after further
investigation a suspense account was produced that
showed 2 clear loss from local suspense entries. We
have now cleared this by clearing gain from local
suspense, which should clear the gain in the office."
Then it says {F/630/3} immediately under that, the
first green box at the top:
"The solution we thought we had for Hucclecote ...
has not resolved the problem, but has actually doubled
the discrepancy. The original figures in suspense were
clear loss from local suspense ... of £998.81 which was
the original loss in for the branch and this shows
twice. We have entered a clear gain from local suspense
but has doubled the discrepancy that was showing on AA
su from £1,997.62 to £3,995.24."
So that's one where they think they have fixed and
what in fact happens is it has not been fixed at all.
In fact, it has doubled the discrepancy.
Then if we go forward to May 2017.
MR JUSTICE FRASER: Different PEAK now.
MR GREEN: My Lord, it is a different document entirely. It
is {F/1794.1/1}, which is the agenda for the Post Office
operations board of 23rd May 2018. The point 1 refers
to the fact that 1794 was the redacted version and the
version we are going to look at is the version with less
redactions on.
MR JUSTICE FRASER: Fewer redactions.
MR GREEN: Fewer.
If we look at page {F/1794.1/34} of that document.
It says:
"There has been 3 reported instances."
This is SAP, GUI Bureau Value Issues.
2018.
"There has been 3 reported instances this week where
the sterling value of remittances of bureau to branches
has doubled when the delivery has been booked into the
branch, For example, Aylesbury GPO were sent a bureau
rem with a total value of £6.219,81, however when they
booked it into branch it populated as £12,439.62"
That passage was previously redacted but we can see
it in 1794.1 finally, if we go to {F/1857/1} which is
hot off the press. This is a document very properly
recently disclosed because it is recent and it is
referring to the date of an incident on 21 June, 2019.
It is surpassing the doubling problem with
a tripling problem, namely the triplication of
transaction acknowledgements into branches for Lottery
transactions in the order of 2 and a half million
pounds.
"TAs for lottery TAs (Transaction Acknowledgements)
for Lottery transactions taken on 20/6/2019 have been
triplicated in branch causing discrepancies.
"Accenture loaded the file 2 times in error, once
correctly."
That appears to be, my Lord, three times in total
because we get a triplication of the sums involved.
MR JUSTICE FRASER: Two times an error, once correctly.
MR GREEN: Exactly. It is believe this is due to the
introduction of new RPOS system into pilot.
MR JUSTICE FRASER: What's RPOS?
MR GREEN: I think it is a new bit of script but I am not
sure.
MR JUSTICE FRASER: All right.
MR GREEN: "All Lottery branches are unable to reconcile
their stock and cash. GL accounts in the Finance systems
have triplicated data. If a technical fix is not in
place then FSC will need to issue Transaction
Corrections (TCs) to enable branches to balance.
"If a technical fix is not in place and FSC do not
issue TCs, then emergency suspense accounting will cause
significant issues operationally for FSC to manage.
"Communication sent to branches for the issues and
timeframe to resolve:
"ATOS/FUJITSU/ACCENTURE is engaged in finding the
root cause and developing a fix. Conference call
arranged to discuss and if a fix is not in place TCs
will be issued from 25/06/19."
Then there is a note which we have asked in
correspondence for an explanation of because we only got
this very recently. Which says:
"Please note:
"Materiality - the threshold would be anything with
a financial impact of over £50,000 and/or a major
adverse reputational or regulatory reaction; the 'Daily
Mail test'. At this stage, it is more helpful to
overreport than to miss the opportunity."
We have enquired where that materiality filter is
applied, whether it is prior to this report being made
or whether it is who that gets sent onto. But the short
point, my Lord, is that there we have a picture of what
Post Office explained it knew in both 2010 and 2012,
where doubling was in issue in a very serious case,
Jo Hamilton's case --
MR DE GARR ROBINSON: My Lord, is my learned friend seeking
to -- inviting your Lordship to make findings about
a criminal case that's before the CCRC?.
MR GREEN: I'm not.
MR DE GARR ROBINSON: Then why does my learned friend keep
talking about Jo Hamilton? I do find it striking that
sometimes he appears to be addressing a jury rather than
a judge.
MR GREEN: My Lord, it may be uncomfortable for the
Post Office to be faced with documents that show they
well knew that there was an issue about doubling, but it
doesn't matter whose cases they arise in --
MR JUSTICE FRASER: What you have been doing in the last
however many minutes, and I didn't stop you with the
references to Jo Hamilton at the beginning, you used
that as a springboard to take me chronologically through
the sequence.
MR GREEN: Indeed.
MR JUSTICE FRASER: I have already made clear I think on at
least three or four occasions this court's position on
the criminal cases.
MR GREEN: Of course my Lord. My Lord, I'm absolutely not
trying to invite your Lordship to make any finding
whatsoever about merits one way or the other. The only
point I'm making is that this is a matter which
Second Sight had identified had been raised by multiple
SPOs in the Second Sight report and on any view was
known to be an issue by Post Office from the documents
we have seen. That's the only point and it is not right
for my learned friend to try to seek to distract from
the stinging nature of the underlying documents by
objecting.
MR DE GARR ROBINSON: My Lord, there are rules about the
conduct of commercial litigation.
MR JUSTICE FRASER: There are.
MR DE GARR ROBINSON: One of those rules is that one doesn't
say things incautiously that might have an impact on
evaluations being done in another place in relation to
different proceedings.
Another one of those rules is not to conflate a
whole series of issues, all of which are very different,
bureau rems in, we have got other kinds of rems, we have
got transfers between stock units, all sorts of
different issues which my learned friend picks up
randomly over a period of 20 years and seeks to promote
them to your Lordship as a single problem which
Post Office has known about for all that time.
Your Lordship will have seen from the PEAKs that
that's not the case. There were individual instances of
individual problems, as far as I can tell, some of these
documents I have not seen before of course, which itself
is extraordinary given the nature of this trial and that
we are at the end of it.
But my Lord what's happened is my learned friend is
seeking to jumble things up and then create
an impression. If I may say so, he has given the game
away in this particular exercise over the last 45
minutes and I would invite him to stop.
MR GREEN: My Lord, doubtless my learned friend can amplify
tomorrow. What I was about to say was specifically to
identify the fact that, on any view, Second Sight had
known and we have seen the other documents. This was
an issue and what we see in the PEAKs is not only the
situation where there is an attempted correction, where
there's doubling, but we see a variety of different
causes other. I was not going to conflate them at all.
I wanted the court to see we have found a wide range
of strands of causes of doubling issues in the accounts
of SPMs. We don't want to conflate them. We want to
identify that, not only are there cases where a SPM --
for example, the regional line manager puts it into
suspense and it doubles, Hucclecote -- your Lordship has
got the point.
MR JUSTICE FRASER: I have been following your analysis of
the PEAKs very carefully.
MR GREEN: I'm most grateful, my Lord.
MR JUSTICE FRASER: I think Mr de Garr Robinson's point is
that you are going rather wider than the confines of
this trial by referencing, firstly, criminal cases over
which I have made it clear I have no coverage and
jurisdiction and they are not part of the Horizon trial.
MR GREEN: My Lord, I agree.
MR JUSTICE FRASER: And he has also suggested that you are
giving a misleading impression by conflating different
technical issues all into one single issue as though it
started in 2000 and has run uninterrupted or unbroken
through to 2019.
MR GREEN: Yes and I'm not seeking to do that. What I'm
seeking to do is identify by way of one problem which
might be faced by SPMs without the insight that you get
from the PEAKs, as to: they have a problem where
a figure doubles in their accounts and they don't
understand why.
Not only is it correct that one of the problems
Second Sight reported about the number of people
complaining, which is when you seek help it doubles, not
only do we find that, we also find a whole load of other
reasons why figures might be doubling in their accounts.
MR JUSTICE FRASER: Understood.
MR GREEN: I don't want to conflate them at all. The short
point is, stepping well back from this, what we see as
an overall picture in relation to the documents the
court has was fairly reflected by Mr Coyne when he
identified that of course the documents we have got are
the ones that successfully got through, made it through
to the third layer, a PEAK was created and is then
ultimately followed through and so we have more
documentation where more investigation was done, less in
documentation where less investigation was done and we
can also see examples in the documents where
subpostmasters, even in the cases that do ultimately
reach a PEAK, subpostmasters either at their or at
intervention logs can be seen to be being bounced back
to NBSC.
My Lord, against that background we have answered
the Horizon Issues as we have sought to do in a matter
of a few words in our closing submissions. We have done
so, we respectfully submit, on a correct formulation of
those issues without any glosses which would distract
from the issues as they were properly formulated by the
court and the way in which those issues fall to be
answered by this court are totally unrecognisable from
the positions that Post Office has taken in relation to
them in a number of material respects, including into
the course of this very trial.
My Lord unless I have anything on which I can assist
you further, those are my submissions.
MR JUSTICE FRASER: Well, I have some questions. Just give
me a second.
Some of them are just minor sweep up points. There
have been various references to a figure for the
claimants' accounting losses generally as about
£18 million.
MR GREEN: My Lord, yes.
MR JUSTICE FRASER: And that is mentioned in one of the
expert statements. I don't know what the origin of that
figure is.
MR GREEN: My Lord it is totted up from the SOCIs, the
schedules of claimant information, that were identified.
MR JUSTICE FRASER: All right. If it is the arithmetic
total of that then I don't need any references.
You mentioned Mr Singh.
MR GREEN: Yes.
MR JUSTICE FRASER: What's your position on Mr Singh?
MR GREEN: The position that we didn't call Mr Singh --
MR JUSTICE FRASER: I am aware of that that is rather why
I'm asking you.
MR GREEN: The position on the evidence is that he said he
had this bizarre huge sum on a phantom transaction.
Mrs Van Den Bogerd says this could have happened,
I can't say it is possible or not possible.
MR JUSTICE FRASER: No, but you haven't called him as
a witness?
MR GREEN: We have not called him as a witness.
MR JUSTICE FRASER: His evidence is not admitted?
MR GREEN: His evidence is not admitted?
MR JUSTICE FRASER: And you showed me the notice to admit.
MR GREEN: And we showed the notice to submit and the basis
upon which it was refused.
MR JUSTICE FRASER: That was my understanding that you did
not call him.
MR GREEN: We did not call him and his evidence is not
before your Lordship.
MR JUSTICE FRASER: That's rather why I was asking the
question.
At the end of the oral evidence when I think the
question of disclosure was being discussed, you stood up
and from memory referred to a solicitor's letter which
you said you thought was 5th August of last year. But
you weren't sure. Now often when counsel says that they
know jolly well it is 5th August and they are
demonstrating their memory recall.
MR GREEN: I'm not sure I was on that occasion.
MR JUSTICE FRASER: But when I go to the H bundle and look
for a letter of 5th August I can't find it. I'm not
saying you have to give me a reference for it, but if
there is a reference for the document that you were
referring to, I would just like to know what it is
because it will save me some time. The authorities
bundle is an agreed bundle I think, yes?
MR GREEN: To which we haven't contributed.
MR JUSTICE FRASER: It is an agreed bundle of authorities?
MR GREEN: It is an agreed bundle.
MR JUSTICE FRASER: You have made various submissions about
Mr Jenkins.
MR GREEN: Indeed.
MR JUSTICE FRASER: And you have repeated them today or
expanded on them today.
MR GREEN: Indeed.
MR JUSTICE FRASER: I'm not inviting you to do this, I'm
just drawing your attention, there is a lot of law or
there's some law on how the court approaches matters if
it is being invited to draw an adverse inference from
someone's absence. I'm proceeding on the basis I'm not
invited to draw an adverse inference because none of the
law relating to it is in the agreed authorities bundle.
MR GREEN: My Lord, the position in relations to that is
reasonably well known but the only -- the way we have
put it is that it is unhelpful not to have called
Mr Jenkins and the reason for it doesn't seem
satisfactory.
MR JUSTICE FRASER: I understand that.
MR GREEN: We have said throughout to Post Office and
I think also to the court, that therefore the weight to
be attached to what's been passed down is very low
indeed.
MR JUSTICE FRASER: I understand that too. That is a rather
different point.
MR GREEN: We also respectfully submit that it feeds into
the impression of Post Office being less than
forthcoming about the true position in relation to --
MR JUSTICE FRASER: I understand that too.
MR GREEN: Beyond that we don't go further.
MR JUSTICE FRASER: That's also a different point.
The next two points really are practical, pragmatic
points, or the next one certainly is, for me which I am
going to raise now because they also affect
Mr de Garr Robinson's and his team's closing.
I happen to take the view that appendix 2 is very,
very useful.
MR DE GARR ROBINSON: Of?
MR JUSTICE FRASER: Of your closing submissions. It
collates a large amount of information in one place. It
is effectively a very useful narrative. You have,
however, made certain submissions this afternoon that
some passages of it, you say, aren't reflected in the
evidence.
MR GREEN: Indeed.
MR JUSTICE FRASER: Now, that may or may not be right and
I have no position on it one way or the other. But in
view of the number of bugs, the length of the appendix
and the quantity of the documents, it is a nettle I have
to grasp now about how it is going to be dealt with.
MR GREEN: Indeed.
MR JUSTICE FRASER: Just to put Mr de Garr Robinson's mind
at rest, I'm not going to ask for any sort of exercise
that has to be done before tomorrow and I'm not going to
ask for any exercise at all other than a purely
referencing one; but it seems to me that for my purposes
in writing the judgment I have to know which passages in
there you say are effectively submission rather than
evidence, so that the Post Office has an opportunity to
direct my attention to wherever it is in the evidence if
it is in the evidence or it might be a submission or it
might be a point of basic computing knowledge which one
is expected to have.
So I would like the two of you, please, just to
address your minds as to how you are going to go about
that. I envisage notification of paragraph numbers from
you to the Post Office where you say that, for example,
paragraph X on page Y, which isn't footnoted, is not
anywhere in the evidence and that then gives the
Post Office the opportunity simply to say the reference
is here or it is a submission.
MR GREEN: My Lord, our approach, given that where there is
a document it appears to be footnoted, where insertions
are made for which no reference is given, we have not
been able quickly to identify what the foundation for
that is.
MR JUSTICE FRASER: Mr Green, I'm not requesting or ordering
anything to be done particularly quickly.
MR GREEN: I'm grateful.
MR JUSTICE FRASER: It is simply -- and I want to make two
points absolutely crystal clear; this judgment is not
going to be out at any time in the next couple of weeks
for obvious reasons and we are going to address
housekeeping tomorrow.
MR GREEN: Indeed.
MR JUSTICE FRASER: When I say housekeeping I mean in terms
of the shape of the rest of 2019; not to conduct a CMC,
just to discuss the point I mentioned to both of you two
weeks ago, what are we going to do about the CMC in July
and the one in September, is it sensible to have either
and/or one of them, etc?
But, as you were taking me through the bugs which
you chose to address this afternoon, there were a number
of paragraphs which you said in relation to those issues
they are not in the evidence or we don't know where that
information is from.
If that, in respect of any particular paragraph, is
the case, then it is very unsatisfactory for me to be
left in a grey area where I don't know has it come from
the evidence or is it a submission? I just need to
know. It is very simple. It is a mechanical task. It
won't be very complicated and it doesn't need doing any
time in the next 7 days, but it does need doing.
MR GREEN: My Lord, yes. I mean, I understand. The basis
on which we have approached --
MR JUSTICE FRASER: I know you have explained that.
MR GREEN: So we should identify --
MR JUSTICE FRASER: I think as a matter of fairness to both
of you, you have to identify a paragraph that you say is
not taken from the evidence if you say it is not taken
froth evidence and then the Post Office has to have the
opportunity simply to give me a reference or say it is
a submission. It is not massively complicated.
MR GREEN: If they have got sources in mind for things --
MR JUSTICE FRASER: Mr Green, I'm not saying --
MR GREEN: But anyway I appreciate your Lordship says it is
a task that has to be done.
MR JUSTICE FRASER: It is a task that has to be done, it is
not designed to impose an enormously expensive or
onerous burden on the legal advisers and it doesn't
necessarily have to be done, if I can put it this way,
to the nth degree for every separate sentence in
a paragraph that might have ten sentences. But it does
have to be done. Because the only other way I'm going
to be in a position to know: is that paragraph based on
any evidence? is by going through the whole of the
evidence myself, which I will be doing but I won't be
doing separately in respect of each paragraph in
appendix 2.
MR GREEN: I see.
MR JUSTICE FRASER: It is fairly straightforward.
MR GREEN: I understand.
MR JUSTICE FRASER: I don't imagine -- when you have had
time to look at it, it might be more obvious that either
there are a list of however long it is or a much shorter
list, but it is not going to be -- and if the answer
from the Post Office is: well, yes you need to look at
this paragraph of Dr Worden 2 or you need to look at
this witness statement of Mr X or Ms Y, that's the only
response I need.
I'm not asking for extra submissions. I would just
like to know because at the moment you are saying it is
not based on the evidence. So just have a talk between
yourselves about the mechanics of that.
MR GREEN: I'm grateful.
MR JUSTICE FRASER: I think that's everything save for one
point. Mr de Garr Robinson, at the end of the afternoon
when there was the exchange between counsel about what
Mr Green was and wasn't doing in mentioning the criminal
cases, you said that one of the rules in commercial
litigation is that one doesn't say things incautiously
that might have an impact on evaluations being done in
proceedings elsewhere.
By that do I take it to mean in relation to
proceedings before the CCRC or are you talking about
something else?
MR DE GARR ROBINSON: I had the CCRC in mind my Lord but
I made the point advisedly as a wider application but my
objection that I made to your Lordship was with respect
to the CCRC.
MR JUSTICE FRASER: Because there are two issues and it is
something that I just need to address now with you
because of the way in which it has arisen in the last
half an hour.
Firstly, if there are any criminal proceedings
currently underway as of today, then, obviously, there
are potential jury issues in terms of any sort of
publication of today's proceedings. So that is a point
which seems to me does have some potential concern
because those sorts of proceedings could be influenced
by anything that was reported about this case.
I'm not sure the CCRC could because that's not
a jury environment and I am sure they would be
sufficiently trained to exclude from their
considerations anything that was reported about this
case rather than looking at the actual judgment itself.
But because of the fact the point has been raised
I wanted to raise it full square with you now.
MR DE GARR ROBINSON: I'm very much obliged to
your Lordship, could I take instructions very briefly?
MR JUSTICE FRASER: I think so.
(Pause).
MR DE GARR ROBINSON: I'm obliged to your Lordship for
giving me that moment. My Lord, there are no
prosecutions being brought by Post Office as we speak.
Post Office understands that there are
investigations being made by the police and possibly
other authorities which may lead to criminal
prosecutions but my instructing solicitors and indeed
a legal officer from the Post Office is not aware that
those investigations have actually materialised into
a prosecution yet. That's not to say that they are
certain there isn't one, but they are not aware of one.
MR JUSTICE FRASER: That is entirely understood. My
potential concern was, obviously, if there were
a criminal trial underway this week, then we may well
have to or would have needed to address whether or not
it was necessary to take any sort of measures to prevent
those Crown Court proceedings being potentially derailed
and having to be started again as a result of anything
that's happened in the last 40 minutes. But on the
basis of what you have said, I'm perfectly content and
unless either of you have anything to say about that?
No? Thank you very much.
Final point. When I say we are going to address the
shape of 2019 at the end of tomorrow, that's not
intended to be an enormous exercise. It is just a very
straightforward pragmatic analysis of how long this
judgment might take and the sequence of the litigation
following on from that and it may be that it is a short
point and it is decided we don't need the CMC in July,
we just have the one in September or something like that
but we can address that tomorrow when you have finished.
MR DE GARR ROBINSON: My Lord, yes.
MR JUSTICE FRASER: Anything else?
MR GREEN: My Lord, no.
MR JUSTICE FRASER: Thank you very much. 10.30 tomorrow.
(4.42 pm)
(The court adjourned until 10.30 am on Tuesday,
2nd July 2019)