This is the unperfected transcript of Day 22 - the final day of the Horizon trial.
Mr Anthony de Garr Robinson QC for the claimants presented his closing submission to the judge.
Mr Anthony de Garr Robinson QC for the claimants presented his closing submission to the judge.
Tuesday, 2nd July 2019
(10.30 am)
Closing submissions by MR DE GARR ROBINSON
MR DE GARR ROBINSON: My Lord, good morning. I'm going to
start by seeking to draw your Lordship's attention to
some extraordinary features of the claimants' case as it
has developed before your Lordship during the course of
this trial.
First of all, I would like to remind your Lordship
of some points of which you will be well aware, and if
I can do it by reference first of all to the Horizon
Issues which is at {C1/1/1}.
The first point is that your Lordship ordered
an expert-led trial and your Lordship will see that from
the header to the Horizon Issues. The header is taken
from remarks that your Lordship made at the CMC on
22nd February, and that's at {C8.4/4/1}. Perhaps we
could have a look at that.
Sometimes one does rather miss the old days when one
had files, my Lord.
MR JUSTICE FRASER: That's the document. Which page?
MR DE GARR ROBINSON: It is at page 54. I don't have
a document.
MR JUSTICE FRASER: There is one on the common screen.
MR DE GARR ROBINSON: I'm so sorry. The reason why I said
that rather intemperate, which I now regret, is that
I see nothing on my screen. I wonder if someone could
look at that for me?
MR JUSTICE FRASER: Have you not got a screen now?
MR DE GARR ROBINSON: It is a black screen, my Lord.
Perhaps it is not plugged in.
MR JUSTICE FRASER: Well, that's unsatisfactory for obvious
reasons. It is not the best start. I think you need
a screen, but I can easily rise for a minute or two to
avoid putting pressure on anyone.
MR DE GARR ROBINSON: I would be very grateful to
your Lordship. I do apologise.
MR JUSTICE FRASER: You don't have to apologise. These
things just happen. Whenever they happen they are
always at a very inconvenient time, but I suppose at
least we haven't got into the depths of the day. We may
as well sort it all out at the beginning. 5 minutes?
MR DE GARR ROBINSON: 5 minutes.
(10.36 am)
(A short break)
(10.38 am)
MR DE GARR ROBINSON: My Lord, I'm hoping we can continue at
approximately this pace for the rest of the day. It
will be a much more enjoyable day that way.
We are on the 22nd February hearing and
your Lordship will be familiar with these words but
I will, if I may, read them out. You said:
"It is very unusual in case management to find
oneself having constantly to try and put either one or
other party back on track for cost-effective resolution
of serious disputes. I reminded myself, again by
reference to the actual transcript, of what I said last
time, although I had a pretty clear recollection, that
what I was going to be doing in March was to deal with
expert issues that were present on the pleadings
concerning Horizon which I described generically as the
next big issue. I wanted the parties to agree or each
propose an isolated number of issues on the pleadings
related to Horizon that would involve expert evidence
but not evidence of individual cases."
{C8.4/4/54}
My Lord, it was with those words ringing in their
ears that the parties agreed the Horizon Issues and
offered them to your Lordship for approval.
Disclosure was then given, hundreds of thousands of
documents were disclosed and reviewed by the experts.
They were disclosure of documents of an expert nature.
The claimants criticised Post Office on their disclosure
and Post Office criticises the claimants on their
disclosure, but for present purposes the important point
is a huge amount of disclosure was given by Post Office;
it was directed at informing the expert process and many
of the documents, I would say most of the documents,
called for expert review and comment.
The next important stage of the litigation was vast
expert reports being served. Mr Coyne's two reports run
to well over 500 pages without appendices.
They were many, many, many, hundreds of times
I would say, more detailed than my learned friend's
outlined allegation document that was produced in August
which no one has really referred to, and they took weeks
properly to assimilate. It is fair to say that the
Post Office was still trying to assimilate Mr Coyne's
second report when the trial actually began.
Your Lordship may appreciate how that felt at the time.
Now, the experts co-operated to identify the
principal agreements and disagreements between them, and
that was, in my submission, a very helpful process that
was overseen and encouraged by your Lordship and it led
to four long joint statements which in my submission
were helpful.
JS2 is particularly helpful in that it sets out in
the bug table what the experts say on each side.
Without that we would be digging through the interstices
of these interminable reports. It brought a welcome
measure of focus, and given we don't have detailed
pleadings or schedules in relation to specific bugs it
gave the parties an opportunity to decide how to
organise their cross-examination and proceed.
Your Lordship will be aware that the parties'
cross-examination was limited. I certainly would have
liked more time given the scale of the reports. But
given the other factual evidence that had been called,
I had four days; my learned friend took three.
Now, the Horizon Issues are expert issues and we
have had an expert-led trial as your Lordship ordered.
But anyone who came to court for the first time
yesterday might have formed a very different impression
of what this case was all about.
One expects oral closings to be about the oral
evidence to explain how the case has developed since the
openings now that everybody is much the wiser. But in
my learned friend's submissions there was almost no
reference to Mr Coyne's oral evidence or to Mr Roll's
oral evidence and no reference at all to the important
respects in which that oral evidence changed the picture
as compared with Mr Coyne's reports on the one hand and
Mr Roll's witness statements on the other.
I don't mention the fact that there were these
changes as a criticism of Mr Coyne but merely as a point
about the unusual approach that the claimants have
adopted, I would suggest felt constrained to adopt, in
the final days of this expert-led trial. My learned
friend was constrained to make many of his submissions
largely without reference to the conclusions reached by
the expert whom his clients had instructed and largely
without reference to the very substantial common ground
that was between the experts, both as set out in the
joint statements and was achieved during the course of
the cross-examination and, indeed, your Lordship's own
questions.
Important parts of my learned friend's submissions
were directly contrary to that evidence and those
agreements. For example, your Lordship will have heard
my learned friend say that Horizon is not robust.
My Lord, that, as your Lordship will be well aware, is
directly contrary to the experts' common views.
My learned friend tries to duck that by effectively
suggesting that robustness doesn't really mean anything.
But, again, that's contrary to what the experts say.
And your Lordship will see that from our written closing
at paragraph 361; that's at page 132. I'm not going to
take your Lordship to it now.
Mr Green didn't engage with any of that evidence,
nor did he engage with Mr Coyne's oral evidence on the
likely total number of detected bugs over a 20-year
period. Your Lordship will I am sure have well in mind
that he said it was no more than 40 including transient
bugs. So if one takes into account that of the 29 bugs
in the bug table he says only 22 were lasting, then one
assumes, one infers, that his view is that the number of
lasting bugs that are in the system would be around 30
over 20 years.
Nor did he engage with Mr Coyne's oral evidence on
the number of instances of remote access that he had
seen over that same period. Your Lordship again will
I hope have in mind that Mr Coyne said he had seen about
30 instances of relevant remote access for the purposes
of these proceedings.
Those numbers loomed large in any worthwhile
consideration of the key extent questions raised by the
Horizon Issues. But they weren't addressed by my
learned friend. Instead he gave examples of things. He
told your Lordship little stories and he relied on
snippets from documents. But what he didn't do is he
didn't attempt to draw back and ask the question, after
all the evidence that your Lordship has heard: where do
we end up on the expert evidence that has been given in
this case?
In my respectful submission, the inevitable
inference is that the claimants have considered very
carefully where we do end up on the expert evidence and
their considered view is that they would prefer the
court to look elsewhere. They would prefer
your Lordship to take an impressionistic approach, and
by that I mean the kind of approach under which all bugs
that caused doubling up can be considered as one and the
same thing, as my learned friend remarkably appeared to
be suggesting last night.
That suggestion was apparently made -- and I may be
being unfair to him, but it was apparently made to
provide some colourable basis on which to call into
question the criminal prosecution. My Lord, that's all
headline-grabbing stuff but it is nothing to do with
Horizon Issues and it is nothing to do with the
expert-led trial that your Lordship ordered and on which
the parties have spent so much time, money and, if I may
say so, sweat.
Now, I propose to focus on where we are left after
the trial. Most notably, in terms of Mr Coyne's
evidence on Horizon Issues 1, 3, 4 and 6 and on
Mr Roll's oral evidence as well. And your Lordship will
see that my submissions, I should say our submissions,
the submissions of myself, Mr Henderson, Mr Draper and
Ms Keating are set out quite fully and at length, and
I'm sorry for the length of the submissions we have
submitted to your Lordship.
That brings me to another feature of the case which
I submit is also extraordinary. In their written
submissions, their closing, the claimants say, I think
more than once but they say it triumphantly, that
Mr Roll was right. What they mean is the Mr Roll that
was portrayed in the witness statement that he would
have had drafted for him was right. But, my Lord, in my
submission a very different Mr Roll came through when he
gave oral evidence, and just as the claimants make no
reference to Mr Coyne's oral evidence during the course
of this trial, they similarly seek to distract, if I may
say so, certainly pay no attention to, the oral evidence
that Mr Roll gave.
Just to remind ourselves quite how far we have come,
perhaps I could ask your Lordship to have a quick look
just at his first witness statements at {E1/7/1}.
Rather than reading out long chunks perhaps I could ask
your Lordship to read particular paragraphs; first of
all, paragraph 7 {E1/7/2}. Then the last sentence of
paragraph 8. The first sentence of paragraph 10. The
last sentence of paragraph 11. And then, my Lord,
finally paragraph 19 {E1/7/3}.
That's a brief selection, but they are what could be
described as headline-grabbing statements made by
Mr Roll.
In my respectful submission, your Lordship will have
heard Mr Roll and will have seen our written closings,
and from our written closings I respectfully submit that
your Lordship should conclude that all of those claims
have effectively gone. For your Lordship's note these
are addressed in pages 38 to 63 of our closing
submissions.
Here are some highlights. Mr Roll accepted that his
recollection of 70% of the work he did could be wrong
because one tends to remember non-mundane things more
clearly than mundane things.
As regards time pressure on the SSC, that was
a feeling from a long time ago and he said his
recollection was quite hazy and he also said that he had
that feeling not very often. He said that when he said
he spent 70% of his time looking for faults on data
stores he was not looking for software bugs. Remarkably
he accepted that coding errors causing financial impact
on branch accounts in his experience were extremely
rare. He accepted that paragraph 19 was wrong, that he
and other colleagues were not routinely working on
coding issues causing financial discrepancies, and even
more remarkably he couldn't recall ever working on
a coding issue which caused a financial impact on
a branch.
Be that as it may, in relation to bugs causing
branch shortfalls, his evidence has been, if I may say
so, over taken by the KELs and the PEAKs that have been
disclosed in this case and the in-depth analysis that
has been performed by the experts on those documents.
Your Lordship will be well aware of Mr Coyne's oral
evidence that he and his team, having read almost all
the KELs and having done innumerable intelligent
searches through all the KELs and all the PEAKs, his
judgment was there are not more than 40 branch-affecting
bugs over 20 years.
Similar points can be made about remote access.
Mr Coyne accepted that any work on branch data is
typically recorded in PEAKs, and for your Lordship's
note that's at paragraph 740 of our closing
{Day16/28:1}. Mr Coyne said that there had been no more
than 30 occasions of relevant remote access; that's
paragraph 746 of our closings. And Mr Coyne said that
the chances of remote access adversely affecting branch
accounts was small. My Lord, that's paragraph 748 of
our closings.
My Lord, all those points are consistent with
Mr Roll's oral evidence. But you would get no hint of
that anywhere in the claimants' oral or written
closings. That, in my submission, is a graphic
illustration of how the claimants would like to distract
the court from the oral evidence given by their two most
important witnesses. What they want to do is tell
little stories of individual incidents instead and to
give the court the impression that those individual
incidents are somehow symptomatic or representative of
a wider phenomenon of which there is no evidence
whatsoever.
My Lord, they would also like to stop the court from
deciding the most practically useful issue in this
entire issue, the Horizon Issue trial. And that is the
extent of the risk or likelihood of Horizon bugs causing
branch shortfalls for which SPMs are held liable.
They do this in a number of ingenious ways. Indeed,
they do it rather beautifully by accusing Post Office of
seeking to rewrite Horizon Issues 1 and 3. Let me deal
with those arguments very briefly.
Horizon Issue 1 first of all. They make three
assertions. First of all, the threshold for satisfying
the enquiry required by Horizon Issue 1 is merely that
there should be a potential of causing discrepancies or
shortfalls, the logic of their position being if some
potential is found, however slight, the court stops
there. And they follow up that submission by saying in
no circumstances, Post Office's attempt to draw
attention to the kind of bug that doesn't have a lasting
impact is an impermissible exercise. They say that
Post Office is introducing a false concept which isn't
written into the Horizon Issue itself.
Thirdly, they argue that the discrepancies and
shortfalls referred to in Issue 1 need only relate to
branch accounts, they don't need to be in branch
accounts.
As to these three arguments, potential first of all.
My Lord, it is important not to forget the words
"extent" and "likelihood" at the beginning of Issue 1.
I don't know if your Lordship has the issues with you?
To what extent was it possible or likely for bugs of the
nature alleged in the GPOC and referred to in the
generic defence to have the potential to cause apparent
or alleged discrepancies?
The essential enquiry is an enquiry as to extent of
likelihood, and I won't invite your Lordship to do it
but if you look at the assertions made in paragraphs 23
and 24 of the generic particulars of claim and the
relevant paragraphs of the generic defence, that is how
the case is put.
These are bugs that caused these things.
"Potential" is just a word that's emphasising that what
the court is doing is assessing extent of likelihood or
risk or potential; it just reinforces the essential
enquiry that the court is required to undertake.
Critically, the court is not required to stop if it
identifies some potential. It actually needs to assess
risk or likelihood in a sensible way and, my Lord,
your Lordship will be well aware that I submit that that
sensible way requires a sense of scale or perspective,
another thing that the claimants would rather not talk
about.
So, my Lord, that deals with potential. Let's now
deal with the so-called impermissible concept of
lasting. In my submission it is not impermissible, it
is fundamental, otherwise the determination called for
by Issue 1 would be practically pointless. I'm not
saying, and the Post Office doesn't submit, that whether
a transient discrepancy was caused is irrelevant. Of
course not. Issue 1(a) refers to apparent or alleged
discrepancies but it also refers to apparent or alleged
shortfalls. I ask forensically: what's a shortfall?
What is the difference between a discrepancy and the
shortfall and why was it necessary for the Horizon
Issues to distinguish between them?
A shortfall is that for which an SPM is held liable.
A discrepancy doesn't really matter at the end of the
day. I hope no one quotes me out of context. But if
your Lordship understands what I'm submitting to you, if
a discrepancy occurs during the course of a month, if it
doesn't actually result in a shortfall at the end of the
month and the postmaster doesn't therefore have to make
that shortfall good, that is not quite the same -- that
is much less important than the question whether
a liability is impermissibly foisted on a subpostmaster.
That is what Issue 1 was trying to get at.
So by all means your Lordship should consider
whether and to what extent there was a likelihood of
bugs causing transient impacts. I'm not suggesting you
should ignore them. But what I do suggest is, by the
same token, nor should your Lordship ignore whether
those bugs caused discrepancies that were caught by
countermeasures either in the short or medium or longer
term, or, on the other hand, were lasting discrepancies
that became shortfalls and that required a subpostmaster
to put his hand into his pocket.
That is why the word "shortfall" is contained in
Issue 1. And by the way, it is part of the overall
judgment of robustness that's called for pursuant to
Issue 3.
My Lord, the final point relating to branch
accounts. This is the most ingenious argument of them
all on Issue 1. In my submission it is sophistry. The
purpose I think, or I suspect, is to allow the claimants
to shoehorn into Issue 1 a consideration of whether bugs
in Post Office's back end systems and whether and to
what extent the bugs in those systems took place.
But there is a problem with that essential approach,
which is that Issue 1 is about bugs in Horizon. The
back end systems which may result in TCs being issued,
those back end systems are not part of the Horizon
system and the TC process was deliberately excluded from
the Horizon Issues because if there were to be a trial
of all Post Office's reconciliation processes over the
last 20 years that would involve millions of documents
and take a six-month trial.
Now, in my submission, the drafters of the Horizon
Issue, of whom Mr Green may well have been one at the
time, could never have dreamed that the investigation
called for by Horizon Issue 1 would be as stunted as the
claimants are now seeking to achieve.
That brings me to Issue 3. Here, two arguments are
being run for which there's no hint in their written
openings. There, they appear to accept that the Horizon
system was relatively robust. Indeed, their suggestion
appeared to be that they had always accepted that it was
relatively robust.
If we could look, please, at {A/1/10}. This is the
second half of paragraph 17.1 of the claimants' written
opening. We don't need to look at the first half. It
says at the top:
"This reflects language pleaded in the GDef, and
indeed 'robustness' has been one of Post Office's
'narrative boxes' and a favoured term in Post Office’s
public relations pronouncements ... Coincidentally or
otherwise, it has also featured in the NFSP's defence of
Post Office, relied upon by Mrs van den Bogerd.
However, as the Claimants made clear in their GReply,
whereas the Claimants'," I think it must be claimants'
case, "is that it is relatively robust and has become
more robust over time – but not so as to be an answer to
the Claim (and in so far as 'robustness' has, in this
case, a sufficiently clear meaning – addressed further
herein)."
So they are toying with whether robustness has
a meaning, but there they purport to say that relative
robustness was actually accepted in their reply. That's
always been their position.
That is the complete opposite of what the claimants
are now saying. They are now trying to ditch their own
expert whilst still purporting to laud him to the skies
in their written and oral submissions.
Their first argument is that Issue 3 raises one
point, not two: robust and extremely unlikely. It seems
to be an attempt to suggest that robustness can have no
meaning other than that. Well, my Lord, whether
robustness has a meaning and what meaning it has is not
in issue between the experts. In JS3 at paragraph 3.1
they both agree that it's relatively robust.
We have a meaning from Mr Coyne. He said
"relatively robust" means performing well compared to
similar big systems. And those similar big systems
require transactions to be handled properly in the
overwhelming proportion of cases, and that they require
the system to have lasting errors of a fraction of
a percentage.
My Lord, what that means is that when faced with any
given transaction, the generic likelihood of that
transaction being erroneous is extremely small. In
other words, it is extremely unlikely to be the cause of
shortfalls.
Clearly I'm not submitting to your Lordship that
means that in any given case your Lordship should reject
any suggestion that any accounts are wrong. I'm not
suggesting that for a moment. I'm simply saying this is
a trial of generic issues, and at a generic level,
absent particular circumstances, and there may be
several, any given set of accounts is overwhelmingly
likely to be reliable.
Now, it is worth pausing for a moment to note some
other things that Mr Coyne agreed. And if we could pick
it up at {Day14/18:1}, he agreed at the bottom of
page 18. Picking it up at line 20 and going over the
page to page {Day14/19:1} -- I said:
"Question: No, I'm not asking you, Mr Coyne -- I'm
not suggesting to you that you have been asked to decide
on whether any particular claimants' claim is right or
not, what I'm suggesting to you is that the context in
which these -- given the context in which these issues
arose -- were drafted, and given the pleadings by
reference to which they were drafted, it was obvious
that the purpose of those issues was to assist the court
so that it could use the judgment that will be produced
in this trial as a basis for making ultimate decisions
in ultimate breach claims by claimants?"
Mr Coyne said:
"Answer: In a later trial?"
I said:
"Question: Yes."
He said:
"Answer: Yes, I was aware of that."
I said:
"Question: Isn't that the main reason why we are
here?"
And he said:
"Answer: Well, it is certainly a reason why we are
here, yes.
"Question: To enable the court to make useful
findings as to the general likelihood of any transaction
being wrongly recorded in a particular case?
"Answer: Yes."
Then if we move on to page {Day14/25:9}, there was
a passage of evidence from line 9 through to page
{Day14/26:15}. Perhaps I could ask your Lordship to
read that.
MR JUSTICE FRASER: From line 9 of page 25?
MR DE GARR ROBINSON: 25. (Pause)
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: So what Mr Coyne accepts there is that
robustness involves two things. One is preventing
errors from arising in the first place, but the other is
having countermeasures to deal with errors when they do
arise, hence the concept of lasting impacts.
At the bottom of page {Day14/26:1} I put to him:
"Question: So robustness is the very concept which
underlies the issues we have been discussing for the
last half hour, yes?"
And he says:
"Answer: Yes."
He further accepted that the concept of robustness
was a mature subject which was the subject of study, and
that's at page {Day14/31:18}. Picking it up at line 18,
I say:
"Question: You agree robustness is an important
concept?"
He says:
"Answer: Yes.
"Question: You agree it is deployed very frequently
in the IT industry?
"Answer: Yes.
"Question: It is a subject of academic study,
isn't it?
"Answer: Yes."
Then picking it up on page {Day14/32:10}:
"Question: ... isn't it obvious, doesn't it follow
as night follows day, that the ultimate question being
wrestled with by the concept of robustness is how well
are the risks faced by a system guarded against? In
other words, what are the risks remaining after you have
taken the countermeasures into account?"
And he says:
"Answer: Yes ..."
And I invite your Lordship to read the rest of his
sentence. (Pause)
My Lord, I won't take your Lordship to the
transcript but he went on to accept that it is possible
to benchmark robustness against other systems, and
against those other systems that are similar, that are
comparable, it compares well. My Lord, that's at
paragraph 365 of our submissions but also {Day14/41:23}
to page {Day14/42:6}.
Critically, comparable systems require the
overwhelming majority of transactions to be handled
properly. They require that there should only be a tiny
proportion of lasting errors of a fraction of
a percentage. My Lord, that's at {Day14/56:1} to
{Day14/59:1}, and in our closing submissions it is
paragraphs 378 to 380.
So all in all Mr Coyne accepts the importance of the
concept of robustness, the reality of the concept as
a concept that's the subject of academic study and
practical application in the IT industry, and he accepts
that a consideration of countermeasures is an important
aspect of forming a view on robustness. And my Lord,
your Lordship will see that point made at paragraph 390
of our closing submissions.
At paragraph 391 we point out that he has sufficient
information to form a judgment on robustness. So that
was something of a lemma when discussing the first
argument that the claimants run in relation to Horizon
Issue 3.
I now come to the second argument. This is the
suggestion that the drafters of Issue 3 had in mind some
sort of conditional probability. It is so bold as to be
impressive. Let me see if I can find ... (Pause)
If your Lordship looks at Horizon Issue 3 {C1/1/1}:
"To what extent and in what respects is the Horizon
System 'robust' and extremely unlikely to be the cause
of shortfalls in branches?"
Now, this is an argument of which a Platonic scholar
would be proud. They don't say extremely unlikely to
cause shortfalls in branches, they say to be the cause
of shortfalls in branches.
So Issue 3 is not about the probability of causing
shortfalls but the probability of something being the
cause of a shortfall once a shortfall has arisen.
My Lord, I would respectfully submit that that
subtlety would obviously not have occurred to anyone at
the time, and bearing in mind the first argument that
the claimants are running, namely that robustness and
extremely unlikely need to be put together, it would be
a very curious argument to be running in any event.
My Lord, the claimants should have ten out of ten
for ingenuity with this argument. Their obvious
intention is to avoid their essential problem that with
3 million sets of branch accounts, ignoring for the
moment that up to 2005 accounts were weekly, that the
evidence shows that relatively speaking there were
hardly any bug impacts and the evidence suggests that
there are even fewer, infinitely fewer remote access
impacts, and that's a big problem for them. So what
they want to say is there's some smaller number of
shortfalls which, very conveniently, have not been the
subject of evidence so no finding can be made about
them.
The essential endeavour in this argument, all this
time and money having been spent in preparing for this
trial, is to put a block to the determination of Issue 3
because neither party has thought to engage in any
statistical analysis of the number of shortfalls versus
the number of bug branch impacts.
My Lord, in my respectful submission that is simply
not an argument that it is proper to run at the end of
a trial with no advance notice. If they were going to
run that kind of argument they should have raised it
before the evidence was even prepared so that the
parties could have addressed the point. And I rather
suggest that if the point had come up your Lordship
would very easily have simply applied exactly the same
language that's used in Issue 1: to cause apparent or
alleged discrepancies or shortfalls in branch accounts.
One of the curiosities of this argument is that they
are suggesting that the mental process of the person
that drafted Horizon Issue 3 was completely different
from the mental process of the person who drafted
Horizon Issue 1.
MR JUSTICE FRASER: Now, these issues I seem to remember,
but I might be misremembering, were agreed by both the
parties, weren't they? I don't recall being involved in
the detailed drafting of the issues.
MR DE GARR ROBINSON: I don't believe your Lordship was.
I wasn't here then but I don't believe your Lordship
was.
MR JUSTICE FRASER: I can go back and read the transcript of
the hearings but I seem to remember I was presented with
the proposed issues and I approved them. Is that right?
MR DE GARR ROBINSON: I think that is right. But, my Lord,
overall my submission is that this is a transparent and
desperate attempt to frustrate the plain intention of
the people that drafted Horizon Issues 1 and 3 to
prevent clear findings being made of obvious practical
utility and to replace them with findings that have no
practical meaning whatsoever.
I would urge your Lordship not to be distracted by
last-minute swerves of this sort. Your Lordship will
already understand that my submission is that the
claimants are trying to sideline the most important
issues and to distract the reader with isolated examples
put forward in the hope of giving the impression that
they have rather more widespread impact than is
justified by the evidence.
Let me give some examples of making that submission
good. The first one, in their written closing they
start on page 1 with remote access. That's their first
point even though remote access is clearly a second
order issue. Mr Coyne effectively admitted in his
cross-examination, and I will be taking your Lordship to
it in due course, that the impact of remote access on
branch accounts was de minimis. He didn't use that
phrase. That's where we got to in my submission.
But it said in the closing submissions, and my
learned friend repeated roundly yesterday morning, that
there is an open back door to the system. That's said
in paragraph 2. And in paragraph 5 they build on that
by saying that there is unfettered remote access.
Now, it is intended, I surmise that it is intended,
to be a headline-catching metaphor, but in my submission
it is an illuminating and quite useful one, and I am
sure this isn't intended. The concept or the metaphor,
the picture, of an open back door, it is not an open
back door onto the street, it is an open back door, if
you want to call it that, onto a walled garden,
consisting only of an elite group of highly trained
professionals, the SSC, who require months of training
before they are even allowed in.
Now the experts may agree that APPSUP privileges,
that privileged user privileges, should not be
permanently available to that many people and that when
people in the SSC wish to use it there should be
a process they go through to get temporary
authorisation.
That may well be the position, but that doesn't mean
that APPSUP was actually misused or that there were any
harmful events as a result of the fact that the facility
was there available for the SSC and then, as time went
on, it became available only for the most senior members
of the SSC, not to the new members.
There isn't evidence, my Lord, of a single harmful
event that was the result of any APPSUP use. Indeed,
there's no evidence of any use of APPSUP to change or
delete transaction data. None at all. But the
important point is that the people to whom the door was
left open were trustworthy professionals. They weren't
burglars waiting to get in through the back door once
the homeowner had gone to bed.
They are not going to run away with any money and
they are not going to start using APPSUP as a means of
doing creative vandalism in the branch network. And to
the extent that it is suggested otherwise, there is no
evidential basis for making that suggestion.
Another example of distraction with little stories
is the rather elaborate business we had yesterday
afternoon in which an impression was given that there is
a longstanding problem of duplicate losses in the system
of which Post Office has always been aware. That's not
what the evidence showed at all. But you speak quickly
and you go to particular sentences in particular
documents, and hey presto, you have created that
impression. That is the essential endeavour that the
claimants are engaging in in this trial.
Another example, the most extraordinary reliance
that my learned friend at the last minute places on some
internal Post Office documents from 2016 and 2017,
largely from those years. What the claimants seem to be
wanting to do is to substitute the opinions of the
experts on matters on which they agree by reference to
some internal management papers about how they are
planning to improve and upgrade their own back end
processes and the IT systems more generally, including
Horizon.
But I ask forensically: if that's all you need, why
have so many millions been spent on experts at all? The
important thing about those documents is none of them
say anything about whether bugs create discrepancies in
branch accounts. They are not about that at all. They
don't purport to grapple with Horizon Issues, and it is
only if you do not consider them carefully that you
might be confused for a moment into thinking that they
are even relevant.
Let me make that point good by just reference to one
document {F/1161/87}. It is a document with intent to
supply attached to a board report. I'm afraid I don't
have page 1 to hand.
MR JUSTICE FRASER: It has not come up yet, I'm afraid,
I think. Can we go to F/1161?
MR DE GARR ROBINSON: I'm sorry, I misread it. It is my
fault. It is {F/1611/87}, I do apologise.
MR JUSTICE FRASER: The dreaded spreadsheet. Where are we
going?
MR DE GARR ROBINSON: 16/11. Is it possible to become
dyslexic in your 50s? I don't know.
MR JUSTICE FRASER: Well, the F bundle generally is not easy
to use because it is so big, I think. Or the F folder.
That is a PEAK.
MR DE GARR ROBINSON: This isn't it. 1611. I'm looking
at -- page 87? 1611.
MR JUSTICE FRASER: I think this is 16/11 and we want 1611.
MR DE GARR ROBINSON: We need to agree a protocol for this.
1611, page 87, please. {F/1611/87}.
This is a board agenda, and if we go to page 87 this
is a technology strategy update. I don't want to put
words into my learned friend's mouth, but the author is
Rob Houghton who is the chief technology officer at
Post Office, and so this may be one of their golden
documents, one of my learned friend's favourite
documents. But let's just have a look at it:
"Context.
"This document forms an update to the IT Strategy
approved in July 2016 by the PO Board. In July we
outlined that IT was not fit for purpose, expensive and
difficult to change."
Stopping there. If your Lordship were a journalist
you would think that looks really bad, but let's read
on. There is a list of questions addressed in this
report, and then:
"Conclusion.
"Since reporting on the IT strategy in July; the
landscape continues to shift with increasing cyber
threat, deteriorating market conditions and several
significant service outages to our back office systems
..."
Your Lordship will note back office and not Horizon:
"... and digital infrastructure. These reinforce
and give further urgency to, rather than change ..."
There must be a missing word there:
" ... the strategic roadmap."
Then:
"In the 5-year overlays, we have aligned our IT
strategy to the business strategies and developed
roadmaps for the digital, infrastructure and Horizon
journeys. Approval of the overlays is critical to
improve our control environment, reduce costs and
achieve our business strategy."
So improving the control environment, reducing costs
and achieving a business strategy. Then two bullet
points down:
"Our view of the operational service risk has
matured and we remain outside of our risk appetite zone
in key operational areas. The infrastructure related
change programmes focus on reducing these risks over
time."
So your Lordship will see the general points being
made.
If one goes to page {F/1611/88}, the first bullet
point says:
"Supplier negotiations are underway."
Second bullet point says:
"There are a 'critical few' dependencies that drive
reduction of the cost base and future business support
in the most material way ..."
So your Lordship has seen what they want is to save
costs, they want to upgrade the system, there are all
sorts of objectives they think they can achieve:
"We need to quickly rationalise and resolve
misaligned contracts enacted to support legacy IT,
obsolescence and lack of PO technical competence
particular focus on Fujitsu and Accenture.
"'Thin Client' delivery success.
"Appetite for reduced service levels and number of
branches/users. We are developing a hypothesis to offer
two/three tier service offering to branches to reduce
cost and target service delivery."
Then on page {F/1611/89}:
"The IT strategy outlined a view of the current
state of technology within PO as failing to meet PO
aspirations on any assessment lens (cost, risk, delivery
or service).
"The strategy laid out key interventions:
"Improve IT - cost reduction, operating model
changes, supplier renegotiations, capability hires,
communications and engagement."
"Reduce operational and security risk - IT
infrastructure programmes, Back office transformation,
Security Transformation programmes and improved
Environment Monitoring/Management, collectively to take
back control of the management of IT Services."
That's a reference to outsourcing; they are planning
to bring things clearly back in.
Then, my Lord, over the page {F/1611/90},
paragraph 4 at the top says:
"Technology sits at the heart of the wider business
strategy set out in our 5 Year plan ...
Then the last sentence before the first bullet:
"Ultimately our technology strategy should be judged
in terms of the benefits it drives for each of our key
customer groupings:
"for end consumers ..."
Then it talks about them. Then for host retailers,
that is postmasters:
" ... technology is critical to simplifying our
offer, removing the complexities which are frequently
cited as the biggest deterrent to running a post office.
Our network IT strategy is working towards the
development of smaller point of sales solutions
(including EPOS integration) that are more easily
integrated alongside the main retail till, driving
operational benefits for the retailer and quicker
transactions for customers."
Stopping there. So the main branch, well, the first
branch benefit or objective that's identified, nothing
to do with the reliability of Horizon, it is to do with
enabling people running their own retail business to
plug the Post Office IT system into their existing EPOS
systems to make it easier for them, less of a barrier
for them to agree to become a Post Office branch.
Then it goes on to say:
"Stability of service is also vital for our
postmasters, ensuring they do not suffer the significant
business detriment that can arise from prolonged outages
..."
So there's a problem that is identified: prolonged
outages. Your Lordship may think it appropriate to note
that that's the only problem that's relating to branches
that is identified in this entire paper, as far as I'm
aware.
Then, my Lord, page {F/1611/98}. I will take
your Lordship to this because this is quoted by my
learned friend in his submissions.
At paragraph 25, it says:
"There are tensions in each contract. Our approach
and risk appetite around public Contract Regulations are
likely to also inform outcome ..."
First bullet point:
"Fujitsu - a 6 year fixed contract signed with PO
which continues to invest in legacy and obsolescence
where FJs own strategy globally is to move to Cloud.
Success is to re-contract to prime UK Cloud partner,
align innovation, re-architect and position as a true
retail partner."
So there's the word obsolescence. My learned friend
loves the word obsolescence that's quoted. But what is
actually being talked about? Moving to the cloud. It
is an improvement, it is not the fixing of a defect.
Then, my Lord, over the page {F/1611/100}, I'm
afraid this is in management speak, I find it hard to
read with a straight face, but paragraph 30:
"Our target state is to deliver an elastic, agile,
secure, future proof, low cost model for the Post Office
without undue reliance on specific suppliers."
Then in 34 it says:
"For Back Office: A Back Office system that provides
a stable and robust financial system and MI/Analytics
capability."
Then "For Retail", it says:
"The Horizon (HNGX) platform is at the end of its
life and needs replacing. Previous attempts to move
away from HNGX platform, specifically with IBM, have
been unsuccessful.
"The current plan is to transition from HNGX to the
updated HNGA (part of the Branch Technology programme)
platform from April 2017 will incur (budgeted) capex of
£16m."
So they are spending £16 million on an improved
version of the existing Horizon system.
Then over the page that's made clear with the next
bullet point {F/1611/101}:
"However, whilst HNGA runs on updated Windows
software, fundamentally its architecture is the same.
This means that it remains a 'closed' and inflexible
platform that cannot support the business' strategic
objectives and is a drain on respective PL's. HNGA will
work and support our business BUT it will not enable us
to compete and succeed with Clients retailers or
Customers."
Then there is an explanation of this thin client
concept, and I don't need to take your Lordship to that.
So I ask forensically: what does this document show
us about how good the Horizon system is? What does it
show us about its likelihood of creating false
discrepancies or false shortfalls in branch accounts?
What does it show us about the robustness of the
existing system and the reliability of the figures in
it? Answer: absolutely nothing.
It is not about those questions, my Lord. And this
demonstrates in really graphic terms the dangers of
taking little snippets from documents and pulling them
together and seeking to create an impression that's far
larger than the document from which the snippets are
drawn.
By relying on documents such as this, in my
respectful submission, the claimants are inviting
your Lordship to make a category error. They are
relying on documents with a completely different focus
in order to establish a position in relation to the
Horizon Issues which is inconsistent with the agreement
between the experts. They are trying to wallpaper over
that inconvenient agreement between the experts by
focusing on different points, by focusing on the
position now, when of course it must be remembered that
most of the claims made in this case relate to
a position many years ago.
My Lord, that entire approach, in my submission, is
the very antithesis of what this trial is supposed to
be. If there is to be an analysis of the document it
should look and see what particular point in time the
document operates, what precisely the document says,
what can properly be drawn from the document and what
cannot properly be drawn from that document. But that
is not an endeavour which, in my submission, the
claimants are interested in doing. Just as, I say this
with some regret, it was not an endeavour which Mr Coyne
was particularly interested in doing in his two reports.
When he came to give evidence the position changed.
If documents like this do my learned friend's job as
he seems to think, I do ask rhetorically why have we
just spent four weeks trying this case?
Another aspect of these documents which really is
very striking is the attempt that my learned friend
makes to criticise Dr Worden for not being aware of
them. There was a clear implication in the tone of my
learned friend's questions of Dr Worden. You are not
aware of these questions, the implication being that he
really ought to be as a reputable expert. And that
implication is made explicit in the claimants' closing
submissions as if the documents represented exactly what
any expert should be looking at instead of looking at
the architecture, looking at the countermeasures and
looking at the operation of the system and seeing how it
worked, which is what Dr Worden has actually done.
But, my Lord, there is a problem with that
criticism. The problem with that criticism is that
Mr Coyne doesn't refer to any of these documents either.
They weren't mentioned in his reports. They weren't put
to him in examination-in-chief. They certainly weren't
put to him in cross-examination. I was blissfully
unaware of them until they were pulled like a rabbit out
of the hat when Dr Worden was giving evidence.
So perhaps conveniently for the claimants, I wasn't
able to ascertain Mr Coyne's views on those documents.
I certainly didn't have a realisation that that was
something I ought to be doing.
Now, my Lord, in my submission, the inevitable
inference is one of two things. Either Mr Coyne wasn't
aware of them either, so by criticising Mr Coyne my
learned friend is also criticising his own expert, or,
which may be more likely --
MR JUSTICE FRASER: You mean by criticising Dr Worden?
MR DE GARR ROBINSON: I'm so sorry, I do this too often. Or
Mr Coyne didn't think these documents were sufficiently
material to the Horizon Issues to be worth referring to.
My Lord, either way this last-minute tactic by the
claimants to change the narrative in my submission is
completely inappropriate. They don't undermine
Dr Worden's view on the Horizon Issues any more than
they undermine Mr Coyne's views that Horizon is
relatively robust and, indeed, is more robust now than
it has ever been before.
My Lord, I would like now to say a few words about
bugs affecting branches. Mr Coyne says that he and
Dr Worden together found 29 potential branch-affecting
bugs in PEAKs and KELs, and in his oral evidence he said
that given the reading he had done and the searches that
he and his team had performed, he didn't think there
were more than 40 bugs overall in PEAKs and KELs.
Now, of the 29 that he has found, he now says that
22 are lasting; in other words, 22 are of a type that
wouldn't, or might not ordinarily be caught by
countermeasures. Pro rata that suggests that the
overall number of lasting bugs in PEAKs and KELs would
be around 30, although I have to say I didn't put that
to him. I didn't realise that there would be this shift
from 29 to 22 when we discussed the point.
Now, the claimants absolutely hate that logic. So
they attack the concept of a lasting impact. In his
written closing and in his oral submission yesterday my
learned friend said that Mr Coyne's use of the term
"lasting impact" in the joint statement -- that's
paragraph 115 of JS2. Perhaps we could look at that.
It is at {D1/2/29}, paragraph 1.15. This is agreed:
"The number of distinct bugs, for which the experts
have seen strong evidence of the bug causing a lasting
discrepancy in branch accounts, is between 12 and 29."
Now, that is something which Mr Coyne has agreed.
It is language that he has used. It is obvious what he
meant. He means not transient, and your Lordship will
see the reference to transience in paragraph 1.10 at the
top of the page. That is an RW statement.
These are bugs whose impacts are not caught by some
countermeasure and so one can't expect that they would
necessarily be corrected.
Now, Mr Coyne accepted all of that, he accepted what
"lasting" meant in cross-examination. And I frankly
don't understand the attempt to suggest now that he
somehow didn't.
The claimants themselves have deployed the concept
of lasting in their own opening. My Lord, that's at
{A/1/35}, paragraph 108:
"As noted above, in Joint 2, the experts have
resolved some but not all points of difference
(eg Dr Worden now accepts that there is strong evidence
of a lasting discrepancy on branch accounts from 12 of
the 29 bugs which Mr Coyne identifies as having such an
effect."
So there we have two statements in that paragraph.
First of all, they themselves are deploying the
concept of lasting effect. Secondly, they are clearly
saying that they clearly thought at that time that
Mr Coyne was saying that there were 29 bugs which had
a lasting effect. And your Lordship will recall the
cross-examination on his last day of evidence about
that.
So against that background it is difficult to
understand what the claimants mean when they say that
the use of the word in paragraph 1.15 was a compromise.
Unless they are seeking to go behind it in some
impermissible way, an agreement was made in without
prejudice discussions between the experts. I can't
imagine they are doing that, but I can't imagine what
else they are doing.
My Lord, what one gets from paragraph 108 is that
while the claimants were entirely comfortable with the
concept which was fully deployed and explained in
Dr Worden's first report, as the evidence has come out
and as Mr Coyne's own evidence on these points has come
out, they have had second thoughts. Now that Mr Coyne's
views have been properly explored, they want to
backtrack.
In my respectful submission, that this process is
being engaged in is actually a good illustration of what
cross-examination can achieve. Nowadays it is
fashionable to suggest that what one gets by
cross-examining witnesses is very limited and shouldn't
we have trial mainly by reference to the documents? In
my submission, this trial is a good example on both
sides of points being established in cross-examination
that wouldn't be established without it.
I make the same submission about Mr Roll's
cross-examination. My learned friend is entitled to
make the same submission about some of my witnesses.
But be that as it may, it is important to recognise the
significance of Mr Coyne's evidence on the number of
bugs to be found in PEAKs and KELs and the proportion of
lasting bugs that are liable not to be caught by
countermeasures. He and Dr Worden disagree on
particular bugs, and your Lordship has our submission on
the bugs in appendix 2 to our closing, but let's take
Mr Coyne's views at face value.
At {Day15/94:1}, he said if you get a bug with
a branch impact there will be a KEL for it. My Lord,
that is addressed at paragraph 352 of our closing
submissions.
At page 122, unless your Lordship would like me to
I'm going to go too fast to allow the reading of the
transcript --
MR JUSTICE FRASER: No, but the transcript I think you are
just referring to has just come up and that's him
agreeing to your question that:
"Question: ... generally speaking, if you get a bug
... once it is detected there will be a KEL ...?"
MR DE GARR ROBINSON: Yes. My Lord, {Day15/122:1} Mr Coyne
says he has now reviewed between 5 and 6,000 KELs. He
says at line 24:
"Answer: ... probably ... between 5 and 6,000."
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: At page {Day15/123:1}, he says that
his team will have reviewed an additional 1,000 KELs.
So between he and his team there has been a review of
between 6 and 7,000 KELs of the 9,500-odd that have been
disclosed.
At page {Day15/123:1} he says that all the bugs that
have been found come to 29. And he goes on to say that
the total number of bugs likely to be found in the KELs
is likely to be no more than 40.
My Lord, at page {Day15/128:16} Mr Coyne agrees that
his search processes are reliable and will have
successfully identified the majority of bugs identified
by the SSC during the operation of the PEAK and KEL
system.
So there aren't that many more bugs to be found even
if he had had more time to look.
Your Lordship will also recall, I'm afraid I don't
have the reference in my note, Mr Coyne saying that it
is possible to identify the number of impacts of
a particular bug by looking at the PEAKs, and that's
quite a reliable system too. And he then shamed me by
indicating that he had given an indication in JS2 as to
the number of impacts.
My Lord, on this side of the court we have done our
best to do a calculation --
MR JUSTICE FRASER: Was it in JS2? I thought it was in his
report. Was that when you took him to the graphic?
MR DE GARR ROBINSON: It was JS2. Second column.
MR GREEN: It is actually both.
MR DE GARR ROBINSON: My learned friend helpfully says it is
also in his report.
MR JUSTICE FRASER: No, when you said he had shamed you
I thought that might be when you said you'd put
a question to him and he then pointed to a graphic.
MR DE GARR ROBINSON: I think I had missed it and he pulled
me up on it, very properly.
On this side of the court we have done our best to
work out how many impacts are suggested, and I'm not
suggesting for a moment that this is accepted but it
appears to be in the region of 545/550 branch impacts in
very round numbers.
So that would suggest less than 20 impacts per bug,
and I would ask your Lordship to bear that statistic in
mind.
I wonder whether this would be a convenient moment?
MR JUSTICE FRASER: I am sure it would, yes.
When you say on your side you have identified it,
that's the one in your closing? The 545/550, that's in
your closing submissions?
MR DE GARR ROBINSON: It is in a footnote to our closing.
I can give you the reference if your Lordship would like
after the break.
MR JUSTICE FRASER: I do not think it is in any of the
evidence, is it?
MR DE GARR ROBINSON: No, it isn't. What has been done --
I'm afraid I haven't done the process, but what has been
done is you look at the column, you and try and work out
what that implies. It is not a simple counting process,
one has to use a degree of judgment. It is footnote 794
of our closing submissions, my Lord.
MR JUSTICE FRASER: Thank you very much. All right. We
will have ten minutes. We will come back in at
11.55 am.
(11.47 am)
(A short break)
(11.55 am)
MR DE GARR ROBINSON: My Lord, talking about bug numbers and
bug impacts, it may be helpful at this point to look at
one of Dr Worden's calculations. If I could ask
your Lordship to go to {D3/8/1}.
Thank you very much. Your Lordship will have seen
this before. It is worth spending a few minutes so that
we all understand it.
This is a calculation which involves all sorts of
financial impacts and also involves an estimate of the
number of bugs detected and undetected in a system over
a period of 20 years.
It starts at row 3, but your Lordship will see the
label in column B. So I'm working behind those column B
labels. So row A is "mean number of branches in PO
network, 1999-2018".
Your Lordship may recall Dr Worden being taken to
task because his number of 13,560 was from 1999, not
from 2000. I think his report did say 2,000, but in
actual fact, as is clear from his spreadsheet, it was
from 1999. So that is the number of branches, mean
number of branches over that period.
Then row B is years of Horizon. 19 years. Total
branch months in C. Nearly 3.1 million in branch
months. So that is 3.1 million sets of accounts, as it
were. But Dr Worden, your Lordship will be aware, has
rounded that down to 3 million. In his reports he talks
about 3 million. He could of course have said 3.1, but
he said 3 because that was sufficient precision for him.
Then there is the scaling factor in relation to
which some time was taken in Dr Worden's
cross-examination, and your Lordship will see --
actually, this is a good place to see. Column C is
central estimate. That is Dr Worden's central estimate,
what he thinks is likely to be the case, and D is his
conservative estimate, because as your Lordship will
recall, he deliberately makes assumptions in favour of
the claimants to increase likelihood or reduce the
likelihood of things happening so as to increase the
number of bugs and so on. And that's his conservative
estimate. He explains all this in his report.
Your Lordship will see that there is a scaling
factor of 0.45 central and he just pushes it up by just
over 10% to 0.5 conservative. We needn't worry about
scaling for these purposes.
Row E is another scaling number. Then row F is the
total number of claimed shortfall. And as your Lordship
ascertained yesterday, that's taken from the schedules
of information and is explained in Dr Worden's first
report.
Row G is:
"Total claimant branch months (sets of monthly
branch accounts for claimants)"
Or, indeed, any other SPMs with similar
characteristics. That is 52,000. So of the
3 million-odd monthly branch accounts, ignoring for the
moment until 2005 they were weekly, 52,000 was occupied
as it were by claimants.
Then H is "scaling factor". I'm not going to talk
to your Lordship about that. Then here's an assumption
in row L:
"Maximum number of KELs with potential impact on
branch accounts, based on limited sampling of KELs."
And his central estimate is 100. Does your Lordship
see that?
MR JUSTICE FRASER: Mm.
MR DE GARR ROBINSON: And his conservative estimate is 200.
It is quite important to note that those two estimates
are explicitly based upon limited sampling of KELs.
Dr Worden didn't do what Mr Coyne did. What Mr Coyne
did is he looked at virtually all the KELs; he and his
team looked at between 6 and 7,000 of the 9,500. That's
not what Dr Worden did. What Dr Worden did is he looked
very carefully at a limited number of KELs. He did
a number of passes through the KEL system, one of which
was he looked at 200 KELs at random, and another pass
that he did was he looked at KELs that referred to
a pound sign, and a third pass that he did was he looked
at KELs that had been referred to in Mr Coyne's expert
report.
The result of those enquiries are set out in
appendix D to his first report, and then of course in
his supplemental report he has looked at more KELs, so
there is an expanded appendix for the 200 KELs that he
looked at. But on any view he had only looked at
a small proportion of KELs; not statistically
insignificant, but hardly occupying the whole ground.
It is important to note, and I will be coming back
to this, that Mr Coyne has now looked at the whole
ground or more or less at the whole ground. He has
looked at over 6,000, over two-thirds, of the KELs.
What's more, he has done intelligent searches through
all of them. What's more, he's done intelligent
searches through the 220,000 PEAKs.
As I think he may have demonstrated to your Lordship
already, his opinion is that he has found most of the
bugs, potentially branch-affecting bugs that are out
there to be found.
I hope I'm not oversimplifying his evidence when
I say that.
And he has found -- his view is that there wouldn't
be more than 40.
MR JUSTICE FRASER: Mr Coyne's view?
MR DE GARR ROBINSON: Mr Coyne. Am I saying --
MR JUSTICE FRASER: No, no, you switched to "his", so
I wanted to be clear.
MR DE GARR ROBINSON: Mr Coyne's view is that there are not
more than 40, and that is compared with 100 in row L,
conservative Dr Worden and the 100 central, 200
conservative.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: Those have been superseded, I would
respectfully submit, by Mr Coyne's own analysis.
If one takes into account the revelation on the last
day of his cross-examination that of the 29 bugs that he
has found that scale up to 40, he is now of the view
that only 22 of them are lasting, so if one were
interested in only lasting bugs, the figures to put in
row L would be 30, rather than 100 and 200.
If one does that, my Lord -- your Lordship may,
I fear, and I'm sorry to say that your Lordship is
probably better at this than I am, but even I can do
this -- that one can insert the figures into the
spreadsheet, one could put the figure of 30 or 40 in
those two columns, C and D, and then see where you get
to. And where you get to is that at the bottom of the
page, in row E2, maximum possible number of bugs, that
changes from 145; if the figure is 30, it becomes 44.
And would it be helpful if I talked you through how
the calculation works?
MR JUSTICE FRASER: I think I can tell how it works.
MR DE GARR ROBINSON: You have it. It is to do with various
assumptions about KEL efficiency, and so on.
MR JUSTICE FRASER: Well, L -- yes -- L is going to change
and that's the figure that's currently in N.
MR DE GARR ROBINSON: One very annoying typographical error
that your Lordship will see in E2 at the bottom of the
page, you see that E2 is defined as R/X?
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: Actually that should be L/X. That's
a typo.
MR JUSTICE FRASER: On the basis there isn't a R.
MR DE GARR ROBINSON: Well, on the basis that what it should
be doing is it should be starting with the number of
bugs in the KELs and then doing something with them in
order to scale up to an estimate of a total number of
bugs to include undetected bugs.
MR JUSTICE FRASER: So E2 equals L/X.
MR DE GARR ROBINSON: E2 equals L/X. And on Dr Worden's
central approach, if there were 30 lasting bugs
detected, one would end up with 44 lasting bugs in
total. That's detected and undetected. The figure
would be higher, significantly higher, in D, but much
lower than 672.
Now, in the claimants' closing, it is quite
interesting, it is suggested that there is a reasonable
evidential basis for assuming that your average bug will
have 40 or 48 affected branches, and that's worth
noting, my Lord. That's {A/5/177}, paragraph 1515. We
don't need to go to it.
But on the basis of the results of Mr Coyne's
enquiries in JS2 in the second column, where he has
found 440- impacts, or 450-odd, one would imagine it is
rather less than 40 or 48, could be closer to 20. And
just for the sake of illustration, if one took E2, the
central estimate of 44 bugs, that's on the basis of 30
lasting in the KELs, and one multiplied by 20, one would
get a total of 880 branch impacts over 3 million sets of
branch accounts over 20 years. I say 20, I should
say 19.
That is a vivid illustration, in my respectful
submission, of the kind of scale of the extent that we
are really talking about when we talk about bugs in
Horizon affecting branch accounts.
But that's not the calculation that the claimants
put to Dr Worden. They rather like, although they
refute his entire methodology, the figure that's
currently in row E2 in column D. They like the 672
figure. Your Lordship will see that in
cross-examination they put to him that if you take 40 or
48 branch impacts per bug and you multiply by 672,
I think if you take 48 branch impacts you get something
like 32,000 branch impacts. Does your Lordship recall
that evidence?
MR JUSTICE FRASER: I do.
MR DE GARR ROBINSON: And by that means Dr Worden, with some
heroic mental arithmetic that I would not have been able
to do, worked out that you would end up with one branch
impact per claimant on average.
Now, I am sure that my learned friend was quite
pleased with that result and I am sure if you were
a journalist you would think "Crikey", but it should be
remembered that there are 561 claimants who are claiming
£19.7 million in shortfalls. On average that's about
£33,000 each. So if you have one claimant with one bug
impact in their tenure, how do you get to a £33,000
shortfall? In my submission, that calculation on its
own is another vivid illustration of the scale of the
judgments that are really called for by the evidence in
this case.
But, my Lord, it goes further than that. As is
explained in our closing at pages 157 to 158, and
perhaps I could ask your Lordship to look at that. It
is {A/6/157}.
MR JUSTICE FRASER: Just while that's coming up I think you
said 19.7 million in shortfalls, but you mean 18.7.
MR DE GARR ROBINSON: Oh gosh, I do. Absolutely.
Thank you.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: If we could pick it up at
paragraph 439:
"If no scaling factor for small claimant branches is
applied at all ..."
So the table in paragraph 637 of Worden 1, and
that's the same table we have been looking at, label D
becomes 1, then label E -- your Lordship will recognise
the 3.1 million-odd figure, H becomes 193,000 and so on,
and then what that produces. And that's explained in
paragraph 440 on the next page {A/6/158}, 440.1:
"For a bug such as the Suspense Account bug which
occurs 16 times with a mean financial impact of £1,000
..."
And your Lordship will recall Mr Coyne accepting
this is one of the big ones. Or your Lordship may
recall. I should not assume your Lordship has perfect
recall of every single piece of evidence that's given in
the case:
" ... in order for there to be a 1 in 10 chance that
a shortfall of £1,000 would be caused in a claimant's
branch in any given month, there would need to be 19,000
similar bugs ..."
Because each bug is assumed to have 16 impacts or
occurrences, that would mean 304,000 occurrences, or
incidents they are called there. So you are looking at
19,000 bugs to give a 10% chance of causing a £1,000
loss and you are looking at 304,000 branch impacts in
order to have that result.
And if the mean financial impact was smaller, let's
assume it is £500, there would have to be 38,000 similar
bugs and 608,000 incidents. And if the mean financial
impact were £100, there would have to be 190,000 similar
bugs with over 3 million incidents.
Now, my Lord, these are just illustrations but what
they illustrate, or what they are intended to do is to
give your Lordship a sense of perspective, a sense of
scale as to what the evidence demonstrates as to the
likelihood of bugs causing shortfalls in branch
accounts.
In my respectful submission, a useful touchstone,
a useful yardstick for measuring scale is the claimants'
claim. That's not to suggest -- I'm absolutely not
suggesting -- that this is a way of disproving the
claimants' claim. Indeed, this calculation that we are
looking at here ignores any scaling factor, it just
looks at average branches of the same sort of tenure as
the claimants. That's all.
But it gives your Lordship a handle on the sense of
how big the bug problem would need to be in order to
have a material chance of generating the sort of losses
that are being -- in fact, a small proportion of the
sort of losses that are being claimed in this case. And
your Lordship may find that useful as a touchstone. It
certainly shouldn't be your only touchstone.
But what touchstone do the claimants use? What do
they say about extent in relation to these Horizon
Issues? They say, and your Lordship will see that from
the appendix to their closings where they very helpfully
set out in a table what they say the answers are to each
issue. In relation to Horizon 1, Issue 1, 3, 4 and 6,
those I hope are the extent of the issues that are
extent questions. They say material and significant
risk.
Now, that begs an important question: material
compared with what? Significant in comparison with
what? And in my submission the complete absence of any
sense of scale is a complete giveaway. To say material
insignificance is not actually an answer to the Horizon
Issues, it is an attempt to avoid answering them.
My Lord, the point is just as stark. Perhaps it is
even more stark with remote access. If I could go to
our closing submissions. That's {A/6/247}. At
paragraph 740 of our closing, perhaps I could invite
your Lordship to read it.
Mr Coyne accepted that any remote access work
involving branch data is typically recorded in PEAKs.
So if there is any remote access happening you will see
it in a PEAK. You may not have full details but you
will see that it happened, typically.
The relevant reference is {Day16/28:1}. Then if one
moves to the next page {A/6/249}, actually,
paragraph 746:
"Mr Coyne fairly accepted that he had identified
relatively few instances of remote access being used to
affect branch accounts, especially compared to the vast
number of branch accounts over the life of Horizon."
So I ask:
"Question: I would be right in thinking, wouldn't
I, that of the PEAKs you have seen you found relatively
few examples of remote access having been exercised?
Would the answer to my question be right?
"Answer: I don't know exactly what the number will
be, but it is tens, twenties --
"Question: Looking at your report it would be low
tens, wouldn't it? You haven't found hundreds?
"Answer: No, I haven't found evidence of hundreds,
no."
I ask:
"Question: So you have found, as I say, a
relatively small number; relative to the fact that we
are talking about 3 million branch accounts over the
last 20 years, all you have actually found is a very
small number which is less than 20 or 30, let's call it
less than 30, would you agree with that?
And his answer was:
"Answer: Yes."
So when considering remote access, which is a big
subject and by its very nature it is bound to be the
subject of forensic interest, it is easy for lawyers to
be considering that kind of issue and it is fun for the
lawyers to do it because there is the Select Committee
meeting at which Post Office gave evidence, and
Post Office's response to the Panorama programme, all
the things which engage the press and are in the public
eye. These things naturally focus attention, and not
surprisingly -- I don't blame my learned friend for
doing it, I am sure I would do it were I in his
position -- that goes on page 2 of my learned friend's
closing submissions.
But if you take your eyes away from the pyrotechnics
to stage left and you actually look at the Horizon
Issues, you see that to call remote access a second
order issue actually overstates its importance. It is
actually, if you think about it, a third order issue.
You would need some remote access to change branch
accounts. Mr Coyne says he hasn't found more than 30.
I respectfully submit the number is far lower than that.
Your Lordship has seen what we say in our
submissions but let's take him at his highest. He says
there is not more than 30. Then you need error or
malice which might have an adverse impact on branch
accounts. That, on any view, is going to be rare
because we know from the evidence given by Mr Roll and
the evidence given by Mr Parker that these things were
done carefully.
And no examples have been found of any remote access
actually causing a discrepancy, and certainly not
a shortfall in branch accounts. Only one candidate has
been put forward; that's {F/432/1}. That's the $1,000
bug. Your Lordship has lengthy submissions about that
in our closing submissions.
We say as a matter of principle it is clear that the
error was made in that case, if it was an error -- no,
it wasn't an error. The change to data that was made in
that case which it is suggested had an impact on branch
accounts was actually a change in back end data; it was
an OCR making a change to data that was in the TPS
system. It wasn't any change to branch account data at
all, it couldn't be because it was in the TPS system.
So there is not one single, we submit, example given
of any exercise of any remote access facility which has
actually been shown to adversely have affected the
branch account.
So, on any view, the proportion of cases in which
that happens is going to be small. As I have already
submitted to your Lordship, the remote access when it is
done, it is done reluctantly, it is done professionally.
There is a four eyes principle. My learned friend has
perfectly fairly drawn attention to what I think he
describes as many examples of OCPs where the monitoring
individual seems to be the same as the individual doing
the change, from which it is inferred, and I don't know
whether this is right or not, I'm not in a position to
dispute it, that the four eyes principle has become two
eyes in those particular cases. That doesn't mean
mistakes have actually been made, but whether they were
or not, it is quite interesting to note that he says
many. So far as we are aware on this side of the case
it is only four examples.
They have plainly been looking very hard. That
gives an idea of the smallness of the problem that we
are looking at. And there's the third requirement. Not
only is it necessary for there to be some relevant
remote access in the first place and for there to be
an error, carelessness or malice, one can't imagine what
malice there would ever be, but there would also have to
be the sort of situation where the postmaster himself
doesn't know that it is happening.
My Lord, it is very important to recognise that in
all the PEAKs that have been looked at and all the OCPs
and the OCRs and the MSCs that have been looked at, so
far as I'm aware only one example has been found of
an SPM not being involved in the process by which his
branch data was worked on.
And I discussed that, as I recall, in
cross-examining Mr Roll who frankly admitted -- this is
from memory so I need to be careful, but my recollection
is that he frankly admitted that it would be suicidal to
muck about with a postmaster's branch accounts without
involving him because then all sorts of problems could
occur.
I do hope my memory is correct. And I will check it
to make sure.
So where do we get to with all of this? 30, taking
Mr Coyne's figure at its highest, multiplied by a small
fraction to reflect a possible error rate, multiplied by
a small fraction to reflect the SPM not knowing what has
happened results in an absolutely tiny number. It would
be surprising if it were more than one in the real
world.
What scale are we talking about? The scale is over
3 million branch accounts. So the chance of a remote
access problem adversely affecting a branch account is
absolutely tiny. It is 1 in 3 million. 2 in 3 million
maybe. It is being hit by lightning territory.
Even Mr Coyne accepted that the chance of remote
access adversely affecting accounts is reasonably small.
My Lord, that's referred to in our submissions at
paragraph 748 at the bottom of this page {A/6/249}. He
didn't want to say vanishingly small, but he couldn't
resist saying it was small.
Yet in their answers to the Horizon Issues regarding
remote access, what do the claimants say? They say
material and significant risk. In closing they put
remote access at the forefront of their case. In my
respectful submission, that they seem to think that
remote access is their best point might be thought
revealing about the overall merits of their case.
My Lord, I'm going to move on and deal briefly now
with some criticisms that were made of Dr Worden. I'm
going to do this as quickly as I possibly can.
The suggestion is made that he is biased. I refute
that suggestion. Reliance was made on the first joint
statement which, in my submission, was entirely
misplaced. If we could go to JS1 at {D1/1/1}, please.
I don't have the page reference. Would
your Lordship give me a moment?
MR JUSTICE FRASER: Is this the part where they set out
balanced approach?
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: I think it is the first page after this,
if we could just go forward.
MR DE GARR ROBINSON: It is page {D1/1/3}. So:
"Each expert's approach to writing his report, and
to this joint memorandum which foreshadows their
reports, could broadly be one of three possible
approaches:
"(a) to focus mainly on negative points ...
"(b) To focus mainly on those aspects of Horizon
which were intended to achieve robustness ... and the
evidence implying that they succeeded."
And (c):
"To provide the court with a clear foundation for
understanding the design and operation of Horizon; then,
building on that foundation, to provide a balanced
assessment of the ways in which Horizon succeeded,
whilst addressing any disclosed issues where Horizon may
have fallen short."
What Mr Coyne said he was going to do is at the
bottom of the page, two lines from the bottom:
"Whilst my report will take a balanced approach, it
is the case that many of the issues require a deep focus
on the occurrences of bugs, errors and defects as well
as the potential for modification of transactional
data."
It is interesting to note that his deep focus was on
bugs and his deep focus was on potential for
modification. But I ask rhetorically, how many
instances did he find of bugs? 29. How many impacts?
500 and something. And how many instances of remote
access? Not more than 30.
My Lord, both of those numbers, as your Lordship
will be aware, are disputed on this side of the court.
But that was the result of his deep focus.
Then Dr Worden, he says:
"I intend to take the balanced approach (c)."
My Lord, in my respectful submission, that's exactly
what he did. He looked at the architecture to see how
Horizon worked. He looked at the countermeasures within
Horizon to see whether they were properly configured and
designed in order to achieve the robustness which they
were designed to achieve. He looked at testing to see
how they were tested to make sure that they were tested
properly.
There is an entire section on testing in his second
report, for example, where he enhanced his review of the
testing process. Then he reviewed how the
countermeasures had operated during the entire life of
Horizon. Your Lordship will see that from his tables in
appendix D of his first report and then the updated 200
bug table in his second report.
My Lord, in my respectful submission, that was the
right thing to do. My learned friend criticises him for
taking a top down approach, but I would respectfully
observe that Horizon is a massive system. If you don't
start by seeing how it is configured and how it works
from the top, and if you don't then delve into the
system to see how the support systems work, to see how
bugs are dealt with and so on, then you are likely just
to get an incoherent catalogue of problems which aren't
properly organised and are just a wall of problems which
are very difficult to manage.
In my respectful submission, that's exactly what
Mr Coyne's reports produced.
I do refute the suggestion, which I am sure my
learned friend has made, that Dr Worden only stayed at
the top level. He went down, he looked down into the
KELs, he looked at PEAKs, he looked at dozens of PEAKs
at the time of his first report and looked at many, many
more by the time of his second report and by the time of
the trial.
The exercise that he performed was not a biased one.
He looked for good and bad. There were sections in his
report where he criticised Post Office and Fujitsu. It
is worth noting, my Lord, that of the 29 bugs that are
currently in the bug list, nine of them were actually
identified by Dr Worden. So if it hadn't been for him,
we would be looking at 20 bugs in all probability.
Your Lordship will get that. It is paragraph 253 of
our closing at footnote 383, page 100. We needn't go
to it.
Dr Worden made some mistakes. He, for example, it
is just an example, made a mistake when doing his
scaling calculation first time round, dividing by 461
instead of a smaller figure. 561 instead of 496,
I think it was. My Lord, that was a mistake. Actually,
as he explained in cross-examination, he dealt with the
mistake. The mistake wasn't repeated when he redid the
calculation for the purposes of his second report, but
it wasn't his only mistake. He did make mistakes.
But there's no basis for saying that he was biased
and there's no basis for suggesting that his mistakes,
if that's what they were, had any substantial impact on
the validity of his ultimate conclusions.
The second suggestion that's made was that he
assumed Horizon was working properly and worked back
from that. My Lord, that's simply not the case. If you
look at all his assumptions, if you look at the
financial calculation we looked at previously, the
spreadsheet, look at all his assumptions about rate of
imperfections. A fortiori look at all his conservative
assumptions which push things radically into the
claimants' direction.
Criticisms are made of his scaling approach.
My Lord, there were some mistakes, but his evidence was
clear. Most of the points that were put to him were
points which misunderstood his approach to undertaking
calculations of that kind. Approximations are not
mistakes. Engineering estimates always involve
approximations. The important question is what
precision do you need to arrive at a usable answer and
what precision is available in the light of the
information you have?
You have to beware searching for spurious precision.
My Lord, I freely accept it doesn't make my job easier
that he did take approximations of figures, he took
48 million instead of 47 million for one of his
calculations, and your Lordship will remember the GAAP
figures. He didn't take the GAAP figures in another one
of his calculations. But, my Lord, all of these things
are small beer compared with the assumptions he makes in
the claimants' favour in {D3/8/1}, 0.45 to 0.5, the
row Y calculations and so on.
Dr Worden is criticised because he didn't consider
Dalmellington at all in his first report and he didn't
consider it properly in his second report. Well,
my Lord, he did consider it in his first report. It is
considered twice actually in his appendix D3 where he
goes through the KELs that Mr Coyne had relied on.
For your Lordship's note -- we don't need to go to
it -- it is {D3/2/117} and 132. Both of them consider
the KEL acha621P. That's the Dalmellington KEL,
I think. So he did consider it but he didn't consider
it, address it specifically in the text. He did address
it specifically in his second report, but there he is
criticised for not having specifically mentioned, for
example, that there were 102 occurrences of the
Dalmellington bug before it was detected.
The reason why there were so many, of course,
your Lordship will be aware, is because the
Dalmellington bug mimicked, it caused human error. So
it looked exactly like human error when it occurred.
And what's more, it was fixed as and when it occurred by
reason of the ordinary countermeasures that apply in
Horizon.
So it never caused any lasting harm. It just looked
like human error and it only got picked up. That's why
it lasted as long as it did and that's why there were so
many impacts. It is not a representative bug in that
respect.
But he deals with it quite fully in his second
report at paragraphs 144 to 163. Mr Green seems to be
aggrieved because he doesn't specifically mention the
112 occurrences, but what he does do is he refers to
Mr Godeseth's second statement, paragraph 55 onwards,
which does contain that information.
I do find myself asking forensically why on earth
does Dr Worden have to repeat the same information
a second time? My Lord, I'm now going to address you on
what's loomed large in this litigation. It is not
a Horizon Issue, but it is a matter to which the
claimants keep going back. It is the question of
disclosure. I'm going to do that, if your Lordship will
permit me by reference to our closing submissions.
If I could ask your Lordship to go to {A/6/371},
picking it up at paragraph 1119. There's
your Lordship's order of 2nd February requiring Model C
disclosure.
Paragraph 1120 and 1121 draws your Lordship's
attention to the practice direction dealing with the
pilot scheme disclosure, and notes that:
"... It is for the party requesting Extended
Disclosure to show that what is sought is appropriate,
reasonable and proportionate (as defined ..."
It is described also as "request-led, search-based
disclosure".
Paragraph 8 of the practice direction also provided
over the page at {A/6/372}:
"(1) The court may order a party to give disclosure
of particular documents or narrow classes of documents
relating to a particular Issue for Disclosure, by
reference to requests set out in or to be set out in
Section 1B of the Disclosure Review document or
otherwise defined by the court."
So what does that establish? It established the
requirement of a request:
"(2) If the parties cannot agree that disclosure
should be given, or the disclosure to be given, pursuant
to a request, then the requesting party must raise the
request at the case management conference …"
Now, I ask forensically, why does it use the word
"must"? The answer is simple. The fundamental shift
that has been achieved between ordinary, if I may say
so, sounding like a dinosaur, standard disclosure and
modern Model C disclosure is that with standard
disclosure there is a default obligation to provide
documents that adversely affect your case or support
your opponent's case.
There is no such requirement in relation to Model C.
All there is is a requirement to disclose known adverse
documents, and a critical word there is of course
"known". You have to know about, and if you don't know
about it then you are not under an obligation to
disclose it.
So in order to be in a position to complain about
not having had a document, it is incumbent upon you as
the requesting party if you are not satisfied by the
response to go to court to get an order. That's how it
is supposed to work. Ironically, it means that more
work now needs to be done by the court than hitherto.
Although it is supposed to reduce costs, I rather wonder
whether in the long run it will increase them.
So that's the regime that we work under nowadays in
these courts.
If I could pick it up at paragraph 1125:
"By the Court’s Fourth CMC Order 1368 Post Office
was ordered to disclose the documents set out in
Schedule 1 to the Order. Schedule 1 was largely agreed
by the parties (mainly before the CMC but some of it
after). It focused on various documents (in particular
reports and briefings) being provided to various
categories of custodians which were defined in the
Schedule itself."
So that's Model C working as intended.
The parties come together, they have a conversation,
they reach an agreement and an order is made. There was
no outstanding issue at that stage, it would seem.
Certainly there was no application for anything further.
So we say at 1126:
"This was an example of Model C disclosure working
broadly [as intended]."
If one goes over the page {A/6/373}, paragraph 1127
adverts to paragraph 13 of the third CMC order where
your Lordship reminded the parties that the experts had
the right to come to court to seek directions.
If I could pick it up at paragraph 1130. Well,
perhaps at 1129, the fourth CMC order:
"... the Court, by paragraphs 8 and 9 of the Order
ordered the experts to provide an Error Codes List and
jointly to compile a list of information which either or
both considered they required."
1130:
"This joint report was duly produced on
26th June 2018."
It should be, it was a request:
" ... for a huge amount of information, explanation
and documentation."
My Lord, stopping there, what often happens in cases
of this kind is that when issues have arisen, when the
car crash has already arisen, you look back with the
benefit of hindsight and you look back at a particular
decision or letter and you look at it through
a microscope and you think "Isn't that awful, I can't
imagine why on earth that was done". Of course at the
time when a party is faced with 30 or 40 pages of
requests which go up hill and down dale which involve
all sorts of different things, and remember these are
technical requests which often the lawyers won't
entirely understand, the lawyers don't act for Fujitsu,
they act for Post Office, that can put a recipient,
a receiving party in a difficult position. And it put,
in my submission, Post Office in a difficult position.
Then over the page {A/6/374} at paragraph 1131,
Mr Coyne on 20th July sent an email to Freeths and WBD
requesting yet further material. And at 1132, it says:
"Some mechanism was clearly required in order that
(i) each side's position on the various requests being
generated by Mr Coyne was properly set out and could be
fully considered and (ii) the requirements of Model C
were met."
And 1133:
"The parties agreed a sensible mechanism which the
Court approved in the Fifth CMC Order ..."
Perhaps I could invite your Lordship to read the two
paragraphs quoted there, the order's at C7, tab 22, but
we needn't look at it.
MR JUSTICE FRASER: The two paragraphs at 1133?
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: My Lord, if I was a trust lawyer
I would say those two paragraphs gave effect to the
spirit and intendment of the pilot scheme for
disclosure, Model C disclosure. In particular, it
created a mechanism by which, if there were any
objections, the claimants would explain why the
documents were needed and then the recipient,
Post Office in this case, would be required to elect as
to whether to continue its objection or not, with the
benefit of the explanation as to relevance and why it is
reasonable and proportionate.
That is how Model C is supposed to work in my
submission.
This mechanism would have ensured that by the end of
August, in fact by the middle of August, the parties
would have known where they stood and it would have been
open to the claimants to raise the matter with
your Lordship and seek an order if need be, if agreement
couldn't be reached.
That is how Model C disclosure is supposed to work.
That's effectively what's said at paragraph 1134. But
your Lordship will see over the page at 1135 {A/6/375}
or 1136, Post Office's response on 8th August:
"[It] provided its response, as required by the
Fifth CMC Order, on 8 August 2018. In many cases, it
provided the information requested and additional
documentation. In other cases it did not. Post Office
was genuinely unsure of what Cs' case was in relation to
many of the Horizon Issues – neither the Outline
Document nor Mr Coyne's reports had been served at this
point – and many of the requests sought not categories
of documents but lengthy analyses and explanations of
matters which Post Office did not have and which
appeared to cut across the carefully negotiated
categories of documentation which the Court had ordered
at the Fourth CMC."
So following this impasse, the claimants were put to
their election pursuant to the fifth CMC order, they
should have served an explanation by 18th August if they
maintained a claim to any of these requests together
with the necessary explanation.
Did they do that? No, they didn't. They in fact
sabotaged this entire exercise by doing nothing. I'm
not here to blame them for doing nothing, I am sure they
had plenty of other things to do at the time. But,
my Lord, I'm here defending constant criticisms made of
my clients, including during my cross-examination of
witnesses. My learned friend kept popping up and giving
your Lordship "That was disclosed on such-and-such
a date". The reason why we are having this conversation
is because Post Office has been consistently attacked by
the claimants for failing to give proper disclosure in
circumstances where those criticisms are based upon
a fundamental misunderstanding of the way that
disclosure is supposed to operate in the modern world,
in my submission.
If one were to look at the claimants' account of
disclosure in their own closing submissions, it is quite
revealing. We needn't go to it, but that is
paragraph 40 of their closing, {A/5/15}. They suddenly
say Mr Coyne made some requests in December or renewed
his requests in December. But, my Lord, by then it was
the ground rush to trial.
Two points that it is important to make clear about
their complaint about disclosure. The first point is in
actual fact, in my respectful submission, the claimants
have had the disclosure that they could legitimately
expect to have. Their real complaint is not that they
have not been given the disclosure they need, their
complaint is that it was disclosed late.
But, my Lord, that brings me to my second point,
which is why was it disclosed late? The answer is it
was disclosed late because they didn't take the
opportunity that they had agreed that they would take in
the terms of the fifth CMC order.
If, as I say, the claimants' only objection is one
of delay, at whose door should responsibility for that
delay be laid? In my submission, it should not be laid
at the door of Post Office. Post Office tried to comply
with the order. It did comply with the order and it got
no response from the claimants, and it does not lie in
the claimants' mouths in those circumstances to say that
between August and December Post Office should have been
busying itself voluntarily disclosing documents it
thought were a dead letter because it had no response.
My Lord, there is the third submission that it is
important to bear in mind as well, because although in
my submission the only real complaint that can be made
about disclosure is one of delay and, as I say, if
a complaint is to be made it is to be made of the
claimants, not of the Post Office, but at the same time
the claimants also wish to maintain a suggestion that
the reason why this has been happening, the reason why
they didn't get these documents when they needed them,
was because Post Office is exhibiting a lack of candour
about its case. That's lawyer's speak for trying to
conceal the truth from the court.
My Lord, nothing can be further from the truth. In
my respectful submission, it is an outrageous submission
for the claimants to make in circumstances where they
had the power in their own hands, an agreed power, to
ventilate this issue in a way that they were required to
do pursuant to the pilot scheme practice direction and
they chose not to take it for whatever reason.
My Lord, just a final point on lack of candour. It
really is important to note that the documents that have
been disclosed, the KELs were disclosed in May last
year, they have had the KELs for well over a year. As
soon as the Horizon Issue trial was ordered, KELs were
agreed to be disclosed.
No order of disclosure for PEAKs has ever been made,
but the PEAKs were identified in the Post Office's EDQ
back in December 2017 and steps were taken. It was
difficult to extract the PEAKs from the proprietary
database on which they are held by Fujitsu and software
had to be written in order to allow them to be extracted
so that they could be used in a usable way. And that
cost quite a bit of money, as I understand it, and that
took time, but there was no resistance to giving them
the PEAKs. No order has actually been made but PEAKs
were provided as soon as they became available, and they
became available in the September.
It is in the submissions. I'm going too fast.
My Lord, OCPs, OCRs, MSCs, I have to confess at the time
I didn't really understand what these documents were.
I'm speaking for me. But, again, once Mr Coyne's report
had been served and it became clear what the
significance of these documents were, urgent steps were
taken actually to secure that these documents were
available. And the dates on which they became available
were set out later on in these submissions.
But, my Lord, this is not a story of Post Office
seeking to hold back material that's unhelpful to its
case. If that had been Post Office's objective, it
wouldn't have run around like a headless chicken trying
to get these documents for the claimants without any
even threat of a court application.
What's more, the effect of these documents has not
been unhelpful to Post Office. In fact, they have been
helpful. The PEAKs, its OCRs and the OCPs and the MSCs
all show, first of all, how few bugs there have been
that are relevant for the purposes of at least Horizon
Issues 1(a) and 3, and secondly, how minuscule the
exercises of remote access have been in the context of
a system of this size.
Those documents are actually helpful. There has
been no smoking gun. There has been nothing that has
been revealed that changes the whole complexion of the
case, save insofar as it reveals things and has been
useful for the arguments I'm now making to
your Lordship.
So I would invite your Lordship to view with some
scepticism an entirely understandable and, I have to
say, entirely typical submission from my learned friend
to the effect that Post Office is engaging in a process
that is inappropriate and deserves judicial criticism in
relation to this particular instance. In my submission,
such a suggestion is grotesquely unfair.
Then, my Lord, just going on through the
submissions. It is worth adverting to, but only very
briefly, paragraph 1146: the decision of the Court of
Appeal in Serafin v Malkiewicz. A very different case:
a question of procedural unfairness to a party. And the
Court of Appeal naturally drew attention to the fact
that the judge made criticisms of the party for not
disclosing documents that that party had not been
ordered to disclose. And I would invite your Lordship
similarly to refrain from making criticisms of that
sort.
MR JUSTICE FRASER: I am glad you have mentioned that case
because that was one of the questions I had for you.
That is under the heading "The Court's interventions on
disclosure".
That case concerned demands and criticisms by the
judge during the trial. Now, I'm unaware of making any
criticisms or demands during this trial.
MR DE GARR ROBINSON: I'm going to interrupt your Lordship
with the greatest of deference.
MR JUSTICE FRASER: Well, I'm interrupted regularly by both
of you, so ...
MR DE GARR ROBINSON: I want to make it clear that this case
is not cited for the purpose of affecting your Lordship
in any way, or suggesting to your Lordship in any way
that that has happened during the case of this trial.
And I would like to reassure your Lordship of that fact.
MR JUSTICE FRASER: I'm not necessarily seeking reassurance,
I'm just seeking clarification because 1146 {A/6/377}
says that that passage is relevant and it is under
a heading "interventions on disclosure". Now, when
I read it I went back and reminded myself not only of my
approach to disclosure during this trial, but Common
Issues trial as well, and I did in the Common Issues
Judgment make certain observations in respect of
managing the group litigation generally.
But so far as this trial is concerned, I think
I have only been involved on three occasions. One was
to invite you to perform a redaction review.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: One was to require a witness statement
of explanation in respect of Royal Mail.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: One was to require a witness statement
of explanation in respect of a large quantity of
documents which came, I was told, originally in May, May
of this year. But I do not think I have criticised
anybody. I certainly haven't intervened. But if I have
I would like you to identify those occasions to me.
MR DE GARR ROBINSON: My Lord, I do not think you have
criticised. No, I would go further. If you had I would
remember, and you haven't. The purpose of my reliance
on this case -- this is a terrible thing, I'm throwing
my learned friend under a bus here --
MR JUSTICE FRASER: No, please don't throw anyone under
a bus.
MR DE GARR ROBINSON: It is that the importance of this
case, this paragraph -- it is a paragraph -- is simply
the principle that my learned friend is making
criticisms which are based upon the very error that was
identified by the Court of Appeal in this case.
That's all.
MR JUSTICE FRASER: Well, that's rather my point. But it is
not necessary to spend too long on it. But what
occurred in that case were interventions during the
trial in respect of disclosure said to be deficient, in
respect of which there was no order.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: What Mr Green did and his team did in
their opening is they made a number of complaints in
respect of the known error log, which we call the KELs,
the PEAKs etc. And as you rightly point out there have
been no applications in respect of that. It is more
what could be said to be a generic complaint about the
Post Office's approach generally, which I understand
your submission to be is effectively made for forensic
reasons, is what it comes down to.
Let me put it slightly differently. I'm generally
unimpressed with a party complaining about not receiving
disclosure if they haven't made a specific application.
Equally, during this trial there have been certain
features of disclosure that have developed, for example,
Royal Mail and the recent bulk disclosure when Fujitsu,
I think you explained, had forgotten that there was some
archive documents --
MR DE GARR ROBINSON: This is the OCRs?
MR JUSTICE FRASER: Yes. Which so far as the court is
concerned I'm always interested in having an explanation
in a witness statement. But I'm not aware I have
criticised the Post Office.
MR DE GARR ROBINSON: Nor am I, my Lord, and I'm happy, I'm
glad your Lordship has given me the opportunity to make
that clear.
The simple point that I make, and I'm glad I have
come to this so we have had this discussion, the simple
point that I make is that just as it would have been
inappropriate were your Lordship to have done that, by
the same token it is inappropriate for my learned friend
to make those criticisms in the hope that your Lordship
is going to echo them.
MR JUSTICE FRASER: I understand. I mean, I also
appreciate, so far as disclosure is concerned generally,
that the Post Office has been reliant on Fujitsu to
a large extent.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: But now that we have the Fujitsu
Post Office contract, it is obvious that Fujitsu has got
a contractual obligation in respect of assistance so far
as documents are concerned, but those things take time
obviously.
MR DE GARR ROBINSON: Yes, and whether that obligation --
I was going to raise the question of what that meant
about control, but my Lord, I'm not in a position to do
that.
MR JUSTICE FRASER: I do not think it is necessary, to be
honest. But the short point is, and I think it is
mentioned once or twice in this passage in your closing,
that the Post Office was reliant upon Fujitsu.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: And as you say, for example, the PEAKs,
they had to develop the certain piece of software to
extract them.
MR DE GARR ROBINSON: Yes, my Lord. I believe that's also
the case with OSPs, and also MSCs have to be -- a lot of
work had to be done in order to -- your Lordship will
have heard Mr Coyne complain about the form in which
MSCs were disclosed. Well, they would have been
completely unusable if effort hadn't been made to put
them into spreadsheet form. And actually, in relation
to that it is worth mentioning that it is possible, it
was possible on the spreadsheet to actually download
individual documents and drop them onto paper in the way
that I did, or was done at my request, so that one could
see precisely what the documents showed.
My Lord, I see that it is 12.58. I have still got
perhaps 10 or 15 minutes on disclosure.
MR JUSTICE FRASER: Do you want to stop now?
MR DE GARR ROBINSON: This may be a convenient moment.
MR JUSTICE FRASER: All right. We will come back at
1.58 pm. Thank you very much.
(12.59 pm)
(The short adjournment)
(1.58 pm)
MR DE GARR ROBINSON: My Lord, two matters of housekeeping
before I proceed.
First of all, I have handed up a series of
typographical -- they are purely typographical --
corrections to our closing submissions. Some of the
internal paragraph references hadn't been updated when
the document was served and there were some false
references. An amended version of the submissions will
be uploaded onto Magnum, but I imagine that
your Lordship will have already started marking up the
version that you have got, so a copy of just the
amendments that have been made has been provided and
that's available to your Lordship.
MR JUSTICE FRASER: I actually got two hard copies. I was
working off the one that came from your solicitors,
although I did get one from you. Yours was in two
volumes and theirs was in one, which is why I used
theirs not yours. But I assume they are both exactly
the same.
MR DE GARR ROBINSON: I would be very surprised if they
weren't.
MR JUSTICE FRASER: That's why I mentioned it. These are
corrections to the submissions?
MR DE GARR ROBINSON: Yes. They are purely typographical.
It is things like footnote 5 should be GDCC, and things
like that.
MR JUSTICE FRASER: Thank you very much.
MR DE GARR ROBINSON: My Lord, the second housekeeping
matter is something I have been taken to task on over
the short adjournment.
When addressing your Lordship on what the impact
would be of having a trial of the reconciliation and
transaction correction processes in Post Office that
have existed over the last 20 years, they have changed
quite a bit during the course of those years, as
I understand it, I told your Lordship it would be over
1 million documents and could take six months. I have
to say that was a product of my brain.
My instructing solicitors have very properly
indicated to me that rather than just firing off the hip
with a prediction of that sort one should be rather more
careful, and really my submission should have been if
those things were to be included in the Horizon trial it
would not have been possible to have covered them within
the timetable that was available by any means.
MR JUSTICE FRASER: All right.
MR DE GARR ROBINSON: So I stand corrected in relation to
that.
My Lord, disclosure. I was on the question of
disclosure and I addressed your Lordship on the
standards that I submitted the claimants were, in my
submission, inappropriately seeking to impose on
Post Office.
Before leaving the question, I would like to compare
the standards that the claimants have been seeking to
impose with the standards that they are willing to
accept for themselves.
My Lord, if we could pick it up in the disclosure
section of our written submissions, and if we could pick
it up, please, at page 382, paragraph 1171 {A/6/382}:
"By paragraph 5 of the Fourth CMC Order, Cs were
ordered to provide disclosure of documents upon which
they intend to rely at Horizon Issues Trial ..."
At that time, it pointed out in 1172:
"... that the Fourth CMC Order provided for witness
statements to be limited to 'any witness of fact whose
generic evidence (in distinction to Claimant-specific
evidence) they wish to rely upon for the purposes of
determining the Horizon Issues'."
So those two orders went together.
The reason why there was a limitation in the order
that was applicable to the claimants was because the
claimants weren't expected to be given the same sort of
evidence that the Post Office would be giving.
That point is made in paragraph 1173 {A/6/383}, and
I invite your Lordship to read that.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: Then picking it up at {A/6/383},
paragraph 1174:
"On 17th July ... Cs provided disclosure of 45
documents. At this time it was not known to Post Office
that Cs would be serving extensive claimant-specific
witness statements."
But certain slightly ominous sounding statements
were made and Post Office sought to explore that in
1175, explaining that:
"... 'additional relevant documents may come to
their attention, and most likely as a result of the work
of experts'."
So no mention was made of claimant-specific evidence
that was to come, but they did say it was not their
intention to make another round of disclosure.
Then there was correspondence in which my
instructing solicitors sought to press my learned
friends on the scope of the disclosure which had been
given, and your Lordship will see the submission that's
made in paragraph 1178 on page {A/6/384}.
There is a suggestion that there was a desire to
avoid a repeat of late disclosure of documents close to
trial which had happened at the Common Issues trial.
My Lord, I don't know about that.
Then we then come to 28th September where the
claimants serve nine witness statements, including
several claimant-specific witnesses. Those witness
statements for the first time exhibited 17 new
documents -- that's in paragraph 1180 -- and a total of
30 new documents were disclosed at that time.
MR JUSTICE FRASER: Is this with the witness statements that
means?
MR DE GARR ROBINSON: My Lord, I believe so, yes. At the
end of paragraph 1180.
Then, your Lordship will see quite a lot of
correspondence where my instructing solicitors are
pressing for information as to the scope of the
disclosure that has been taken.
Then at 1182 there is a reference to the letter of
14th January:
"... 'Given the nature of the Horizon Issues Trial
there was good reason for the court to seek to limit the
ambit and cost of disclosure from the Claimants
themselves. Hence, the disclosure order made here; the
obvious good sense of that has not changed.'"
Well, my Lord, your Lordship will anticipate what my
response to that is. Everything changed when the
claimants served substantial evidence of their own which
they weren't supposed to, and that's the point that's
made in paragraph 1183.
At paragraph 1184 it is pointed out that as
claimants had flouted the order that your Lordship had
made about evidence, Post Office sought to understand
what searches, if any, had been conducted by claimants
so that it could understand whether further disclosure
would be required.
Several requests were made.
1185:
"This lack of response by the Cs should be viewed in
light of the approach adopted by Post Office. One
example of the further disclosure that Post Office has
agreed to give to Cs which is outside that ordered by
the Court is the requests for disclosure made by Cs on
18 December 2018 in which the Cs sought ..."
Your Lordship will see the following categories of
documents including:
"... 'disclosure of the documents that were
responsive to searches by, and collated by the Defendant
in respect of the operation of branches by Angela Burke,
Aakash Patny, Anup Patny, Jayesh Tank, Setpal Singh and
Adrees Latif. We would expect such documents to include
but not be limited to."
Then there is a whole series of documents which the
claimants were demanding.
So your Lordship will see the double standards that
are being applied by the claimants. On the one hand
they are standing on the orders, or rather the lack of
orders, that have been made against them, but at the
same time although no order has been made against
Post Office either, they are demanding all manner of
further documents.
Further requests for claimant-specific disclosure on
4th February. Post Office responds on the 11th and
provides disclosure on the 20th:
"The failure by [claimants] to provide their own
claimant-specific disclosure suggests that they clearly
intended disclosure to be a one-sided exercise."
{A/6/386}
My Lord, there is a further observation I would
respectfully make about that. The observation being
Post Office are criticised for only producing documents
on 20th February. My Lord, in my submission, the double
standards explicit in that criticism takes the breath
away.
On the one hand claimants refuse to give disclosure
themselves or explain what disclosure has been given
further to the few documents they have or provided. On
the other hand they are demanding more and more
documents from Post Office. And one can understand why.
I'm not criticising them for that. What I'm criticising
them for is for applying double standards between
themselves and Post Office.
Then at 1187 the claimants write to Post Office
indicating there is a document which is helpful to the
claimants, Mr Tank's Yahoo -- what is it called? --
Yahoo group forum document which he has found. He has
found a post. The point is made just for consistency,
notwithstanding the claimants' complaints about delay,
it is pointed out in paragraph 1187 that there is no
explanation of delay by the claimants. This is a double
standards point, not a complaint point.
My Lord, there is then a series of paragraphs
dealing with various documents that appear to exist. At
page {A/6/387} there is a long part of the transcript of
Mr Tank's cross-examination.
About halfway down the page, where Mr Tank says:
"Answer: I kept all my Post Office sort of related
paperwork in a box file and that's -- when I was asked
to look for evidence I went strictly to that box file
and that's where I sourced all my information from."
So when asked "How do you know all this?" he says
there is a box file that's got this information.
Similar points arise at the top of page {A/6/388}.
A lot of reliance on this box file. Amongst other
things, 1192:
"... it was Horizon generated receipts, print-outs,
with hand-written dates and reference numbers on them."
My Lord, a similar summary of the evidence of
Mr Latif where he discusses the records that he had
access to to help him change his evidence. My Lord,
that led to a request for disclosure which your Lordship
will see on paragraph 1196 on page {A/6/390}.
My Lord, just to cut a long story short,
your Lordship will see at paragraph 1198, {A/6/391}, the
claimants refused:
"... to provide further disclosure of the requested
documents on the basis that they were either outside Cs'
control, no longer existed or were not relevant to the
[Claimants'] evidence and not relied upon at trial."
My Lord, if one compares the approach that
Post Office adopted to providing disclosure relevant to
the allegations made by the claimants against them with
the approach that the claimants have adopted in relation
to the same issues, one does see distinct divergence, in
my submission.
The claimants appear happy to stand behind the fact
that no order has been made against them, whereas they
won't for a moment countenance the possibility of the
fact that no order has been made as being at all
relevant to Post Office's position.
To be fair, Post Office hasn't stood on the orders
either. Virtually all of the disclosure about which the
claimants make so much complaint is actually all
voluntary. But still Post Office is accused of lack of
candour.
My Lord, before leaving the question of disclosure,
and in particular disclosure by the claimants, I would
respectfully submit that that reinforces the point that
I hope your Lordship already has loud and clear from our
submissions in relation to how inappropriate it would be
for your Lordship to make findings that
claimant-specific witnesses were the victims of
undetected bugs or other defects in Horizon causing them
trouble.
In principle --
MR JUSTICE FRASER: You mean in a Horizon Issues judgment?
MR DE GARR ROBINSON: Yes. My Lord, in principle that is
a matter for --
MR JUSTICE FRASER: None of the Horizon Issues go to that
at all.
MR DE GARR ROBINSON: Well, my Lord, the reason why
I mention it is because your Lordship is invited to make
all sorts of findings by the claimants and I'm quite
anxious that your Lordship should understand, and
I imagine you already do, that we say that would be
an entirely inappropriate --
MR JUSTICE FRASER: Well, I'm trying the Horizon Issues and
I am answering the Horizon Issues. There will obviously
be a lot of detail in the judgment prior to arriving at
the answers to those issues. The idea that a specific
finding could be made on the evidence before this court
on a specific shortfall for a specific claimant in
a specific amount allocating it to a specific bug seems
to me to go outside the Horizon Issues.
MR DE GARR ROBINSON: I'm obliged to hear that from
your Lordship.
MR JUSTICE FRASER: But I'm surprised it is a point of
enormous difference between you given the orders, given
the CMC orders about what the trial is about.
MR DE GARR ROBINSON: My Lord, I agree. It is surprising.
But if your Lordship were to go to the claimants'
closing submissions, you will see that you are being
invited to make all sorts of findings. And one of the
most remarkable ones I think is in relation to Mr Latif.
Your Lordship will recall there is said to be
a difficulty about a stock unit transfer in or around
July 2015.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: The evidence demonstrates, in my
submission quite clearly, if your Lordship were to make
findings on it, which I say you shouldn't, the evidence
demonstrates quite clearly that there was no such
problem during the period identified by Mr Latif.
The claimants are suggesting your Lordship should
make a finding that it happened at some other
unspecified time that's not covered by any disclosure,
it is not covered by any evidence, in circumstances
where Mr Latif himself didn't say it happened at
a different time.
There is an Alice in Wonderland reaction to those
submissions, but it is right that your Lordship should
understand -- I can see that your Lordship does -- our
response to the submissions of that kind. In my
submission, it would be entirely inappropriate.
MR JUSTICE FRASER: Well, the way in which I deal in the
judgment with claimant-specific issues is obviously
something I am going to have to consider carefully and
deal with carefully. But so far as the findings on the
Horizon Issues are concerned, it is very clear on the
face of the order and the discussion that led to the
order that they are not going to be claimant-specific.
They are generic.
MR DE GARR ROBINSON: I'm grateful for your Lordship's
indication.
MR JUSTICE FRASER: That's not to say that I have
a concluded view now that I should ignore all of the
claimant-specific factual evidence, but I'm fairly
clear, as with the Common Issues in fact, that the
answers arrived at in the judgment are the answers to
the issues that I am trying. I would have thought
that's a fairly elementary approach.
MR DE GARR ROBINSON: To the extent that -- the experts --
Mr Coyne doesn't base any conclusions on the witness
evidence, as far as I'm aware. He refers to the witness
evidence as being consistent with his views. But
nowhere -- I hope I'm correct, I am sure I am correct in
saying. The fact that I'm saying I am sure I am correct
in saying means I'm not sure. But I think I am correct
in saying that nowhere does Mr Coyne conclude that
a particular bug existed because a particular
claimant-specific witness has said something.
At most, what he does is he says, well, I noticed
Mrs Burke had problems with reconciliation, or I see
that Mr Latif says this, that and the other. But he
doesn't, first of all, base any conclusions on the basis
of their evidence, nor should he. Secondly, nor does he
suggest that their assertions flow from a bug that he
has identified that could be the source of
a particular -- I think I'm right in saying he doesn't
specifically say bug 13. This was in existence at that
time and that's when we had Mr Latif's problem.
My Lord, that's not how his evidence is constructed,
as I recall. In those circumstances it would be
highly -- your Lordship as got my submissions. There
hasn't been proper disclosure. If there were to be some
findings about what the claimants did, what they
suffered, there would need to be an investigation of
wider issues about how the relevant branches were
operated. One would need to go into the history. One
would need to investigate all sorts of things.
These snapshot witness statements, most of which are
almost impossible to understand because they are so
short and bereft of detail, one would need to go into
far more material in order to be able to be in
a position to make findings at all. If your Lordship
were to make findings, in my submission the only proper
findings would be to dismiss the claims that are made.
But my primary submission is that your Lordship should
not entertain them in the first place.
My Lord, that completes my submissions on
disclosure.
Let me now come as briefly as I can to the bugs that
my learned friend addressed you on yesterday. What
I would like to do as briefly as possible is to talk
about bug 11, bug 12, bug 13 and bug 15.
Bug 11 is Girobank. Your Lordship will recall that
Post Office contends that there's no evidence of
financial impact from the relevant bugs. Your Lordship
may or may not -- I suspect not -- be alive to this
already, but bug 11 raised in appendix 2 raises six
issues. My learned friend didn't deal with all of those
six issues in his submissions yesterday. He only
addressed Issues 1 and 2 and he referred to a principal
PEAK which is at {F/25/1}, and perhaps I could ask
your Lordship to go to that very briefly.
My Lord, this is opened on 5th May. The claimants
argue that Post Office hadn't grappled with the concept
of what a discrepancy is, and I'm informed that that's
in the transcript at {Day21/101:1}. If that is their
submission, in my submission it is not correct.
Post Office has grappled with what a discrepancy is,
but the word "discrepancy" means more than merely
a comparison between one thing and something else, as
the claimants suggest.
The thing and the something else are relevant to the
Horizon Issues. They can't be any two things. Here
what's required is a discrepancy in branch accounts.
Now, the PEAK notes that there was a £505
discrepancy between a branch's cash account, and the
cash account at this period of time means its own
accounts, and its daily reports.
Your Lordship will see that from the bottom of
page {F/25/1}. 5th May at 15.02:
"New evidence added - Full message store.
"... Response:
"This difference (£505.72) between the Cash Account
and the Daily reports is explained by," a particular
KEL.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: The discrepancy there is not
a discrepancy in the branch's accounts. It is
a discrepancy between the accounts and the report that
was sent to Girobank every night. The issues discussed
in the PEAK are reporting issues, they are not account
issues.
Indeed, Mr Coyne agreed the following in
cross-examination, that the detection and investigation
of Issues 1 and 2 in this PEAK demonstrated the good and
effective operation of robustness countermeasures in
Horizon. My Lord, that was at {Day17/52:1} to 56, and
pages 63 to 64.
He also accepted that the PEAK was not evidence of
a transaction correction or error notice being issued to
the subpostmaster in such a way as to subject him to
a risk of loss. That is {Day17/65:1}. My Lord, that
summarises my submissions on bug 11.
If we could move on to bug 12: counter replacement
causing one-sided transaction. The claimants raise
a specific issue, namely that the KEL at F421, that's
J Ballantyne 5328R, does not make reference to issuing a
transaction correction, or error notice, or a BIM
report.
If we could go to that, it is at {F/421/1}. It is
true that the KEL does not make reference to a BIMS or
to an error notice or transaction correction. However,
usually KELs don't do that. They don't include specific
instructions in relation to BIMS or TCs. It is
a standard process which goes beyond what's required in
one specific KEL.
If a SPM needs to be made good it is a standard
process, and in paragraph 12 of his third witness
statement Mr Parker said:
"This process of identifying a bug, then identifying
its effects and then remedying those effects is not
special to Horizon. It is a standard part of any IT
support practice." {E2/13/3}
Having said all that though, look at the bottom at
page 1, just before "Evidence" {F/421/1}. About three
lines from the bottom, it says:
"For a multi-counter outlet ... need to retrieve the
messagestore from another counter, as well as the
affected counter. For MCO (and SCO), transaction
numbers for the RiposteVersionString messages should
reveal the original transactions. When you have
identified any missing transactions attach the details
to the PinICL," that is an old name for PEAK, "and route
to MSU."
Now, the reference to MSU, there has been quite
a bit of evidence about it but for present purposes
I can simply rely on Parker 1. If I could ask
your Lordship to go to {E2/11/12}. It is paragraph 44.
Starting at the second sentence, Mr Parker says:
"There was (and is) a process run by the Management
Support Unit (MSU) which involves examination of various
system reporting and may result in Business Incident
Management Service (BIMS) entries going to Post Office.
An incident may also be raised by MSU with the SSC to
provide support to the MSU in resolution of the BIMS."
My Lord, the MSU is the body within Fujitsu which is
involved in the production of BIMS.
So the fact that the MSU was being involved by the
SSC here actually is an indication that what could end
up happening is that a BIMS report will go from MSU to
Post Office.
My Lord, secondly in the context of this very bug,
it is worth looking at {F/77.1/1}, one of the other
PEAKs. Same phenomenon. If one goes over, this is
24th November 2000, so a similar period. One goes to
page {F/77.1/3} of that PEAK. The bottom green box
starts at 5th December 2000 at 10.43:
"Response:
"4/12/2000 ... By John Moran - MSU."
It is worth noting.
MR JUSTICE FRASER: I can't see where you are.
MR DE GARR ROBINSON: It is halfway down the page.
MR JUSTICE FRASER: Have you still got black and white or
have you got colour?
MR DE GARR ROBINSON: No, I have green now. It is the green
box. If I didn't make that clear, I do apologise.
5th December at 10.43.
MR JUSTICE FRASER: Yes, I've got that. Oh, I see, yes.
MR DE GARR ROBINSON: So MSU is involved.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: "This incident has the same cause as
one to be recorded on the SIL ..."
I'm afraid I don't know what SIL means:
"... under BIM."
And there is a number. You see MSU is involved in
BIMS. Then if one goes down to the bottom of the box,
about two-thirds of the way down there is a line that
starts "These two credits". Does your Lordship see
that?
MR JUSTICE FRASER: Yes, equal £167.12.
MR DE GARR ROBINSON: "As they were written over they were
not added to the cash account.
"Final BIM issued.
"Please close this call.
"Closing as Reconciliation resolved."
Here you have evidence first of all of BIMS being
sent in relation to precisely this bug, and second you
have evidence of the MSU being involved. So that
disposes of the submission made by my learned friend
about that.
Withdrawn stock next. That is bug 13. If I could
pick it up at {F/765/1}. Your Lordship may recall that
this is a problem where stock is withdrawn. It is no
longer being sold by Post Office. Branches are supposed
to rem out the relevant stock and send it back to
Post Office, but what sometimes happened is that they
don't rem it out, they just send it to Post Office and
the problem is Horizon doesn't actually have a pair of
eyes. If you don't record the rem, then Horizon doesn't
know that the stock has been removed, and then there is
a problem because you have stock which Horizon thinks is
held at the branch, and it is historic stock, it's stock
which Post Office no longer sells. And that creates
problems in itself and what to do about that stock and
so on.
This was a case where the SPM did not follow the
correct procedure. The underlying issue explained of in
the PEAK is not a bug in withdrawn stock, but it is the
SPM not following correct procedure in remming out the
stock that's being withdrawn.
One gets that, my Lord, if one looks at page 1 of
this PEAK halfway down the big yellow box:
"This office physically held 137 £5 Post Office
stamps and did not rem them out before the date the rem
out icon disappeared."
That will be when, I presume, reference data ceased
to be applicable to in relation to that particular
stock:
"The office physically returned the stamps to
Transaction Processing as advised and the office then
did a Trading Period balance on 17 November 2010 and
showed this value as a loss."
Now, what my learned friend did yesterday was go
like a magnet to the second line and said: look, look,
transaction processing advised them to do what they did.
It must be a bug. I don't know if your Lordship recalls
that?
I think the quote from the transcript is:
"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."
That is {Day21/110:14}. But your Lordship will see
from this very sentence that the SPM had not done what
he was supposed to do. He had not remmed out the stamps
in the first place. So it is quite right that the SPM
had been advised to return the stamps, but they had to
also rem them out. That is fundamental to the way that
the Post Office business works and it is commonsense
really because, as it says, Horizon doesn't have eyes.
So, my Lord, again, a very brief submission in
relation to that.
Phantom transactions is more complicated. There
were three documents that my learned friend referred to
yesterday. The first was -- well, a PEAK at F/97, then
at PEAK at F/100.1 and then a PEAK at F/88.2. These
PEAKs are relied on with a view to casting doubt on the
conclusion set out in those PEAKs that the problems were
not probably attributable to bugs in Horizon.
If we go to {F/97/1} first, your Lordship may recall
that I cross-examined Mr Coyne about this, who would
only rely upon the Romec engineer point. Does
your Lordship recall the Romec engineer point?
MR JUSTICE FRASER: I do.
MR DE GARR ROBINSON: That was the point he had referred to
in his report and I had taken him to the end of the PEAK
in which Patrick Carroll had set out his overall
conclusion.
My learned friend very fairly went to a different
page of the PEAK, page {F/97/7}. Before going there, it
is worth noting that this is a master PEAK that doesn't
just relate to one particular branch, it covers a number
of different incidents with different branches. And
that's an important part of the context.
If one goes to page {F/97/7}, that's relevant to
what's said in the box that my learned friend referred
to. It is 19th June, 15.17, about two-thirds of the way
down. My learned friend understandably relies on the
sentence:
"This has been seen at Old Iselworth (OI) and Wawne
... 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 ends Wendy is
arranging for installation of Kit at OI on Friday ..."
So it is the Old Isleworth site 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 ..."
So this is 19th June, the installation took place on
22nd June. This entry followed up on 20th July at
13.40. That's at page {F/97/8}. It says:
"Comtest readings have been correlated with perfmon
stats and a recommendation to install resitive screens
at Old Iselworth has been made."
So that is the last entry relating to this site on
this PEAK. But it is picked up in {F/100.1/1}, if
I could ask your Lordship to go to {F/100.1/1}. I'm
sorry to jump about like this. Your Lordship will see
the last entry was 22nd July. This picks up at the top
of the page, 25th July 2001 at 9.39.
MR JUSTICE FRASER: 25, yes.
MR DE GARR ROBINSON: Does your Lordship see that in the top
green box?
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: Third line down:
"24/07/01:
"PM reports that he has been having phantom
transactions ..."
Over the page, here is Patrick Carroll's conclusion
about all of this {F/100.1/2}. 5th September:
"Following a significant amount of monitoring ..."
Stopping there. That clearly is the case. Mr Coyne
himself I think accepted that Fujitsu had worked quite
hard to get to the bottom of what these reported
problems were:
"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 phanton Txns,
these pointed to a touch screen problem and as a result
the screen was replaced with a Resitive model. As this
produced no measurable improvement it has to be assumed
that the problems were user related."
So that's Mr Carroll's considered opinion having
done all the tests and changed all the equipment that
could be changed.
If one goes back to the previous master PEAK at
{F/97/9}, his overall conclusion at 12th November 2001
at 9.48 at the bottom of the page:
"Phantom Txns have not been proven in circumstances
which preclude user error.
"In all cases where these have occurred a user error
related cause can be attributed to the phenomenon.
"I am therefore closing this call as no fault in
product."
My Lord, that's the conclusion of Mr Carroll who is
on site and who is the experienced Fujitsu operative.
In my submission, that opinion should not be rejected.
So far as is relevant to the Old Isleworth site it
is also worth noting something about the postmaster
there. If one goes to {F/88.2/1}, this is
15th February 2001. So we were in July 2002, weren't
we? I'm so sorry, we were in September 2001. That was
the last entry for Isleworth in the previous F/100.1,
but this started in 15th February 2001.
If one goes to the green box at the top about
halfway down:
"Outlet went live ... requested further training,
the PM was referred to his RNM ... It would appear that
there is a training issue here which needs to be
addressed."
Then PON actions:
"Has PM completed and passed his training?
"When, where and with whom did PM complete training?
"Has further training been considered?"
One sees that the postmaster seems in need of help.
Then at page {F/88.2/3}, 20th September, 12.07, the
main box:
"PON have written to the RNM to address the training
issue, see text below.
"From RNM - 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 txns
(discrepancies?) 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."
So the changes made didn't seem to change anything
but then miraculously the transactions stop. What's
more, it seems that this postmaster is the kind of
person that needs more support than perhaps you would
expect for more postmasters.
MR JUSTICE FRASER: What does "PON" stand for?
MR DE GARR ROBINSON: My Lord, I would have to take
instructions.
MR JUSTICE FRASER: Don't worry, I will give you a list of
three-letter acronyms later on.
MR DE GARR ROBINSON: I shall look forward to that.
MR JUSTICE FRASER: I will give it to both of you so you are
not the only person who has to ...
MR DE GARR ROBINSON: Unless I can assist your Lordship
further, those would be my submissions on the PEAKs that
my learned friend went through yesterday.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: Then, with your Lordship's permission,
I propose to deal with a number of miscellaneous points.
First of all, number of criticisms made of Mrs Van
Den Bogerd's evidence. First of all, she was criticised
for her treatment of the Helen Rose report and she
accepted some criticism in her evidence, and my learned
friend places some emphasis on that in his submissions.
In my respectful submission, her witness statement
was clear and fair. It wasn't right to criticise her.
If I could ask your Lordship to go to her witness
statement. It is {E2/5/34}. This is to do with the
Helen Rose report.
Your Lordship will recall that Mr Coyne in his first
report made a number of criticisms based on the
Helen Rose report, including the criticism that
a transaction correction had wrongly been sent to the
postmaster in that case.
At the bottom of page 34 Mrs Van Den Bogerd refers
to paragraph 5.175 of his report. She says:
"The extracts taken from the report by Helen Rose
... are taken out of context and mistakenly claim that
the relevant reversal was issued in error by Horizon,
not the Subpostmaster. The Rose report makes it clear
that:
"154.1. The concerns were based on the fact that
reversals were not being shown on the particular data
sets reviewed / reports typically run by Subpostmasters
in branch on Horizon.
"154.2. Transaction reversal data can be extracted
from Horizon;
"154.3. The issue was therefore surrounding how the
transaction reversals were displayed / accessible in
branch and that there was no issue with Horizon itself."
Then 155:
"There is therefore no indication that the reversal
was not notified to the subpostmaster. When recovery
was carried out a discontinued session receipt would
have been printed and messages would have been clearly
displayed to the User in branch during the recovery
process."
My Lord, in those paragraphs Mrs Van Den Bogerd is
not saying that the postmaster reversed the transaction.
She is saying that what happened in the branch was part
of the reversal process and was correct. And she was
saying that the SPM knew about it.
In my submission, the evidence shows that both those
points were true.
MR JUSTICE FRASER: Both which points, sorry?
MR DE GARR ROBINSON: First of all, that the reversal was
part of the recovery process operating normally, and
secondly, the SPM knew that the recovery process was
operating because he received the printed receipts.
MR JUSTICE FRASER: You are saying that's what's on the face
of her witness statement? Are we at the point of her
witness statement now rather than her cross-examination?
MR DE GARR ROBINSON: Yes, I'm talking about her witness
statement. I'm not going to go to her
cross-examination, my Lord. What I'm submitting to
your Lordship is that there was no basis for criticising
her --
MR GREEN: My Lord, I'm sorry to interrupt but she agreed
that was wrong in her cross-examination.
MR DE GARR ROBINSON: What she said in her cross-examination
was that she didn't intend to suggest that the SPM was
responsible for the reversal and that's not what she
says in this witness statement. So she was right to
have that intention.
MR JUSTICE FRASER: All right.
MR DE GARR ROBINSON: So, my Lord, that's point 1 on Mrs Van
Den Bogerd.
Point number 2, paragraph 183 of the same witness
statement {E2/5/42}, she says:
"Before the change to Horizon Online, a cash check
was completed in branch by the HFSOs. Branches were
notified in advance that this cash check would be
carried out. I recall that this mandatory cash check
across the entire network caused a temporary spike in
declared losses. I suspect that this was due to
branches tidying up their accounts before the cash check
and therefore losses coming to the surface that had
previously been ignored or covered up."
My Lord, it's always dangerous for counsel to say
this, but I don't understand what the criticism of
Mrs Van Den Bogerd is here. She is speaking from her
own recollection and she remembers a spike in declared
losses during the period of the mandatory cash check,
which of course is before the migration to
Horizon Online.
She is not purporting to state a considered view
that having investigated the matter this must be the
cause. She specifically qualifies what she says by
saying:
"I suspect that this was due to branches tidying up
their accounts."
So she lived through the process and she suggests
a possible reason. My Lord, that's not speculation on
which she needed to be challenged by reference to
Fujitsu documents to which my learned friend sought to
take her.
My Lord, furthermore it is worth noting that
Mr Coyne himself agreed, and had already agreed, that it
was a likely factor in the spike along with a possible
increase in bugs during migration to Horizon.
My Lord, that's in JC2, {D2/4.1/219}. Perhaps we
could go to that.
Paragraph 5.345. Mr Coyne has referred to certain
statements made by Dr Worden in the previous paragraph,
you can see the end of it is in italics, and he says:
"In my experience, a major change to a platform will
almost always lead to an increase in bugs, errors and
other issues. Therefore, I do not agree that the
mandatory cash check was 'more likely' to be the cause
of the spike in 2010. The most likely scenario is that
both of these were factors."
So my learned friend's own expert says it was
a factor.
Mrs Van Den Bogerd is talking about the period
before migration when the mandatory cash checks were
being done. My Lord, in those circumstances it was not
necessary to take Mrs Van Den Bogerd to any documents to
challenge her evidence. Her evidence was perfectly
understandable and needn't have been challenged at all.
And that's my very brief submission about that.
My Lord, a third point about Mrs Van Den Bogerd's
evidence is that in paragraph 35 of the claimants'
submissions there is a suggestion that she is basing her
evidence on a document that she didn't actually have
because it didn't come into existence until after her
witness statement was made.
I don't criticise the claimants for doing this but
they have done this quite a lot. I can tell
your Lordship on instructions why it was that the cash
declaration that they refer to in paragraph 35 -- it is
not necessary to go to it -- was dated 1st March when
her witness statement was dated the previous year, in
November the previous year.
MR JUSTICE FRASER: 1st March this year then, you mean?
MR DE GARR ROBINSON: Yes. She couldn't explain that in
cross-examination.
My Lord, on instructions I'm in a position to
explain it now. During preparation for the
cross-examination of Mr Patny, my instructing solicitors
realised that they didn't have a copy of that particular
declaration and so they requested it from Post Office.
That was on 1st March. My Lord, that was the document
that was then hyperlinked to Mrs Van Den Bogerd's
statement. She doesn't specifically say "I refer to
a cash declaration". The fact is she had seen a cash
declaration. She had seen it a long time before, but it
was the same data just obtained on a different date.
So the fact that the date on that particular
document post-dates the date of her witness statement
doesn't mean that she didn't have sight of the same
information previously; she certainly did have. That's
why one sees it. She makes a number of statements about
it which are correct.
My Lord, a similar claim made in the claimants'
written closings about the evidence of Mr Membery, who,
as your Lordship is aware, was unable to give evidence.
It is at paragraph 123(c) of their submissions. It is
to the effect that Mr Membery refers to a document that
has been produced to him by Mr Lenton but the metadata
on the document suggests that the document was produced
by someone called BH.
My Lord, I can tell your Lordship on instructions
that the version of that document received by my
instructing solicitors from Fujitsu recorded that it had
last been modified by Mr Lenton. How BH's name got on
it and who BH is they have no idea, and my instructing
solicitors are happy to provide that document in native
form if my learned friend would wish to see it.
My Lord, another miscellaneous point. Criticism is
made of Mr Johnson in my learned friend's closing
submissions. He is criticised because he didn't know
the source of a screenshot. My Lord, it did not matter
where the screenshot came from. What mattered was
whether it was a true screenshot. It was not from the
Horizon guide, but Mr Johnson did not claim that it came
from the Horizon guide.
If I could ask your Lordship to go to {E2/4/2}. At
paragraph 10 you will see that Mr Johnson says:
"The screenshots that appear in this statement are
primarily taken from a document called Post Office
Onboarding Counter Guide ... Where a screenshot has been
taken from another document I refer to that document."
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: My Lord, Mr Johnson didn't know where
the particular screenshot had come from but he did know
that it was a proper screenshot, and that's the short
point.
And indeed it was. My Lord, another criticism made
of the claimants' witnesses relates to the
cross-examination of Mrs Burke. It was suggested in the
claimants' closing submissions that because Mrs Van Den
Bogerd's witness statement was only amended shortly
before she went into the box, that enabled my learned
friend Mr Draper to criticise Mrs Burke on the basis of
a false apprehension of the facts to be collected from
the unamended witness statement.
My Lord, this related to the fact that in Mrs Burke
had undertaken three transactions for several different
customers all in one basket, which of course is not
proper procedure. And she explained why. She very
fairly explained why in her witness statement: it was
because Horizon was playing up.
But what was suggested to her was, first of all,
that was a breach of procedure as it evidently was, and
secondly, it was suggested to her that by doing that she
increased the risks of unfortunate things happening.
For example, if there was a failure it might well be
that the failure would relate to a transaction in
relation to which the customer was long gone.
My Lord, that was all that was put. No criticism
was made of Mrs Burke and, more importantly, no
suggestion was put to her that her having undertaken
a series of different transactions all within one
basket, that had a causative effect on the issue that
she then faced when there was a system outage on
9th May.
My learned friend's suggestion that a false case was
improperly put to Mrs Burke on the basis of an unamended
witness statement of Mrs Van Den Bogerd's is simply
unfounded, and it is an unfair criticism and
I resist it.
My Lord, I'm going to finish with some very brief
submissions in relation to audits. A number of
submissions are made about audits in the claimants'
closings. I can pick them up really at paragraph 620 of
the claimants' closings, which is at page {A/5/217}.
Picking it up at (c) at page {A/5/220}:
"The purpose of these audits is to provide
assurances to Post Office and its auditors about the
risk of material misstatements in Post Office’s
financial statements. The audits themselves expressly
state that they 'should not be used by anyone other than
these specified parties'."
So there is a claim.
Your Lordship will recall that there were a number
of financial audits done by Ernst & Young for
Post Office, including in 2011 and 2013. They are the
documents that spring to my mind as being relevant.
MR JUSTICE FRASER: I think you just read from page 217 and
you said page 217, but I followed it perfectly
adequately.
MR DE GARR ROBINSON: I'm grateful. I read from page 220
and I may have given the wrong page number.
MR JUSTICE FRASER: No, you said 220 but in mine it is 217.
But that could be because I'm using the hard copy.
MR DE GARR ROBINSON: I see. My hard copy version seems to
be different, I'm so sorry.
MR JUSTICE FRASER: Don't worry. Just pause one second.
On the common screen we have got page 217, which
starts with paragraph (a) and runs down to (g) and has
621(a) and (b), which is what my version has. Have you
got a different version?
MR DE GARR ROBINSON: My version has page 220.
MR JUSTICE FRASER: I do not think it has come up before so
I do not think it is an issue. But I have got the right
paragraph and I followed you reading it.
MR DE GARR ROBINSON: I'm grateful.
But in subparagraph (c) {A/5/217}, a factual claim
is made, which is that there are Post Office's financial
audits is one thing and then there are the service
audits that were done for Fujitsu. And the suggestion
is being made that the service audits were just about
Post Office's financial statements.
My Lord, in my submission that is demonstrably
wrong, and if we could look at the document referred to
in the submission, {F/1041/9}, please:
"Intended use.
"This report, including the description of tests of
controls and results thereof in the Description of Tests
and Results, is intended solely for the information and
use of Fujitsu, POL as the user of the IT support
processes and controls used by and on behalf of Fujitsu
to support the HNG-X and POLSAP applications during some
or all of the period ... and the independent auditors of
POL, who have a sufficient understanding to consider it,
along with other information including information about
controls implemented by user entities themselves, when
assessing the risks of material misstatements of user
entities' financial statements."
So your Lordship will see two things. First of all,
the intended users are not just the auditors when
auditing. The intended users include Fujitsu itself and
Post Office as the user of the IT support processes
provided by Fujitsu. That's point 1, and point 1
demonstrates that the submission made in (c) that we saw
before was incorrect.
Point 2 is wider really, which is there is
an interesting, how can I put it, tension in the
claimants' attitude because on the one hand they are
very happy to trumpet the reliability of Ernst & Young's
financial audit when referring to the 2001 E&Y
management letter, but when it comes to a document which
is helpful to Post Office they suddenly say, well, it is
just financial misstatements, it's got nothing to do
with the price of fish.
So, my Lord, those are the two submissions I derive
from that page.
Finally, it is worth adverting in this context to
the evidence which Mr Coyne gave on {Day16/188:1}. We
needn't go to it, but on that page Mr Coyne accepted
that the service audits are more specific than general
financial audits.
My Lord, if we can go back to the closings, that
will be at {A/5/217}, I believe, we see another claim
that's made by the claimants. (d):
"The control objectives and controls are selected by
Fujitsu rather than the auditor ..."
What the claimants are trying to do here is they are
faced with a series of audits that essentially are very
positive about Fujitsu's approach to the IT support that
it is giving, all the way from 2012 through to 2017.
So here they are trying to think of points which
will somehow undermine the value of those audits. And
what they are suggesting in (d) is that somehow Fujitsu
is marking its own homework. It has self-developed the
test which it is to be inferred therefore has no real
value and cannot be taken as a justification for
anything.
But, my Lord, first of all, there is no proper basis
for suggesting that Fujitsu would have developed tests
that had no value in this context. Secondly, if one
goes back to the document, so this is {F/1041/8},
please -- I'm so sorry, I'm afraid I haven't marked up
this document. If your Lordship will give me a moment.
MR JUSTICE FRASER: Don't worry.
MR DE GARR ROBINSON: Yes. Top of the page. This is
Ernst & Young describing the work it was doing:
"An assurance engagement to report on a description
of a service organisation's system and the suitability
of the design and operating effectiveness of the service
organisation's controls to achieve the related control
objectives stated in the description involves performing
procedures to obtain evidence about the fairness of the
presentation of the description and the suitability of
the design and operating effectiveness of those controls
to achieve the related control objectives stated in the
description."
My word, that is a six-line sentence:
"Our procedures included assessing the risks that
the Description is not fairly presented and that the
controls were not suitably designed or operating
effectively to achieve the related control objectives
stated in the Description."
So your Lordship will see the auditors weren't
simply taking a test that was given by Fujitsu and
blindly applying it, they were also assessing the risks
involved in the relevant functions and assessing the
suitability of the design of the controls.
Then it goes on:
"Our procedures also included testing the operating
effectiveness of those controls that we consider
necessary to provide reasonable assurance that the
related control objectives stated in the Description
were achieved. A reasonable assurance engagement of
this type also includes evaluating the overall
presentation of the Description, the suitability of the
control objectives stated therein and the suitability of
the criteria specified by the service organisation and
described in the Assertion."
So what Ernst & Young are saying there is that not
only did they audit the processes to ensure that the
control objectives were met, but they also evaluated the
suitability of the control objectives in the first
place.
So this wasn't simply them jumping over a hurdle
that Fujitsu had very conveniently placed for itself.
It was Ernst & Young evaluating the value of that very
hurdle itself. I hope I put that point clearly.
MR JUSTICE FRASER: Yes. As I understand it, Fujitsu
specify the control objectives and the service auditors'
responsibilities, which actually start on the previous
page, on page {F/1041/7}, are described in detail in
those two full paragraphs?
MR DE GARR ROBINSON: Yes. Those functions include
evaluating the objectives in the first place; in other
words, making sure that the objectives that they have
been told to test for are appropriate objectives to
test for.
My Lord, that's my submission on the suggestion that
appears to be made in paragraph (d) in paragraph 620 of
the closing submissions, that there's something wrong
with the objectives because Fujitsu selected them.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: My Lord, (f) in paragraph 620 on
page 217 of the closing submissions:
"Much of the content of these audits has been simply
cut and pasted from one year to the next. (Eg for
control objective 10, which Dr Worden relied upon, the
tables within the audits for 2013 and 2014 are
identical, and likewise for 2015 and 2016 ..."
My Lord, difficult to know what to make of that
submission.
It is added to in (g), where it is said:
"Dr Worden said he had noticed that quite a lot of
the wording was very similar or identical from one year
to the next ..."
There appears to be an intention, I may be wrong,
but so far as I can tell it may be intended to achieve
the implication that somehow Ernst & Young weren't doing
a proper job, it was simply repeating words it had said
previous years. If that is the suggestion, there's
absolutely no basis for it. Given that they were every
year evaluating and assessing similar objectives it is
not surprising to see them use the same language if
their views remain the same.
My Lord, if that's not being suggested I don't
understand what the relevance of the submission is and
I invite you not to understand it as well.
My Lord, that leaves a submission in (e), which is
that:
"The audit expressly excludes the auditor giving any
opinion on application processing and application
controls."
My Lord, difficult to know what that submission
means. In my submission, the control objectives and the
conclusions drawn in relation to those control
objectives mean what they say. It wouldn't be right to
somehow construe Ernst & Young's opinions as somehow
having no application to Horizon system or the like if
that's the intended implication.
My Lord, finally, paragraph 621 {A/5/217}. This is
where the claimants cut to the chase.
They say:
"The only one of the EY audits which significantly
advances the parties' understanding of the issues in
this litigation is in fact the 2011 EY audit ..."
My Lord, all these comments in relation to the
service audits which actually are more specific than the
financial audit in 2011, all these arguments are
designed somehow to dismiss the value of the service
audits. And in my submission it wouldn't be right to do
that.
Secondly, it wouldn't be right also to suggest that
all that matters is the 2011 EY audit. What should also
be borne in mind, for example, is the 2013 EY audit.
I needn't go to the document, but for your Lordship's
note the reference is {F/1138/1}.
At page 4 of that document it records that
management action has addressed many of the issues
raised. At page 7 of that document it praises the
efforts taken to strength the control environment.
My Lord, in my submission, that advances the parties'
understanding of the issues in this litigation.
My Lord, your Lordship may be really rather
exhausted by the miscellaneous points/submissions that
I have to make. I do have time to make one final
submission and then I will stop, you will be pleased to
know.
MR JUSTICE FRASER: I'm not remotely exhausted by them.
MR DE GARR ROBINSON: Perhaps it is just me.
My Lord, it is a criticism that's made quite
forcefully in the closing to the effect that Post Office
didn't behave with candour because when it had
corrections for its witness statements it would produce
a document setting out those corrections rather than
producing a red-lined version, an amended witness
statement.
I say this some trepidation because your Lordship's
experience may be completely different from my own, but
I have to tell your Lordship that neither I nor members
of my team have ever encountered an amended witness
statement that has been amended in the way the
claimants' witness statements are amended. I have to
say personally I feel a certain amount of discomfort,
just because of habit, I think, of amending witness
statements that way.
So suggesting that Post Office is at fault for not
adopting a practice which to me is an innovation is, in
my submission, rather unfair. It is inappropriate in my
submission to make criticisms of Post Office because its
legal team did things in the traditional way.
MR JUSTICE FRASER: Well, I have seen them done in three or
four different ways and the way you have adopted was not
unconventional.
MR DE GARR ROBINSON: I'm grateful.
My Lord, I would finish on this point by suggesting,
by speculating really, what would Mr Roll's witness
statement look like if it had been amended to reflect
the evidence he gave in open court.
Unless I can assist your Lordship further, those are
my submissions.
MR JUSTICE FRASER: Your closing submissions in toto rather
than the end of the miscellaneous section?
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: All right. Well, I have got some
questions for you. We have also got a bit of
housekeeping, but it is not going to be very much.
MR DE GARR ROBINSON: Actually, on housekeeping matters,
there has been some discussion between Mr Green and
Mr Henderson about the suggestion that your Lordship
made yesterday about a mechanism for identifying the
source for various statements contained in appendix 2.
My Lord, we would suggest the simplest route would
be for the claimants to provide us with a version of
appendix 2 which marks in highlight particular passages
for which they don't understand the support relied upon.
If that's provided by a certain date then my team
can respond by a certain date by explaining the source
with a footnote.
MR GREEN: My Lord, I did mention to my learned friend
Mr Henderson that we are already partway through
producing overnight, the efforts of the entire team,
a table which is confined only to those points
your Lordship mentioned, which is where no evidential
support is identified.
MR JUSTICE FRASER: All right, I'm not going to deal with
this now. We are going to have a short break for the
shorthand writers. We will come back, I won't deal with
housekeeping straightaway because I have some questions
on your submissions.
MR DE GARR ROBINSON: I should not have --
MR JUSTICE FRASER: Which I will deal with then and then we
will deal with housekeeping.
There's that point plus another one and then we are
going to deal with the rest of 2019.
MR GREEN: My Lord, we have a couple of references by way of
factual correction only to hand up to the court.
MR JUSTICE FRASER: We do. Or I do, not using the royal
plural -- these breaks are for the writers, they are not
for me or for you, although I know most people look
forward to them, particularly people who are not on
their feet. But I don't think it is going to take
longer than about 20 minutes. But I think out of
fairness I will break for them and then we will have the
20 minutes.
MR GREEN: Of course.
MR JUSTICE FRASER: In view of the fact we are so gloriously
ahead of schedule, let's come back at 3.25 pm.
(3.15 pm)
(A short break)
(3.25 pm)
Housekeeping
MR JUSTICE FRASER: The first point I actually confirmed
with you during your submissions. It was that the
Horizon Issues were drafted and agreed by the parties
and approved by me. I do not think they were imposed on
the parties. But you confirmed that this morning.
I'm fairly sure I know the answer to this question
but I just want to be very clear just because the nature
of the word might, or has been so contentious. Insofar
as the place I go for a benchmark definition of
robustness, I intend, unless you tell me I should go
somewhere else, to go to your pleading because you
define it in your pleading.
MR DE GARR ROBINSON: Sorry, I'm not sure I understand
your Lordship's question.
MR JUSTICE FRASER: Well, robustness, the experts have
agreed a particular meaning. I asked Mr Green where
I should go for his meaning yesterday and he told me,
made submissions.
I would ordinarily, and I have looked at your
pleading quite carefully. Where you do define
robustness?
MR DE GARR ROBINSON: Do you know, I'm afraid I need to
remind myself. Could we go to my pleading, my Lord?
It is a slightly impertinent thing to ask the judge.
MR JUSTICE FRASER: I am sure it is your pleading, but
I might be wrong.
My private screen seems to have stopped working for
some reason. It will be in the C bundle somewhere.
MR DE GARR ROBINSON: {C3/3/1}. It will be referred to in
the Horizon Issues, won't it? I'm so sorry, my Lord.
MR JUSTICE FRASER: Not at all.
MR DE GARR ROBINSON: Mr Draper is suggesting paragraph 16
on page 5, so he'll get all the blame if he's wrong.
MR JUSTICE FRASER: So {C3/3/5}, paragraph 16.
MR GREEN: My Lord, yes.
MR JUSTICE FRASER: I didn't bring the reference to your
pleading with me but I did look at it very recently, so
I'm pretty sure that's likely to be right.
MR DE GARR ROBINSON: So what's pleaded in paragraph 16 is:
"Like any other IT system, Horizon is not perfect,
but Post Office maintains that it is robust and that it
is extremely unlikely to be the cause of losses in
branches. Its design and technical controls, 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."
Is that the paragraph your Lordship had in mind?
MR JUSTICE FRASER: On the basis I haven't got that
paragraph you have just read in front of me on either
screen. Just to put your mind at rest, it is not
designed to be a trick question.
MR DE GARR ROBINSON: No, and I'm not trying to be clever
either, my Lord. Every time I do try I always come a
cropper.
MR JUSTICE FRASER: I have now lost it off the common screen
as well. Give me one second.
The short point is that the experts in different
places discuss robustness in more or less expert terms.
They also agree robustness in one of the joint
statements. I asked Mr Green the specific question
yesterday where I go for his benchmark definition of
what robustness is and I wanted just a similar reference
from you.
MR DE GARR ROBINSON: My Lord, I would not say that
paragraph 16 is a definition. I would submit that
a system that is robust is extremely unlikely to be the
cause of losses in branches.
MR JUSTICE FRASER: Quite.
MR DE GARR ROBINSON: It has the features referred to in the
rest of that paragraph {C3/3/5}.
My Lord, my submission would be that it is
consistent with the answers that Mr Coyne gave when
I asked him to unpack what he meant by "relatively
robust" when he identified comparable systems, and then
he identified the condition under which comparable
systems needed to operate and he said Horizon compared
well with those conditions.
MR JUSTICE FRASER: I know, and I have gone through that in
some detail.
MR DE GARR ROBINSON: I'm grateful.
MR JUSTICE FRASER: In fact, probably the best thing to do,
and it is really just a request for a reference, I would
just be grateful for a reference from you for the best
place to go for -- I used the phrase loosely
yesterday -- a benchmark definition of robustness. If
one is describing robustness in specific terms this is
what it is.
MR DE GARR ROBINSON: My Lord, I will --
MR JUSTICE FRASER: It is just going to be a reference.
I do not think it is this passage in your pleading
actually because I was looking in your pleading, I was
looking in all the pleadings, and that is not the one
that jumped out at me. But I'm not going to tell you
where the best place is, that's rather why I was asking
the question.
MR DE GARR ROBINSON: My Lord, I will try to resist the
inevitable urge that barristers are subject to --
MR JUSTICE FRASER: I don't want a fresh definition.
MR DE GARR ROBINSON: Exactly. I will try and avoid all of
that. But it may be I will be referring to a number of
different documents.
MR JUSTICE FRASER: That's perfectly in order, but I just
want somewhere to go to so when I'm explaining, probably
quite early on in the judgment, what robustness is,
I can explain by reference to a specific -- the way in
which the defendant construes the word "robustness",
because it is obviously different from what the
claimants are suggesting.
MR DE GARR ROBINSON: I need to remind myself of where the
claimants' definition is, but that's my problem.
MR JUSTICE FRASER: I do not think I got one, but I got
an explanation of why there wasn't one.
So that was supposed to be an easy starter, that
one. We will see about that.
The next point is essentially -- I suppose it could
be seen as an extension of the appendix 2 exercise, but
yesterday Mr Green took me to your closings at page 68
and paragraph 147.4.
MR DE GARR ROBINSON: 68?
MR JUSTICE FRASER: Yes. Paragraph 147.4.
MR DE GARR ROBINSON: My Lord, I have it. {A/6/68}
MR JUSTICE FRASER: Which what it says about the adoption of
section 1.4 of the relevant document and when it was
adopted. I beg your pardon, when it was implemented.
Now, that's either evidence or it is a submission.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: He says it is a submission, it wasn't in
any of the evidence.
MR DE GARR ROBINSON: My Lord, it is my instructions but,
my Lord, it is based upon a previous version of the
document and then an amended version of the document.
And I will undertake to give your Lordship the two
references.
MR JUSTICE FRASER: And the dates on documents.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: If you could give me those references
that would be great.
MR DE GARR ROBINSON: If my understanding of those documents
is incorrect I will withdraw the --
MR JUSTICE FRASER: All right. That's helpful, thank you.
The next point is -- and you won't be able to give
me this now, but just if it could be added to the
references I'm going to be sent -- could I have
a reference, please, to the ruling that I made at the
PTR about Mr Henderson -- the witness, not counsel --
and the Second Sight report. Because within the
transcript of that hearing when one gets to my ruling
there is just a bracket that says "see separate ruling".
I am sure it is there somewhere, I just don't know
where it is and I would like to know.
MR DE GARR ROBINSON: We will look for it.
MR JUSTICE FRASER: Appendix 2 we will come onto in
a moment.
There are the following acronyms I would just like
to know what they stand for. Two are from today, one is
yesterday. Yesterday's was RPOS. So R-P-O-S, RPOS.
Today's were SIL and the one this afternoon, PON.
I would just like to know what they are.
MR DE GARR ROBINSON: My Lord, yes. It will be something
like retail point of sale --
MR JUSTICE FRASER: I was tempted to guess and I decided
I wouldn't and I would like to know what they are,
please.
MR DE GARR ROBINSON: That will be done.
MR JUSTICE FRASER: Then the appendix 2 exercise, and this
again is not supposed to be controversial but I think
from your side, Mr de Garr Robinson, you said can they
highlight it with highlighter and they may have started
doing the exercise in a different way. Is that right,
Mr Green?
MR GREEN: Yes, we have a table just identifying roman
paragraphs.
MR JUSTICE FRASER: I do not think anyone is going to go to
the stake on the difference.
MR DE GARR ROBINSON: If they want to do unnecessary work
I won't stop them. I was anxious that there shouldn't
be any hidden submissions, but my learned friend has
assured me that isn't the position so it is just
a matter of dates.
MR JUSTICE FRASER: Of dates, yes. Well, actually, there is
more than just dates, there's something I want to make
clear because I'm very anxious to do two things. One is
not to create unnecessary expensive work and the other
is not to lead to fresh rounds of submissions.
When you are responding to that, Mr de Garr Robinson
and your team, it seems to me there are one of three
available alternatives: a reference to a witness
statement; a reference to an expert's report; or just
the word "submission" to demonstrate that it is
a submission.
MR DE GARR ROBINSON: My Lord, what about a reference to
a document or a reference to commonsense?
MR JUSTICE FRASER: All right. Well, let's have a look at
appendix 2.
By reference to commonsense, do you mean a statement
of the blindingly obvious?
MR DE GARR ROBINSON: Yes. Or words to that effect.
MR JUSTICE FRASER: All right. Well, then we will add the
following alternatives: reference to a document and then
in brackets it has to be said if that document was put
to anyone or not.
MR DE GARR ROBINSON: Okay.
MR JUSTICE FRASER: And then the fifth one, I will adopt
your nomenclature and we will say commonsense.
MR DE GARR ROBINSON: I'm grateful.
MR JUSTICE FRASER: So those are the five alternatives.
MR DE GARR ROBINSON: I understand.
MR JUSTICE FRASER: The ones we were shown yesterday, I'm
not necessarily sure any of them jump out at me as
commonsense, but as I said yesterday some of them, for
example, the notion of slave units and a master unit,
that is commonsense within the IT world and I have
a degree of commonsense in that respect. And if,
following the commonsense answers, I need any more
information, then I will ask.
MR DE GARR ROBINSON: I see. So, for example, slave and
master, I am sure there are documents about that because
I'm familiar with the concept, yes.
MR JUSTICE FRASER: All right?
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: I would just like to make this clear for
everyone's benefit. This is not designed to be a new,
expensive, lengthy grinding exercise. It is just so
I have everything marshalled in one place.
Right, well, that's everything from me. Mr Green,
you were proffering a piece of paper and I asked you to
take your seat.
MR GREEN: My Lord, just a couple of points of
clarification.
MR JUSTICE FRASER: Thank you.
MR GREEN: Yesterday my learned friend said that the double
trouble document references, he thought that many of
them he hadn't seen at all before. So they are all the
references to where they are in the evidence in these
proceedings, including in the Post Office's own
submissions. That's the first page.
MR JUSTICE FRASER: Yes.
MR GREEN: Drafting of Horizon Issues 1 and 3, the short
point is the words were carefully chosen. We were
asking for caused as a separate issue, and potentially
caused. That was resisted by the defendant. In the end
we got a careful agreement on what the wording actually
is and my learned friend's speculation about potential
being what he said is just not available when one looks
at how that happened. And (c) is 18.7 should be million
in shortfalls. We are not able to, at this stage,
disaggregate the extent to which they were caused by
bugs, errors and defects, erroneous TAs or TCs, helpline
advice etc at this stage, but that is the figure in the
SOCIs.
And disclosure, my learned friend said regarding the
disclosure issues, a dead letter from August to
December, there are the intervening references and then
we just added this afternoon in handwriting in relation
to EY audits, Dr Worden, his understanding of
application processing meaning SSC was out of scope and
the reference to the transcript where he said that.
MR JUSTICE FRASER: Yes.
MR GREEN: In the ISAE audits and then over the page just to
give the context to the KELs' complaint, the history of
them said not to be in Post Office's control. There are
lots of other references, we only put four there. There
was obviously an RFI about it as well. It was
identified as an issue in the pleadings and maintained
in the CMC in October 2017 when your Lordship was
specifically asked about it.
MR DE GARR ROBINSON: Why is my learned friend -- just going
through this list, obviously I will need to look at the
drafting of Horizon Issues 1 and 3, but your Lordship
has my submissions on what the document means.
My Lord, application processing. My submission
about that is that what was excluded from the service
audit was any kind of audit by Ernst & Young of the
operation of Horizon itself, the software and how good
Horizon itself was. What wasn't excluded was the
controls that was applied by and on Fujitsu in relation
to the support it provided in relation to Horizon.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: It is important your Lordship has that
submission. If it is suggested that my learned friend
or that Dr Worden suggested otherwise, I will have
a look at that. But I would be surprised if that's what
he said.
Then there is this paragraph (f). I don't know why
my learned friend has even mentioned it. Is there some
objective being achieved by it?
MR GREEN: My Lord, it was just in relation to the
disclosure history which was an issue my learned friend
responded to.
MR JUSTICE FRASER: Can one of you just give me the
reference to the contract between Fujitsu and the
Post Office which I know was produced and put to
Dr Worden in re-examination?
MR GREEN: My Lord, the contract itself is at {F/1659.2/868}
I think is the one with paragraphs 25.8 and 25.10 and so
forth about court assist. Court support.
MR JUSTICE FRASER: I am sure if I consider it necessary to
look at it I will look at it. If I consider I need any
submissions about it, I will ask for them in writing.
Does that seem a sensible way forward,
Mr de Garr Robinson?
MR DE GARR ROBINSON: My Lord, yes. If your Lordship is
minded to make some kind of finding on the basis of the
contract, I would welcome the opportunity to make
submissions, although I have no idea what those
submissions would be.
MR JUSTICE FRASER: All right.
MR DE GARR ROBINSON: If you understand what I'm saying.
MR JUSTICE FRASER: I do. In other words, if I'm going to
look at it for any material reason I should give you
both, I would give you both the opportunity to --
MR DE GARR ROBINSON: I'm grateful.
MR JUSTICE FRASER: Because the document is there, I have
looked at it on a speed read basis. Only one part of it
was put to Dr Worden in relation, I think, to the cost
of making ARQ requests.
MR DE GARR ROBINSON: That was by me.
MR JUSTICE FRASER: By you. Well, I think you said when you
introduced the document you had had it uploaded for that
purpose.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: I didn't even know if it was in play
before that, to be honest.
Right. Just remind me, Mr Green, what does TA
stand for?
MR GREEN: The transaction acknowledgements, which are the
automatic ones from National Lottery.
MR JUSTICE FRASER: Because TCs used to be called error
notes for the first four years.
MR GREEN: They did, and then they came in, and then TAs are
the ones that you can't do anything about at all.
MR JUSTICE FRASER: So far as Horizon Issues, other than the
fact that there's a judgment to be written, is that
everything?
MR DE GARR ROBINSON: My Lord, I do hope so.
MR JUSTICE FRASER: I think it is. Apart from obviously
appendix 2 and the reference point. That then brings us
onto the second part of 2019.
At the moment there are the following dates in 2019
which are in the diary and have already been subject to
orders by me. 23rd July; 18th September; they are both
the CMCs. The 17th October, which is down as
a pre-trial review, and then a trial starting on
4th November on the limitation issues.
It does not seem to me that those dates are
necessarily either should stay there or are achievable
certainly, so far as a limitation trial is concerned,
because they were set when this trial was supposed to
finish at the very end of -- well, originally early
April and now it is obviously 2nd July, therefore,
I'll -- is this now your part of ship Mr Henderson?
Mr Green, what do you have to say about it, if
anything?
MR GREEN: I understand the solicitors on both sides have
discussed it and rather taken up your Lordship's
suggestion of maybe having a CMC in September rather
than July when we will know more, we are not sure how
much more, but we will know more and --
MR JUSTICE FRASER: When you say you will know more though,
the only piece of information I think you will know is
how far I'm on in writing the judgment because I do not
think I will have been able to finish the judgment by
the middle of September.
MR GREEN: My Lord, that's slightly why I am putting it
lightly in those terms.
MR JUSTICE FRASER: Right.
MR GREEN: We might have an idea of the timeline on that.
In those circumstances the parties solicitors' have had
discussions and subject to obviously your Lordship's
view, are agreeable to the November listing coming out
if it can't be used for proportionality reasons.
Your Lordship should know there are also discussions in
parallel about how holding a mediation --
MR JUSTICE FRASER: You don't need to tell me anything about
that. Well, at one point I think before the Common
Issues trial when we were scoping the group litigation
generally, you asked me for a three-month stay and
I told you I didn't have one.
MR GREEN: I think there may have been a slight
misunderstanding but I know what your Lordship means.
MR JUSTICE FRASER: When I say don't tell me about the
mediation what I mean is unless you are asking me to fix
a timetable around periods of time, you don't need to
trouble me with that.
MR GREEN: My Lord only to have in mind that it would be
helpful not to have anything too compressed happening in
late October, that we can get to that in September.
MR JUSTICE FRASER: All right. When you said the November
trial coming out.
MR GREEN: My Lord, yes?
MR JUSTICE FRASER: There is another trial date in 2020.
MR GREEN: Precisely, for which we can consider directions
in September.
MR JUSTICE FRASER: All right. Okay Mr Henderson?
MR HENDERSON: My Lord, I'm grateful for the indication in
relation to the November trial and certainly the parties
are agreed that that's not feasible really to use that
date at the moment. The question really is when
your Lordship lists the further issues trial, which is
referred to in the orders.
If we are to use the current date which
your Lordship has put aside for the fourth trial, which
is four weeks commencing on 2nd March, the question that
really arises is: what steps need to be taken in advance
of that? Can I just ask your Lordship, I appreciate
your Lordship is, in light of the dates your Lordship
has just recounted to the court, you have probably
already looked at the seventh CMC order, but if I could
just ask your Lordship briefly to look at it. It is at
{C7/39/1}.
What your Lordship will see is the various steps
from paragraph 3 of that order. Does your Lordship have
that on the screen now?
MR JUSTICE FRASER: I have page 1. Is that what you are
asking me?
MR HENDERSON: Actually if you go to step 2 you will see the
steps that have been ordered. 3.1 and 2 have taken
place, but from 3.4 all the way up to 20.2 is currently
the subject of a stay. So where we --
MR JUSTICE FRASER: And that's the stay that was agreed
I think in the interregnum period?
MR HENDERSON: Exactly so my Lord, on 12th April. If
your Lordship casts an eye down that you will see that,
absent that stay, we would now be in the position of
having completed the first round of disclosure;
pleadings would be very nearly completed; principal
pleadings would be incompleted; we would have particular
claims and defences; replies would be shortly due and we
would also be about to embark upon the second round of
disclosure referred to as extended disclosure.
We would also be about to convene a costs and case
management conference on 23rd July. So there is a great
deal of activity and you will see that if you carry on
in those directions you will also see -- I appreciate
your Lordship has these points in mind -- that there is
also a directions for the identification of the issues
that would form part of the fourth trial and that's in
paragraphs 13 to 17.{C7/39/4}
All of that is currently stayed. We would
respectfully support the suggestion that the further
issues trial be vacated and re-listed for the 2nd March,
which is currently the slot that's being held for the
fourth trial, so in other words we use that slot for the
third trial.
The only question is as to how we arrive at detailed
alternative, updated directions for all of this. My
concern -- our concern is that if we use the
18th September CMC we might not be leaving ourselves
enough time because if nothing has happened and the stay
remains in place and nothing has happened and we come
before your Lordship on 18th September, we have only got
a relatively compressed period to do an enormous amount
of work; and my respectful suggestion would be that the
parties liaise now, try to agree those steps, but in
default of agreement or in default of the court's
approval of any such agreement that we arrive at, we use
the 23rd July appointment to try and --
MR JUSTICE FRASER: At the moment everything is stayed
except the Horizon Issues. You are not asking me to
lift the stay?
MR HENDERSON: I'm not asking you to lift the stay.
MR JUSTICE FRASER: But you would like the 23rd July kept
in?
MR HENDERSON: I am only asking that the November trial be
vacated and re-listed for 2nd March; that the parties be
directed to liaise to try and agree directions leading
up to that trial and in default of agreement, we come
back on 23rd July so that we can have a timetable in
place in July rather than in September leading up to
that --
MR JUSTICE FRASER: You mean have a timetable ordered in
July to cover the period July to March rather than wait
until September?
MR HENDERSON: Indeed so, my Lord.
MR JUSTICE FRASER: By definition, therefore, one of the
things you are going to be seeking either on 23rd July
by asking me or before that by seeking agreement of the
claimants is the lifting of the stay?
MR HENDERSON: Yes, I think it would be part of the
agreement.
MR JUSTICE FRASER: Unless I suppose you could both decide
that the stay will be lifted on date X and this
direction will start after that date?
MR HENDERSON: Absolutely. My learned friend has already
indicated that there is discussion of a possible
mediation, that needs to be borne in mind. There are
other things happening. Your Lordship obviously is
going to be preparing the Horizon Issues judgment.
There is the application for permission to appeal in
front of the Court of Appeal. The Court of Appeal has
indicated we won't get a decision on permission until
September/October probably.
MR JUSTICE FRASER: But that doesn't factor into anything
that is happening at first instance.
MR HENDERSON: I'm not suggesting that it does certainly for
the time being. So we need to get directions in place,
in our submission, leading up to, if your Lordship
approves, the 2nd March. Really I think the only point
of difference, and it is a slight one, is that we think
that using the September date rather than the July date
may be to save up problems for ourselves.
MR JUSTICE FRASER: Mr Green.
MR GREEN: My Lord, the only thing, the parties' solicitors
have agreed, subject to your Lordship's view, that it is
sensible for the pleadings for the further issues trial
to be done with the benefit of receipt of the Horizon
Issues judgment for obvious reasons.
There's also the question of whether the Court of
Appeal grants permission on the obligations, breach of
which and concealment of a breach of which --
MR JUSTICE FRASER: But that is from the Common Issues you
are talking about?
MR GREEN: My Lord, yes. Just highlighting what feeds into
what will have to be pleaded. For that reason the only
actual step --
MR JUSTICE FRASER: I don't follow that point at all but it
doesn't matter for the moment. Go on.
MR GREEN: Rather than trying to get the parties to agree
everything going forward between now and March now,
we'll come back on the 23rd --
MR JUSTICE FRASER: I'm not going to ask you to agree
anything now. I'm not going to make any orders now
other than varying any dates I have to order.
MR GREEN: I mean prior to the 23rd, my Lord. The only
actual substantial step, I think, prior to the pleadings
being done is the giving of stage 4 disclosure by
Post Office, which was to be done prior to the
pleadings. So if that can be completed in good time
prior to --
MR JUSTICE FRASER: All right, I think --
MR GREEN: It just avoids us coming back on the 23rd.
MR JUSTICE FRASER: Mr Henderson has got a valid point which
is, if you can't agree everything sensibly and you wait
until 18th September, you have only given yourself six
months.
MR GREEN: My Lord, the only problem on my learned friend's
submission, with respect, is that there's giving stage
four disclosure and then the pleadings start, which the
parties have agreed in discussions between solicitors
shouldn't happen until after the Horizon judgment.
MR JUSTICE FRASER: But Mr Green if it is as straightforward
as that and you can agree all of those directions
between the two sides in the next week or so, you do not
even have to come on the 23rd July.
MR GREEN: I was just trying to short cut it, my Lord. I'm
in your Lordship's hands.
MR JUSTICE FRASER: I'm going to tell you what orders I'm
going to make now and then each of you can tell me if
I have missed anything else and then there's another
point I have to address.
Order
The pre-trial review on 17th October for the further
issues trial has to be vacated. The trial date on 4th
November for the further issues trial has to be vacated.
The further issues trial will be tried in the current
time set aside starting on 2nd March with the same time
estimate as it currently has. The parties are to seek
to agree directions for the further issues trial by noon
on 21 July and failing agreement they are to attend for
a CMC at 10.30 on 23rd July.
I would like to hope that that is not going to be
necessary. It would be far cheaper for everyone if it
were not needed. Right. So is that 2019 tidied up?
MR HENDERSON: I think the only outstanding matter is the
18th September hearing.
MR JUSTICE FRASER: I'm leaving that in because until I know
what's happening on 23rd July -- I assume one of the
directions you might seek to agree between yourselves,
if you can agree everything else, is that the 18/9 can
come out but I am going to leave it in for the moment.
Right. The only other point which goes to or arises
from judgment number 5 is, as the managing judge and as
the parties have got cost management orders, I made
certain observations about costs and also said it would
have to be revisited at the end of the Horizon Issues
trial. Since then there has been one further
notification on costs which was from the Post Office,
who I think notified in a sum of 13.9 million it is now,
which is about £1 million more than it was when I made
the comments that I did in judgment no. 5.
I'm not going to make any orders or anything like
that but I would like the two of you, please, when
discussing and considering your draft directions in
advance of 23rd July to direct your minds to what, if
any, further costs in case management conferences and/or
costs management orders you might be considering and the
point at which this year that can be addressed in the
absence of agreement. Because obviously if there is
going to be a contested CCMC there are certain steps you
have got to take in advance for 28 days, 14 days, etc.
Right. Is that everything?
MR GREEN: Yes.
MR JUSTICE FRASER: Is that everything?
MR HENDERSON: I believe so. Two minor points. We have
agreed between counsel dates for the bug appendix point.
I do not think that needs to be subject to an order
but my learned friend has indicated he will serve his
comments by 9th July, a week today, and we are going to
aim to respond to that by 19th July. But I do not think
that needs to be in any order.
MR JUSTICE FRASER: No, but thank you for telling me, it
means I know when I'm going to get it.
MR HENDERSON: Presumably on the order that your Lordship
has just made it would be the usual common costs in case
and liberty to apply?
MR JUSTICE FRASER: Yes. Although they would be Horizon
Issue costs anyway given it has only been 25 minutes,
but by all means and could someone draw up the order
please. I hadn't seen Mr Warwick appear from behind the
screen.
Right. Is that everything?
MR GREEN: My Lord, it is.
MR JUSTICE FRASER: Mr de Garr Robinson, does that seem to
be everything?
MR DE GARR ROBINSON: It is.
MR JUSTICE FRASER: Mr Henderson, so far as the directions
are concerned I think that's everything?
MR HENDERSON: My Lord, yes.
MR JUSTICE FRASER: Thank you all very much.
(4.00 pm)
(The court adjourned)
(10.30 am)
Closing submissions by MR DE GARR ROBINSON
MR DE GARR ROBINSON: My Lord, good morning. I'm going to
start by seeking to draw your Lordship's attention to
some extraordinary features of the claimants' case as it
has developed before your Lordship during the course of
this trial.
First of all, I would like to remind your Lordship
of some points of which you will be well aware, and if
I can do it by reference first of all to the Horizon
Issues which is at {C1/1/1}.
The first point is that your Lordship ordered
an expert-led trial and your Lordship will see that from
the header to the Horizon Issues. The header is taken
from remarks that your Lordship made at the CMC on
22nd February, and that's at {C8.4/4/1}. Perhaps we
could have a look at that.
Sometimes one does rather miss the old days when one
had files, my Lord.
MR JUSTICE FRASER: That's the document. Which page?
MR DE GARR ROBINSON: It is at page 54. I don't have
a document.
MR JUSTICE FRASER: There is one on the common screen.
MR DE GARR ROBINSON: I'm so sorry. The reason why I said
that rather intemperate, which I now regret, is that
I see nothing on my screen. I wonder if someone could
look at that for me?
MR JUSTICE FRASER: Have you not got a screen now?
MR DE GARR ROBINSON: It is a black screen, my Lord.
Perhaps it is not plugged in.
MR JUSTICE FRASER: Well, that's unsatisfactory for obvious
reasons. It is not the best start. I think you need
a screen, but I can easily rise for a minute or two to
avoid putting pressure on anyone.
MR DE GARR ROBINSON: I would be very grateful to
your Lordship. I do apologise.
MR JUSTICE FRASER: You don't have to apologise. These
things just happen. Whenever they happen they are
always at a very inconvenient time, but I suppose at
least we haven't got into the depths of the day. We may
as well sort it all out at the beginning. 5 minutes?
MR DE GARR ROBINSON: 5 minutes.
(10.36 am)
(A short break)
(10.38 am)
MR DE GARR ROBINSON: My Lord, I'm hoping we can continue at
approximately this pace for the rest of the day. It
will be a much more enjoyable day that way.
We are on the 22nd February hearing and
your Lordship will be familiar with these words but
I will, if I may, read them out. You said:
"It is very unusual in case management to find
oneself having constantly to try and put either one or
other party back on track for cost-effective resolution
of serious disputes. I reminded myself, again by
reference to the actual transcript, of what I said last
time, although I had a pretty clear recollection, that
what I was going to be doing in March was to deal with
expert issues that were present on the pleadings
concerning Horizon which I described generically as the
next big issue. I wanted the parties to agree or each
propose an isolated number of issues on the pleadings
related to Horizon that would involve expert evidence
but not evidence of individual cases."
{C8.4/4/54}
My Lord, it was with those words ringing in their
ears that the parties agreed the Horizon Issues and
offered them to your Lordship for approval.
Disclosure was then given, hundreds of thousands of
documents were disclosed and reviewed by the experts.
They were disclosure of documents of an expert nature.
The claimants criticised Post Office on their disclosure
and Post Office criticises the claimants on their
disclosure, but for present purposes the important point
is a huge amount of disclosure was given by Post Office;
it was directed at informing the expert process and many
of the documents, I would say most of the documents,
called for expert review and comment.
The next important stage of the litigation was vast
expert reports being served. Mr Coyne's two reports run
to well over 500 pages without appendices.
They were many, many, many, hundreds of times
I would say, more detailed than my learned friend's
outlined allegation document that was produced in August
which no one has really referred to, and they took weeks
properly to assimilate. It is fair to say that the
Post Office was still trying to assimilate Mr Coyne's
second report when the trial actually began.
Your Lordship may appreciate how that felt at the time.
Now, the experts co-operated to identify the
principal agreements and disagreements between them, and
that was, in my submission, a very helpful process that
was overseen and encouraged by your Lordship and it led
to four long joint statements which in my submission
were helpful.
JS2 is particularly helpful in that it sets out in
the bug table what the experts say on each side.
Without that we would be digging through the interstices
of these interminable reports. It brought a welcome
measure of focus, and given we don't have detailed
pleadings or schedules in relation to specific bugs it
gave the parties an opportunity to decide how to
organise their cross-examination and proceed.
Your Lordship will be aware that the parties'
cross-examination was limited. I certainly would have
liked more time given the scale of the reports. But
given the other factual evidence that had been called,
I had four days; my learned friend took three.
Now, the Horizon Issues are expert issues and we
have had an expert-led trial as your Lordship ordered.
But anyone who came to court for the first time
yesterday might have formed a very different impression
of what this case was all about.
One expects oral closings to be about the oral
evidence to explain how the case has developed since the
openings now that everybody is much the wiser. But in
my learned friend's submissions there was almost no
reference to Mr Coyne's oral evidence or to Mr Roll's
oral evidence and no reference at all to the important
respects in which that oral evidence changed the picture
as compared with Mr Coyne's reports on the one hand and
Mr Roll's witness statements on the other.
I don't mention the fact that there were these
changes as a criticism of Mr Coyne but merely as a point
about the unusual approach that the claimants have
adopted, I would suggest felt constrained to adopt, in
the final days of this expert-led trial. My learned
friend was constrained to make many of his submissions
largely without reference to the conclusions reached by
the expert whom his clients had instructed and largely
without reference to the very substantial common ground
that was between the experts, both as set out in the
joint statements and was achieved during the course of
the cross-examination and, indeed, your Lordship's own
questions.
Important parts of my learned friend's submissions
were directly contrary to that evidence and those
agreements. For example, your Lordship will have heard
my learned friend say that Horizon is not robust.
My Lord, that, as your Lordship will be well aware, is
directly contrary to the experts' common views.
My learned friend tries to duck that by effectively
suggesting that robustness doesn't really mean anything.
But, again, that's contrary to what the experts say.
And your Lordship will see that from our written closing
at paragraph 361; that's at page 132. I'm not going to
take your Lordship to it now.
Mr Green didn't engage with any of that evidence,
nor did he engage with Mr Coyne's oral evidence on the
likely total number of detected bugs over a 20-year
period. Your Lordship will I am sure have well in mind
that he said it was no more than 40 including transient
bugs. So if one takes into account that of the 29 bugs
in the bug table he says only 22 were lasting, then one
assumes, one infers, that his view is that the number of
lasting bugs that are in the system would be around 30
over 20 years.
Nor did he engage with Mr Coyne's oral evidence on
the number of instances of remote access that he had
seen over that same period. Your Lordship again will
I hope have in mind that Mr Coyne said he had seen about
30 instances of relevant remote access for the purposes
of these proceedings.
Those numbers loomed large in any worthwhile
consideration of the key extent questions raised by the
Horizon Issues. But they weren't addressed by my
learned friend. Instead he gave examples of things. He
told your Lordship little stories and he relied on
snippets from documents. But what he didn't do is he
didn't attempt to draw back and ask the question, after
all the evidence that your Lordship has heard: where do
we end up on the expert evidence that has been given in
this case?
In my respectful submission, the inevitable
inference is that the claimants have considered very
carefully where we do end up on the expert evidence and
their considered view is that they would prefer the
court to look elsewhere. They would prefer
your Lordship to take an impressionistic approach, and
by that I mean the kind of approach under which all bugs
that caused doubling up can be considered as one and the
same thing, as my learned friend remarkably appeared to
be suggesting last night.
That suggestion was apparently made -- and I may be
being unfair to him, but it was apparently made to
provide some colourable basis on which to call into
question the criminal prosecution. My Lord, that's all
headline-grabbing stuff but it is nothing to do with
Horizon Issues and it is nothing to do with the
expert-led trial that your Lordship ordered and on which
the parties have spent so much time, money and, if I may
say so, sweat.
Now, I propose to focus on where we are left after
the trial. Most notably, in terms of Mr Coyne's
evidence on Horizon Issues 1, 3, 4 and 6 and on
Mr Roll's oral evidence as well. And your Lordship will
see that my submissions, I should say our submissions,
the submissions of myself, Mr Henderson, Mr Draper and
Ms Keating are set out quite fully and at length, and
I'm sorry for the length of the submissions we have
submitted to your Lordship.
That brings me to another feature of the case which
I submit is also extraordinary. In their written
submissions, their closing, the claimants say, I think
more than once but they say it triumphantly, that
Mr Roll was right. What they mean is the Mr Roll that
was portrayed in the witness statement that he would
have had drafted for him was right. But, my Lord, in my
submission a very different Mr Roll came through when he
gave oral evidence, and just as the claimants make no
reference to Mr Coyne's oral evidence during the course
of this trial, they similarly seek to distract, if I may
say so, certainly pay no attention to, the oral evidence
that Mr Roll gave.
Just to remind ourselves quite how far we have come,
perhaps I could ask your Lordship to have a quick look
just at his first witness statements at {E1/7/1}.
Rather than reading out long chunks perhaps I could ask
your Lordship to read particular paragraphs; first of
all, paragraph 7 {E1/7/2}. Then the last sentence of
paragraph 8. The first sentence of paragraph 10. The
last sentence of paragraph 11. And then, my Lord,
finally paragraph 19 {E1/7/3}.
That's a brief selection, but they are what could be
described as headline-grabbing statements made by
Mr Roll.
In my respectful submission, your Lordship will have
heard Mr Roll and will have seen our written closings,
and from our written closings I respectfully submit that
your Lordship should conclude that all of those claims
have effectively gone. For your Lordship's note these
are addressed in pages 38 to 63 of our closing
submissions.
Here are some highlights. Mr Roll accepted that his
recollection of 70% of the work he did could be wrong
because one tends to remember non-mundane things more
clearly than mundane things.
As regards time pressure on the SSC, that was
a feeling from a long time ago and he said his
recollection was quite hazy and he also said that he had
that feeling not very often. He said that when he said
he spent 70% of his time looking for faults on data
stores he was not looking for software bugs. Remarkably
he accepted that coding errors causing financial impact
on branch accounts in his experience were extremely
rare. He accepted that paragraph 19 was wrong, that he
and other colleagues were not routinely working on
coding issues causing financial discrepancies, and even
more remarkably he couldn't recall ever working on
a coding issue which caused a financial impact on
a branch.
Be that as it may, in relation to bugs causing
branch shortfalls, his evidence has been, if I may say
so, over taken by the KELs and the PEAKs that have been
disclosed in this case and the in-depth analysis that
has been performed by the experts on those documents.
Your Lordship will be well aware of Mr Coyne's oral
evidence that he and his team, having read almost all
the KELs and having done innumerable intelligent
searches through all the KELs and all the PEAKs, his
judgment was there are not more than 40 branch-affecting
bugs over 20 years.
Similar points can be made about remote access.
Mr Coyne accepted that any work on branch data is
typically recorded in PEAKs, and for your Lordship's
note that's at paragraph 740 of our closing
{Day16/28:1}. Mr Coyne said that there had been no more
than 30 occasions of relevant remote access; that's
paragraph 746 of our closings. And Mr Coyne said that
the chances of remote access adversely affecting branch
accounts was small. My Lord, that's paragraph 748 of
our closings.
My Lord, all those points are consistent with
Mr Roll's oral evidence. But you would get no hint of
that anywhere in the claimants' oral or written
closings. That, in my submission, is a graphic
illustration of how the claimants would like to distract
the court from the oral evidence given by their two most
important witnesses. What they want to do is tell
little stories of individual incidents instead and to
give the court the impression that those individual
incidents are somehow symptomatic or representative of
a wider phenomenon of which there is no evidence
whatsoever.
My Lord, they would also like to stop the court from
deciding the most practically useful issue in this
entire issue, the Horizon Issue trial. And that is the
extent of the risk or likelihood of Horizon bugs causing
branch shortfalls for which SPMs are held liable.
They do this in a number of ingenious ways. Indeed,
they do it rather beautifully by accusing Post Office of
seeking to rewrite Horizon Issues 1 and 3. Let me deal
with those arguments very briefly.
Horizon Issue 1 first of all. They make three
assertions. First of all, the threshold for satisfying
the enquiry required by Horizon Issue 1 is merely that
there should be a potential of causing discrepancies or
shortfalls, the logic of their position being if some
potential is found, however slight, the court stops
there. And they follow up that submission by saying in
no circumstances, Post Office's attempt to draw
attention to the kind of bug that doesn't have a lasting
impact is an impermissible exercise. They say that
Post Office is introducing a false concept which isn't
written into the Horizon Issue itself.
Thirdly, they argue that the discrepancies and
shortfalls referred to in Issue 1 need only relate to
branch accounts, they don't need to be in branch
accounts.
As to these three arguments, potential first of all.
My Lord, it is important not to forget the words
"extent" and "likelihood" at the beginning of Issue 1.
I don't know if your Lordship has the issues with you?
To what extent was it possible or likely for bugs of the
nature alleged in the GPOC and referred to in the
generic defence to have the potential to cause apparent
or alleged discrepancies?
The essential enquiry is an enquiry as to extent of
likelihood, and I won't invite your Lordship to do it
but if you look at the assertions made in paragraphs 23
and 24 of the generic particulars of claim and the
relevant paragraphs of the generic defence, that is how
the case is put.
These are bugs that caused these things.
"Potential" is just a word that's emphasising that what
the court is doing is assessing extent of likelihood or
risk or potential; it just reinforces the essential
enquiry that the court is required to undertake.
Critically, the court is not required to stop if it
identifies some potential. It actually needs to assess
risk or likelihood in a sensible way and, my Lord,
your Lordship will be well aware that I submit that that
sensible way requires a sense of scale or perspective,
another thing that the claimants would rather not talk
about.
So, my Lord, that deals with potential. Let's now
deal with the so-called impermissible concept of
lasting. In my submission it is not impermissible, it
is fundamental, otherwise the determination called for
by Issue 1 would be practically pointless. I'm not
saying, and the Post Office doesn't submit, that whether
a transient discrepancy was caused is irrelevant. Of
course not. Issue 1(a) refers to apparent or alleged
discrepancies but it also refers to apparent or alleged
shortfalls. I ask forensically: what's a shortfall?
What is the difference between a discrepancy and the
shortfall and why was it necessary for the Horizon
Issues to distinguish between them?
A shortfall is that for which an SPM is held liable.
A discrepancy doesn't really matter at the end of the
day. I hope no one quotes me out of context. But if
your Lordship understands what I'm submitting to you, if
a discrepancy occurs during the course of a month, if it
doesn't actually result in a shortfall at the end of the
month and the postmaster doesn't therefore have to make
that shortfall good, that is not quite the same -- that
is much less important than the question whether
a liability is impermissibly foisted on a subpostmaster.
That is what Issue 1 was trying to get at.
So by all means your Lordship should consider
whether and to what extent there was a likelihood of
bugs causing transient impacts. I'm not suggesting you
should ignore them. But what I do suggest is, by the
same token, nor should your Lordship ignore whether
those bugs caused discrepancies that were caught by
countermeasures either in the short or medium or longer
term, or, on the other hand, were lasting discrepancies
that became shortfalls and that required a subpostmaster
to put his hand into his pocket.
That is why the word "shortfall" is contained in
Issue 1. And by the way, it is part of the overall
judgment of robustness that's called for pursuant to
Issue 3.
My Lord, the final point relating to branch
accounts. This is the most ingenious argument of them
all on Issue 1. In my submission it is sophistry. The
purpose I think, or I suspect, is to allow the claimants
to shoehorn into Issue 1 a consideration of whether bugs
in Post Office's back end systems and whether and to
what extent the bugs in those systems took place.
But there is a problem with that essential approach,
which is that Issue 1 is about bugs in Horizon. The
back end systems which may result in TCs being issued,
those back end systems are not part of the Horizon
system and the TC process was deliberately excluded from
the Horizon Issues because if there were to be a trial
of all Post Office's reconciliation processes over the
last 20 years that would involve millions of documents
and take a six-month trial.
Now, in my submission, the drafters of the Horizon
Issue, of whom Mr Green may well have been one at the
time, could never have dreamed that the investigation
called for by Horizon Issue 1 would be as stunted as the
claimants are now seeking to achieve.
That brings me to Issue 3. Here, two arguments are
being run for which there's no hint in their written
openings. There, they appear to accept that the Horizon
system was relatively robust. Indeed, their suggestion
appeared to be that they had always accepted that it was
relatively robust.
If we could look, please, at {A/1/10}. This is the
second half of paragraph 17.1 of the claimants' written
opening. We don't need to look at the first half. It
says at the top:
"This reflects language pleaded in the GDef, and
indeed 'robustness' has been one of Post Office's
'narrative boxes' and a favoured term in Post Office’s
public relations pronouncements ... Coincidentally or
otherwise, it has also featured in the NFSP's defence of
Post Office, relied upon by Mrs van den Bogerd.
However, as the Claimants made clear in their GReply,
whereas the Claimants'," I think it must be claimants'
case, "is that it is relatively robust and has become
more robust over time – but not so as to be an answer to
the Claim (and in so far as 'robustness' has, in this
case, a sufficiently clear meaning – addressed further
herein)."
So they are toying with whether robustness has
a meaning, but there they purport to say that relative
robustness was actually accepted in their reply. That's
always been their position.
That is the complete opposite of what the claimants
are now saying. They are now trying to ditch their own
expert whilst still purporting to laud him to the skies
in their written and oral submissions.
Their first argument is that Issue 3 raises one
point, not two: robust and extremely unlikely. It seems
to be an attempt to suggest that robustness can have no
meaning other than that. Well, my Lord, whether
robustness has a meaning and what meaning it has is not
in issue between the experts. In JS3 at paragraph 3.1
they both agree that it's relatively robust.
We have a meaning from Mr Coyne. He said
"relatively robust" means performing well compared to
similar big systems. And those similar big systems
require transactions to be handled properly in the
overwhelming proportion of cases, and that they require
the system to have lasting errors of a fraction of
a percentage.
My Lord, what that means is that when faced with any
given transaction, the generic likelihood of that
transaction being erroneous is extremely small. In
other words, it is extremely unlikely to be the cause of
shortfalls.
Clearly I'm not submitting to your Lordship that
means that in any given case your Lordship should reject
any suggestion that any accounts are wrong. I'm not
suggesting that for a moment. I'm simply saying this is
a trial of generic issues, and at a generic level,
absent particular circumstances, and there may be
several, any given set of accounts is overwhelmingly
likely to be reliable.
Now, it is worth pausing for a moment to note some
other things that Mr Coyne agreed. And if we could pick
it up at {Day14/18:1}, he agreed at the bottom of
page 18. Picking it up at line 20 and going over the
page to page {Day14/19:1} -- I said:
"Question: No, I'm not asking you, Mr Coyne -- I'm
not suggesting to you that you have been asked to decide
on whether any particular claimants' claim is right or
not, what I'm suggesting to you is that the context in
which these -- given the context in which these issues
arose -- were drafted, and given the pleadings by
reference to which they were drafted, it was obvious
that the purpose of those issues was to assist the court
so that it could use the judgment that will be produced
in this trial as a basis for making ultimate decisions
in ultimate breach claims by claimants?"
Mr Coyne said:
"Answer: In a later trial?"
I said:
"Question: Yes."
He said:
"Answer: Yes, I was aware of that."
I said:
"Question: Isn't that the main reason why we are
here?"
And he said:
"Answer: Well, it is certainly a reason why we are
here, yes.
"Question: To enable the court to make useful
findings as to the general likelihood of any transaction
being wrongly recorded in a particular case?
"Answer: Yes."
Then if we move on to page {Day14/25:9}, there was
a passage of evidence from line 9 through to page
{Day14/26:15}. Perhaps I could ask your Lordship to
read that.
MR JUSTICE FRASER: From line 9 of page 25?
MR DE GARR ROBINSON: 25. (Pause)
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: So what Mr Coyne accepts there is that
robustness involves two things. One is preventing
errors from arising in the first place, but the other is
having countermeasures to deal with errors when they do
arise, hence the concept of lasting impacts.
At the bottom of page {Day14/26:1} I put to him:
"Question: So robustness is the very concept which
underlies the issues we have been discussing for the
last half hour, yes?"
And he says:
"Answer: Yes."
He further accepted that the concept of robustness
was a mature subject which was the subject of study, and
that's at page {Day14/31:18}. Picking it up at line 18,
I say:
"Question: You agree robustness is an important
concept?"
He says:
"Answer: Yes.
"Question: You agree it is deployed very frequently
in the IT industry?
"Answer: Yes.
"Question: It is a subject of academic study,
isn't it?
"Answer: Yes."
Then picking it up on page {Day14/32:10}:
"Question: ... isn't it obvious, doesn't it follow
as night follows day, that the ultimate question being
wrestled with by the concept of robustness is how well
are the risks faced by a system guarded against? In
other words, what are the risks remaining after you have
taken the countermeasures into account?"
And he says:
"Answer: Yes ..."
And I invite your Lordship to read the rest of his
sentence. (Pause)
My Lord, I won't take your Lordship to the
transcript but he went on to accept that it is possible
to benchmark robustness against other systems, and
against those other systems that are similar, that are
comparable, it compares well. My Lord, that's at
paragraph 365 of our submissions but also {Day14/41:23}
to page {Day14/42:6}.
Critically, comparable systems require the
overwhelming majority of transactions to be handled
properly. They require that there should only be a tiny
proportion of lasting errors of a fraction of
a percentage. My Lord, that's at {Day14/56:1} to
{Day14/59:1}, and in our closing submissions it is
paragraphs 378 to 380.
So all in all Mr Coyne accepts the importance of the
concept of robustness, the reality of the concept as
a concept that's the subject of academic study and
practical application in the IT industry, and he accepts
that a consideration of countermeasures is an important
aspect of forming a view on robustness. And my Lord,
your Lordship will see that point made at paragraph 390
of our closing submissions.
At paragraph 391 we point out that he has sufficient
information to form a judgment on robustness. So that
was something of a lemma when discussing the first
argument that the claimants run in relation to Horizon
Issue 3.
I now come to the second argument. This is the
suggestion that the drafters of Issue 3 had in mind some
sort of conditional probability. It is so bold as to be
impressive. Let me see if I can find ... (Pause)
If your Lordship looks at Horizon Issue 3 {C1/1/1}:
"To what extent and in what respects is the Horizon
System 'robust' and extremely unlikely to be the cause
of shortfalls in branches?"
Now, this is an argument of which a Platonic scholar
would be proud. They don't say extremely unlikely to
cause shortfalls in branches, they say to be the cause
of shortfalls in branches.
So Issue 3 is not about the probability of causing
shortfalls but the probability of something being the
cause of a shortfall once a shortfall has arisen.
My Lord, I would respectfully submit that that
subtlety would obviously not have occurred to anyone at
the time, and bearing in mind the first argument that
the claimants are running, namely that robustness and
extremely unlikely need to be put together, it would be
a very curious argument to be running in any event.
My Lord, the claimants should have ten out of ten
for ingenuity with this argument. Their obvious
intention is to avoid their essential problem that with
3 million sets of branch accounts, ignoring for the
moment that up to 2005 accounts were weekly, that the
evidence shows that relatively speaking there were
hardly any bug impacts and the evidence suggests that
there are even fewer, infinitely fewer remote access
impacts, and that's a big problem for them. So what
they want to say is there's some smaller number of
shortfalls which, very conveniently, have not been the
subject of evidence so no finding can be made about
them.
The essential endeavour in this argument, all this
time and money having been spent in preparing for this
trial, is to put a block to the determination of Issue 3
because neither party has thought to engage in any
statistical analysis of the number of shortfalls versus
the number of bug branch impacts.
My Lord, in my respectful submission that is simply
not an argument that it is proper to run at the end of
a trial with no advance notice. If they were going to
run that kind of argument they should have raised it
before the evidence was even prepared so that the
parties could have addressed the point. And I rather
suggest that if the point had come up your Lordship
would very easily have simply applied exactly the same
language that's used in Issue 1: to cause apparent or
alleged discrepancies or shortfalls in branch accounts.
One of the curiosities of this argument is that they
are suggesting that the mental process of the person
that drafted Horizon Issue 3 was completely different
from the mental process of the person who drafted
Horizon Issue 1.
MR JUSTICE FRASER: Now, these issues I seem to remember,
but I might be misremembering, were agreed by both the
parties, weren't they? I don't recall being involved in
the detailed drafting of the issues.
MR DE GARR ROBINSON: I don't believe your Lordship was.
I wasn't here then but I don't believe your Lordship
was.
MR JUSTICE FRASER: I can go back and read the transcript of
the hearings but I seem to remember I was presented with
the proposed issues and I approved them. Is that right?
MR DE GARR ROBINSON: I think that is right. But, my Lord,
overall my submission is that this is a transparent and
desperate attempt to frustrate the plain intention of
the people that drafted Horizon Issues 1 and 3 to
prevent clear findings being made of obvious practical
utility and to replace them with findings that have no
practical meaning whatsoever.
I would urge your Lordship not to be distracted by
last-minute swerves of this sort. Your Lordship will
already understand that my submission is that the
claimants are trying to sideline the most important
issues and to distract the reader with isolated examples
put forward in the hope of giving the impression that
they have rather more widespread impact than is
justified by the evidence.
Let me give some examples of making that submission
good. The first one, in their written closing they
start on page 1 with remote access. That's their first
point even though remote access is clearly a second
order issue. Mr Coyne effectively admitted in his
cross-examination, and I will be taking your Lordship to
it in due course, that the impact of remote access on
branch accounts was de minimis. He didn't use that
phrase. That's where we got to in my submission.
But it said in the closing submissions, and my
learned friend repeated roundly yesterday morning, that
there is an open back door to the system. That's said
in paragraph 2. And in paragraph 5 they build on that
by saying that there is unfettered remote access.
Now, it is intended, I surmise that it is intended,
to be a headline-catching metaphor, but in my submission
it is an illuminating and quite useful one, and I am
sure this isn't intended. The concept or the metaphor,
the picture, of an open back door, it is not an open
back door onto the street, it is an open back door, if
you want to call it that, onto a walled garden,
consisting only of an elite group of highly trained
professionals, the SSC, who require months of training
before they are even allowed in.
Now the experts may agree that APPSUP privileges,
that privileged user privileges, should not be
permanently available to that many people and that when
people in the SSC wish to use it there should be
a process they go through to get temporary
authorisation.
That may well be the position, but that doesn't mean
that APPSUP was actually misused or that there were any
harmful events as a result of the fact that the facility
was there available for the SSC and then, as time went
on, it became available only for the most senior members
of the SSC, not to the new members.
There isn't evidence, my Lord, of a single harmful
event that was the result of any APPSUP use. Indeed,
there's no evidence of any use of APPSUP to change or
delete transaction data. None at all. But the
important point is that the people to whom the door was
left open were trustworthy professionals. They weren't
burglars waiting to get in through the back door once
the homeowner had gone to bed.
They are not going to run away with any money and
they are not going to start using APPSUP as a means of
doing creative vandalism in the branch network. And to
the extent that it is suggested otherwise, there is no
evidential basis for making that suggestion.
Another example of distraction with little stories
is the rather elaborate business we had yesterday
afternoon in which an impression was given that there is
a longstanding problem of duplicate losses in the system
of which Post Office has always been aware. That's not
what the evidence showed at all. But you speak quickly
and you go to particular sentences in particular
documents, and hey presto, you have created that
impression. That is the essential endeavour that the
claimants are engaging in in this trial.
Another example, the most extraordinary reliance
that my learned friend at the last minute places on some
internal Post Office documents from 2016 and 2017,
largely from those years. What the claimants seem to be
wanting to do is to substitute the opinions of the
experts on matters on which they agree by reference to
some internal management papers about how they are
planning to improve and upgrade their own back end
processes and the IT systems more generally, including
Horizon.
But I ask forensically: if that's all you need, why
have so many millions been spent on experts at all? The
important thing about those documents is none of them
say anything about whether bugs create discrepancies in
branch accounts. They are not about that at all. They
don't purport to grapple with Horizon Issues, and it is
only if you do not consider them carefully that you
might be confused for a moment into thinking that they
are even relevant.
Let me make that point good by just reference to one
document {F/1161/87}. It is a document with intent to
supply attached to a board report. I'm afraid I don't
have page 1 to hand.
MR JUSTICE FRASER: It has not come up yet, I'm afraid,
I think. Can we go to F/1161?
MR DE GARR ROBINSON: I'm sorry, I misread it. It is my
fault. It is {F/1611/87}, I do apologise.
MR JUSTICE FRASER: The dreaded spreadsheet. Where are we
going?
MR DE GARR ROBINSON: 16/11. Is it possible to become
dyslexic in your 50s? I don't know.
MR JUSTICE FRASER: Well, the F bundle generally is not easy
to use because it is so big, I think. Or the F folder.
That is a PEAK.
MR DE GARR ROBINSON: This isn't it. 1611. I'm looking
at -- page 87? 1611.
MR JUSTICE FRASER: I think this is 16/11 and we want 1611.
MR DE GARR ROBINSON: We need to agree a protocol for this.
1611, page 87, please. {F/1611/87}.
This is a board agenda, and if we go to page 87 this
is a technology strategy update. I don't want to put
words into my learned friend's mouth, but the author is
Rob Houghton who is the chief technology officer at
Post Office, and so this may be one of their golden
documents, one of my learned friend's favourite
documents. But let's just have a look at it:
"Context.
"This document forms an update to the IT Strategy
approved in July 2016 by the PO Board. In July we
outlined that IT was not fit for purpose, expensive and
difficult to change."
Stopping there. If your Lordship were a journalist
you would think that looks really bad, but let's read
on. There is a list of questions addressed in this
report, and then:
"Conclusion.
"Since reporting on the IT strategy in July; the
landscape continues to shift with increasing cyber
threat, deteriorating market conditions and several
significant service outages to our back office systems
..."
Your Lordship will note back office and not Horizon:
"... and digital infrastructure. These reinforce
and give further urgency to, rather than change ..."
There must be a missing word there:
" ... the strategic roadmap."
Then:
"In the 5-year overlays, we have aligned our IT
strategy to the business strategies and developed
roadmaps for the digital, infrastructure and Horizon
journeys. Approval of the overlays is critical to
improve our control environment, reduce costs and
achieve our business strategy."
So improving the control environment, reducing costs
and achieving a business strategy. Then two bullet
points down:
"Our view of the operational service risk has
matured and we remain outside of our risk appetite zone
in key operational areas. The infrastructure related
change programmes focus on reducing these risks over
time."
So your Lordship will see the general points being
made.
If one goes to page {F/1611/88}, the first bullet
point says:
"Supplier negotiations are underway."
Second bullet point says:
"There are a 'critical few' dependencies that drive
reduction of the cost base and future business support
in the most material way ..."
So your Lordship has seen what they want is to save
costs, they want to upgrade the system, there are all
sorts of objectives they think they can achieve:
"We need to quickly rationalise and resolve
misaligned contracts enacted to support legacy IT,
obsolescence and lack of PO technical competence
particular focus on Fujitsu and Accenture.
"'Thin Client' delivery success.
"Appetite for reduced service levels and number of
branches/users. We are developing a hypothesis to offer
two/three tier service offering to branches to reduce
cost and target service delivery."
Then on page {F/1611/89}:
"The IT strategy outlined a view of the current
state of technology within PO as failing to meet PO
aspirations on any assessment lens (cost, risk, delivery
or service).
"The strategy laid out key interventions:
"Improve IT - cost reduction, operating model
changes, supplier renegotiations, capability hires,
communications and engagement."
"Reduce operational and security risk - IT
infrastructure programmes, Back office transformation,
Security Transformation programmes and improved
Environment Monitoring/Management, collectively to take
back control of the management of IT Services."
That's a reference to outsourcing; they are planning
to bring things clearly back in.
Then, my Lord, over the page {F/1611/90},
paragraph 4 at the top says:
"Technology sits at the heart of the wider business
strategy set out in our 5 Year plan ...
Then the last sentence before the first bullet:
"Ultimately our technology strategy should be judged
in terms of the benefits it drives for each of our key
customer groupings:
"for end consumers ..."
Then it talks about them. Then for host retailers,
that is postmasters:
" ... technology is critical to simplifying our
offer, removing the complexities which are frequently
cited as the biggest deterrent to running a post office.
Our network IT strategy is working towards the
development of smaller point of sales solutions
(including EPOS integration) that are more easily
integrated alongside the main retail till, driving
operational benefits for the retailer and quicker
transactions for customers."
Stopping there. So the main branch, well, the first
branch benefit or objective that's identified, nothing
to do with the reliability of Horizon, it is to do with
enabling people running their own retail business to
plug the Post Office IT system into their existing EPOS
systems to make it easier for them, less of a barrier
for them to agree to become a Post Office branch.
Then it goes on to say:
"Stability of service is also vital for our
postmasters, ensuring they do not suffer the significant
business detriment that can arise from prolonged outages
..."
So there's a problem that is identified: prolonged
outages. Your Lordship may think it appropriate to note
that that's the only problem that's relating to branches
that is identified in this entire paper, as far as I'm
aware.
Then, my Lord, page {F/1611/98}. I will take
your Lordship to this because this is quoted by my
learned friend in his submissions.
At paragraph 25, it says:
"There are tensions in each contract. Our approach
and risk appetite around public Contract Regulations are
likely to also inform outcome ..."
First bullet point:
"Fujitsu - a 6 year fixed contract signed with PO
which continues to invest in legacy and obsolescence
where FJs own strategy globally is to move to Cloud.
Success is to re-contract to prime UK Cloud partner,
align innovation, re-architect and position as a true
retail partner."
So there's the word obsolescence. My learned friend
loves the word obsolescence that's quoted. But what is
actually being talked about? Moving to the cloud. It
is an improvement, it is not the fixing of a defect.
Then, my Lord, over the page {F/1611/100}, I'm
afraid this is in management speak, I find it hard to
read with a straight face, but paragraph 30:
"Our target state is to deliver an elastic, agile,
secure, future proof, low cost model for the Post Office
without undue reliance on specific suppliers."
Then in 34 it says:
"For Back Office: A Back Office system that provides
a stable and robust financial system and MI/Analytics
capability."
Then "For Retail", it says:
"The Horizon (HNGX) platform is at the end of its
life and needs replacing. Previous attempts to move
away from HNGX platform, specifically with IBM, have
been unsuccessful.
"The current plan is to transition from HNGX to the
updated HNGA (part of the Branch Technology programme)
platform from April 2017 will incur (budgeted) capex of
£16m."
So they are spending £16 million on an improved
version of the existing Horizon system.
Then over the page that's made clear with the next
bullet point {F/1611/101}:
"However, whilst HNGA runs on updated Windows
software, fundamentally its architecture is the same.
This means that it remains a 'closed' and inflexible
platform that cannot support the business' strategic
objectives and is a drain on respective PL's. HNGA will
work and support our business BUT it will not enable us
to compete and succeed with Clients retailers or
Customers."
Then there is an explanation of this thin client
concept, and I don't need to take your Lordship to that.
So I ask forensically: what does this document show
us about how good the Horizon system is? What does it
show us about its likelihood of creating false
discrepancies or false shortfalls in branch accounts?
What does it show us about the robustness of the
existing system and the reliability of the figures in
it? Answer: absolutely nothing.
It is not about those questions, my Lord. And this
demonstrates in really graphic terms the dangers of
taking little snippets from documents and pulling them
together and seeking to create an impression that's far
larger than the document from which the snippets are
drawn.
By relying on documents such as this, in my
respectful submission, the claimants are inviting
your Lordship to make a category error. They are
relying on documents with a completely different focus
in order to establish a position in relation to the
Horizon Issues which is inconsistent with the agreement
between the experts. They are trying to wallpaper over
that inconvenient agreement between the experts by
focusing on different points, by focusing on the
position now, when of course it must be remembered that
most of the claims made in this case relate to
a position many years ago.
My Lord, that entire approach, in my submission, is
the very antithesis of what this trial is supposed to
be. If there is to be an analysis of the document it
should look and see what particular point in time the
document operates, what precisely the document says,
what can properly be drawn from the document and what
cannot properly be drawn from that document. But that
is not an endeavour which, in my submission, the
claimants are interested in doing. Just as, I say this
with some regret, it was not an endeavour which Mr Coyne
was particularly interested in doing in his two reports.
When he came to give evidence the position changed.
If documents like this do my learned friend's job as
he seems to think, I do ask rhetorically why have we
just spent four weeks trying this case?
Another aspect of these documents which really is
very striking is the attempt that my learned friend
makes to criticise Dr Worden for not being aware of
them. There was a clear implication in the tone of my
learned friend's questions of Dr Worden. You are not
aware of these questions, the implication being that he
really ought to be as a reputable expert. And that
implication is made explicit in the claimants' closing
submissions as if the documents represented exactly what
any expert should be looking at instead of looking at
the architecture, looking at the countermeasures and
looking at the operation of the system and seeing how it
worked, which is what Dr Worden has actually done.
But, my Lord, there is a problem with that
criticism. The problem with that criticism is that
Mr Coyne doesn't refer to any of these documents either.
They weren't mentioned in his reports. They weren't put
to him in examination-in-chief. They certainly weren't
put to him in cross-examination. I was blissfully
unaware of them until they were pulled like a rabbit out
of the hat when Dr Worden was giving evidence.
So perhaps conveniently for the claimants, I wasn't
able to ascertain Mr Coyne's views on those documents.
I certainly didn't have a realisation that that was
something I ought to be doing.
Now, my Lord, in my submission, the inevitable
inference is one of two things. Either Mr Coyne wasn't
aware of them either, so by criticising Mr Coyne my
learned friend is also criticising his own expert, or,
which may be more likely --
MR JUSTICE FRASER: You mean by criticising Dr Worden?
MR DE GARR ROBINSON: I'm so sorry, I do this too often. Or
Mr Coyne didn't think these documents were sufficiently
material to the Horizon Issues to be worth referring to.
My Lord, either way this last-minute tactic by the
claimants to change the narrative in my submission is
completely inappropriate. They don't undermine
Dr Worden's view on the Horizon Issues any more than
they undermine Mr Coyne's views that Horizon is
relatively robust and, indeed, is more robust now than
it has ever been before.
My Lord, I would like now to say a few words about
bugs affecting branches. Mr Coyne says that he and
Dr Worden together found 29 potential branch-affecting
bugs in PEAKs and KELs, and in his oral evidence he said
that given the reading he had done and the searches that
he and his team had performed, he didn't think there
were more than 40 bugs overall in PEAKs and KELs.
Now, of the 29 that he has found, he now says that
22 are lasting; in other words, 22 are of a type that
wouldn't, or might not ordinarily be caught by
countermeasures. Pro rata that suggests that the
overall number of lasting bugs in PEAKs and KELs would
be around 30, although I have to say I didn't put that
to him. I didn't realise that there would be this shift
from 29 to 22 when we discussed the point.
Now, the claimants absolutely hate that logic. So
they attack the concept of a lasting impact. In his
written closing and in his oral submission yesterday my
learned friend said that Mr Coyne's use of the term
"lasting impact" in the joint statement -- that's
paragraph 115 of JS2. Perhaps we could look at that.
It is at {D1/2/29}, paragraph 1.15. This is agreed:
"The number of distinct bugs, for which the experts
have seen strong evidence of the bug causing a lasting
discrepancy in branch accounts, is between 12 and 29."
Now, that is something which Mr Coyne has agreed.
It is language that he has used. It is obvious what he
meant. He means not transient, and your Lordship will
see the reference to transience in paragraph 1.10 at the
top of the page. That is an RW statement.
These are bugs whose impacts are not caught by some
countermeasure and so one can't expect that they would
necessarily be corrected.
Now, Mr Coyne accepted all of that, he accepted what
"lasting" meant in cross-examination. And I frankly
don't understand the attempt to suggest now that he
somehow didn't.
The claimants themselves have deployed the concept
of lasting in their own opening. My Lord, that's at
{A/1/35}, paragraph 108:
"As noted above, in Joint 2, the experts have
resolved some but not all points of difference
(eg Dr Worden now accepts that there is strong evidence
of a lasting discrepancy on branch accounts from 12 of
the 29 bugs which Mr Coyne identifies as having such an
effect."
So there we have two statements in that paragraph.
First of all, they themselves are deploying the
concept of lasting effect. Secondly, they are clearly
saying that they clearly thought at that time that
Mr Coyne was saying that there were 29 bugs which had
a lasting effect. And your Lordship will recall the
cross-examination on his last day of evidence about
that.
So against that background it is difficult to
understand what the claimants mean when they say that
the use of the word in paragraph 1.15 was a compromise.
Unless they are seeking to go behind it in some
impermissible way, an agreement was made in without
prejudice discussions between the experts. I can't
imagine they are doing that, but I can't imagine what
else they are doing.
My Lord, what one gets from paragraph 108 is that
while the claimants were entirely comfortable with the
concept which was fully deployed and explained in
Dr Worden's first report, as the evidence has come out
and as Mr Coyne's own evidence on these points has come
out, they have had second thoughts. Now that Mr Coyne's
views have been properly explored, they want to
backtrack.
In my respectful submission, that this process is
being engaged in is actually a good illustration of what
cross-examination can achieve. Nowadays it is
fashionable to suggest that what one gets by
cross-examining witnesses is very limited and shouldn't
we have trial mainly by reference to the documents? In
my submission, this trial is a good example on both
sides of points being established in cross-examination
that wouldn't be established without it.
I make the same submission about Mr Roll's
cross-examination. My learned friend is entitled to
make the same submission about some of my witnesses.
But be that as it may, it is important to recognise the
significance of Mr Coyne's evidence on the number of
bugs to be found in PEAKs and KELs and the proportion of
lasting bugs that are liable not to be caught by
countermeasures. He and Dr Worden disagree on
particular bugs, and your Lordship has our submission on
the bugs in appendix 2 to our closing, but let's take
Mr Coyne's views at face value.
At {Day15/94:1}, he said if you get a bug with
a branch impact there will be a KEL for it. My Lord,
that is addressed at paragraph 352 of our closing
submissions.
At page 122, unless your Lordship would like me to
I'm going to go too fast to allow the reading of the
transcript --
MR JUSTICE FRASER: No, but the transcript I think you are
just referring to has just come up and that's him
agreeing to your question that:
"Question: ... generally speaking, if you get a bug
... once it is detected there will be a KEL ...?"
MR DE GARR ROBINSON: Yes. My Lord, {Day15/122:1} Mr Coyne
says he has now reviewed between 5 and 6,000 KELs. He
says at line 24:
"Answer: ... probably ... between 5 and 6,000."
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: At page {Day15/123:1}, he says that
his team will have reviewed an additional 1,000 KELs.
So between he and his team there has been a review of
between 6 and 7,000 KELs of the 9,500-odd that have been
disclosed.
At page {Day15/123:1} he says that all the bugs that
have been found come to 29. And he goes on to say that
the total number of bugs likely to be found in the KELs
is likely to be no more than 40.
My Lord, at page {Day15/128:16} Mr Coyne agrees that
his search processes are reliable and will have
successfully identified the majority of bugs identified
by the SSC during the operation of the PEAK and KEL
system.
So there aren't that many more bugs to be found even
if he had had more time to look.
Your Lordship will also recall, I'm afraid I don't
have the reference in my note, Mr Coyne saying that it
is possible to identify the number of impacts of
a particular bug by looking at the PEAKs, and that's
quite a reliable system too. And he then shamed me by
indicating that he had given an indication in JS2 as to
the number of impacts.
My Lord, on this side of the court we have done our
best to do a calculation --
MR JUSTICE FRASER: Was it in JS2? I thought it was in his
report. Was that when you took him to the graphic?
MR DE GARR ROBINSON: It was JS2. Second column.
MR GREEN: It is actually both.
MR DE GARR ROBINSON: My learned friend helpfully says it is
also in his report.
MR JUSTICE FRASER: No, when you said he had shamed you
I thought that might be when you said you'd put
a question to him and he then pointed to a graphic.
MR DE GARR ROBINSON: I think I had missed it and he pulled
me up on it, very properly.
On this side of the court we have done our best to
work out how many impacts are suggested, and I'm not
suggesting for a moment that this is accepted but it
appears to be in the region of 545/550 branch impacts in
very round numbers.
So that would suggest less than 20 impacts per bug,
and I would ask your Lordship to bear that statistic in
mind.
I wonder whether this would be a convenient moment?
MR JUSTICE FRASER: I am sure it would, yes.
When you say on your side you have identified it,
that's the one in your closing? The 545/550, that's in
your closing submissions?
MR DE GARR ROBINSON: It is in a footnote to our closing.
I can give you the reference if your Lordship would like
after the break.
MR JUSTICE FRASER: I do not think it is in any of the
evidence, is it?
MR DE GARR ROBINSON: No, it isn't. What has been done --
I'm afraid I haven't done the process, but what has been
done is you look at the column, you and try and work out
what that implies. It is not a simple counting process,
one has to use a degree of judgment. It is footnote 794
of our closing submissions, my Lord.
MR JUSTICE FRASER: Thank you very much. All right. We
will have ten minutes. We will come back in at
11.55 am.
(11.47 am)
(A short break)
(11.55 am)
MR DE GARR ROBINSON: My Lord, talking about bug numbers and
bug impacts, it may be helpful at this point to look at
one of Dr Worden's calculations. If I could ask
your Lordship to go to {D3/8/1}.
Thank you very much. Your Lordship will have seen
this before. It is worth spending a few minutes so that
we all understand it.
This is a calculation which involves all sorts of
financial impacts and also involves an estimate of the
number of bugs detected and undetected in a system over
a period of 20 years.
It starts at row 3, but your Lordship will see the
label in column B. So I'm working behind those column B
labels. So row A is "mean number of branches in PO
network, 1999-2018".
Your Lordship may recall Dr Worden being taken to
task because his number of 13,560 was from 1999, not
from 2000. I think his report did say 2,000, but in
actual fact, as is clear from his spreadsheet, it was
from 1999. So that is the number of branches, mean
number of branches over that period.
Then row B is years of Horizon. 19 years. Total
branch months in C. Nearly 3.1 million in branch
months. So that is 3.1 million sets of accounts, as it
were. But Dr Worden, your Lordship will be aware, has
rounded that down to 3 million. In his reports he talks
about 3 million. He could of course have said 3.1, but
he said 3 because that was sufficient precision for him.
Then there is the scaling factor in relation to
which some time was taken in Dr Worden's
cross-examination, and your Lordship will see --
actually, this is a good place to see. Column C is
central estimate. That is Dr Worden's central estimate,
what he thinks is likely to be the case, and D is his
conservative estimate, because as your Lordship will
recall, he deliberately makes assumptions in favour of
the claimants to increase likelihood or reduce the
likelihood of things happening so as to increase the
number of bugs and so on. And that's his conservative
estimate. He explains all this in his report.
Your Lordship will see that there is a scaling
factor of 0.45 central and he just pushes it up by just
over 10% to 0.5 conservative. We needn't worry about
scaling for these purposes.
Row E is another scaling number. Then row F is the
total number of claimed shortfall. And as your Lordship
ascertained yesterday, that's taken from the schedules
of information and is explained in Dr Worden's first
report.
Row G is:
"Total claimant branch months (sets of monthly
branch accounts for claimants)"
Or, indeed, any other SPMs with similar
characteristics. That is 52,000. So of the
3 million-odd monthly branch accounts, ignoring for the
moment until 2005 they were weekly, 52,000 was occupied
as it were by claimants.
Then H is "scaling factor". I'm not going to talk
to your Lordship about that. Then here's an assumption
in row L:
"Maximum number of KELs with potential impact on
branch accounts, based on limited sampling of KELs."
And his central estimate is 100. Does your Lordship
see that?
MR JUSTICE FRASER: Mm.
MR DE GARR ROBINSON: And his conservative estimate is 200.
It is quite important to note that those two estimates
are explicitly based upon limited sampling of KELs.
Dr Worden didn't do what Mr Coyne did. What Mr Coyne
did is he looked at virtually all the KELs; he and his
team looked at between 6 and 7,000 of the 9,500. That's
not what Dr Worden did. What Dr Worden did is he looked
very carefully at a limited number of KELs. He did
a number of passes through the KEL system, one of which
was he looked at 200 KELs at random, and another pass
that he did was he looked at KELs that referred to
a pound sign, and a third pass that he did was he looked
at KELs that had been referred to in Mr Coyne's expert
report.
The result of those enquiries are set out in
appendix D to his first report, and then of course in
his supplemental report he has looked at more KELs, so
there is an expanded appendix for the 200 KELs that he
looked at. But on any view he had only looked at
a small proportion of KELs; not statistically
insignificant, but hardly occupying the whole ground.
It is important to note, and I will be coming back
to this, that Mr Coyne has now looked at the whole
ground or more or less at the whole ground. He has
looked at over 6,000, over two-thirds, of the KELs.
What's more, he has done intelligent searches through
all of them. What's more, he's done intelligent
searches through the 220,000 PEAKs.
As I think he may have demonstrated to your Lordship
already, his opinion is that he has found most of the
bugs, potentially branch-affecting bugs that are out
there to be found.
I hope I'm not oversimplifying his evidence when
I say that.
And he has found -- his view is that there wouldn't
be more than 40.
MR JUSTICE FRASER: Mr Coyne's view?
MR DE GARR ROBINSON: Mr Coyne. Am I saying --
MR JUSTICE FRASER: No, no, you switched to "his", so
I wanted to be clear.
MR DE GARR ROBINSON: Mr Coyne's view is that there are not
more than 40, and that is compared with 100 in row L,
conservative Dr Worden and the 100 central, 200
conservative.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: Those have been superseded, I would
respectfully submit, by Mr Coyne's own analysis.
If one takes into account the revelation on the last
day of his cross-examination that of the 29 bugs that he
has found that scale up to 40, he is now of the view
that only 22 of them are lasting, so if one were
interested in only lasting bugs, the figures to put in
row L would be 30, rather than 100 and 200.
If one does that, my Lord -- your Lordship may,
I fear, and I'm sorry to say that your Lordship is
probably better at this than I am, but even I can do
this -- that one can insert the figures into the
spreadsheet, one could put the figure of 30 or 40 in
those two columns, C and D, and then see where you get
to. And where you get to is that at the bottom of the
page, in row E2, maximum possible number of bugs, that
changes from 145; if the figure is 30, it becomes 44.
And would it be helpful if I talked you through how
the calculation works?
MR JUSTICE FRASER: I think I can tell how it works.
MR DE GARR ROBINSON: You have it. It is to do with various
assumptions about KEL efficiency, and so on.
MR JUSTICE FRASER: Well, L -- yes -- L is going to change
and that's the figure that's currently in N.
MR DE GARR ROBINSON: One very annoying typographical error
that your Lordship will see in E2 at the bottom of the
page, you see that E2 is defined as R/X?
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: Actually that should be L/X. That's
a typo.
MR JUSTICE FRASER: On the basis there isn't a R.
MR DE GARR ROBINSON: Well, on the basis that what it should
be doing is it should be starting with the number of
bugs in the KELs and then doing something with them in
order to scale up to an estimate of a total number of
bugs to include undetected bugs.
MR JUSTICE FRASER: So E2 equals L/X.
MR DE GARR ROBINSON: E2 equals L/X. And on Dr Worden's
central approach, if there were 30 lasting bugs
detected, one would end up with 44 lasting bugs in
total. That's detected and undetected. The figure
would be higher, significantly higher, in D, but much
lower than 672.
Now, in the claimants' closing, it is quite
interesting, it is suggested that there is a reasonable
evidential basis for assuming that your average bug will
have 40 or 48 affected branches, and that's worth
noting, my Lord. That's {A/5/177}, paragraph 1515. We
don't need to go to it.
But on the basis of the results of Mr Coyne's
enquiries in JS2 in the second column, where he has
found 440- impacts, or 450-odd, one would imagine it is
rather less than 40 or 48, could be closer to 20. And
just for the sake of illustration, if one took E2, the
central estimate of 44 bugs, that's on the basis of 30
lasting in the KELs, and one multiplied by 20, one would
get a total of 880 branch impacts over 3 million sets of
branch accounts over 20 years. I say 20, I should
say 19.
That is a vivid illustration, in my respectful
submission, of the kind of scale of the extent that we
are really talking about when we talk about bugs in
Horizon affecting branch accounts.
But that's not the calculation that the claimants
put to Dr Worden. They rather like, although they
refute his entire methodology, the figure that's
currently in row E2 in column D. They like the 672
figure. Your Lordship will see that in
cross-examination they put to him that if you take 40 or
48 branch impacts per bug and you multiply by 672,
I think if you take 48 branch impacts you get something
like 32,000 branch impacts. Does your Lordship recall
that evidence?
MR JUSTICE FRASER: I do.
MR DE GARR ROBINSON: And by that means Dr Worden, with some
heroic mental arithmetic that I would not have been able
to do, worked out that you would end up with one branch
impact per claimant on average.
Now, I am sure that my learned friend was quite
pleased with that result and I am sure if you were
a journalist you would think "Crikey", but it should be
remembered that there are 561 claimants who are claiming
£19.7 million in shortfalls. On average that's about
£33,000 each. So if you have one claimant with one bug
impact in their tenure, how do you get to a £33,000
shortfall? In my submission, that calculation on its
own is another vivid illustration of the scale of the
judgments that are really called for by the evidence in
this case.
But, my Lord, it goes further than that. As is
explained in our closing at pages 157 to 158, and
perhaps I could ask your Lordship to look at that. It
is {A/6/157}.
MR JUSTICE FRASER: Just while that's coming up I think you
said 19.7 million in shortfalls, but you mean 18.7.
MR DE GARR ROBINSON: Oh gosh, I do. Absolutely.
Thank you.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: If we could pick it up at
paragraph 439:
"If no scaling factor for small claimant branches is
applied at all ..."
So the table in paragraph 637 of Worden 1, and
that's the same table we have been looking at, label D
becomes 1, then label E -- your Lordship will recognise
the 3.1 million-odd figure, H becomes 193,000 and so on,
and then what that produces. And that's explained in
paragraph 440 on the next page {A/6/158}, 440.1:
"For a bug such as the Suspense Account bug which
occurs 16 times with a mean financial impact of £1,000
..."
And your Lordship will recall Mr Coyne accepting
this is one of the big ones. Or your Lordship may
recall. I should not assume your Lordship has perfect
recall of every single piece of evidence that's given in
the case:
" ... in order for there to be a 1 in 10 chance that
a shortfall of £1,000 would be caused in a claimant's
branch in any given month, there would need to be 19,000
similar bugs ..."
Because each bug is assumed to have 16 impacts or
occurrences, that would mean 304,000 occurrences, or
incidents they are called there. So you are looking at
19,000 bugs to give a 10% chance of causing a £1,000
loss and you are looking at 304,000 branch impacts in
order to have that result.
And if the mean financial impact was smaller, let's
assume it is £500, there would have to be 38,000 similar
bugs and 608,000 incidents. And if the mean financial
impact were £100, there would have to be 190,000 similar
bugs with over 3 million incidents.
Now, my Lord, these are just illustrations but what
they illustrate, or what they are intended to do is to
give your Lordship a sense of perspective, a sense of
scale as to what the evidence demonstrates as to the
likelihood of bugs causing shortfalls in branch
accounts.
In my respectful submission, a useful touchstone,
a useful yardstick for measuring scale is the claimants'
claim. That's not to suggest -- I'm absolutely not
suggesting -- that this is a way of disproving the
claimants' claim. Indeed, this calculation that we are
looking at here ignores any scaling factor, it just
looks at average branches of the same sort of tenure as
the claimants. That's all.
But it gives your Lordship a handle on the sense of
how big the bug problem would need to be in order to
have a material chance of generating the sort of losses
that are being -- in fact, a small proportion of the
sort of losses that are being claimed in this case. And
your Lordship may find that useful as a touchstone. It
certainly shouldn't be your only touchstone.
But what touchstone do the claimants use? What do
they say about extent in relation to these Horizon
Issues? They say, and your Lordship will see that from
the appendix to their closings where they very helpfully
set out in a table what they say the answers are to each
issue. In relation to Horizon 1, Issue 1, 3, 4 and 6,
those I hope are the extent of the issues that are
extent questions. They say material and significant
risk.
Now, that begs an important question: material
compared with what? Significant in comparison with
what? And in my submission the complete absence of any
sense of scale is a complete giveaway. To say material
insignificance is not actually an answer to the Horizon
Issues, it is an attempt to avoid answering them.
My Lord, the point is just as stark. Perhaps it is
even more stark with remote access. If I could go to
our closing submissions. That's {A/6/247}. At
paragraph 740 of our closing, perhaps I could invite
your Lordship to read it.
Mr Coyne accepted that any remote access work
involving branch data is typically recorded in PEAKs.
So if there is any remote access happening you will see
it in a PEAK. You may not have full details but you
will see that it happened, typically.
The relevant reference is {Day16/28:1}. Then if one
moves to the next page {A/6/249}, actually,
paragraph 746:
"Mr Coyne fairly accepted that he had identified
relatively few instances of remote access being used to
affect branch accounts, especially compared to the vast
number of branch accounts over the life of Horizon."
So I ask:
"Question: I would be right in thinking, wouldn't
I, that of the PEAKs you have seen you found relatively
few examples of remote access having been exercised?
Would the answer to my question be right?
"Answer: I don't know exactly what the number will
be, but it is tens, twenties --
"Question: Looking at your report it would be low
tens, wouldn't it? You haven't found hundreds?
"Answer: No, I haven't found evidence of hundreds,
no."
I ask:
"Question: So you have found, as I say, a
relatively small number; relative to the fact that we
are talking about 3 million branch accounts over the
last 20 years, all you have actually found is a very
small number which is less than 20 or 30, let's call it
less than 30, would you agree with that?
And his answer was:
"Answer: Yes."
So when considering remote access, which is a big
subject and by its very nature it is bound to be the
subject of forensic interest, it is easy for lawyers to
be considering that kind of issue and it is fun for the
lawyers to do it because there is the Select Committee
meeting at which Post Office gave evidence, and
Post Office's response to the Panorama programme, all
the things which engage the press and are in the public
eye. These things naturally focus attention, and not
surprisingly -- I don't blame my learned friend for
doing it, I am sure I would do it were I in his
position -- that goes on page 2 of my learned friend's
closing submissions.
But if you take your eyes away from the pyrotechnics
to stage left and you actually look at the Horizon
Issues, you see that to call remote access a second
order issue actually overstates its importance. It is
actually, if you think about it, a third order issue.
You would need some remote access to change branch
accounts. Mr Coyne says he hasn't found more than 30.
I respectfully submit the number is far lower than that.
Your Lordship has seen what we say in our
submissions but let's take him at his highest. He says
there is not more than 30. Then you need error or
malice which might have an adverse impact on branch
accounts. That, on any view, is going to be rare
because we know from the evidence given by Mr Roll and
the evidence given by Mr Parker that these things were
done carefully.
And no examples have been found of any remote access
actually causing a discrepancy, and certainly not
a shortfall in branch accounts. Only one candidate has
been put forward; that's {F/432/1}. That's the $1,000
bug. Your Lordship has lengthy submissions about that
in our closing submissions.
We say as a matter of principle it is clear that the
error was made in that case, if it was an error -- no,
it wasn't an error. The change to data that was made in
that case which it is suggested had an impact on branch
accounts was actually a change in back end data; it was
an OCR making a change to data that was in the TPS
system. It wasn't any change to branch account data at
all, it couldn't be because it was in the TPS system.
So there is not one single, we submit, example given
of any exercise of any remote access facility which has
actually been shown to adversely have affected the
branch account.
So, on any view, the proportion of cases in which
that happens is going to be small. As I have already
submitted to your Lordship, the remote access when it is
done, it is done reluctantly, it is done professionally.
There is a four eyes principle. My learned friend has
perfectly fairly drawn attention to what I think he
describes as many examples of OCPs where the monitoring
individual seems to be the same as the individual doing
the change, from which it is inferred, and I don't know
whether this is right or not, I'm not in a position to
dispute it, that the four eyes principle has become two
eyes in those particular cases. That doesn't mean
mistakes have actually been made, but whether they were
or not, it is quite interesting to note that he says
many. So far as we are aware on this side of the case
it is only four examples.
They have plainly been looking very hard. That
gives an idea of the smallness of the problem that we
are looking at. And there's the third requirement. Not
only is it necessary for there to be some relevant
remote access in the first place and for there to be
an error, carelessness or malice, one can't imagine what
malice there would ever be, but there would also have to
be the sort of situation where the postmaster himself
doesn't know that it is happening.
My Lord, it is very important to recognise that in
all the PEAKs that have been looked at and all the OCPs
and the OCRs and the MSCs that have been looked at, so
far as I'm aware only one example has been found of
an SPM not being involved in the process by which his
branch data was worked on.
And I discussed that, as I recall, in
cross-examining Mr Roll who frankly admitted -- this is
from memory so I need to be careful, but my recollection
is that he frankly admitted that it would be suicidal to
muck about with a postmaster's branch accounts without
involving him because then all sorts of problems could
occur.
I do hope my memory is correct. And I will check it
to make sure.
So where do we get to with all of this? 30, taking
Mr Coyne's figure at its highest, multiplied by a small
fraction to reflect a possible error rate, multiplied by
a small fraction to reflect the SPM not knowing what has
happened results in an absolutely tiny number. It would
be surprising if it were more than one in the real
world.
What scale are we talking about? The scale is over
3 million branch accounts. So the chance of a remote
access problem adversely affecting a branch account is
absolutely tiny. It is 1 in 3 million. 2 in 3 million
maybe. It is being hit by lightning territory.
Even Mr Coyne accepted that the chance of remote
access adversely affecting accounts is reasonably small.
My Lord, that's referred to in our submissions at
paragraph 748 at the bottom of this page {A/6/249}. He
didn't want to say vanishingly small, but he couldn't
resist saying it was small.
Yet in their answers to the Horizon Issues regarding
remote access, what do the claimants say? They say
material and significant risk. In closing they put
remote access at the forefront of their case. In my
respectful submission, that they seem to think that
remote access is their best point might be thought
revealing about the overall merits of their case.
My Lord, I'm going to move on and deal briefly now
with some criticisms that were made of Dr Worden. I'm
going to do this as quickly as I possibly can.
The suggestion is made that he is biased. I refute
that suggestion. Reliance was made on the first joint
statement which, in my submission, was entirely
misplaced. If we could go to JS1 at {D1/1/1}, please.
I don't have the page reference. Would
your Lordship give me a moment?
MR JUSTICE FRASER: Is this the part where they set out
balanced approach?
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: I think it is the first page after this,
if we could just go forward.
MR DE GARR ROBINSON: It is page {D1/1/3}. So:
"Each expert's approach to writing his report, and
to this joint memorandum which foreshadows their
reports, could broadly be one of three possible
approaches:
"(a) to focus mainly on negative points ...
"(b) To focus mainly on those aspects of Horizon
which were intended to achieve robustness ... and the
evidence implying that they succeeded."
And (c):
"To provide the court with a clear foundation for
understanding the design and operation of Horizon; then,
building on that foundation, to provide a balanced
assessment of the ways in which Horizon succeeded,
whilst addressing any disclosed issues where Horizon may
have fallen short."
What Mr Coyne said he was going to do is at the
bottom of the page, two lines from the bottom:
"Whilst my report will take a balanced approach, it
is the case that many of the issues require a deep focus
on the occurrences of bugs, errors and defects as well
as the potential for modification of transactional
data."
It is interesting to note that his deep focus was on
bugs and his deep focus was on potential for
modification. But I ask rhetorically, how many
instances did he find of bugs? 29. How many impacts?
500 and something. And how many instances of remote
access? Not more than 30.
My Lord, both of those numbers, as your Lordship
will be aware, are disputed on this side of the court.
But that was the result of his deep focus.
Then Dr Worden, he says:
"I intend to take the balanced approach (c)."
My Lord, in my respectful submission, that's exactly
what he did. He looked at the architecture to see how
Horizon worked. He looked at the countermeasures within
Horizon to see whether they were properly configured and
designed in order to achieve the robustness which they
were designed to achieve. He looked at testing to see
how they were tested to make sure that they were tested
properly.
There is an entire section on testing in his second
report, for example, where he enhanced his review of the
testing process. Then he reviewed how the
countermeasures had operated during the entire life of
Horizon. Your Lordship will see that from his tables in
appendix D of his first report and then the updated 200
bug table in his second report.
My Lord, in my respectful submission, that was the
right thing to do. My learned friend criticises him for
taking a top down approach, but I would respectfully
observe that Horizon is a massive system. If you don't
start by seeing how it is configured and how it works
from the top, and if you don't then delve into the
system to see how the support systems work, to see how
bugs are dealt with and so on, then you are likely just
to get an incoherent catalogue of problems which aren't
properly organised and are just a wall of problems which
are very difficult to manage.
In my respectful submission, that's exactly what
Mr Coyne's reports produced.
I do refute the suggestion, which I am sure my
learned friend has made, that Dr Worden only stayed at
the top level. He went down, he looked down into the
KELs, he looked at PEAKs, he looked at dozens of PEAKs
at the time of his first report and looked at many, many
more by the time of his second report and by the time of
the trial.
The exercise that he performed was not a biased one.
He looked for good and bad. There were sections in his
report where he criticised Post Office and Fujitsu. It
is worth noting, my Lord, that of the 29 bugs that are
currently in the bug list, nine of them were actually
identified by Dr Worden. So if it hadn't been for him,
we would be looking at 20 bugs in all probability.
Your Lordship will get that. It is paragraph 253 of
our closing at footnote 383, page 100. We needn't go
to it.
Dr Worden made some mistakes. He, for example, it
is just an example, made a mistake when doing his
scaling calculation first time round, dividing by 461
instead of a smaller figure. 561 instead of 496,
I think it was. My Lord, that was a mistake. Actually,
as he explained in cross-examination, he dealt with the
mistake. The mistake wasn't repeated when he redid the
calculation for the purposes of his second report, but
it wasn't his only mistake. He did make mistakes.
But there's no basis for saying that he was biased
and there's no basis for suggesting that his mistakes,
if that's what they were, had any substantial impact on
the validity of his ultimate conclusions.
The second suggestion that's made was that he
assumed Horizon was working properly and worked back
from that. My Lord, that's simply not the case. If you
look at all his assumptions, if you look at the
financial calculation we looked at previously, the
spreadsheet, look at all his assumptions about rate of
imperfections. A fortiori look at all his conservative
assumptions which push things radically into the
claimants' direction.
Criticisms are made of his scaling approach.
My Lord, there were some mistakes, but his evidence was
clear. Most of the points that were put to him were
points which misunderstood his approach to undertaking
calculations of that kind. Approximations are not
mistakes. Engineering estimates always involve
approximations. The important question is what
precision do you need to arrive at a usable answer and
what precision is available in the light of the
information you have?
You have to beware searching for spurious precision.
My Lord, I freely accept it doesn't make my job easier
that he did take approximations of figures, he took
48 million instead of 47 million for one of his
calculations, and your Lordship will remember the GAAP
figures. He didn't take the GAAP figures in another one
of his calculations. But, my Lord, all of these things
are small beer compared with the assumptions he makes in
the claimants' favour in {D3/8/1}, 0.45 to 0.5, the
row Y calculations and so on.
Dr Worden is criticised because he didn't consider
Dalmellington at all in his first report and he didn't
consider it properly in his second report. Well,
my Lord, he did consider it in his first report. It is
considered twice actually in his appendix D3 where he
goes through the KELs that Mr Coyne had relied on.
For your Lordship's note -- we don't need to go to
it -- it is {D3/2/117} and 132. Both of them consider
the KEL acha621P. That's the Dalmellington KEL,
I think. So he did consider it but he didn't consider
it, address it specifically in the text. He did address
it specifically in his second report, but there he is
criticised for not having specifically mentioned, for
example, that there were 102 occurrences of the
Dalmellington bug before it was detected.
The reason why there were so many, of course,
your Lordship will be aware, is because the
Dalmellington bug mimicked, it caused human error. So
it looked exactly like human error when it occurred.
And what's more, it was fixed as and when it occurred by
reason of the ordinary countermeasures that apply in
Horizon.
So it never caused any lasting harm. It just looked
like human error and it only got picked up. That's why
it lasted as long as it did and that's why there were so
many impacts. It is not a representative bug in that
respect.
But he deals with it quite fully in his second
report at paragraphs 144 to 163. Mr Green seems to be
aggrieved because he doesn't specifically mention the
112 occurrences, but what he does do is he refers to
Mr Godeseth's second statement, paragraph 55 onwards,
which does contain that information.
I do find myself asking forensically why on earth
does Dr Worden have to repeat the same information
a second time? My Lord, I'm now going to address you on
what's loomed large in this litigation. It is not
a Horizon Issue, but it is a matter to which the
claimants keep going back. It is the question of
disclosure. I'm going to do that, if your Lordship will
permit me by reference to our closing submissions.
If I could ask your Lordship to go to {A/6/371},
picking it up at paragraph 1119. There's
your Lordship's order of 2nd February requiring Model C
disclosure.
Paragraph 1120 and 1121 draws your Lordship's
attention to the practice direction dealing with the
pilot scheme disclosure, and notes that:
"... It is for the party requesting Extended
Disclosure to show that what is sought is appropriate,
reasonable and proportionate (as defined ..."
It is described also as "request-led, search-based
disclosure".
Paragraph 8 of the practice direction also provided
over the page at {A/6/372}:
"(1) The court may order a party to give disclosure
of particular documents or narrow classes of documents
relating to a particular Issue for Disclosure, by
reference to requests set out in or to be set out in
Section 1B of the Disclosure Review document or
otherwise defined by the court."
So what does that establish? It established the
requirement of a request:
"(2) If the parties cannot agree that disclosure
should be given, or the disclosure to be given, pursuant
to a request, then the requesting party must raise the
request at the case management conference …"
Now, I ask forensically, why does it use the word
"must"? The answer is simple. The fundamental shift
that has been achieved between ordinary, if I may say
so, sounding like a dinosaur, standard disclosure and
modern Model C disclosure is that with standard
disclosure there is a default obligation to provide
documents that adversely affect your case or support
your opponent's case.
There is no such requirement in relation to Model C.
All there is is a requirement to disclose known adverse
documents, and a critical word there is of course
"known". You have to know about, and if you don't know
about it then you are not under an obligation to
disclose it.
So in order to be in a position to complain about
not having had a document, it is incumbent upon you as
the requesting party if you are not satisfied by the
response to go to court to get an order. That's how it
is supposed to work. Ironically, it means that more
work now needs to be done by the court than hitherto.
Although it is supposed to reduce costs, I rather wonder
whether in the long run it will increase them.
So that's the regime that we work under nowadays in
these courts.
If I could pick it up at paragraph 1125:
"By the Court’s Fourth CMC Order 1368 Post Office
was ordered to disclose the documents set out in
Schedule 1 to the Order. Schedule 1 was largely agreed
by the parties (mainly before the CMC but some of it
after). It focused on various documents (in particular
reports and briefings) being provided to various
categories of custodians which were defined in the
Schedule itself."
So that's Model C working as intended.
The parties come together, they have a conversation,
they reach an agreement and an order is made. There was
no outstanding issue at that stage, it would seem.
Certainly there was no application for anything further.
So we say at 1126:
"This was an example of Model C disclosure working
broadly [as intended]."
If one goes over the page {A/6/373}, paragraph 1127
adverts to paragraph 13 of the third CMC order where
your Lordship reminded the parties that the experts had
the right to come to court to seek directions.
If I could pick it up at paragraph 1130. Well,
perhaps at 1129, the fourth CMC order:
"... the Court, by paragraphs 8 and 9 of the Order
ordered the experts to provide an Error Codes List and
jointly to compile a list of information which either or
both considered they required."
1130:
"This joint report was duly produced on
26th June 2018."
It should be, it was a request:
" ... for a huge amount of information, explanation
and documentation."
My Lord, stopping there, what often happens in cases
of this kind is that when issues have arisen, when the
car crash has already arisen, you look back with the
benefit of hindsight and you look back at a particular
decision or letter and you look at it through
a microscope and you think "Isn't that awful, I can't
imagine why on earth that was done". Of course at the
time when a party is faced with 30 or 40 pages of
requests which go up hill and down dale which involve
all sorts of different things, and remember these are
technical requests which often the lawyers won't
entirely understand, the lawyers don't act for Fujitsu,
they act for Post Office, that can put a recipient,
a receiving party in a difficult position. And it put,
in my submission, Post Office in a difficult position.
Then over the page {A/6/374} at paragraph 1131,
Mr Coyne on 20th July sent an email to Freeths and WBD
requesting yet further material. And at 1132, it says:
"Some mechanism was clearly required in order that
(i) each side's position on the various requests being
generated by Mr Coyne was properly set out and could be
fully considered and (ii) the requirements of Model C
were met."
And 1133:
"The parties agreed a sensible mechanism which the
Court approved in the Fifth CMC Order ..."
Perhaps I could invite your Lordship to read the two
paragraphs quoted there, the order's at C7, tab 22, but
we needn't look at it.
MR JUSTICE FRASER: The two paragraphs at 1133?
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: My Lord, if I was a trust lawyer
I would say those two paragraphs gave effect to the
spirit and intendment of the pilot scheme for
disclosure, Model C disclosure. In particular, it
created a mechanism by which, if there were any
objections, the claimants would explain why the
documents were needed and then the recipient,
Post Office in this case, would be required to elect as
to whether to continue its objection or not, with the
benefit of the explanation as to relevance and why it is
reasonable and proportionate.
That is how Model C is supposed to work in my
submission.
This mechanism would have ensured that by the end of
August, in fact by the middle of August, the parties
would have known where they stood and it would have been
open to the claimants to raise the matter with
your Lordship and seek an order if need be, if agreement
couldn't be reached.
That is how Model C disclosure is supposed to work.
That's effectively what's said at paragraph 1134. But
your Lordship will see over the page at 1135 {A/6/375}
or 1136, Post Office's response on 8th August:
"[It] provided its response, as required by the
Fifth CMC Order, on 8 August 2018. In many cases, it
provided the information requested and additional
documentation. In other cases it did not. Post Office
was genuinely unsure of what Cs' case was in relation to
many of the Horizon Issues – neither the Outline
Document nor Mr Coyne's reports had been served at this
point – and many of the requests sought not categories
of documents but lengthy analyses and explanations of
matters which Post Office did not have and which
appeared to cut across the carefully negotiated
categories of documentation which the Court had ordered
at the Fourth CMC."
So following this impasse, the claimants were put to
their election pursuant to the fifth CMC order, they
should have served an explanation by 18th August if they
maintained a claim to any of these requests together
with the necessary explanation.
Did they do that? No, they didn't. They in fact
sabotaged this entire exercise by doing nothing. I'm
not here to blame them for doing nothing, I am sure they
had plenty of other things to do at the time. But,
my Lord, I'm here defending constant criticisms made of
my clients, including during my cross-examination of
witnesses. My learned friend kept popping up and giving
your Lordship "That was disclosed on such-and-such
a date". The reason why we are having this conversation
is because Post Office has been consistently attacked by
the claimants for failing to give proper disclosure in
circumstances where those criticisms are based upon
a fundamental misunderstanding of the way that
disclosure is supposed to operate in the modern world,
in my submission.
If one were to look at the claimants' account of
disclosure in their own closing submissions, it is quite
revealing. We needn't go to it, but that is
paragraph 40 of their closing, {A/5/15}. They suddenly
say Mr Coyne made some requests in December or renewed
his requests in December. But, my Lord, by then it was
the ground rush to trial.
Two points that it is important to make clear about
their complaint about disclosure. The first point is in
actual fact, in my respectful submission, the claimants
have had the disclosure that they could legitimately
expect to have. Their real complaint is not that they
have not been given the disclosure they need, their
complaint is that it was disclosed late.
But, my Lord, that brings me to my second point,
which is why was it disclosed late? The answer is it
was disclosed late because they didn't take the
opportunity that they had agreed that they would take in
the terms of the fifth CMC order.
If, as I say, the claimants' only objection is one
of delay, at whose door should responsibility for that
delay be laid? In my submission, it should not be laid
at the door of Post Office. Post Office tried to comply
with the order. It did comply with the order and it got
no response from the claimants, and it does not lie in
the claimants' mouths in those circumstances to say that
between August and December Post Office should have been
busying itself voluntarily disclosing documents it
thought were a dead letter because it had no response.
My Lord, there is the third submission that it is
important to bear in mind as well, because although in
my submission the only real complaint that can be made
about disclosure is one of delay and, as I say, if
a complaint is to be made it is to be made of the
claimants, not of the Post Office, but at the same time
the claimants also wish to maintain a suggestion that
the reason why this has been happening, the reason why
they didn't get these documents when they needed them,
was because Post Office is exhibiting a lack of candour
about its case. That's lawyer's speak for trying to
conceal the truth from the court.
My Lord, nothing can be further from the truth. In
my respectful submission, it is an outrageous submission
for the claimants to make in circumstances where they
had the power in their own hands, an agreed power, to
ventilate this issue in a way that they were required to
do pursuant to the pilot scheme practice direction and
they chose not to take it for whatever reason.
My Lord, just a final point on lack of candour. It
really is important to note that the documents that have
been disclosed, the KELs were disclosed in May last
year, they have had the KELs for well over a year. As
soon as the Horizon Issue trial was ordered, KELs were
agreed to be disclosed.
No order of disclosure for PEAKs has ever been made,
but the PEAKs were identified in the Post Office's EDQ
back in December 2017 and steps were taken. It was
difficult to extract the PEAKs from the proprietary
database on which they are held by Fujitsu and software
had to be written in order to allow them to be extracted
so that they could be used in a usable way. And that
cost quite a bit of money, as I understand it, and that
took time, but there was no resistance to giving them
the PEAKs. No order has actually been made but PEAKs
were provided as soon as they became available, and they
became available in the September.
It is in the submissions. I'm going too fast.
My Lord, OCPs, OCRs, MSCs, I have to confess at the time
I didn't really understand what these documents were.
I'm speaking for me. But, again, once Mr Coyne's report
had been served and it became clear what the
significance of these documents were, urgent steps were
taken actually to secure that these documents were
available. And the dates on which they became available
were set out later on in these submissions.
But, my Lord, this is not a story of Post Office
seeking to hold back material that's unhelpful to its
case. If that had been Post Office's objective, it
wouldn't have run around like a headless chicken trying
to get these documents for the claimants without any
even threat of a court application.
What's more, the effect of these documents has not
been unhelpful to Post Office. In fact, they have been
helpful. The PEAKs, its OCRs and the OCPs and the MSCs
all show, first of all, how few bugs there have been
that are relevant for the purposes of at least Horizon
Issues 1(a) and 3, and secondly, how minuscule the
exercises of remote access have been in the context of
a system of this size.
Those documents are actually helpful. There has
been no smoking gun. There has been nothing that has
been revealed that changes the whole complexion of the
case, save insofar as it reveals things and has been
useful for the arguments I'm now making to
your Lordship.
So I would invite your Lordship to view with some
scepticism an entirely understandable and, I have to
say, entirely typical submission from my learned friend
to the effect that Post Office is engaging in a process
that is inappropriate and deserves judicial criticism in
relation to this particular instance. In my submission,
such a suggestion is grotesquely unfair.
Then, my Lord, just going on through the
submissions. It is worth adverting to, but only very
briefly, paragraph 1146: the decision of the Court of
Appeal in Serafin v Malkiewicz. A very different case:
a question of procedural unfairness to a party. And the
Court of Appeal naturally drew attention to the fact
that the judge made criticisms of the party for not
disclosing documents that that party had not been
ordered to disclose. And I would invite your Lordship
similarly to refrain from making criticisms of that
sort.
MR JUSTICE FRASER: I am glad you have mentioned that case
because that was one of the questions I had for you.
That is under the heading "The Court's interventions on
disclosure".
That case concerned demands and criticisms by the
judge during the trial. Now, I'm unaware of making any
criticisms or demands during this trial.
MR DE GARR ROBINSON: I'm going to interrupt your Lordship
with the greatest of deference.
MR JUSTICE FRASER: Well, I'm interrupted regularly by both
of you, so ...
MR DE GARR ROBINSON: I want to make it clear that this case
is not cited for the purpose of affecting your Lordship
in any way, or suggesting to your Lordship in any way
that that has happened during the case of this trial.
And I would like to reassure your Lordship of that fact.
MR JUSTICE FRASER: I'm not necessarily seeking reassurance,
I'm just seeking clarification because 1146 {A/6/377}
says that that passage is relevant and it is under
a heading "interventions on disclosure". Now, when
I read it I went back and reminded myself not only of my
approach to disclosure during this trial, but Common
Issues trial as well, and I did in the Common Issues
Judgment make certain observations in respect of
managing the group litigation generally.
But so far as this trial is concerned, I think
I have only been involved on three occasions. One was
to invite you to perform a redaction review.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: One was to require a witness statement
of explanation in respect of Royal Mail.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: One was to require a witness statement
of explanation in respect of a large quantity of
documents which came, I was told, originally in May, May
of this year. But I do not think I have criticised
anybody. I certainly haven't intervened. But if I have
I would like you to identify those occasions to me.
MR DE GARR ROBINSON: My Lord, I do not think you have
criticised. No, I would go further. If you had I would
remember, and you haven't. The purpose of my reliance
on this case -- this is a terrible thing, I'm throwing
my learned friend under a bus here --
MR JUSTICE FRASER: No, please don't throw anyone under
a bus.
MR DE GARR ROBINSON: It is that the importance of this
case, this paragraph -- it is a paragraph -- is simply
the principle that my learned friend is making
criticisms which are based upon the very error that was
identified by the Court of Appeal in this case.
That's all.
MR JUSTICE FRASER: Well, that's rather my point. But it is
not necessary to spend too long on it. But what
occurred in that case were interventions during the
trial in respect of disclosure said to be deficient, in
respect of which there was no order.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: What Mr Green did and his team did in
their opening is they made a number of complaints in
respect of the known error log, which we call the KELs,
the PEAKs etc. And as you rightly point out there have
been no applications in respect of that. It is more
what could be said to be a generic complaint about the
Post Office's approach generally, which I understand
your submission to be is effectively made for forensic
reasons, is what it comes down to.
Let me put it slightly differently. I'm generally
unimpressed with a party complaining about not receiving
disclosure if they haven't made a specific application.
Equally, during this trial there have been certain
features of disclosure that have developed, for example,
Royal Mail and the recent bulk disclosure when Fujitsu,
I think you explained, had forgotten that there was some
archive documents --
MR DE GARR ROBINSON: This is the OCRs?
MR JUSTICE FRASER: Yes. Which so far as the court is
concerned I'm always interested in having an explanation
in a witness statement. But I'm not aware I have
criticised the Post Office.
MR DE GARR ROBINSON: Nor am I, my Lord, and I'm happy, I'm
glad your Lordship has given me the opportunity to make
that clear.
The simple point that I make, and I'm glad I have
come to this so we have had this discussion, the simple
point that I make is that just as it would have been
inappropriate were your Lordship to have done that, by
the same token it is inappropriate for my learned friend
to make those criticisms in the hope that your Lordship
is going to echo them.
MR JUSTICE FRASER: I understand. I mean, I also
appreciate, so far as disclosure is concerned generally,
that the Post Office has been reliant on Fujitsu to
a large extent.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: But now that we have the Fujitsu
Post Office contract, it is obvious that Fujitsu has got
a contractual obligation in respect of assistance so far
as documents are concerned, but those things take time
obviously.
MR DE GARR ROBINSON: Yes, and whether that obligation --
I was going to raise the question of what that meant
about control, but my Lord, I'm not in a position to do
that.
MR JUSTICE FRASER: I do not think it is necessary, to be
honest. But the short point is, and I think it is
mentioned once or twice in this passage in your closing,
that the Post Office was reliant upon Fujitsu.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: And as you say, for example, the PEAKs,
they had to develop the certain piece of software to
extract them.
MR DE GARR ROBINSON: Yes, my Lord. I believe that's also
the case with OSPs, and also MSCs have to be -- a lot of
work had to be done in order to -- your Lordship will
have heard Mr Coyne complain about the form in which
MSCs were disclosed. Well, they would have been
completely unusable if effort hadn't been made to put
them into spreadsheet form. And actually, in relation
to that it is worth mentioning that it is possible, it
was possible on the spreadsheet to actually download
individual documents and drop them onto paper in the way
that I did, or was done at my request, so that one could
see precisely what the documents showed.
My Lord, I see that it is 12.58. I have still got
perhaps 10 or 15 minutes on disclosure.
MR JUSTICE FRASER: Do you want to stop now?
MR DE GARR ROBINSON: This may be a convenient moment.
MR JUSTICE FRASER: All right. We will come back at
1.58 pm. Thank you very much.
(12.59 pm)
(The short adjournment)
(1.58 pm)
MR DE GARR ROBINSON: My Lord, two matters of housekeeping
before I proceed.
First of all, I have handed up a series of
typographical -- they are purely typographical --
corrections to our closing submissions. Some of the
internal paragraph references hadn't been updated when
the document was served and there were some false
references. An amended version of the submissions will
be uploaded onto Magnum, but I imagine that
your Lordship will have already started marking up the
version that you have got, so a copy of just the
amendments that have been made has been provided and
that's available to your Lordship.
MR JUSTICE FRASER: I actually got two hard copies. I was
working off the one that came from your solicitors,
although I did get one from you. Yours was in two
volumes and theirs was in one, which is why I used
theirs not yours. But I assume they are both exactly
the same.
MR DE GARR ROBINSON: I would be very surprised if they
weren't.
MR JUSTICE FRASER: That's why I mentioned it. These are
corrections to the submissions?
MR DE GARR ROBINSON: Yes. They are purely typographical.
It is things like footnote 5 should be GDCC, and things
like that.
MR JUSTICE FRASER: Thank you very much.
MR DE GARR ROBINSON: My Lord, the second housekeeping
matter is something I have been taken to task on over
the short adjournment.
When addressing your Lordship on what the impact
would be of having a trial of the reconciliation and
transaction correction processes in Post Office that
have existed over the last 20 years, they have changed
quite a bit during the course of those years, as
I understand it, I told your Lordship it would be over
1 million documents and could take six months. I have
to say that was a product of my brain.
My instructing solicitors have very properly
indicated to me that rather than just firing off the hip
with a prediction of that sort one should be rather more
careful, and really my submission should have been if
those things were to be included in the Horizon trial it
would not have been possible to have covered them within
the timetable that was available by any means.
MR JUSTICE FRASER: All right.
MR DE GARR ROBINSON: So I stand corrected in relation to
that.
My Lord, disclosure. I was on the question of
disclosure and I addressed your Lordship on the
standards that I submitted the claimants were, in my
submission, inappropriately seeking to impose on
Post Office.
Before leaving the question, I would like to compare
the standards that the claimants have been seeking to
impose with the standards that they are willing to
accept for themselves.
My Lord, if we could pick it up in the disclosure
section of our written submissions, and if we could pick
it up, please, at page 382, paragraph 1171 {A/6/382}:
"By paragraph 5 of the Fourth CMC Order, Cs were
ordered to provide disclosure of documents upon which
they intend to rely at Horizon Issues Trial ..."
At that time, it pointed out in 1172:
"... that the Fourth CMC Order provided for witness
statements to be limited to 'any witness of fact whose
generic evidence (in distinction to Claimant-specific
evidence) they wish to rely upon for the purposes of
determining the Horizon Issues'."
So those two orders went together.
The reason why there was a limitation in the order
that was applicable to the claimants was because the
claimants weren't expected to be given the same sort of
evidence that the Post Office would be giving.
That point is made in paragraph 1173 {A/6/383}, and
I invite your Lordship to read that.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: Then picking it up at {A/6/383},
paragraph 1174:
"On 17th July ... Cs provided disclosure of 45
documents. At this time it was not known to Post Office
that Cs would be serving extensive claimant-specific
witness statements."
But certain slightly ominous sounding statements
were made and Post Office sought to explore that in
1175, explaining that:
"... 'additional relevant documents may come to
their attention, and most likely as a result of the work
of experts'."
So no mention was made of claimant-specific evidence
that was to come, but they did say it was not their
intention to make another round of disclosure.
Then there was correspondence in which my
instructing solicitors sought to press my learned
friends on the scope of the disclosure which had been
given, and your Lordship will see the submission that's
made in paragraph 1178 on page {A/6/384}.
There is a suggestion that there was a desire to
avoid a repeat of late disclosure of documents close to
trial which had happened at the Common Issues trial.
My Lord, I don't know about that.
Then we then come to 28th September where the
claimants serve nine witness statements, including
several claimant-specific witnesses. Those witness
statements for the first time exhibited 17 new
documents -- that's in paragraph 1180 -- and a total of
30 new documents were disclosed at that time.
MR JUSTICE FRASER: Is this with the witness statements that
means?
MR DE GARR ROBINSON: My Lord, I believe so, yes. At the
end of paragraph 1180.
Then, your Lordship will see quite a lot of
correspondence where my instructing solicitors are
pressing for information as to the scope of the
disclosure that has been taken.
Then at 1182 there is a reference to the letter of
14th January:
"... 'Given the nature of the Horizon Issues Trial
there was good reason for the court to seek to limit the
ambit and cost of disclosure from the Claimants
themselves. Hence, the disclosure order made here; the
obvious good sense of that has not changed.'"
Well, my Lord, your Lordship will anticipate what my
response to that is. Everything changed when the
claimants served substantial evidence of their own which
they weren't supposed to, and that's the point that's
made in paragraph 1183.
At paragraph 1184 it is pointed out that as
claimants had flouted the order that your Lordship had
made about evidence, Post Office sought to understand
what searches, if any, had been conducted by claimants
so that it could understand whether further disclosure
would be required.
Several requests were made.
1185:
"This lack of response by the Cs should be viewed in
light of the approach adopted by Post Office. One
example of the further disclosure that Post Office has
agreed to give to Cs which is outside that ordered by
the Court is the requests for disclosure made by Cs on
18 December 2018 in which the Cs sought ..."
Your Lordship will see the following categories of
documents including:
"... 'disclosure of the documents that were
responsive to searches by, and collated by the Defendant
in respect of the operation of branches by Angela Burke,
Aakash Patny, Anup Patny, Jayesh Tank, Setpal Singh and
Adrees Latif. We would expect such documents to include
but not be limited to."
Then there is a whole series of documents which the
claimants were demanding.
So your Lordship will see the double standards that
are being applied by the claimants. On the one hand
they are standing on the orders, or rather the lack of
orders, that have been made against them, but at the
same time although no order has been made against
Post Office either, they are demanding all manner of
further documents.
Further requests for claimant-specific disclosure on
4th February. Post Office responds on the 11th and
provides disclosure on the 20th:
"The failure by [claimants] to provide their own
claimant-specific disclosure suggests that they clearly
intended disclosure to be a one-sided exercise."
{A/6/386}
My Lord, there is a further observation I would
respectfully make about that. The observation being
Post Office are criticised for only producing documents
on 20th February. My Lord, in my submission, the double
standards explicit in that criticism takes the breath
away.
On the one hand claimants refuse to give disclosure
themselves or explain what disclosure has been given
further to the few documents they have or provided. On
the other hand they are demanding more and more
documents from Post Office. And one can understand why.
I'm not criticising them for that. What I'm criticising
them for is for applying double standards between
themselves and Post Office.
Then at 1187 the claimants write to Post Office
indicating there is a document which is helpful to the
claimants, Mr Tank's Yahoo -- what is it called? --
Yahoo group forum document which he has found. He has
found a post. The point is made just for consistency,
notwithstanding the claimants' complaints about delay,
it is pointed out in paragraph 1187 that there is no
explanation of delay by the claimants. This is a double
standards point, not a complaint point.
My Lord, there is then a series of paragraphs
dealing with various documents that appear to exist. At
page {A/6/387} there is a long part of the transcript of
Mr Tank's cross-examination.
About halfway down the page, where Mr Tank says:
"Answer: I kept all my Post Office sort of related
paperwork in a box file and that's -- when I was asked
to look for evidence I went strictly to that box file
and that's where I sourced all my information from."
So when asked "How do you know all this?" he says
there is a box file that's got this information.
Similar points arise at the top of page {A/6/388}.
A lot of reliance on this box file. Amongst other
things, 1192:
"... it was Horizon generated receipts, print-outs,
with hand-written dates and reference numbers on them."
My Lord, a similar summary of the evidence of
Mr Latif where he discusses the records that he had
access to to help him change his evidence. My Lord,
that led to a request for disclosure which your Lordship
will see on paragraph 1196 on page {A/6/390}.
My Lord, just to cut a long story short,
your Lordship will see at paragraph 1198, {A/6/391}, the
claimants refused:
"... to provide further disclosure of the requested
documents on the basis that they were either outside Cs'
control, no longer existed or were not relevant to the
[Claimants'] evidence and not relied upon at trial."
My Lord, if one compares the approach that
Post Office adopted to providing disclosure relevant to
the allegations made by the claimants against them with
the approach that the claimants have adopted in relation
to the same issues, one does see distinct divergence, in
my submission.
The claimants appear happy to stand behind the fact
that no order has been made against them, whereas they
won't for a moment countenance the possibility of the
fact that no order has been made as being at all
relevant to Post Office's position.
To be fair, Post Office hasn't stood on the orders
either. Virtually all of the disclosure about which the
claimants make so much complaint is actually all
voluntary. But still Post Office is accused of lack of
candour.
My Lord, before leaving the question of disclosure,
and in particular disclosure by the claimants, I would
respectfully submit that that reinforces the point that
I hope your Lordship already has loud and clear from our
submissions in relation to how inappropriate it would be
for your Lordship to make findings that
claimant-specific witnesses were the victims of
undetected bugs or other defects in Horizon causing them
trouble.
In principle --
MR JUSTICE FRASER: You mean in a Horizon Issues judgment?
MR DE GARR ROBINSON: Yes. My Lord, in principle that is
a matter for --
MR JUSTICE FRASER: None of the Horizon Issues go to that
at all.
MR DE GARR ROBINSON: Well, my Lord, the reason why
I mention it is because your Lordship is invited to make
all sorts of findings by the claimants and I'm quite
anxious that your Lordship should understand, and
I imagine you already do, that we say that would be
an entirely inappropriate --
MR JUSTICE FRASER: Well, I'm trying the Horizon Issues and
I am answering the Horizon Issues. There will obviously
be a lot of detail in the judgment prior to arriving at
the answers to those issues. The idea that a specific
finding could be made on the evidence before this court
on a specific shortfall for a specific claimant in
a specific amount allocating it to a specific bug seems
to me to go outside the Horizon Issues.
MR DE GARR ROBINSON: I'm obliged to hear that from
your Lordship.
MR JUSTICE FRASER: But I'm surprised it is a point of
enormous difference between you given the orders, given
the CMC orders about what the trial is about.
MR DE GARR ROBINSON: My Lord, I agree. It is surprising.
But if your Lordship were to go to the claimants'
closing submissions, you will see that you are being
invited to make all sorts of findings. And one of the
most remarkable ones I think is in relation to Mr Latif.
Your Lordship will recall there is said to be
a difficulty about a stock unit transfer in or around
July 2015.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: The evidence demonstrates, in my
submission quite clearly, if your Lordship were to make
findings on it, which I say you shouldn't, the evidence
demonstrates quite clearly that there was no such
problem during the period identified by Mr Latif.
The claimants are suggesting your Lordship should
make a finding that it happened at some other
unspecified time that's not covered by any disclosure,
it is not covered by any evidence, in circumstances
where Mr Latif himself didn't say it happened at
a different time.
There is an Alice in Wonderland reaction to those
submissions, but it is right that your Lordship should
understand -- I can see that your Lordship does -- our
response to the submissions of that kind. In my
submission, it would be entirely inappropriate.
MR JUSTICE FRASER: Well, the way in which I deal in the
judgment with claimant-specific issues is obviously
something I am going to have to consider carefully and
deal with carefully. But so far as the findings on the
Horizon Issues are concerned, it is very clear on the
face of the order and the discussion that led to the
order that they are not going to be claimant-specific.
They are generic.
MR DE GARR ROBINSON: I'm grateful for your Lordship's
indication.
MR JUSTICE FRASER: That's not to say that I have
a concluded view now that I should ignore all of the
claimant-specific factual evidence, but I'm fairly
clear, as with the Common Issues in fact, that the
answers arrived at in the judgment are the answers to
the issues that I am trying. I would have thought
that's a fairly elementary approach.
MR DE GARR ROBINSON: To the extent that -- the experts --
Mr Coyne doesn't base any conclusions on the witness
evidence, as far as I'm aware. He refers to the witness
evidence as being consistent with his views. But
nowhere -- I hope I'm correct, I am sure I am correct in
saying. The fact that I'm saying I am sure I am correct
in saying means I'm not sure. But I think I am correct
in saying that nowhere does Mr Coyne conclude that
a particular bug existed because a particular
claimant-specific witness has said something.
At most, what he does is he says, well, I noticed
Mrs Burke had problems with reconciliation, or I see
that Mr Latif says this, that and the other. But he
doesn't, first of all, base any conclusions on the basis
of their evidence, nor should he. Secondly, nor does he
suggest that their assertions flow from a bug that he
has identified that could be the source of
a particular -- I think I'm right in saying he doesn't
specifically say bug 13. This was in existence at that
time and that's when we had Mr Latif's problem.
My Lord, that's not how his evidence is constructed,
as I recall. In those circumstances it would be
highly -- your Lordship as got my submissions. There
hasn't been proper disclosure. If there were to be some
findings about what the claimants did, what they
suffered, there would need to be an investigation of
wider issues about how the relevant branches were
operated. One would need to go into the history. One
would need to investigate all sorts of things.
These snapshot witness statements, most of which are
almost impossible to understand because they are so
short and bereft of detail, one would need to go into
far more material in order to be able to be in
a position to make findings at all. If your Lordship
were to make findings, in my submission the only proper
findings would be to dismiss the claims that are made.
But my primary submission is that your Lordship should
not entertain them in the first place.
My Lord, that completes my submissions on
disclosure.
Let me now come as briefly as I can to the bugs that
my learned friend addressed you on yesterday. What
I would like to do as briefly as possible is to talk
about bug 11, bug 12, bug 13 and bug 15.
Bug 11 is Girobank. Your Lordship will recall that
Post Office contends that there's no evidence of
financial impact from the relevant bugs. Your Lordship
may or may not -- I suspect not -- be alive to this
already, but bug 11 raised in appendix 2 raises six
issues. My learned friend didn't deal with all of those
six issues in his submissions yesterday. He only
addressed Issues 1 and 2 and he referred to a principal
PEAK which is at {F/25/1}, and perhaps I could ask
your Lordship to go to that very briefly.
My Lord, this is opened on 5th May. The claimants
argue that Post Office hadn't grappled with the concept
of what a discrepancy is, and I'm informed that that's
in the transcript at {Day21/101:1}. If that is their
submission, in my submission it is not correct.
Post Office has grappled with what a discrepancy is,
but the word "discrepancy" means more than merely
a comparison between one thing and something else, as
the claimants suggest.
The thing and the something else are relevant to the
Horizon Issues. They can't be any two things. Here
what's required is a discrepancy in branch accounts.
Now, the PEAK notes that there was a £505
discrepancy between a branch's cash account, and the
cash account at this period of time means its own
accounts, and its daily reports.
Your Lordship will see that from the bottom of
page {F/25/1}. 5th May at 15.02:
"New evidence added - Full message store.
"... Response:
"This difference (£505.72) between the Cash Account
and the Daily reports is explained by," a particular
KEL.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: The discrepancy there is not
a discrepancy in the branch's accounts. It is
a discrepancy between the accounts and the report that
was sent to Girobank every night. The issues discussed
in the PEAK are reporting issues, they are not account
issues.
Indeed, Mr Coyne agreed the following in
cross-examination, that the detection and investigation
of Issues 1 and 2 in this PEAK demonstrated the good and
effective operation of robustness countermeasures in
Horizon. My Lord, that was at {Day17/52:1} to 56, and
pages 63 to 64.
He also accepted that the PEAK was not evidence of
a transaction correction or error notice being issued to
the subpostmaster in such a way as to subject him to
a risk of loss. That is {Day17/65:1}. My Lord, that
summarises my submissions on bug 11.
If we could move on to bug 12: counter replacement
causing one-sided transaction. The claimants raise
a specific issue, namely that the KEL at F421, that's
J Ballantyne 5328R, does not make reference to issuing a
transaction correction, or error notice, or a BIM
report.
If we could go to that, it is at {F/421/1}. It is
true that the KEL does not make reference to a BIMS or
to an error notice or transaction correction. However,
usually KELs don't do that. They don't include specific
instructions in relation to BIMS or TCs. It is
a standard process which goes beyond what's required in
one specific KEL.
If a SPM needs to be made good it is a standard
process, and in paragraph 12 of his third witness
statement Mr Parker said:
"This process of identifying a bug, then identifying
its effects and then remedying those effects is not
special to Horizon. It is a standard part of any IT
support practice." {E2/13/3}
Having said all that though, look at the bottom at
page 1, just before "Evidence" {F/421/1}. About three
lines from the bottom, it says:
"For a multi-counter outlet ... need to retrieve the
messagestore from another counter, as well as the
affected counter. For MCO (and SCO), transaction
numbers for the RiposteVersionString messages should
reveal the original transactions. When you have
identified any missing transactions attach the details
to the PinICL," that is an old name for PEAK, "and route
to MSU."
Now, the reference to MSU, there has been quite
a bit of evidence about it but for present purposes
I can simply rely on Parker 1. If I could ask
your Lordship to go to {E2/11/12}. It is paragraph 44.
Starting at the second sentence, Mr Parker says:
"There was (and is) a process run by the Management
Support Unit (MSU) which involves examination of various
system reporting and may result in Business Incident
Management Service (BIMS) entries going to Post Office.
An incident may also be raised by MSU with the SSC to
provide support to the MSU in resolution of the BIMS."
My Lord, the MSU is the body within Fujitsu which is
involved in the production of BIMS.
So the fact that the MSU was being involved by the
SSC here actually is an indication that what could end
up happening is that a BIMS report will go from MSU to
Post Office.
My Lord, secondly in the context of this very bug,
it is worth looking at {F/77.1/1}, one of the other
PEAKs. Same phenomenon. If one goes over, this is
24th November 2000, so a similar period. One goes to
page {F/77.1/3} of that PEAK. The bottom green box
starts at 5th December 2000 at 10.43:
"Response:
"4/12/2000 ... By John Moran - MSU."
It is worth noting.
MR JUSTICE FRASER: I can't see where you are.
MR DE GARR ROBINSON: It is halfway down the page.
MR JUSTICE FRASER: Have you still got black and white or
have you got colour?
MR DE GARR ROBINSON: No, I have green now. It is the green
box. If I didn't make that clear, I do apologise.
5th December at 10.43.
MR JUSTICE FRASER: Yes, I've got that. Oh, I see, yes.
MR DE GARR ROBINSON: So MSU is involved.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: "This incident has the same cause as
one to be recorded on the SIL ..."
I'm afraid I don't know what SIL means:
"... under BIM."
And there is a number. You see MSU is involved in
BIMS. Then if one goes down to the bottom of the box,
about two-thirds of the way down there is a line that
starts "These two credits". Does your Lordship see
that?
MR JUSTICE FRASER: Yes, equal £167.12.
MR DE GARR ROBINSON: "As they were written over they were
not added to the cash account.
"Final BIM issued.
"Please close this call.
"Closing as Reconciliation resolved."
Here you have evidence first of all of BIMS being
sent in relation to precisely this bug, and second you
have evidence of the MSU being involved. So that
disposes of the submission made by my learned friend
about that.
Withdrawn stock next. That is bug 13. If I could
pick it up at {F/765/1}. Your Lordship may recall that
this is a problem where stock is withdrawn. It is no
longer being sold by Post Office. Branches are supposed
to rem out the relevant stock and send it back to
Post Office, but what sometimes happened is that they
don't rem it out, they just send it to Post Office and
the problem is Horizon doesn't actually have a pair of
eyes. If you don't record the rem, then Horizon doesn't
know that the stock has been removed, and then there is
a problem because you have stock which Horizon thinks is
held at the branch, and it is historic stock, it's stock
which Post Office no longer sells. And that creates
problems in itself and what to do about that stock and
so on.
This was a case where the SPM did not follow the
correct procedure. The underlying issue explained of in
the PEAK is not a bug in withdrawn stock, but it is the
SPM not following correct procedure in remming out the
stock that's being withdrawn.
One gets that, my Lord, if one looks at page 1 of
this PEAK halfway down the big yellow box:
"This office physically held 137 £5 Post Office
stamps and did not rem them out before the date the rem
out icon disappeared."
That will be when, I presume, reference data ceased
to be applicable to in relation to that particular
stock:
"The office physically returned the stamps to
Transaction Processing as advised and the office then
did a Trading Period balance on 17 November 2010 and
showed this value as a loss."
Now, what my learned friend did yesterday was go
like a magnet to the second line and said: look, look,
transaction processing advised them to do what they did.
It must be a bug. I don't know if your Lordship recalls
that?
I think the quote from the transcript is:
"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."
That is {Day21/110:14}. But your Lordship will see
from this very sentence that the SPM had not done what
he was supposed to do. He had not remmed out the stamps
in the first place. So it is quite right that the SPM
had been advised to return the stamps, but they had to
also rem them out. That is fundamental to the way that
the Post Office business works and it is commonsense
really because, as it says, Horizon doesn't have eyes.
So, my Lord, again, a very brief submission in
relation to that.
Phantom transactions is more complicated. There
were three documents that my learned friend referred to
yesterday. The first was -- well, a PEAK at F/97, then
at PEAK at F/100.1 and then a PEAK at F/88.2. These
PEAKs are relied on with a view to casting doubt on the
conclusion set out in those PEAKs that the problems were
not probably attributable to bugs in Horizon.
If we go to {F/97/1} first, your Lordship may recall
that I cross-examined Mr Coyne about this, who would
only rely upon the Romec engineer point. Does
your Lordship recall the Romec engineer point?
MR JUSTICE FRASER: I do.
MR DE GARR ROBINSON: That was the point he had referred to
in his report and I had taken him to the end of the PEAK
in which Patrick Carroll had set out his overall
conclusion.
My learned friend very fairly went to a different
page of the PEAK, page {F/97/7}. Before going there, it
is worth noting that this is a master PEAK that doesn't
just relate to one particular branch, it covers a number
of different incidents with different branches. And
that's an important part of the context.
If one goes to page {F/97/7}, that's relevant to
what's said in the box that my learned friend referred
to. It is 19th June, 15.17, about two-thirds of the way
down. My learned friend understandably relies on the
sentence:
"This has been seen at Old Iselworth (OI) and Wawne
... 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 ends Wendy is
arranging for installation of Kit at OI on Friday ..."
So it is the Old Isleworth site 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 ..."
So this is 19th June, the installation took place on
22nd June. This entry followed up on 20th July at
13.40. That's at page {F/97/8}. It says:
"Comtest readings have been correlated with perfmon
stats and a recommendation to install resitive screens
at Old Iselworth has been made."
So that is the last entry relating to this site on
this PEAK. But it is picked up in {F/100.1/1}, if
I could ask your Lordship to go to {F/100.1/1}. I'm
sorry to jump about like this. Your Lordship will see
the last entry was 22nd July. This picks up at the top
of the page, 25th July 2001 at 9.39.
MR JUSTICE FRASER: 25, yes.
MR DE GARR ROBINSON: Does your Lordship see that in the top
green box?
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: Third line down:
"24/07/01:
"PM reports that he has been having phantom
transactions ..."
Over the page, here is Patrick Carroll's conclusion
about all of this {F/100.1/2}. 5th September:
"Following a significant amount of monitoring ..."
Stopping there. That clearly is the case. Mr Coyne
himself I think accepted that Fujitsu had worked quite
hard to get to the bottom of what these reported
problems were:
"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 phanton Txns,
these pointed to a touch screen problem and as a result
the screen was replaced with a Resitive model. As this
produced no measurable improvement it has to be assumed
that the problems were user related."
So that's Mr Carroll's considered opinion having
done all the tests and changed all the equipment that
could be changed.
If one goes back to the previous master PEAK at
{F/97/9}, his overall conclusion at 12th November 2001
at 9.48 at the bottom of the page:
"Phantom Txns have not been proven in circumstances
which preclude user error.
"In all cases where these have occurred a user error
related cause can be attributed to the phenomenon.
"I am therefore closing this call as no fault in
product."
My Lord, that's the conclusion of Mr Carroll who is
on site and who is the experienced Fujitsu operative.
In my submission, that opinion should not be rejected.
So far as is relevant to the Old Isleworth site it
is also worth noting something about the postmaster
there. If one goes to {F/88.2/1}, this is
15th February 2001. So we were in July 2002, weren't
we? I'm so sorry, we were in September 2001. That was
the last entry for Isleworth in the previous F/100.1,
but this started in 15th February 2001.
If one goes to the green box at the top about
halfway down:
"Outlet went live ... requested further training,
the PM was referred to his RNM ... It would appear that
there is a training issue here which needs to be
addressed."
Then PON actions:
"Has PM completed and passed his training?
"When, where and with whom did PM complete training?
"Has further training been considered?"
One sees that the postmaster seems in need of help.
Then at page {F/88.2/3}, 20th September, 12.07, the
main box:
"PON have written to the RNM to address the training
issue, see text below.
"From RNM - 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 txns
(discrepancies?) 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."
So the changes made didn't seem to change anything
but then miraculously the transactions stop. What's
more, it seems that this postmaster is the kind of
person that needs more support than perhaps you would
expect for more postmasters.
MR JUSTICE FRASER: What does "PON" stand for?
MR DE GARR ROBINSON: My Lord, I would have to take
instructions.
MR JUSTICE FRASER: Don't worry, I will give you a list of
three-letter acronyms later on.
MR DE GARR ROBINSON: I shall look forward to that.
MR JUSTICE FRASER: I will give it to both of you so you are
not the only person who has to ...
MR DE GARR ROBINSON: Unless I can assist your Lordship
further, those would be my submissions on the PEAKs that
my learned friend went through yesterday.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: Then, with your Lordship's permission,
I propose to deal with a number of miscellaneous points.
First of all, number of criticisms made of Mrs Van
Den Bogerd's evidence. First of all, she was criticised
for her treatment of the Helen Rose report and she
accepted some criticism in her evidence, and my learned
friend places some emphasis on that in his submissions.
In my respectful submission, her witness statement
was clear and fair. It wasn't right to criticise her.
If I could ask your Lordship to go to her witness
statement. It is {E2/5/34}. This is to do with the
Helen Rose report.
Your Lordship will recall that Mr Coyne in his first
report made a number of criticisms based on the
Helen Rose report, including the criticism that
a transaction correction had wrongly been sent to the
postmaster in that case.
At the bottom of page 34 Mrs Van Den Bogerd refers
to paragraph 5.175 of his report. She says:
"The extracts taken from the report by Helen Rose
... are taken out of context and mistakenly claim that
the relevant reversal was issued in error by Horizon,
not the Subpostmaster. The Rose report makes it clear
that:
"154.1. The concerns were based on the fact that
reversals were not being shown on the particular data
sets reviewed / reports typically run by Subpostmasters
in branch on Horizon.
"154.2. Transaction reversal data can be extracted
from Horizon;
"154.3. The issue was therefore surrounding how the
transaction reversals were displayed / accessible in
branch and that there was no issue with Horizon itself."
Then 155:
"There is therefore no indication that the reversal
was not notified to the subpostmaster. When recovery
was carried out a discontinued session receipt would
have been printed and messages would have been clearly
displayed to the User in branch during the recovery
process."
My Lord, in those paragraphs Mrs Van Den Bogerd is
not saying that the postmaster reversed the transaction.
She is saying that what happened in the branch was part
of the reversal process and was correct. And she was
saying that the SPM knew about it.
In my submission, the evidence shows that both those
points were true.
MR JUSTICE FRASER: Both which points, sorry?
MR DE GARR ROBINSON: First of all, that the reversal was
part of the recovery process operating normally, and
secondly, the SPM knew that the recovery process was
operating because he received the printed receipts.
MR JUSTICE FRASER: You are saying that's what's on the face
of her witness statement? Are we at the point of her
witness statement now rather than her cross-examination?
MR DE GARR ROBINSON: Yes, I'm talking about her witness
statement. I'm not going to go to her
cross-examination, my Lord. What I'm submitting to
your Lordship is that there was no basis for criticising
her --
MR GREEN: My Lord, I'm sorry to interrupt but she agreed
that was wrong in her cross-examination.
MR DE GARR ROBINSON: What she said in her cross-examination
was that she didn't intend to suggest that the SPM was
responsible for the reversal and that's not what she
says in this witness statement. So she was right to
have that intention.
MR JUSTICE FRASER: All right.
MR DE GARR ROBINSON: So, my Lord, that's point 1 on Mrs Van
Den Bogerd.
Point number 2, paragraph 183 of the same witness
statement {E2/5/42}, she says:
"Before the change to Horizon Online, a cash check
was completed in branch by the HFSOs. Branches were
notified in advance that this cash check would be
carried out. I recall that this mandatory cash check
across the entire network caused a temporary spike in
declared losses. I suspect that this was due to
branches tidying up their accounts before the cash check
and therefore losses coming to the surface that had
previously been ignored or covered up."
My Lord, it's always dangerous for counsel to say
this, but I don't understand what the criticism of
Mrs Van Den Bogerd is here. She is speaking from her
own recollection and she remembers a spike in declared
losses during the period of the mandatory cash check,
which of course is before the migration to
Horizon Online.
She is not purporting to state a considered view
that having investigated the matter this must be the
cause. She specifically qualifies what she says by
saying:
"I suspect that this was due to branches tidying up
their accounts."
So she lived through the process and she suggests
a possible reason. My Lord, that's not speculation on
which she needed to be challenged by reference to
Fujitsu documents to which my learned friend sought to
take her.
My Lord, furthermore it is worth noting that
Mr Coyne himself agreed, and had already agreed, that it
was a likely factor in the spike along with a possible
increase in bugs during migration to Horizon.
My Lord, that's in JC2, {D2/4.1/219}. Perhaps we
could go to that.
Paragraph 5.345. Mr Coyne has referred to certain
statements made by Dr Worden in the previous paragraph,
you can see the end of it is in italics, and he says:
"In my experience, a major change to a platform will
almost always lead to an increase in bugs, errors and
other issues. Therefore, I do not agree that the
mandatory cash check was 'more likely' to be the cause
of the spike in 2010. The most likely scenario is that
both of these were factors."
So my learned friend's own expert says it was
a factor.
Mrs Van Den Bogerd is talking about the period
before migration when the mandatory cash checks were
being done. My Lord, in those circumstances it was not
necessary to take Mrs Van Den Bogerd to any documents to
challenge her evidence. Her evidence was perfectly
understandable and needn't have been challenged at all.
And that's my very brief submission about that.
My Lord, a third point about Mrs Van Den Bogerd's
evidence is that in paragraph 35 of the claimants'
submissions there is a suggestion that she is basing her
evidence on a document that she didn't actually have
because it didn't come into existence until after her
witness statement was made.
I don't criticise the claimants for doing this but
they have done this quite a lot. I can tell
your Lordship on instructions why it was that the cash
declaration that they refer to in paragraph 35 -- it is
not necessary to go to it -- was dated 1st March when
her witness statement was dated the previous year, in
November the previous year.
MR JUSTICE FRASER: 1st March this year then, you mean?
MR DE GARR ROBINSON: Yes. She couldn't explain that in
cross-examination.
My Lord, on instructions I'm in a position to
explain it now. During preparation for the
cross-examination of Mr Patny, my instructing solicitors
realised that they didn't have a copy of that particular
declaration and so they requested it from Post Office.
That was on 1st March. My Lord, that was the document
that was then hyperlinked to Mrs Van Den Bogerd's
statement. She doesn't specifically say "I refer to
a cash declaration". The fact is she had seen a cash
declaration. She had seen it a long time before, but it
was the same data just obtained on a different date.
So the fact that the date on that particular
document post-dates the date of her witness statement
doesn't mean that she didn't have sight of the same
information previously; she certainly did have. That's
why one sees it. She makes a number of statements about
it which are correct.
My Lord, a similar claim made in the claimants'
written closings about the evidence of Mr Membery, who,
as your Lordship is aware, was unable to give evidence.
It is at paragraph 123(c) of their submissions. It is
to the effect that Mr Membery refers to a document that
has been produced to him by Mr Lenton but the metadata
on the document suggests that the document was produced
by someone called BH.
My Lord, I can tell your Lordship on instructions
that the version of that document received by my
instructing solicitors from Fujitsu recorded that it had
last been modified by Mr Lenton. How BH's name got on
it and who BH is they have no idea, and my instructing
solicitors are happy to provide that document in native
form if my learned friend would wish to see it.
My Lord, another miscellaneous point. Criticism is
made of Mr Johnson in my learned friend's closing
submissions. He is criticised because he didn't know
the source of a screenshot. My Lord, it did not matter
where the screenshot came from. What mattered was
whether it was a true screenshot. It was not from the
Horizon guide, but Mr Johnson did not claim that it came
from the Horizon guide.
If I could ask your Lordship to go to {E2/4/2}. At
paragraph 10 you will see that Mr Johnson says:
"The screenshots that appear in this statement are
primarily taken from a document called Post Office
Onboarding Counter Guide ... Where a screenshot has been
taken from another document I refer to that document."
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: My Lord, Mr Johnson didn't know where
the particular screenshot had come from but he did know
that it was a proper screenshot, and that's the short
point.
And indeed it was. My Lord, another criticism made
of the claimants' witnesses relates to the
cross-examination of Mrs Burke. It was suggested in the
claimants' closing submissions that because Mrs Van Den
Bogerd's witness statement was only amended shortly
before she went into the box, that enabled my learned
friend Mr Draper to criticise Mrs Burke on the basis of
a false apprehension of the facts to be collected from
the unamended witness statement.
My Lord, this related to the fact that in Mrs Burke
had undertaken three transactions for several different
customers all in one basket, which of course is not
proper procedure. And she explained why. She very
fairly explained why in her witness statement: it was
because Horizon was playing up.
But what was suggested to her was, first of all,
that was a breach of procedure as it evidently was, and
secondly, it was suggested to her that by doing that she
increased the risks of unfortunate things happening.
For example, if there was a failure it might well be
that the failure would relate to a transaction in
relation to which the customer was long gone.
My Lord, that was all that was put. No criticism
was made of Mrs Burke and, more importantly, no
suggestion was put to her that her having undertaken
a series of different transactions all within one
basket, that had a causative effect on the issue that
she then faced when there was a system outage on
9th May.
My learned friend's suggestion that a false case was
improperly put to Mrs Burke on the basis of an unamended
witness statement of Mrs Van Den Bogerd's is simply
unfounded, and it is an unfair criticism and
I resist it.
My Lord, I'm going to finish with some very brief
submissions in relation to audits. A number of
submissions are made about audits in the claimants'
closings. I can pick them up really at paragraph 620 of
the claimants' closings, which is at page {A/5/217}.
Picking it up at (c) at page {A/5/220}:
"The purpose of these audits is to provide
assurances to Post Office and its auditors about the
risk of material misstatements in Post Office’s
financial statements. The audits themselves expressly
state that they 'should not be used by anyone other than
these specified parties'."
So there is a claim.
Your Lordship will recall that there were a number
of financial audits done by Ernst & Young for
Post Office, including in 2011 and 2013. They are the
documents that spring to my mind as being relevant.
MR JUSTICE FRASER: I think you just read from page 217 and
you said page 217, but I followed it perfectly
adequately.
MR DE GARR ROBINSON: I'm grateful. I read from page 220
and I may have given the wrong page number.
MR JUSTICE FRASER: No, you said 220 but in mine it is 217.
But that could be because I'm using the hard copy.
MR DE GARR ROBINSON: I see. My hard copy version seems to
be different, I'm so sorry.
MR JUSTICE FRASER: Don't worry. Just pause one second.
On the common screen we have got page 217, which
starts with paragraph (a) and runs down to (g) and has
621(a) and (b), which is what my version has. Have you
got a different version?
MR DE GARR ROBINSON: My version has page 220.
MR JUSTICE FRASER: I do not think it has come up before so
I do not think it is an issue. But I have got the right
paragraph and I followed you reading it.
MR DE GARR ROBINSON: I'm grateful.
But in subparagraph (c) {A/5/217}, a factual claim
is made, which is that there are Post Office's financial
audits is one thing and then there are the service
audits that were done for Fujitsu. And the suggestion
is being made that the service audits were just about
Post Office's financial statements.
My Lord, in my submission that is demonstrably
wrong, and if we could look at the document referred to
in the submission, {F/1041/9}, please:
"Intended use.
"This report, including the description of tests of
controls and results thereof in the Description of Tests
and Results, is intended solely for the information and
use of Fujitsu, POL as the user of the IT support
processes and controls used by and on behalf of Fujitsu
to support the HNG-X and POLSAP applications during some
or all of the period ... and the independent auditors of
POL, who have a sufficient understanding to consider it,
along with other information including information about
controls implemented by user entities themselves, when
assessing the risks of material misstatements of user
entities' financial statements."
So your Lordship will see two things. First of all,
the intended users are not just the auditors when
auditing. The intended users include Fujitsu itself and
Post Office as the user of the IT support processes
provided by Fujitsu. That's point 1, and point 1
demonstrates that the submission made in (c) that we saw
before was incorrect.
Point 2 is wider really, which is there is
an interesting, how can I put it, tension in the
claimants' attitude because on the one hand they are
very happy to trumpet the reliability of Ernst & Young's
financial audit when referring to the 2001 E&Y
management letter, but when it comes to a document which
is helpful to Post Office they suddenly say, well, it is
just financial misstatements, it's got nothing to do
with the price of fish.
So, my Lord, those are the two submissions I derive
from that page.
Finally, it is worth adverting in this context to
the evidence which Mr Coyne gave on {Day16/188:1}. We
needn't go to it, but on that page Mr Coyne accepted
that the service audits are more specific than general
financial audits.
My Lord, if we can go back to the closings, that
will be at {A/5/217}, I believe, we see another claim
that's made by the claimants. (d):
"The control objectives and controls are selected by
Fujitsu rather than the auditor ..."
What the claimants are trying to do here is they are
faced with a series of audits that essentially are very
positive about Fujitsu's approach to the IT support that
it is giving, all the way from 2012 through to 2017.
So here they are trying to think of points which
will somehow undermine the value of those audits. And
what they are suggesting in (d) is that somehow Fujitsu
is marking its own homework. It has self-developed the
test which it is to be inferred therefore has no real
value and cannot be taken as a justification for
anything.
But, my Lord, first of all, there is no proper basis
for suggesting that Fujitsu would have developed tests
that had no value in this context. Secondly, if one
goes back to the document, so this is {F/1041/8},
please -- I'm so sorry, I'm afraid I haven't marked up
this document. If your Lordship will give me a moment.
MR JUSTICE FRASER: Don't worry.
MR DE GARR ROBINSON: Yes. Top of the page. This is
Ernst & Young describing the work it was doing:
"An assurance engagement to report on a description
of a service organisation's system and the suitability
of the design and operating effectiveness of the service
organisation's controls to achieve the related control
objectives stated in the description involves performing
procedures to obtain evidence about the fairness of the
presentation of the description and the suitability of
the design and operating effectiveness of those controls
to achieve the related control objectives stated in the
description."
My word, that is a six-line sentence:
"Our procedures included assessing the risks that
the Description is not fairly presented and that the
controls were not suitably designed or operating
effectively to achieve the related control objectives
stated in the Description."
So your Lordship will see the auditors weren't
simply taking a test that was given by Fujitsu and
blindly applying it, they were also assessing the risks
involved in the relevant functions and assessing the
suitability of the design of the controls.
Then it goes on:
"Our procedures also included testing the operating
effectiveness of those controls that we consider
necessary to provide reasonable assurance that the
related control objectives stated in the Description
were achieved. A reasonable assurance engagement of
this type also includes evaluating the overall
presentation of the Description, the suitability of the
control objectives stated therein and the suitability of
the criteria specified by the service organisation and
described in the Assertion."
So what Ernst & Young are saying there is that not
only did they audit the processes to ensure that the
control objectives were met, but they also evaluated the
suitability of the control objectives in the first
place.
So this wasn't simply them jumping over a hurdle
that Fujitsu had very conveniently placed for itself.
It was Ernst & Young evaluating the value of that very
hurdle itself. I hope I put that point clearly.
MR JUSTICE FRASER: Yes. As I understand it, Fujitsu
specify the control objectives and the service auditors'
responsibilities, which actually start on the previous
page, on page {F/1041/7}, are described in detail in
those two full paragraphs?
MR DE GARR ROBINSON: Yes. Those functions include
evaluating the objectives in the first place; in other
words, making sure that the objectives that they have
been told to test for are appropriate objectives to
test for.
My Lord, that's my submission on the suggestion that
appears to be made in paragraph (d) in paragraph 620 of
the closing submissions, that there's something wrong
with the objectives because Fujitsu selected them.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: My Lord, (f) in paragraph 620 on
page 217 of the closing submissions:
"Much of the content of these audits has been simply
cut and pasted from one year to the next. (Eg for
control objective 10, which Dr Worden relied upon, the
tables within the audits for 2013 and 2014 are
identical, and likewise for 2015 and 2016 ..."
My Lord, difficult to know what to make of that
submission.
It is added to in (g), where it is said:
"Dr Worden said he had noticed that quite a lot of
the wording was very similar or identical from one year
to the next ..."
There appears to be an intention, I may be wrong,
but so far as I can tell it may be intended to achieve
the implication that somehow Ernst & Young weren't doing
a proper job, it was simply repeating words it had said
previous years. If that is the suggestion, there's
absolutely no basis for it. Given that they were every
year evaluating and assessing similar objectives it is
not surprising to see them use the same language if
their views remain the same.
My Lord, if that's not being suggested I don't
understand what the relevance of the submission is and
I invite you not to understand it as well.
My Lord, that leaves a submission in (e), which is
that:
"The audit expressly excludes the auditor giving any
opinion on application processing and application
controls."
My Lord, difficult to know what that submission
means. In my submission, the control objectives and the
conclusions drawn in relation to those control
objectives mean what they say. It wouldn't be right to
somehow construe Ernst & Young's opinions as somehow
having no application to Horizon system or the like if
that's the intended implication.
My Lord, finally, paragraph 621 {A/5/217}. This is
where the claimants cut to the chase.
They say:
"The only one of the EY audits which significantly
advances the parties' understanding of the issues in
this litigation is in fact the 2011 EY audit ..."
My Lord, all these comments in relation to the
service audits which actually are more specific than the
financial audit in 2011, all these arguments are
designed somehow to dismiss the value of the service
audits. And in my submission it wouldn't be right to do
that.
Secondly, it wouldn't be right also to suggest that
all that matters is the 2011 EY audit. What should also
be borne in mind, for example, is the 2013 EY audit.
I needn't go to the document, but for your Lordship's
note the reference is {F/1138/1}.
At page 4 of that document it records that
management action has addressed many of the issues
raised. At page 7 of that document it praises the
efforts taken to strength the control environment.
My Lord, in my submission, that advances the parties'
understanding of the issues in this litigation.
My Lord, your Lordship may be really rather
exhausted by the miscellaneous points/submissions that
I have to make. I do have time to make one final
submission and then I will stop, you will be pleased to
know.
MR JUSTICE FRASER: I'm not remotely exhausted by them.
MR DE GARR ROBINSON: Perhaps it is just me.
My Lord, it is a criticism that's made quite
forcefully in the closing to the effect that Post Office
didn't behave with candour because when it had
corrections for its witness statements it would produce
a document setting out those corrections rather than
producing a red-lined version, an amended witness
statement.
I say this some trepidation because your Lordship's
experience may be completely different from my own, but
I have to tell your Lordship that neither I nor members
of my team have ever encountered an amended witness
statement that has been amended in the way the
claimants' witness statements are amended. I have to
say personally I feel a certain amount of discomfort,
just because of habit, I think, of amending witness
statements that way.
So suggesting that Post Office is at fault for not
adopting a practice which to me is an innovation is, in
my submission, rather unfair. It is inappropriate in my
submission to make criticisms of Post Office because its
legal team did things in the traditional way.
MR JUSTICE FRASER: Well, I have seen them done in three or
four different ways and the way you have adopted was not
unconventional.
MR DE GARR ROBINSON: I'm grateful.
My Lord, I would finish on this point by suggesting,
by speculating really, what would Mr Roll's witness
statement look like if it had been amended to reflect
the evidence he gave in open court.
Unless I can assist your Lordship further, those are
my submissions.
MR JUSTICE FRASER: Your closing submissions in toto rather
than the end of the miscellaneous section?
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: All right. Well, I have got some
questions for you. We have also got a bit of
housekeeping, but it is not going to be very much.
MR DE GARR ROBINSON: Actually, on housekeeping matters,
there has been some discussion between Mr Green and
Mr Henderson about the suggestion that your Lordship
made yesterday about a mechanism for identifying the
source for various statements contained in appendix 2.
My Lord, we would suggest the simplest route would
be for the claimants to provide us with a version of
appendix 2 which marks in highlight particular passages
for which they don't understand the support relied upon.
If that's provided by a certain date then my team
can respond by a certain date by explaining the source
with a footnote.
MR GREEN: My Lord, I did mention to my learned friend
Mr Henderson that we are already partway through
producing overnight, the efforts of the entire team,
a table which is confined only to those points
your Lordship mentioned, which is where no evidential
support is identified.
MR JUSTICE FRASER: All right, I'm not going to deal with
this now. We are going to have a short break for the
shorthand writers. We will come back, I won't deal with
housekeeping straightaway because I have some questions
on your submissions.
MR DE GARR ROBINSON: I should not have --
MR JUSTICE FRASER: Which I will deal with then and then we
will deal with housekeeping.
There's that point plus another one and then we are
going to deal with the rest of 2019.
MR GREEN: My Lord, we have a couple of references by way of
factual correction only to hand up to the court.
MR JUSTICE FRASER: We do. Or I do, not using the royal
plural -- these breaks are for the writers, they are not
for me or for you, although I know most people look
forward to them, particularly people who are not on
their feet. But I don't think it is going to take
longer than about 20 minutes. But I think out of
fairness I will break for them and then we will have the
20 minutes.
MR GREEN: Of course.
MR JUSTICE FRASER: In view of the fact we are so gloriously
ahead of schedule, let's come back at 3.25 pm.
(3.15 pm)
(A short break)
(3.25 pm)
Housekeeping
MR JUSTICE FRASER: The first point I actually confirmed
with you during your submissions. It was that the
Horizon Issues were drafted and agreed by the parties
and approved by me. I do not think they were imposed on
the parties. But you confirmed that this morning.
I'm fairly sure I know the answer to this question
but I just want to be very clear just because the nature
of the word might, or has been so contentious. Insofar
as the place I go for a benchmark definition of
robustness, I intend, unless you tell me I should go
somewhere else, to go to your pleading because you
define it in your pleading.
MR DE GARR ROBINSON: Sorry, I'm not sure I understand
your Lordship's question.
MR JUSTICE FRASER: Well, robustness, the experts have
agreed a particular meaning. I asked Mr Green where
I should go for his meaning yesterday and he told me,
made submissions.
I would ordinarily, and I have looked at your
pleading quite carefully. Where you do define
robustness?
MR DE GARR ROBINSON: Do you know, I'm afraid I need to
remind myself. Could we go to my pleading, my Lord?
It is a slightly impertinent thing to ask the judge.
MR JUSTICE FRASER: I am sure it is your pleading, but
I might be wrong.
My private screen seems to have stopped working for
some reason. It will be in the C bundle somewhere.
MR DE GARR ROBINSON: {C3/3/1}. It will be referred to in
the Horizon Issues, won't it? I'm so sorry, my Lord.
MR JUSTICE FRASER: Not at all.
MR DE GARR ROBINSON: Mr Draper is suggesting paragraph 16
on page 5, so he'll get all the blame if he's wrong.
MR JUSTICE FRASER: So {C3/3/5}, paragraph 16.
MR GREEN: My Lord, yes.
MR JUSTICE FRASER: I didn't bring the reference to your
pleading with me but I did look at it very recently, so
I'm pretty sure that's likely to be right.
MR DE GARR ROBINSON: So what's pleaded in paragraph 16 is:
"Like any other IT system, Horizon is not perfect,
but Post Office maintains that it is robust and that it
is extremely unlikely to be the cause of losses in
branches. Its design and technical controls, 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."
Is that the paragraph your Lordship had in mind?
MR JUSTICE FRASER: On the basis I haven't got that
paragraph you have just read in front of me on either
screen. Just to put your mind at rest, it is not
designed to be a trick question.
MR DE GARR ROBINSON: No, and I'm not trying to be clever
either, my Lord. Every time I do try I always come a
cropper.
MR JUSTICE FRASER: I have now lost it off the common screen
as well. Give me one second.
The short point is that the experts in different
places discuss robustness in more or less expert terms.
They also agree robustness in one of the joint
statements. I asked Mr Green the specific question
yesterday where I go for his benchmark definition of
what robustness is and I wanted just a similar reference
from you.
MR DE GARR ROBINSON: My Lord, I would not say that
paragraph 16 is a definition. I would submit that
a system that is robust is extremely unlikely to be the
cause of losses in branches.
MR JUSTICE FRASER: Quite.
MR DE GARR ROBINSON: It has the features referred to in the
rest of that paragraph {C3/3/5}.
My Lord, my submission would be that it is
consistent with the answers that Mr Coyne gave when
I asked him to unpack what he meant by "relatively
robust" when he identified comparable systems, and then
he identified the condition under which comparable
systems needed to operate and he said Horizon compared
well with those conditions.
MR JUSTICE FRASER: I know, and I have gone through that in
some detail.
MR DE GARR ROBINSON: I'm grateful.
MR JUSTICE FRASER: In fact, probably the best thing to do,
and it is really just a request for a reference, I would
just be grateful for a reference from you for the best
place to go for -- I used the phrase loosely
yesterday -- a benchmark definition of robustness. If
one is describing robustness in specific terms this is
what it is.
MR DE GARR ROBINSON: My Lord, I will --
MR JUSTICE FRASER: It is just going to be a reference.
I do not think it is this passage in your pleading
actually because I was looking in your pleading, I was
looking in all the pleadings, and that is not the one
that jumped out at me. But I'm not going to tell you
where the best place is, that's rather why I was asking
the question.
MR DE GARR ROBINSON: My Lord, I will try to resist the
inevitable urge that barristers are subject to --
MR JUSTICE FRASER: I don't want a fresh definition.
MR DE GARR ROBINSON: Exactly. I will try and avoid all of
that. But it may be I will be referring to a number of
different documents.
MR JUSTICE FRASER: That's perfectly in order, but I just
want somewhere to go to so when I'm explaining, probably
quite early on in the judgment, what robustness is,
I can explain by reference to a specific -- the way in
which the defendant construes the word "robustness",
because it is obviously different from what the
claimants are suggesting.
MR DE GARR ROBINSON: I need to remind myself of where the
claimants' definition is, but that's my problem.
MR JUSTICE FRASER: I do not think I got one, but I got
an explanation of why there wasn't one.
So that was supposed to be an easy starter, that
one. We will see about that.
The next point is essentially -- I suppose it could
be seen as an extension of the appendix 2 exercise, but
yesterday Mr Green took me to your closings at page 68
and paragraph 147.4.
MR DE GARR ROBINSON: 68?
MR JUSTICE FRASER: Yes. Paragraph 147.4.
MR DE GARR ROBINSON: My Lord, I have it. {A/6/68}
MR JUSTICE FRASER: Which what it says about the adoption of
section 1.4 of the relevant document and when it was
adopted. I beg your pardon, when it was implemented.
Now, that's either evidence or it is a submission.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: He says it is a submission, it wasn't in
any of the evidence.
MR DE GARR ROBINSON: My Lord, it is my instructions but,
my Lord, it is based upon a previous version of the
document and then an amended version of the document.
And I will undertake to give your Lordship the two
references.
MR JUSTICE FRASER: And the dates on documents.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: If you could give me those references
that would be great.
MR DE GARR ROBINSON: If my understanding of those documents
is incorrect I will withdraw the --
MR JUSTICE FRASER: All right. That's helpful, thank you.
The next point is -- and you won't be able to give
me this now, but just if it could be added to the
references I'm going to be sent -- could I have
a reference, please, to the ruling that I made at the
PTR about Mr Henderson -- the witness, not counsel --
and the Second Sight report. Because within the
transcript of that hearing when one gets to my ruling
there is just a bracket that says "see separate ruling".
I am sure it is there somewhere, I just don't know
where it is and I would like to know.
MR DE GARR ROBINSON: We will look for it.
MR JUSTICE FRASER: Appendix 2 we will come onto in
a moment.
There are the following acronyms I would just like
to know what they stand for. Two are from today, one is
yesterday. Yesterday's was RPOS. So R-P-O-S, RPOS.
Today's were SIL and the one this afternoon, PON.
I would just like to know what they are.
MR DE GARR ROBINSON: My Lord, yes. It will be something
like retail point of sale --
MR JUSTICE FRASER: I was tempted to guess and I decided
I wouldn't and I would like to know what they are,
please.
MR DE GARR ROBINSON: That will be done.
MR JUSTICE FRASER: Then the appendix 2 exercise, and this
again is not supposed to be controversial but I think
from your side, Mr de Garr Robinson, you said can they
highlight it with highlighter and they may have started
doing the exercise in a different way. Is that right,
Mr Green?
MR GREEN: Yes, we have a table just identifying roman
paragraphs.
MR JUSTICE FRASER: I do not think anyone is going to go to
the stake on the difference.
MR DE GARR ROBINSON: If they want to do unnecessary work
I won't stop them. I was anxious that there shouldn't
be any hidden submissions, but my learned friend has
assured me that isn't the position so it is just
a matter of dates.
MR JUSTICE FRASER: Of dates, yes. Well, actually, there is
more than just dates, there's something I want to make
clear because I'm very anxious to do two things. One is
not to create unnecessary expensive work and the other
is not to lead to fresh rounds of submissions.
When you are responding to that, Mr de Garr Robinson
and your team, it seems to me there are one of three
available alternatives: a reference to a witness
statement; a reference to an expert's report; or just
the word "submission" to demonstrate that it is
a submission.
MR DE GARR ROBINSON: My Lord, what about a reference to
a document or a reference to commonsense?
MR JUSTICE FRASER: All right. Well, let's have a look at
appendix 2.
By reference to commonsense, do you mean a statement
of the blindingly obvious?
MR DE GARR ROBINSON: Yes. Or words to that effect.
MR JUSTICE FRASER: All right. Well, then we will add the
following alternatives: reference to a document and then
in brackets it has to be said if that document was put
to anyone or not.
MR DE GARR ROBINSON: Okay.
MR JUSTICE FRASER: And then the fifth one, I will adopt
your nomenclature and we will say commonsense.
MR DE GARR ROBINSON: I'm grateful.
MR JUSTICE FRASER: So those are the five alternatives.
MR DE GARR ROBINSON: I understand.
MR JUSTICE FRASER: The ones we were shown yesterday, I'm
not necessarily sure any of them jump out at me as
commonsense, but as I said yesterday some of them, for
example, the notion of slave units and a master unit,
that is commonsense within the IT world and I have
a degree of commonsense in that respect. And if,
following the commonsense answers, I need any more
information, then I will ask.
MR DE GARR ROBINSON: I see. So, for example, slave and
master, I am sure there are documents about that because
I'm familiar with the concept, yes.
MR JUSTICE FRASER: All right?
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: I would just like to make this clear for
everyone's benefit. This is not designed to be a new,
expensive, lengthy grinding exercise. It is just so
I have everything marshalled in one place.
Right, well, that's everything from me. Mr Green,
you were proffering a piece of paper and I asked you to
take your seat.
MR GREEN: My Lord, just a couple of points of
clarification.
MR JUSTICE FRASER: Thank you.
MR GREEN: Yesterday my learned friend said that the double
trouble document references, he thought that many of
them he hadn't seen at all before. So they are all the
references to where they are in the evidence in these
proceedings, including in the Post Office's own
submissions. That's the first page.
MR JUSTICE FRASER: Yes.
MR GREEN: Drafting of Horizon Issues 1 and 3, the short
point is the words were carefully chosen. We were
asking for caused as a separate issue, and potentially
caused. That was resisted by the defendant. In the end
we got a careful agreement on what the wording actually
is and my learned friend's speculation about potential
being what he said is just not available when one looks
at how that happened. And (c) is 18.7 should be million
in shortfalls. We are not able to, at this stage,
disaggregate the extent to which they were caused by
bugs, errors and defects, erroneous TAs or TCs, helpline
advice etc at this stage, but that is the figure in the
SOCIs.
And disclosure, my learned friend said regarding the
disclosure issues, a dead letter from August to
December, there are the intervening references and then
we just added this afternoon in handwriting in relation
to EY audits, Dr Worden, his understanding of
application processing meaning SSC was out of scope and
the reference to the transcript where he said that.
MR JUSTICE FRASER: Yes.
MR GREEN: In the ISAE audits and then over the page just to
give the context to the KELs' complaint, the history of
them said not to be in Post Office's control. There are
lots of other references, we only put four there. There
was obviously an RFI about it as well. It was
identified as an issue in the pleadings and maintained
in the CMC in October 2017 when your Lordship was
specifically asked about it.
MR DE GARR ROBINSON: Why is my learned friend -- just going
through this list, obviously I will need to look at the
drafting of Horizon Issues 1 and 3, but your Lordship
has my submissions on what the document means.
My Lord, application processing. My submission
about that is that what was excluded from the service
audit was any kind of audit by Ernst & Young of the
operation of Horizon itself, the software and how good
Horizon itself was. What wasn't excluded was the
controls that was applied by and on Fujitsu in relation
to the support it provided in relation to Horizon.
MR JUSTICE FRASER: Yes.
MR DE GARR ROBINSON: It is important your Lordship has that
submission. If it is suggested that my learned friend
or that Dr Worden suggested otherwise, I will have
a look at that. But I would be surprised if that's what
he said.
Then there is this paragraph (f). I don't know why
my learned friend has even mentioned it. Is there some
objective being achieved by it?
MR GREEN: My Lord, it was just in relation to the
disclosure history which was an issue my learned friend
responded to.
MR JUSTICE FRASER: Can one of you just give me the
reference to the contract between Fujitsu and the
Post Office which I know was produced and put to
Dr Worden in re-examination?
MR GREEN: My Lord, the contract itself is at {F/1659.2/868}
I think is the one with paragraphs 25.8 and 25.10 and so
forth about court assist. Court support.
MR JUSTICE FRASER: I am sure if I consider it necessary to
look at it I will look at it. If I consider I need any
submissions about it, I will ask for them in writing.
Does that seem a sensible way forward,
Mr de Garr Robinson?
MR DE GARR ROBINSON: My Lord, yes. If your Lordship is
minded to make some kind of finding on the basis of the
contract, I would welcome the opportunity to make
submissions, although I have no idea what those
submissions would be.
MR JUSTICE FRASER: All right.
MR DE GARR ROBINSON: If you understand what I'm saying.
MR JUSTICE FRASER: I do. In other words, if I'm going to
look at it for any material reason I should give you
both, I would give you both the opportunity to --
MR DE GARR ROBINSON: I'm grateful.
MR JUSTICE FRASER: Because the document is there, I have
looked at it on a speed read basis. Only one part of it
was put to Dr Worden in relation, I think, to the cost
of making ARQ requests.
MR DE GARR ROBINSON: That was by me.
MR JUSTICE FRASER: By you. Well, I think you said when you
introduced the document you had had it uploaded for that
purpose.
MR DE GARR ROBINSON: Yes.
MR JUSTICE FRASER: I didn't even know if it was in play
before that, to be honest.
Right. Just remind me, Mr Green, what does TA
stand for?
MR GREEN: The transaction acknowledgements, which are the
automatic ones from National Lottery.
MR JUSTICE FRASER: Because TCs used to be called error
notes for the first four years.
MR GREEN: They did, and then they came in, and then TAs are
the ones that you can't do anything about at all.
MR JUSTICE FRASER: So far as Horizon Issues, other than the
fact that there's a judgment to be written, is that
everything?
MR DE GARR ROBINSON: My Lord, I do hope so.
MR JUSTICE FRASER: I think it is. Apart from obviously
appendix 2 and the reference point. That then brings us
onto the second part of 2019.
At the moment there are the following dates in 2019
which are in the diary and have already been subject to
orders by me. 23rd July; 18th September; they are both
the CMCs. The 17th October, which is down as
a pre-trial review, and then a trial starting on
4th November on the limitation issues.
It does not seem to me that those dates are
necessarily either should stay there or are achievable
certainly, so far as a limitation trial is concerned,
because they were set when this trial was supposed to
finish at the very end of -- well, originally early
April and now it is obviously 2nd July, therefore,
I'll -- is this now your part of ship Mr Henderson?
Mr Green, what do you have to say about it, if
anything?
MR GREEN: I understand the solicitors on both sides have
discussed it and rather taken up your Lordship's
suggestion of maybe having a CMC in September rather
than July when we will know more, we are not sure how
much more, but we will know more and --
MR JUSTICE FRASER: When you say you will know more though,
the only piece of information I think you will know is
how far I'm on in writing the judgment because I do not
think I will have been able to finish the judgment by
the middle of September.
MR GREEN: My Lord, that's slightly why I am putting it
lightly in those terms.
MR JUSTICE FRASER: Right.
MR GREEN: We might have an idea of the timeline on that.
In those circumstances the parties solicitors' have had
discussions and subject to obviously your Lordship's
view, are agreeable to the November listing coming out
if it can't be used for proportionality reasons.
Your Lordship should know there are also discussions in
parallel about how holding a mediation --
MR JUSTICE FRASER: You don't need to tell me anything about
that. Well, at one point I think before the Common
Issues trial when we were scoping the group litigation
generally, you asked me for a three-month stay and
I told you I didn't have one.
MR GREEN: I think there may have been a slight
misunderstanding but I know what your Lordship means.
MR JUSTICE FRASER: When I say don't tell me about the
mediation what I mean is unless you are asking me to fix
a timetable around periods of time, you don't need to
trouble me with that.
MR GREEN: My Lord only to have in mind that it would be
helpful not to have anything too compressed happening in
late October, that we can get to that in September.
MR JUSTICE FRASER: All right. When you said the November
trial coming out.
MR GREEN: My Lord, yes?
MR JUSTICE FRASER: There is another trial date in 2020.
MR GREEN: Precisely, for which we can consider directions
in September.
MR JUSTICE FRASER: All right. Okay Mr Henderson?
MR HENDERSON: My Lord, I'm grateful for the indication in
relation to the November trial and certainly the parties
are agreed that that's not feasible really to use that
date at the moment. The question really is when
your Lordship lists the further issues trial, which is
referred to in the orders.
If we are to use the current date which
your Lordship has put aside for the fourth trial, which
is four weeks commencing on 2nd March, the question that
really arises is: what steps need to be taken in advance
of that? Can I just ask your Lordship, I appreciate
your Lordship is, in light of the dates your Lordship
has just recounted to the court, you have probably
already looked at the seventh CMC order, but if I could
just ask your Lordship briefly to look at it. It is at
{C7/39/1}.
What your Lordship will see is the various steps
from paragraph 3 of that order. Does your Lordship have
that on the screen now?
MR JUSTICE FRASER: I have page 1. Is that what you are
asking me?
MR HENDERSON: Actually if you go to step 2 you will see the
steps that have been ordered. 3.1 and 2 have taken
place, but from 3.4 all the way up to 20.2 is currently
the subject of a stay. So where we --
MR JUSTICE FRASER: And that's the stay that was agreed
I think in the interregnum period?
MR HENDERSON: Exactly so my Lord, on 12th April. If
your Lordship casts an eye down that you will see that,
absent that stay, we would now be in the position of
having completed the first round of disclosure;
pleadings would be very nearly completed; principal
pleadings would be incompleted; we would have particular
claims and defences; replies would be shortly due and we
would also be about to embark upon the second round of
disclosure referred to as extended disclosure.
We would also be about to convene a costs and case
management conference on 23rd July. So there is a great
deal of activity and you will see that if you carry on
in those directions you will also see -- I appreciate
your Lordship has these points in mind -- that there is
also a directions for the identification of the issues
that would form part of the fourth trial and that's in
paragraphs 13 to 17.{C7/39/4}
All of that is currently stayed. We would
respectfully support the suggestion that the further
issues trial be vacated and re-listed for the 2nd March,
which is currently the slot that's being held for the
fourth trial, so in other words we use that slot for the
third trial.
The only question is as to how we arrive at detailed
alternative, updated directions for all of this. My
concern -- our concern is that if we use the
18th September CMC we might not be leaving ourselves
enough time because if nothing has happened and the stay
remains in place and nothing has happened and we come
before your Lordship on 18th September, we have only got
a relatively compressed period to do an enormous amount
of work; and my respectful suggestion would be that the
parties liaise now, try to agree those steps, but in
default of agreement or in default of the court's
approval of any such agreement that we arrive at, we use
the 23rd July appointment to try and --
MR JUSTICE FRASER: At the moment everything is stayed
except the Horizon Issues. You are not asking me to
lift the stay?
MR HENDERSON: I'm not asking you to lift the stay.
MR JUSTICE FRASER: But you would like the 23rd July kept
in?
MR HENDERSON: I am only asking that the November trial be
vacated and re-listed for 2nd March; that the parties be
directed to liaise to try and agree directions leading
up to that trial and in default of agreement, we come
back on 23rd July so that we can have a timetable in
place in July rather than in September leading up to
that --
MR JUSTICE FRASER: You mean have a timetable ordered in
July to cover the period July to March rather than wait
until September?
MR HENDERSON: Indeed so, my Lord.
MR JUSTICE FRASER: By definition, therefore, one of the
things you are going to be seeking either on 23rd July
by asking me or before that by seeking agreement of the
claimants is the lifting of the stay?
MR HENDERSON: Yes, I think it would be part of the
agreement.
MR JUSTICE FRASER: Unless I suppose you could both decide
that the stay will be lifted on date X and this
direction will start after that date?
MR HENDERSON: Absolutely. My learned friend has already
indicated that there is discussion of a possible
mediation, that needs to be borne in mind. There are
other things happening. Your Lordship obviously is
going to be preparing the Horizon Issues judgment.
There is the application for permission to appeal in
front of the Court of Appeal. The Court of Appeal has
indicated we won't get a decision on permission until
September/October probably.
MR JUSTICE FRASER: But that doesn't factor into anything
that is happening at first instance.
MR HENDERSON: I'm not suggesting that it does certainly for
the time being. So we need to get directions in place,
in our submission, leading up to, if your Lordship
approves, the 2nd March. Really I think the only point
of difference, and it is a slight one, is that we think
that using the September date rather than the July date
may be to save up problems for ourselves.
MR JUSTICE FRASER: Mr Green.
MR GREEN: My Lord, the only thing, the parties' solicitors
have agreed, subject to your Lordship's view, that it is
sensible for the pleadings for the further issues trial
to be done with the benefit of receipt of the Horizon
Issues judgment for obvious reasons.
There's also the question of whether the Court of
Appeal grants permission on the obligations, breach of
which and concealment of a breach of which --
MR JUSTICE FRASER: But that is from the Common Issues you
are talking about?
MR GREEN: My Lord, yes. Just highlighting what feeds into
what will have to be pleaded. For that reason the only
actual step --
MR JUSTICE FRASER: I don't follow that point at all but it
doesn't matter for the moment. Go on.
MR GREEN: Rather than trying to get the parties to agree
everything going forward between now and March now,
we'll come back on the 23rd --
MR JUSTICE FRASER: I'm not going to ask you to agree
anything now. I'm not going to make any orders now
other than varying any dates I have to order.
MR GREEN: I mean prior to the 23rd, my Lord. The only
actual substantial step, I think, prior to the pleadings
being done is the giving of stage 4 disclosure by
Post Office, which was to be done prior to the
pleadings. So if that can be completed in good time
prior to --
MR JUSTICE FRASER: All right, I think --
MR GREEN: It just avoids us coming back on the 23rd.
MR JUSTICE FRASER: Mr Henderson has got a valid point which
is, if you can't agree everything sensibly and you wait
until 18th September, you have only given yourself six
months.
MR GREEN: My Lord, the only problem on my learned friend's
submission, with respect, is that there's giving stage
four disclosure and then the pleadings start, which the
parties have agreed in discussions between solicitors
shouldn't happen until after the Horizon judgment.
MR JUSTICE FRASER: But Mr Green if it is as straightforward
as that and you can agree all of those directions
between the two sides in the next week or so, you do not
even have to come on the 23rd July.
MR GREEN: I was just trying to short cut it, my Lord. I'm
in your Lordship's hands.
MR JUSTICE FRASER: I'm going to tell you what orders I'm
going to make now and then each of you can tell me if
I have missed anything else and then there's another
point I have to address.
Order
The pre-trial review on 17th October for the further
issues trial has to be vacated. The trial date on 4th
November for the further issues trial has to be vacated.
The further issues trial will be tried in the current
time set aside starting on 2nd March with the same time
estimate as it currently has. The parties are to seek
to agree directions for the further issues trial by noon
on 21 July and failing agreement they are to attend for
a CMC at 10.30 on 23rd July.
I would like to hope that that is not going to be
necessary. It would be far cheaper for everyone if it
were not needed. Right. So is that 2019 tidied up?
MR HENDERSON: I think the only outstanding matter is the
18th September hearing.
MR JUSTICE FRASER: I'm leaving that in because until I know
what's happening on 23rd July -- I assume one of the
directions you might seek to agree between yourselves,
if you can agree everything else, is that the 18/9 can
come out but I am going to leave it in for the moment.
Right. The only other point which goes to or arises
from judgment number 5 is, as the managing judge and as
the parties have got cost management orders, I made
certain observations about costs and also said it would
have to be revisited at the end of the Horizon Issues
trial. Since then there has been one further
notification on costs which was from the Post Office,
who I think notified in a sum of 13.9 million it is now,
which is about £1 million more than it was when I made
the comments that I did in judgment no. 5.
I'm not going to make any orders or anything like
that but I would like the two of you, please, when
discussing and considering your draft directions in
advance of 23rd July to direct your minds to what, if
any, further costs in case management conferences and/or
costs management orders you might be considering and the
point at which this year that can be addressed in the
absence of agreement. Because obviously if there is
going to be a contested CCMC there are certain steps you
have got to take in advance for 28 days, 14 days, etc.
Right. Is that everything?
MR GREEN: Yes.
MR JUSTICE FRASER: Is that everything?
MR HENDERSON: I believe so. Two minor points. We have
agreed between counsel dates for the bug appendix point.
I do not think that needs to be subject to an order
but my learned friend has indicated he will serve his
comments by 9th July, a week today, and we are going to
aim to respond to that by 19th July. But I do not think
that needs to be in any order.
MR JUSTICE FRASER: No, but thank you for telling me, it
means I know when I'm going to get it.
MR HENDERSON: Presumably on the order that your Lordship
has just made it would be the usual common costs in case
and liberty to apply?
MR JUSTICE FRASER: Yes. Although they would be Horizon
Issue costs anyway given it has only been 25 minutes,
but by all means and could someone draw up the order
please. I hadn't seen Mr Warwick appear from behind the
screen.
Right. Is that everything?
MR GREEN: My Lord, it is.
MR JUSTICE FRASER: Mr de Garr Robinson, does that seem to
be everything?
MR DE GARR ROBINSON: It is.
MR JUSTICE FRASER: Mr Henderson, so far as the directions
are concerned I think that's everything?
MR HENDERSON: My Lord, yes.
MR JUSTICE FRASER: Thank you all very much.
(4.00 pm)
(The court adjourned)