page   1
 1                                  Wednesday, 18th November 2009
 2   (9.37 am)
 3                     PAUL CHAMBERLAIN, sworn
 4                     Examined by MR MOYNIHAN
 5   THE CHAIRMAN:  Your full names for the record, please.
 6   A.  Paul Chamberlain.
 7   THE CHAIRMAN:  Take a seat, please, Mr Chamberlain.
 8   MR MOYNIHAN:  Good morning, Mr Chamberlain.
 9   A.  Good morning.
10   Q.  You have helpfully provided for the Inquiry a statement.
11       Do you have a copy of that?
12   A.  I do indeed, yes.
13   Q.  For others the statement is FI0136 but we will not need
14       to bring it up.  It is a statement you have signed and
15       you are therefore content that what you have set out was
16       the truth as best you could in that statement?
17   A.  Yes, indeed.
18   Q.  What we will do is as we proceed I may make some
19       references to your statement.  If there are matters that
20       require adjustment in your statement, please just
21       indicate so because what we are interested in is the
22       accurate evidence that you can give today?
23   A.  Okay.
24   Q.  Your own background, you come now from a company, that's
25       the Forensic Science Services Limited?
page   2
 1   A.  Yes, the Forensic Science Service Limited is
 2       a Government-owned company, a major supplier of Forensic
 3       Services in England and Wales and to other investigative
 4       agencies throughout the world.
 5   Q.  You yourself are a forensic scientist but you practice
 6       exclusively in relation to fingerprints?
 7   A.  That's correct.  My specialism is fingerprints.
 8   Q.  Do you have both a research and a practical interest?
 9   A.  I have a role to play in research which is mostly around
10       taking research and deploying it into practical
11       casework.  I don't actually undertake research myself
12       and my other roles are round quality management and I
13       still undertake casework on a regular basis.
14   Q.  Do you still, on a regular basis, give evidence in
15       court?
16   A.  As required, yes.
17   Q.  What I actually want to do is to look, with your
18       assistance, at some of the practices and procedures
19       concerning fingerprints.
20           In your own statement, you mention membership of
21       committees and other organisations if I can just deal
22       with those, first of all, you mention a European Network
23       for Forensic Science.
24           Are you a member of that European Network?
25   A.  Yes.  The ENFSI (the European Network of Forensic
page   3
 1       Science Institutes) is actually an organisation for
 2       directors of forensic science labs throughout the
 3       European Union.  In order to be a member, in effect, the
 4       members are those directors.  The institutes need to
 5       meet certain criteria, for example, they have to have a
 6       research and development programme.  Within ENFSI there
 7       are 16 specialist working groups, one of which is
 8       dedicated to fingerprints and I'm currently the chair of
 9       that working group.
10   Q.  So far as the work of that organisation is it sponsored
11       by the European Union?
12   A.  Recently in the last year, yes.  It has become what the
13       European Union call a monopoly organisation, which means
14       that it the first port of call for the European Union
15       with regard to any forensic issues.  There is some
16       funding now being provided by the European Union.
17   Q.  I will come on in due course to the practices but at the
18       European level is there any work being done at present
19       in relation to standardisation of fingerprint
20       techniques?
21   A.  The strategic plan of the working group does mention a
22       move towards what we've termed a universal standard for
23       fingerprint identification but at this moment in time
24       there is no ongoing work.
25   Q.  Is there, if one looks across the membership, across the
page   4
 1       European Union, are there variations in practice in
 2       relation to fingerprint identification?
 3   A.  There is.  The majority of members have a numerical
 4       standard so they set a numerical threshold generally
 5       around 12 minutiae.  The minority of members have
 6       holistic or non-numeric practices, as we have in England
 7       and Wales.
 8   Q.  Does the fact that the majority of the members have a
 9       numerical standard around 12, is that itself a
10       dis-inhibitor to the search for some universal standard?
11   A.  Perhaps inhibitor is the wrong word but, yes, at the end
12       of the day, they have a system that they believe works
13       successfully and, therefore, they are, needless to say,
14       not willing to move away from that necessarily.
15   Q.  If I go across the Atlantic to America, we are aware of
16       a National Academy of Sciences report that came out at
17       the beginning of this year, I understand, through the
18       International Association for Identification, IAI, that
19       you are on a working party?
20   A.  Yes.  The Standardisation Committee, this is a new
21       committee but it's called the Standardisation Committee
22       2 because there was an initial committee way back in the
23       1970s.  The aim of the Committee is really to address
24       two resolutions of the IAI.  The first that states that
25       there is no scientific basis for a numerical threshold
page   5
 1       and the second, which is a resolution which prohibits
 2       quality of reporting or probabilistic reporting by IAI
 3       members.  That Committee has finished its final meeting
 4       and the report will be published in January.
 5   Q.  I don't want you to breach confidentiality in relation
 6       to that organisation but we can understand then that the
 7       two topics that are delayed now until January, I think
 8       your statement has said the autumn but delayed now until
 9       January, are firstly the question of whether there is an
10       empirical basis for a numerical standard; that's the
11       first?
12   A.  That's correct, yes.
13   Q.  Then, secondly, a topic that I will deal with towards
14       the end of my examination.  It will also cover the
15       relevance of probability-based evidence?
16   A.  Correct.
17   Q.  Do I take it from what you have said the current
18       position of the IAI precludes a Fingerprint Examiner
19       giving any evidence based on a probability analysis?
20   A.  That's correct.
21   Q.  We will come to that a little later.
22           Before I get too far, one of the things in addition
23       to your statement that I have found useful in preparing
24       to ask you some questions is a chapter in a book and the
25       book is by Professor Fraser and is it Mr Williams?
page   6
 1   A.  (Nodded)
 2   Q.  The Handbook of Forensic Science, a relatively recent
 3       textbook, in which there is a chapter on fingerprints
 4       which you co-author with Professor Champod?
 5   A.  That's correct.
 6   Q.  What is your working relationship with
 7       Professor Champod?
 8   A.  I worked with Professor Champod at the Forensic Science
 9       Service some ten years ago when he was one of the lead
10       scientists there and we have worked on a number of
11       projects since then.
12   Q.  I will use this, as I've said, as a helpful guide
13       through some of the topics that I am looking at and I
14       will be conscious of the fact that I may sometimes be
15       reading things that are penned by you, sometimes reading
16       things penned by Professor Champod so if I am straying
17       into his area, then please just say because he will give
18       evidence next week.
19           Before I go too far, what I should pick up -- and it
20       is perhaps worth just putting on the overhead
21       projector -- is a section in the chapter that begins at
22       page 74 and actually mentions the McKie case so we just
23       actually want to see what you are on record as having
24       said about the McKie case.  (Pause)
25           Page 74 is simply brought up to give the context.
page   7
 1       There's a section in the chapter relating to bias and
 2       errors in fingerprint identification and it mentions the
 3       work of, among others, Dr Dror to whom we have had
 4       reference in the Inquiry.
 5           In that context if I go to the next page, page 75,
 6       we will see that you mention in the second paragraph on
 7       the page the case of McNamee, again something we have
 8       discussed in the Inquiry.  You then mention the McKie
 9       case, which I will come back to just in a second.
10       Finally, you mention the Brandon Mayfield case.
11   A.  Yes.
12   Q.  What I actually want to do is simply to bring to the
13       fore what is written about the McKie case.  I am not
14       myself going to ask you anything about the McKie case,
15       but it is just as well to put on the record what you
16       have said in a paper that's co-authored with
17       Professor Champod.
18           You say:
19           "The second case is the supposed misattribution
20       (with allegedly 16 points in agreement) of a mark found
21       in a scene in Scotland to the thumbprint of Shirley
22       McKie.  The McKie case is covered in a recent book
23       (McKie and Russell) and a full inquiry of the Scottish
24       Criminal Record Office and Scottish Fingerprint Service
25       has been undertaken by the Justice 1 Committee of the
page   8
 1       Scottish Parliament.  In this case, there is not a
 2       complete consensus.  A minority of latent print
 3       examiners continue to claim that McKie was indeed the
 4       source of the disputed mark, whereas the majority have
 5       declared an exclusion in this case.  This lack of
 6       consistency between examiners is a worrying fact."
 7           So far as that is written it has not stated a
 8       position one way or the other on the identification or
 9       mis-identification of Y7.
10   A.  That's correct, yes.
11   Q.  So far as you yourself are concerned, have you actually
12       ever on any formal basis studied original material
13       relating to Y7?
14   A.  No.  I've seen no original material.
15   Q.  You have, however, looked at material on the Internet?
16   A.  Yes, indeed, yes.
17   Q.  I don't myself wish to ask you your opinion about Y7
18       and, therefore, I will not take matters further than
19       that but it is just as well to know that you have at
20       least studied some material relating to Y7?
21   A.  Yes.
22   Q.  What I want to do is to look at practice and procedure
23       more generally and I will come ultimately to probability
24       analysis.  Plainly I will be asking the practical
25       questions, primarily from the standpoint of your own
page   9
 1       organisation the FSS but if we stray beyond that to look
 2       at practice more widely, then so be it but I am looking
 3       primarily at the way that you practice and I will also
 4       use the textbook as one of my lead sources in relation
 5       to this.
 6           If I begin then with a general understanding of the
 7       methodology ACE-V, you and your organisation subscribe
 8       to the ACE-V methodology?
 9   A.  We do indeed, yes.
10   Q.  So far as the initial A, the analysis, is concerned,
11       what we are interested in understanding is whether there
12       is a practice, first of all, in relation to the mark of
13       beginning by an initial analysis of the mark and do I
14       understand that you would do that?
15   A.  Yes.  The initial action of the examination is to look
16       at the mark, the latent print, before the actual known
17       prints are seen and another word rather than analysis
18       phase is the information phase, which probably is a
19       little clearer, is to look at the image of that mark and
20       to record all of the observations that can be make.  So
21       this is the number and type of minutiae, other features
22       that might be present.
23   Q.  Plainly, as I proceed through this I will be discussing
24       each of these initials as a separate exercise.  Perhaps
25       if we could put up page 68 of the article, it might
page  10
 1       provide a helpful basis for the discussion.  You are
 2       discussing ACE-V.  What you say is:
 3           "In general, most examiners subscribe to the ACE-V
 4       methodology or a comparable protocol.  ACE-V is an
 5       acronym that stands for analysis, comparison, evaluation
 6       and verification and implies four distinct stages to the
 7       comparison between a mark and a print.  In the UK
 8       practice (and elsewhere), it is common to find little
 9       distinction or clear-cut separation between its stages."
10   A.  Yes, I think that's true.  Although ACE-V is practised
11       by the majority of examiners, looking at case notes,
12       looking at the way in which they undertake their
13       examination, it's sometimes difficult to split it into
14       four separate stages.
15   Q.  In particular, when I am looking at the analysis, I am
16       interested in the distinction between it and the
17       comparison phase.
18           As you practise, do you make a comprehensive
19       assessment of the mark before you carry out any
20       comparison?
21   A.  I think there is a difference here between perhaps what
22       theoretically one would want to do and what one can do
23       pragmatically.  Ideally you would make a comprehensive
24       assessment of that latent print and I think where the
25       mark is of poor quality that note-taking tends to be
page  11
 1       very extensive and the examination very extensive.
 2           As the quality of the mark becomes better, then I
 3       think some of the note-taking tends to be briefer.  From
 4       a pragmatic stance, particularly in my own organisation,
 5       time is of essence so the current process that we use
 6       for recording analysis is perhaps briefer than one would
 7       want to see.
 8   Q.  So if we begin, I suppose, the way that I am primarily
 9       interested in, which is practice.  First of all, the FSS
10       is a commercial albeit a Government-owned and controlled
11       organisation?
12   A.  That's correct.
13   Q.  Do the commercial practicalities impact on the manner in
14       which you practise.
15   A.  They do.  We work -- obviously the amount of fingerprint
16       comparison work that we do within England and Wales is
17       quite small compared to the police agencies and so,
18       therefore, our customers are not looking for us to
19       perhaps provide or charge for more expensive note-taking
20       over and above those police agencies.
21   Q.  Approaching it in stages, as you practise in practice,
22       if you are dealing with a mark that is of good quality
23       (and by that I mean it has a substantial number of
24       readily observable, discernible characteristics, any
25       number off the top of my head, minimum of 20, very clear
page  12
 1       observable characteristics), would you in practice
 2       expect to take much in the way of any note at the
 3       analysis stage?
 4   A.  Well, my expectation -- in a sense there would be a note
 5       that that is the case.  So in our current way of
 6       recording this, the sort of minimum I expect to see is
 7       some recognition of the general pattern, if that was
 8       available, and some comment about the fact that there
 9       are 20 readily visible minutiae within that mark and
10       that would be the extent of the note-taking.
11   Q.  Assume a complex mark.  Again, I suppose that requires
12       definition.  Can I begin, before I infect things with my
13       misconceptions perhaps, how would you characterise or
14       define a complex mark?
15   A.  Well, a complex mark, a difficult mark, would be one
16       where the minutiae in the features are not very distinct
17       within the image that you're inspecting, there is
18       clearly a lot of distortion, there might be overlays, a
19       number of factors that essentially distort the image
20       that you are trying to analyse.
21   Q.  With a complex mark, can you explain in practice what
22       form of note-taking would you undertake if you are
23       analysing a complex mark?
24   A.  Well, for complex marks the note-taking would be more
25       extensive.  By definition it needs to be.  So, again, we
page  13
 1       would be recording the general pattern, the number and
 2       type of minutiae and other features that are seen but we
 3       would need also to document any distortion, any
 4       artefacts that were visible within that image and
 5       ideally look for some mechanism by which that distortion
 6       has taken place.
 7   Q.  Let us assume, for sake of argument, again, a number,
 8       let us say there are 10 observable characteristics.
 9       First of all, is that something you would regard as a
10       complex mark or a clear mark?
11   A.  I would probably drop down a bit lower, perhaps around
12       six, seven or eight as being into the area of complex.
13   Q.  That is fine, six, seven or eight.  If you are in the
14       realms of six, seven or eight, would you in practice,
15       given the constraints you operate under, would you
16       actually list each of the six, seven or eight
17       characteristics and give a description, for example,
18       there is a possible ridge ending here, a possible
19       bifurcation there or do you simply note down that you
20       have observed in total possibly six, seven or eight
21       characteristics?
22   A.  If it's a mark where there are that number of
23       characteristics and no great deal of distortion or
24       other, I'll use the term red flags, something in that
25       image that makes you concentrate more and wonder about
page  14
 1       how it was actually deposited, then the recording would
 2       be, yes, there are six minutiae clearly visible on it
 3       and that would be the result of it.
 4           It really is driven by that particular mark.
 5       Sometimes looking at the mark, even if you have clear
 6       features, there will be something there that says I need
 7       to look more closely, I need to analyse this more
 8       closely, maybe the ridge flow is distorted or there is
 9       some aspect of it that requires closer attention.
10   Q.  So you may be more specific depending on how challenging
11       the mark actually is found to be?
12   A.  Yes.  The more challenging the mark, the more I would
13       expect to see the analysis increase in both time and
14       recording.
15   Q.  Is this summary that you keep, the note you keep, is it
16       in narrative rather than visual form at this stage?
17   A.  No, we -- the notes will be taken contemporaneously.  We
18       have a form, although examiners can use plain paper or
19       type it out on a computer should they wish, but the
20       notes are taken contemporaneously and kept within the
21       case file.
22   Q.  I am grateful that they are contemporaneous, for that
23       observation.  What I meant is it someone simply
24       narrating, "I have seen, say, six features", in
25       narrative form rather than actually having a picture of
page  15
 1       the mark on which you would record where the features
 2       were?
 3   A.  I'm sorry.  Yes, it's narrative form.
 4   Q.  If I then take that up and -- well, first of all, can
 5       there be a conclusion or an end to the process at the
 6       analysis if insufficient characteristics are reserved?
 7   A.  Yes.  Part of that process -- and the American's term
 8       this sufficiency -- is to determine whether or not that
 9       particular mark has any value within the examination.
10       So certainly at the analysis stage the decision could be
11       made that it's not comparable or that it's -- well, that
12       it's not comparable.
13   Q.  We have also heard some evidence that there may be an
14       intermediate stage, a mark may not be comparable to this
15       extent, that it may not be capable of ultimate
16       identification to an individual but if the mark is
17       sufficiently important it could be compared against, for
18       example, the prime suspect to see if he could be
19       excluded.
20   A.  Within the process that we use we determine whether the
21       mark is comparable and, therefore, that means it can be
22       compared against any set of known fingerprints that are
23       supplied.  I don't think myself I can make a distinction
24       between it's excludable and it's identifiable.  At that
25       stage of analysis what I am determining is can I compare
page  16
 1       that mark.
 2   Q.  So would there, therefore, be a relatively low threshold
 3       for that or not?
 4   A.  Yes.  By looking at the casework that we do, the vast
 5       majority of marks go through to the comparison stage.
 6       We actually discard very few at the analysis stage.
 7   Q.  First of all, to look at the scale, how many examiners
 8       do you have, Fingerprint Examiners?
 9   A.  We currently have six Fingerprint Examiners.
10   Q.  Can you give an indication of the sort of volume of work
11       that you deal with?
12   A.  Yes.  All examiners, in order to maintain competence,
13       have to deal with at least 12 cases in a year and most
14       examiners at the moment are dealing with somewhere
15       between 12 and 30 cases.
16           I should explain they work within different teams to
17       different customers and, therefore, to different
18       requirements and our examiners carry out the full
19       fingerprint process so we undertake the comparison but
20       we also undertake all the detection within the
21       laboratory as well.
22   Q.  So, in fact, you will develop marks, for example, on an
23       exhibit and then carry out the comparison yourself?
24   A.  That's correct.
25   Q.  Would you, therefore, accept -- and this has been a
page  17
 1       suggestion -- that your company is not as fully engaged
 2       in fingerprints as a police bureau, at least in terms of
 3       volume of work?
 4   A.  In terms of volume, that's absolutely correct.  We do a
 5       very small volume of work compared to a police
 6       Fingerprint Bureau.  Absolutely correct.
 7   Q.  Do you see that as having any relevance to the
 8       methodologies that you apply, the fact that you are a
 9       small bureau whereas police bureaux are much larger?
10   A.  Yes, I think it's fair to say it does because,
11       obviously, I've worked in a police fingerprint bureau.
12       I'm not under the same work loads and demands as a
13       fingerprint bureau.  So note-taking, for example, there
14       is, in theory at least, more time for that to take
15       place.  But then I do have those other constraints of
16       needing to meet very tight deadlines for customers and
17       needing to keep costs down, so, yes, different pressures
18       because I'm doing a slightly different job.
19   Q.  However -- and this is where I'm interested just in your
20       background with various committees that you sit in and,
21       as you say, you have had a police background before this
22       company that you work for now -- so far as the benefits
23       of note-taking applied to the different environment of a
24       police bureau, a much larger bureau, do you commend
25       note-taking as a standard methodology?
page  18
 1   A.  Yes, absolutely.  I think that one of the key issues
 2       around forensic science, and I think the NAS report
 3       tries to make this clear, is this concept of
 4       transparency, the ability to show to a court what you
 5       have actually done and how you achieved the results that
 6       you are presenting.  Part of that is contemporaneous
 7       note-taking and I think that is absolutely essential to
 8       the scientific process.
 9   Q.  There has also been some discussion, and I just ask for
10       your comment on this, that contemporaneous note-taking
11       can actually be helpful to the reasoning processes of
12       the examiner.  In other words, he has observed something
13       and then note-taking is a helpful check, can he actually
14       record it in a way that to him is persuasive.
15           Do you recognise that as a collateral benefit?
16   A.  Yes, I do indeed.  I think it is a very useful way of
17       articulating to yourself, for want of a better phrase,
18       what you have actually seen and what that actually
19       means.  I think it's very easy to rely on subconscious
20       thought or as some people describe the black book
21       process which doesn't actually force you to really
22       explain what it is that you are seeing and what that
23       actually means.
24   Q.  If I move from analysis into comparison -- before I do,
25       perhaps I should ask you something else and, please,
page  19
 1       feel free, if this is an area that's more
 2       Professor Champod.  Written in the analysis section we
 3       can see in italics that the word "tolerances" is used.
 4           "Analysis requires the examiner to examine a mark
 5       without the print in order to assess and locate the
 6       friction ridge skin features that may be used for
 7       further comparison.  The analysis also allows an
 8       assessment of the quality of the mark and the factors
 9       such as distortion, pressure, amount of residue and the
10       nature of the substrate and of detection technique used.
11       Taking these factors into account will help to set the
12       tolerances or boundaries that the examiner will have to
13       allow for considering a potential correspondence."
14           Can you explain what is meant by tolerances in that
15       context?
16   A.  Yes, I think it's very much Professor Champod's word but
17       I believe what we are applying here is that you can, as
18       you were saying before, make some decisions about how
19       useful that particular mark is going to be, whether it's
20       identifiable, whether it's searchable at that analysis
21       stage.
22   Q.  I will take you on then to comparison and, in
23       particular, what I am going to do is turn to page 69.
24       It is the second line of page 69.  What is written is:
25           "The comparison should always entail observation of
page  20
 1       features in the mark followed by their location in the
 2       print.  The reverse process, from the print to the mark,
 3       should only be made with care (and being documented)."
 4           As I read that, first of all, what you are saying is
 5       that, if I understand it, you can work in the reverse
 6       manner, it is possible to do it but you must do it with
 7       care?
 8   A.  Yes, it's certainly possible to do it.  I think it's
 9       allowable to do it but you do need to document that that
10       has taken place and be very clear in your own mind that
11       you are now seeing something in the mark that you didn't
12       see before because you have exemplar alongside it.
13   Q.  First of all, I just wish to be clear as far as accepted
14       practice is concerned.  We have heard ranges of opinion.
15       Perhaps the purest would be when you enter the
16       comparison stage you should only work with the
17       characteristics that you had noted at the initial
18       analysis stage.  That's the purest.  Then there are
19       ranges of view on the other side, perhaps the pragmatic
20       view, that there will sometimes be some event that
21       becomes more discernible in the light of the known print
22       so that you can go back to the latent in light of the
23       known and observe a characteristic not previously
24       observed and take that into account in the comparison.
25           Do I understand, therefore, in practice you will
page  21
 1       have occasion to work in the reverse manner?
 2   A.  Yes, in practice we do but with that stipulation that we
 3       are very clear and we have document that that is what
 4       we've done.
 5   Q.  Obviously, we have heard the danger here is some form of
 6       unconscious/subconscious bias that someone is now, in
 7       effect, perhaps at its crudest, seeing what he wants to
 8       see in the crime scene mark.
 9           Is that the bias that you are trying to protect
10       against?
11   A.  Absolutely.  I think it's a very easy route to follow,
12       to be influenced by the print and to start to see
13       features you didn't see before.  To guard against that,
14       you need to work in that direction, between mark and
15       print.  But from a practical standpoint, most examiners
16       I think would admit that if you do that then sometimes
17       you do miss genuine features.  So it's a practical
18       application.
19   Q.  I suppose it might be inferred that what you are saying
20       is that the protection that you have in place to guard
21       against the subconscious bias is, in fact, you do it
22       consciously because you are writing a note that you
23       have, in fact, done this in the reverse manner?
24   A.  Absolutely and, as an example, if the majority of
25       corresponding features were found only after looking at
page  22
 1       the print, clearly that would indicate to me, another
 2       one of those red flags, something is perhaps wrong and,
 3       therefore, I'd be very cautious in terms of formulating
 4       a finding on that comparison.
 5   Q.  The article proceeds to mention that:
 6           "Instead of focussing on minutiae during the
 7       comparison process, it is advised to compare every ridge
 8       and furrow (assessing comparatively their length and
 9       sequence) and then the shape of pores and edges if the
10       quality of the mark allows."
11           Is that set out as something that is more the ideal
12       than the way you would work in practice?
13   A.  No, I think there are two methodologies really.  The
14       more common methodology is to look for those minutiae
15       and associate the minutiae together in a configuration
16       by counting the ridges between them.  The other
17       methodology is to, as it says, trace each ridge.  So
18       even if there is not a minutiae there, look at the
19       ridge, look at the flow of that ridge.
20           That is quite common practice in the United States.
21       I wouldn't say the majority practice but common practice
22       in the United States and it does have some benefits.  If
23       you apply that technique, which is one that I've
24       adopted, you find that, in some respects, you notice
25       more discrepancies than if you just merely centre on the
page  23
 1       minutiae points, the events.
 2   Q.  An alternative methodology we heard is that an examiner,
 3       having carried out a comprehensive analysis of the mark,
 4       will visualise a target group within the mark, an area
 5       that has a prominent series of characteristics, and will
 6       then start analysis of potential matches by looking for
 7       a target group and then, if he finds the target group,
 8       will then expand out going between the two.
 9           Do you practice in that manner?
10   A.  Yes, it's actually quite a logical -- you need a
11       starting point.  In some respects you can put an S
12       between the A and the C, as a sort of search area
13       whereby you have your latent, you have some features in
14       that latent, you're looking for the known print which is
15       a potential match, and you are doing that by taking some
16       notable feature within the mark and looking for it in
17       the print and that's the process that you take.
18           Once you find that starting point, you can then
19       undertake the comparison by building out from that point
20       or, indeed, select another point and build out from that
21       point.
22   Q.  If you are doing your examination in that manner, as you
23       say, with the intermediate S, the search, is that a
24       stage where you or is that a situation where you would
25       see note-taking as essential, given the risks of being
page  24
 1       influenced as you proceed from one back to the other?
 2   A.  Interesting, we probably wouldn't take any notes at that
 3       point.  Essentially, all we're doing there is taking
 4       that target group and looking for it.  What you should
 5       remember, in most comparisons you have not got a mark
 6       against one particular set of ten prints.  You might
 7       have several sets of ten prints so you're looking for
 8       the particular finger or thumb which is going to match
 9       within that group.  That's the S type process.
10           Once you have found that potential, then you move
11       into the comparison stage and start, again, with
12       a starting point and then start tracing ridges and
13       moving out.
14   Q.  So the note-taking would be done at the end of the
15       analysis phase.  You might use an intermediate stage, a
16       search, to narrow down so that you then know the
17       particular marks you are going to look at more
18       carefully, but then you practice what you have mentioned
19       earlier, where you would do a more comprehensive
20       analysis of even the print, tracing ridges and the likes
21       because you feel that that gives more potential to see
22       differences as well as similarities?
23   A.  Yes, that's correct.
24   Q.  Also in the article you mention, at the second stage at
25       the end of the comparison stage, the benefits of
page  25
 1       documentation.  There seems to be a hint here again, of
 2       a practical slant to this because you:
 3           "Documentation of this crucial stage is sometimes
 4       sparse due to time constraints and it is not common
 5       practice to produce any form of chart to record the
 6       information."
 7           Is that --
 8   A.  In the comparison stage?
 9   Q.  Yes.
10   A.  Yes, we do actually produce some documentation there.
11       If we exclude a mark, then actually we don't document
12       why we have excluded it.  We merely state it's excluded
13       but if we make an identification, if we make some
14       association, then we will document why that association
15       has been made.  That can be in narrative form but more
16       often is in the form of a digital image of the mark and
17       the print showing those features that actually
18       correspond.
19   Q.  But you do accept that that is not, across the
20       fingerprint practising community, common practice?
21   A.  It's not.  Obviously, if we go back 20 or 30 years it
22       was common practice for each identification to be
23       charted and produced for court but as courts have not
24       required that, then that practice has ceased.
25   Q.  The text does, however, gone on to say:
page  26
 1           "The absence of documentation may lead to
 2       difficulties in disputed cases", and you say "we", that
 3       is you and Professor Champod, "would advise systematic
 4       documentation of the comparison stage."
 5           So in a sense you have committed yourself to the
 6       benefits of documentation at the comparison stage,
 7       conscious that it could always come into a dispute in
 8       court?
 9   A.  Yes, I think the guiding principle here is that in each
10       of these stages of this examination, as the forensic
11       scientist, what I am doing is recording my observations
12       on which I will base my findings.  So if there's a
13       disputed identification at a later date I can go back to
14       contemporaneous notes and show how I formulated my
15       decision.
16   Q.  Again, just to follow this through to give a
17       perspective, have you yourself had to give evidence in a
18       case that involved a disputed identification?
19   A.  No, I haven't.
20   Q.  Have you heard -- other than the examples that are
21       mentioned in the text -- have you heard of examples of
22       cases going to court with a dispute?
23   A.  Being disputed in court?
24   Q.  Yes.
25   A.  Rarely, and not very many from the United Kingdom.
page  27
 1   Q.  We will come on to some other features of that a little
 2       bit later.
 3           What I want to look at in the next context of
 4       evaluation that has been carried out, I will break this
 5       down and what I want to ask you about is a concept I
 6       discussed with you yesterday about unexplained
 7       differences.
 8           As you have said, when you are carrying out your
 9       comparison you are as attuned to differences between the
10       mark and the print as you are to similarities and we
11       understand that if a difference is observed it's not
12       necessarily fatal to a conclusion of match if some
13       persuasive reason is present for the difference.
14           Is that fair?
15   A.  That's absolutely fair, yes.  I think if there is a
16       discrepancy, if that can be explained or some
17       proposition put forward for why that discrepancy occurs,
18       then that's valid.
19   Q.  Perhaps this is helpful, you would use the word
20       "discrepancy".  If it has an explanation, a persuasive
21       explanation, then it is not regarded as a difference; is
22       that fair?
23   A.  Yes, that's fair.
24   Q.  But if there is no ready explanation for the discrepancy
25       are you beginning to look at it as a difference?
page  28
 1   A.  Again, it's a warning signal.  If you cannot see a
 2       reason for it, then you need to be very cautious from
 3       that point on.  I think if you have one discrepancy and
 4       you cannot find a persuasive argument for the one
 5       discrepancy, how you proceed is probably based on the
 6       other information that you can see.  But if you have a
 7       number of discrepancies and you cannot actually find
 8       persuasive mechanisms for them, then really you are
 9       moving towards exclusion.
10   Q.  Can I take it from the answer that you have given that
11       you could, at least in theory at the moment -- we'll
12       come back to practice -- you could in theory contemplate
13       a situation where, with a discrepancy for which you have
14       no clear explanation, you might nonetheless conclude
15       that there's a match if there is a sufficiency of
16       similarities present.
17   A.  I think that is absolutely possible, with the caveat
18       that in any report that is produced I would be saying
19       that there is this discrepancy so that it would be very
20       clear to courts and to others and to defence examiners
21       that I'm aware there's a discrepancy there but I can't
22       find a mechanism but because of the other information
23       I'm willing to provide a finding.
24   Q.  Have you had experience of doing just exactly that,
25       reaching a conclusion in the presence of an unexplained
page  29
 1       discrepancy?
 2   A.  Actually, I don't recall that I have, no.  I think this
 3       is a very theoretical argument.
 4   Q.  That is why I am just interested in the theory, then
 5       coming back to the practice.  So though you can conceive
 6       of this in theory, in practice you've never actually
 7       done it?
 8   A.  Personally, no.
 9   Q.  Can you explain the difference really between the two?
10       Is there an in-built direction in which you would err,
11       so to speak?
12   A.  I don't quite understand.  In terms of ...?
13   Q.  Without me putting words in your mouth, can you give us
14       an explanation why it would be that despite the
15       theoretical possibility that you might be able to
16       express a conclusion in the presence of an unexplained
17       discrepancy, that in fact you never have?
18   A.  I've just never come across that.  Yes, that sounds -- I
19       think in most cases where I have had discrepancies I
20       have been able to put forward a proposition for those
21       occurring.  It's just the casework that I've seen.
22   Q.  If I move beyond the question of evaluation now to
23       verification, I understand verification -- and your
24       article confirms this -- to be an essential, an
25       important part of the methodology.  Is that fair?
page  30
 1   A.  That's correct, yes.
 2   Q.  If I understand it correctly, standing back, the ACE
 3       part is an individual assessment with a high subjective
 4       content; would that be fair?
 5   A.  That's correct.
 6   Q.  The verification is, in effect, a quality assurance that
 7       a different individual of competence and qualification
 8       will himself reach the same conclusion, so it's a
 9       cross-check?
10   A.  Yes.  I think it's fair to say verification has -- there
11       are a number of different schools of thought around
12       verification.  We take the view that verification is
13       essentially a quality assurance, it's providing
14       assurance that the correct answer has been found by the
15       initial examiner.
16           There is another school of thought which tends to go
17       by the term of "blind checking", which is essentially a
18       re-examination.  So its two independent examinations
19       arriving at the same result, both of which are aimed at
20       providing assurance.  One is more labour intensive than
21       the other.
22   Q.  Also we have heard some questions raised as between the
23       two schools that in a smaller organisation, obviously
24       yours is on the smaller end with six examiners, it may
25       be, frankly, impractical to expect entirely blind
page  31
 1       verification to proceed?
 2   A.  Yes.  I mean, I've been party to the same discussions
 3       and with ourselves, for example, blind verification
 4       would be an issue, essentially it's going to raise costs
 5       and we do have a limited number of staff.
 6           I think there is, again, perhaps an academic
 7       argument that if that's the right thing to do that's
 8       what should be done and the resources should follow the
 9       correct practice and not the other way round.
10   Q.  Do you see, with all the imperfections as you said of
11       the cost constraints, do you see a necessity for
12       entirely blind verification or are you satisfied that in
13       the appropriate climate with individuals suitably
14       trained and acting with integrity, obviously, that blind
15       verification may be an unnecessary complication?
16   A.  My personal view is blind verification is a step too
17       far.  I think, as you say, with the correct training, in
18       the correct environment, assurance by an independent
19       check is purely sufficient.  I don't see the need to
20       move to blind verification.
21   Q.  In this context, you have obviously used the word
22       "independent", which is the key to this.  The key is
23       what is meant by independence.  In a blind verification
24       one can say there's a high degree of independence, the
25       Second Examiner does not know what his has preceded and
page  32
 1       is doing an entirely fresh analysis.  In a sense, that
 2       is the highest degree of independence.
 3           Before we really debate what is the interpretation
 4       you would apply to independence, if I come back into
 5       your own work and I have an example that I will give,
 6       again, if we were looking at a complex mark let's use
 7       the range of six, seven or eight features that you have
 8       mentioned is indicative of a complex mark for you.  The
 9       example I discussed with you yesterday is that a First
10       Examiner, let us say you, you have reached a conclusion
11       of a unique identification based on let us say the most
12       challenging number of six because you have found
13       something peculiar about that combination of six that to
14       your eye is indicative of uniqueness.  Is that fair?
15   A.  That's fair, yes.
16   Q.  I am hypothetically the Second Examiner.  I also see
17       those same six but I am just not myself seeing a
18       sufficient peculiarity in the six.  So I am happy that
19       there are six characteristics there but I am not myself
20       tending to the view that it's unique.
21           Would it be, in practice, could there be a
22       discussion between the two of us to see if a consensus
23       could be arrived at?
24   A.  Within the FSS, yes, that form of peer review discussion
25       takes place and we're quite happy with that.  Again,
page  33
 1       a caveat is because we have well established examiners
 2       with lots of experience, which makes me comfortable with
 3       that process.  But what we do is we do not have, for
 4       want of a better word, a conference of all Fingerprint
 5       Examiners discussing a particular comparison.  So that
 6       first and Second Examiner may have a discussion.  They
 7       may formulate a consensus opinion, again, which would be
 8       documented, as will any of their findings or they may
 9       still disagree, in which case it will go on to another
10       process through myself as the Lead Examiner to come to a
11       conclusion as to what we should report in that
12       particular case.
13   Q.  Again, if I am looking in purely theoretical way at what
14       might be understood by the word "independence" in this
15       process, one might suggest that if the two examiners
16       have conferred, they have had a conversation that
17       has resulted in the second examiner (in this case me)
18       concluding that there is indeed something persuasively
19       distinctive about this combination of six so that I can
20       now sign off on an identification, it might be suggested
21       that that conversation having taken place has
22       compromised my independence.
23           Equally, the converse argument is that I am still an
24       examiner of skill and competence, ultimately the test
25       for me is my own integrity and professionalism and I
page  34
 1       know I'm responsible in court for that conclusion.  The
 2       independence really consists in my skill and integrity.
 3           What view do you take on that?
 4   A.  I think the independence is the independence of the
 5       examination, so if we take that scenario, the First
 6       Examiner is independent in formulating their findings.
 7       The Second Examiner has been independent in formulating
 8       the findings.  The discussion is after, if you like,
 9       that process of independence so the Second Examiner
10       hasn't been influenced directly.  The conversation is
11       taking place on the difference of findings as like a
12       separate stage.
13           Yes, it's fair.  It's not a perfect system and in
14       the forming of the word "independence", it probably
15       isn't there but it is a practical, again, solution to
16       those issues.
17   Q.  You have already anticipated by one of your questions,
18       the next stage and you do actually have a formal process
19       if there is a difference of view between two examiners
20       for the resolution of that, it comes to you as the
21       Senior Examiner to resolve the difference of view?
22   A.  It does and, again, I take a very pragmatic view.  The
23       first set of questions is what value is this particular
24       identification within the case.  If it has no probative
25       value within that particular case then it's more an
page  35
 1       academic issue as to whether or not we have made a
 2       match.
 3           Then, obviously, I have full written reports of the
 4       examiners who have looked at that before, and more often
 5       than not what I will do is recruit in another examiner
 6       who has had nothing to do with case to make some
 7       judgments for me.
 8   Q.  Have you had experience, before we become too
 9       theoretical, have you had experience of actually
10       reporting out as an identification through this
11       arbitration process something where there had been at an
12       earlier stage a difference of view amongst some of your
13       colleagues with one of them not yet persuaded that there
14       truly is an identification?  Have you had experience of
15       reporting that out?
16   A.  Again, no, but I think that's part and parcel of, as has
17       been pointed out, the low level of casework that we
18       have.  Most of the thinking, as it were, and the way in
19       which I now carry out my role has developed whilst I
20       have been with the Forensic Science Service.  It would
21       be interesting to apply those concepts in a busier
22       Fingerprint Bureau to see if they actually work.
23   Q.  With that caveat, therefore, that you have not actually
24       done this in practice, if we can just take the benefit
25       of the thinking you have had in relation to the
page  36
 1       processes, could you conceive of a situation in which
 2       the arbitration process would result in an
 3       identification being reported, for example, to the
 4       police despite a doubt in the mind of one of your
 5       colleagues?
 6   A.  Again, from a theoretical point of view, yes, but I
 7       think the point there is that the court, the police,
 8       should be made well aware that there is this difference
 9       of view between examiners.  I think the underlying thing
10       there is that all the examinations have been competently
11       carried out but I have no issue in producing a report to
12       court that says, "Three examiners believe this is a
13       match, for these reasons ... and one examiner believes
14       it's not, for this reason ..."  I think that's
15       fulfilling the role as a forensic scientist.
16   Q.  The next part I want to do is just pick up something
17       that is in the text, moving to a separate part of the
18       evaluation.  I have talked about unexplained
19       differences.  I have gone through verification.  I want
20       to come back now and look at the standards that are
21       being applied.
22           Your Article goes through, and I don't need to
23       rehearse the history with you of the 16-point standard.
24       It then comes to the practice within the UK of the
25       non-numeric approach and you have a formulation that's
page  37
 1       on this page, page 69, of the non-numeric standard that
 2       I am interested in just looking at with you.  It is the
 3       penultimate paragraph on the page.  At least to me the
 4       text is a little blurred but we will do our best:
 5           "In the absence of dissimilarities, the examiner
 6       will weigh the corresponding features in reference to
 7       the standards for identification.  In a nutshell, the
 8       individualisation will be reached when the examiner
 9       observes a level agreement (across the three levels of
10       legible features) that exceeds the highest level of
11       correspondence he observed through his/her training and
12       experience in comparisons involving non-matching
13       entities."
14           Then it says:
15           "An identification is then concluded when the mark
16       shows sufficient ridge quality (clarity) of friction
17       ridges in agreement with the print so that the
18       probability for such a match to happen if a print from
19       another person is submitted is deemed to be impossible."
20           I am interested in that in two different ways.  If I
21       go at it in the reverse way, as we will see later on in
22       the article you are, in fact, an advocate for the use in
23       fingerprint examination of probability material?
24   A.  That's correct.
25   Q.  Writing from that perspective -- and it's clear just to
page  38
 1       make that perspective -- what you are, in fact, saying
 2       is that conventional fingerprint evidence has implicit
 3       within it a probability analysis.  It's simply that the
 4       examiner has concluded from his observation that the
 5       probability of a match is, in fact, so high that any
 6       other individual presenting those same features is, in
 7       fact, impossible?
 8   A.  Absolutely right.  Conventional fingerprints is
 9       probabilistic by basis, a form of subjective
10       probability.
11   Q.  Albeit they might think the threshold is one of
12       100 per cent before they would express an opinion?
13   A.  That's right.  It's like a caveat being put over the top
14       of that that, yes, it's probabilistic but only when I
15       get into the very high likelihoods of the match am I
16       willing to report it.
17   Q.  We will look at that a little bit later, in a moment.
18           What I want to do is reverse back to the preceding
19       sentence because that implies that an examiner would be
20       able to say to me, if asked, that perhaps through
21       working with an AFIS system that throws up potential
22       matches, an examiner would be able to say to me,
23       "Mr Moynihan, I have only ever, in all the AFIS
24       researches I have done, I have only ever have found two
25       individuals who happen to share five, six, seven
page  39
 1       characteristics", whatever it may be, "and, therefore,
 2       the threshold I work to in practice is that highest
 3       number.  I have seen two individuals' prints who do not
 4       match and let us say for sake of argument they had five
 5       features in common so, therefore, my working rule is I
 6       must have a minimum of five before I identify".
 7           Is that as I should be reading that particular
 8       sentence?
 9   A.  Yes.  Essentially, what we are saying there is that each
10       individual examiner is formulating a personal threshold
11       based on their previous observations.  It's more than
12       just the number of minutiae.  It's about the
13       configurations of minutiae as well.  Any particular
14       configuration of five minutiae that I have seen many,
15       many times before I would treat with less rigour in a
16       comparison than someone who has only seen that
17       configuration once before.
18   Q.  In fact, before I ask you to comment on the evidence we
19       have had from others, do you yourself have a personal
20       threshold that you apply consistently across all
21       examinations or does your threshold vary?
22   A.  I don't have a numerical threshold but undoubtedly, yes,
23       my examinations are based on my experience as a
24       fingerprint examiner.
25   Q.  Can you tell me -- again, I know that you wouldnot be
page  40
 1       expressing your conclusion numerically but as you work
 2       and you're checking for corresponding features, do you
 3       have a minimum number at all that you would be working
 4       to before you would express a conclusion of identity?
 5   A.  Not consciously, no.  I wouldn't say that I set a
 6       threshold at any particular number but again probably
 7       subconsciously there is a number in my head somewhere
 8       but consciously I would say I have no number.
 9   Q.  If it is in your head somewhere, it's not popping out
10       today, fair enough.
11           What this then comes back to is the answer you have
12       given is much the same as the other examiners who have
13       given evidence.  We have not had any examiner who says,
14       "Yes, there is a minimum number I apply consistently",
15       that leads me to wonder whether the examiners are, in
16       fact, carrying out a more intuitive judgment because
17       they are not saying, "Mr Moynihan, I have seen two
18       individuals with five characteristics in common,
19       therefore, my minimum number is five", they are simply
20       saying depending on the particular configuration or
21       constellation of minutiae in a mark I will either see an
22       identity or not.
23           Is that fair, it is a more intuitive process?
24   A.  I think that is fair comment.  I think there's probably
25       a number of reasons why Fingerprint Examiners are loath
page  41
 1       to state a number and one of the reasons for that is we
 2       know there is no number.  We mostly accept that so I
 3       think there's a nervousness to state a number even
 4       though perhaps some Fingerprint Examiners are perhaps
 5       consciously using a number.  But, yes, as an intuitive
 6       process that's correct.
 7   Q.  If I bring up pages 73 and 74 of the article, perhaps if
 8       we could begin to page 73 and then move pretty sharply
 9       on to page 74, page 73 -- if we just move the text up
10       because it is the last two paragraphs I am interested
11       in -- you observe, as you have told us earlier on when
12       talking about standards, that in continental Europe a
13       fair number of countries use a standard, 12 being quite
14       common, apparently Italy still being at 16 or 17?
15   A.  Yes, that's correct.
16   Q.  However, then you go on, along with Professor Champod,
17       at the foot of page 73 in fact, to argue, as you say,
18       from a logical perspective that there are reasons for
19       not insisting on a numerical standard?
20   A.  Yes.
21   Q.  So, therefore, if I am looking at the perspective from
22       which you are giving evidence, you are someone who is
23       persuaded there should be no numerical standard?
24   A.  I am absolutely persuaded of that, yes.
25   Q.  If we look then at page 74 and I will give you a
page  42
 1       chance -- and I must apologise, I hadn't forewarned you
 2       I was going to use this article to the extent that I am
 3       today -- if I give you a chance to read it, along with
 4       others in the hall, the reasons you set out as 1, 2 and
 5       3, and the question will be whether you are satisfied
 6       that that gives your own reasoning for firmly advocating
 7       the absence of a numerical standard.  (Pause)
 8   A.  Yes.
 9   Q.  You are content with that?
10   A.  Yes.
11   Q.  As we say, in looking at -- and this is not intended in
12       any way to trap you -- what you have actually written is
13       that you can, in fact, if it is done appropriately, use
14       even as few as three minutiae and arrive at a reliable
15       conclusion.  The question we will come to ultimately is
16       how you express that conclusion.
17   A.  Yes, you can come to a reliable conclusion and, you're
18       absolutely right, how you express that conclusion is
19       perhaps the issue to discuss.
20   Q.  If I look at this just now, what I am interested in is
21       moving through, before we come to probabilities, in
22       other words, how you advocate things are expressed, if I
23       take a step back and look at current practice.  We have
24       had examiners give evidence and, indeed, even current
25       trainers give evidence that the training for Fingerprint
page  43
 1       Examiners is to express themselves only in terms of
 2       100 per cent certainty.  In the absence of 100 per cent
 3       certainty, there will be no expression of opinion about
 4       identity.
 5           Do I understand you, from what you have said about
 6       even IAI's current position where they do not permit
 7       views of degrees of probability that you would accept
 8       that that is the orthodox, in fact, universal practice
 9       of Fingerprint Examiners?
10   A.  Yes, it is the universal practice.  I think particularly
11       with the IAI statement, for example, and I can't really
12       do this but it's necessary to look at the context in
13       which this was actually brought into force so it's not a
14       clear-cut rejection of probabilities as more a response
15       to events that were taking place in the early 1980s.  I
16       think also there's a large sociological driver to this.
17       I think there is a perception amongst the fingerprint
18       community that the customers, the courts and police
19       forces require them to produce 100 per cent evidence
20       each time and that is what is driving the process.
21   Q.  So there's a perception by the fingerprint community, as
22       you say, universally that the courts are expecting of
23       them only -- only -- 100 per cent certainty as to
24       identification and nothing else?
25   A.  I believe that to be the case, yes.
page  44
 1   Q.  Before we look at possible variants to all of this, from
 2       a scientific perspective, what is your reaction to the
 3       proposition that someone can be 100 per cent certain
 4       based on the observation of minutiae in what is, after
 5       all, a sometimes distorted, potentially partial
 6       impression of a fingerprint?
 7   A.  I think from a purely scientific perspective,
 8       identification or individualisation is not achievable,
 9       certainly on a scientific basis, but I think within
10       again that pragmatic use of fingerprints what we are
11       doing is dismissing the likelihood that the source could
12       be another because it's so diminishingly small and,
13       secondly, as experts what we're giving is a personal
14       opinion that it can only have come from one particular
15       individual.  Identification for me is my opinion rather
16       than something fundamentally founded on a scientific
17       principle.
18   Q.  Because you will come ultimately, as we will see, to
19       advocate probability as a way forward, do you have from
20       the benefit of the work you have been doing any
21       misgivings about evidence being presented in court based
22       on 100 per cent certainty coupled with the absence of a
23       numerical threshold so that we have what I was
24       discussing earlier as an intuitive judgment being formed
25       by an examiner?
page  45
 1   A.  I think it's fair to say, yes, I do have some small
 2       misgivings.  I think the current system does allow for
 3       some overstating of the evidence.  I can't produce
 4       examples of this but I'm pretty confident that some
 5       identifications, the actual configurations of the
 6       minutiae, the basis on which that identification was
 7       given is not as strong as the examiner believes it to
 8       be.
 9   Q.  Indeed, if I look at what has been written by you, I go
10       back to page 72, page 72 is written:
11           "Fingerprint identification in the UK and around the
12       world is understood to mean that a mark has been
13       attributed to a particular individual to the exclusion
14       of all others, although it is seldom articulated in this
15       way. 'Others' refers often to any human in the world,
16       living or dead.  In fact such a claim may be unnecessary
17       as in all but a few scenarios the suspect (or source)
18       may be from a much smaller or restricted population.  It
19       is therefore interesting that examiners have felt the
20       need to make a much greater claim, presumably to
21       increase the perceived evidential value of the
22       identification."
23   A.  Correct, yes.
24   Q.  In fact, what I am interested in, as you write there, in
25       effect it is a two-edged sword.  First of all, by the
page  46
 1       claim of 100 per cent certainty the examiner may be
 2       overstating the reliability of his conclusion in one
 3       particular case -- may be, not necessarily is.  Correct?
 4   A.  Correct.
 5   Q.  Secondly, this is the other edge of the sword, by
 6       addressing such a standard Fingerprint Examiners, the
 7       community universally may be denying to the courts a
 8       body of evidence?
 9   A.  I believe that to be the case, yes.
10   Q.  If I pick up in relation to your own statement -- I
11       don't have the particular reference here -- do you have
12       experience of giving reports and, therefore, giving
13       evidence to the judicial system, where you are not
14       concluding a unique identity but you are in fact
15       offering a view that is, I might suggest, a consistency
16       between a mark and an individual but not to the degree
17       of individualisation?
18   A.  Yes, in my career I have actually produced evidence on
19       two occasions, where on one particular occasion, the
20       examination had been inconclusive in terms of
21       associating the particular mark.  I was asked to, if you
22       like, give a measure of how inconclusive that was and
23       that is what I did, based on my observations.  So it
24       wasn't an identification but it was an attempt to
25       provide the court with some idea of how likely it was
page  47
 1       that that mark came from a particular individual.
 2   Q.  We have heard and we may hear some more evidence about
 3       this next week that there is perhaps or there has been
 4       evidence, starting in England, that looks at it perhaps
 5       from a way that might be more comfortable for a
 6       Fingerprint Examiner expressed in a way that he is
 7       incapable of excluding an individual.
 8           Is that a more relaxed way for a Fingerprint
 9       Examiner to be looking at it?
10   A.  I think that's nicely put.  I think it's a very
11       comfortable way.  If your career has developed with this
12       duality of identification/not identification to be able
13       to say, "I actually can't make a decision one way or the
14       other", is a comfortable option.
15   Q.  So, in other words, by incapable of excluding what I
16       understand this to mean, and please correct me if I am
17       wrong, is the Fingerprint Examiner is saying the overall
18       pattern between the mark and the print is the same.
19       It's a whorl or a loop or whatever.  Secondly, there is
20       a limited number of observable characteristics which
21       happen to be in common between the mark and the print.
22       There are no unexplained differences.  So far so good?
23   A.  Yes.
24   Q.  But the examiner is saying the level of similarity is
25       not so high as to persuade me as to unique identity?
page  48
 1   A.  Yes, in a sense, as I mentioned about personal
 2       thresholds, it hasn't exceeded the personal threshold of
 3       that examiner.
 4   Q.  Without the work you have been engaged in most recently
 5       on probability analysis, how have you found yourself
 6       able to express this intermediate position when giving
 7       evidence in the past?
 8   A.  This has been purely subjective and reported as such.
 9       What I have --
10                    (Interruption: car alarm)
11   Q.  It has been more a subjective judgment that you have
12       expressed?
13   A.  Yes.  The one particular case that I'm thinking of, we
14       had only a particular number of individuals who were
15       known to be or for it to have been possible for those
16       individuals to have left the mark.  So I had a
17       population with which to work to.  So there were
18       immediately some judgments that could be made around
19       general pattern, for example.  But it was intuitive and
20       it was very much based on my opinion as to how likely
21       that was to occur.
22   Q.  I am grateful to you.  What you have explained is it
23       really fits in with the text, if we put in an artificial
24       constraint and tell you it can only be a limited number
25       of individuals, the number of variables is, therefore,
page  49
 1       at least in theory, constrained by the population that
 2       you are being asked to consider.  Is that fair?
 3   A.  Yes.
 4   Q.  Therefore, working within this artificial constraint
 5       that you are assuming it can only be one of a limited
 6       number of individuals you, therefore, might need a lower
 7       number of points in agreement to discriminate among that
 8       population?
 9   A.  Yes.  Professor Champod in an article actually made this
10       point, if you imagine hypothetical situations of cruise
11       ships in the middle of the Atlantic with a fixed number
12       of individuals then the number of correspondences you
13       would need between the mark on the murder weapon and a
14       particular individual will reduce because you can look
15       at the entire population.
16   Q.  That, in a sense, explains and really introduces what
17       you are talking about with this, in fact, implicit
18       probability analysis that if you are looking at it from
19       an objective point of view, standing back, the scenario
20       that you are given can give you parameters for your
21       probability analysis so if the scenario is the murderer
22       must have been someone on that ship and you have
23       captured them all at a point in time then the
24       probabilities assist you in saying that you may need a
25       fewer number of variables to identify the culprit?
page  50
 1   A.  In that particular hypothetical situation, yes, that's
 2       right.
 3   Q.  The alternative example we have had, which is not
 4       theoretical, it is practical, is identifying individuals
 5       who have died in a plane crash by reference to
 6       fingerprints.  Again, it would be the same probability
 7       scenario that is at play, there is a limited number of
 8       potential individuals.  You are simply discriminating
 9       among them, you might need very few characteristics to
10       do that?
11   A.  Absolutely.
12   Q.  What I would like to do is just in looking at this in
13       your own statement, paragraph 53 you mention -- perhaps
14       I should bring it on screen FI0136.  It is paragraph 53
15       which is at page 12, paragraph 53, page 12.
16           You are looking at likelihood ratios and I will come
17       to that after we are all fortified by some coffee to
18       assist but what you say is:
19           "However, the likelihood ratio for different
20       configurations of the same number of minutiae varies.
21       In other words, two different matching marks and prints
22       with the same number of corresponding minutiae may have
23       a different weight of evidence."
24           If -- and I do say if -- I understand this
25       correctly, what you are saying is simply to say that
page  51
 1       there are, let us say, eight points that correspond,
 2       simply eight in the abstract, and I have two sets of
 3       matching prints, each with eight points matching, you
 4       are, in fact, saying from a statistical point of view,
 5       one may in fact be more probable than the other, even
 6       though they both have the same abstract number or
 7       concrete number of matching features of eight?
 8   A.  Yes, because what we're talking about is a configuration
 9       of a set number of minutiae so a particular
10       configuration of eight minutiae could be quite rare and
11       another configuration of eight minutiae could be very,
12       very common.  So when measuring those probalistically,
13       statistically, you are going to get different weights of
14       evidence.
15   Q.  Before we move into probability -- and this is the point
16       for concluding this particular section -- I discussed
17       with you this morning me, I think the discussion this
18       morning will help you to understand where I am coming
19       from.  We have heard evidence that Fingerprint Officers,
20       Fingerprint Examiners, may think in terms of events.  Is
21       that a term you are familiar with?
22   A.  Yes.
23   Q.  An event can cover something that is in the background
24       substrate, a pimple in wood or something of that sort,
25       or it could cover a ridge characteristic and if it's a
page  52
 1       ridge characteristic it could be a ridge ending or a
 2       bifurcation.
 3           Do you follow?
 4   A.  Yes, I do.
 5   Q.  Do you accept that?
 6   A.  Yes.
 7   Q.  We have heard some evidence that it may not matter to a
 8       Fingerprint Examiner what the precise nature of the
 9       event is, so long as he finds an event in the same
10       location in mark and print.
11           First of all, just thinking of this in terms of
12       conventional fingerprint evidence, what is your comment
13       on the proposition that the precise nature of the event
14       does not matter provided there is some event in the
15       corresponding location between mark and print?
16   A.  No, I think whether it's a bifurcation or a ridge ending
17       is very important.  I think if you considered two known
18       fingerprints with three bifurcations lying alongside
19       each other and you had another known fingerprint with
20       three clear ridge endings lying alongside each other,
21       they will be different.  I don't think anyone would say
22       that they are matching events so the type of minutiae or
23       bifurcation is very significant.
24   Q.  Let us just assume, for sake of argument, that the
25       precise discrimination in mark and print is doubtful, so
page  53
 1       that it is open to interpretation, and let us say an
 2       examiner, conscious of this and he has reached perfectly
 3       genuinely the conclusion of a match between the two so,
 4       therefore, he is going to have to call the judgments the
 5       same way in mark and print.  Let us say he is doing it
 6       perfectly genuine and you have use the example of three
 7       so I will use that.  He, therefore, has three events in
 8       sequence in mark and print.  He is going to call them
 9       either three matching bifurcations or three matching
10       ridge endings and plainly in mark and print he must call
11       them both the same way.
12           At least in theory, starting in theory and then
13       we'll look at the practicalities, would there be a
14       difference from the studies that you are aware of be a
15       difference in the probability of a match if it's called
16       three matching ridge endings as opposed to calling it as
17       three matching bifurcations in a row?
18   A.  It's interesting.  We've not -- or personally, I haven't
19       actually carried out that experiment using our
20       mathematical model that but my opinion would be, yes,
21       potentially there could be.  This is really the thing
22       with measuring the probabilities of a particular
23       configuration, it is possible they would have the same
24       weight of evidence but my expectation is it would be
25       different.
page  54
 1   Q.  Why would you have that expectation?  I appreciate you
 2       have not been asked to carry out the experiment.
 3   A.  I am just basing that on my experience so, therefore, it
 4       might not be the strongest piece of evidence I can put
 5       forward but my experience tells me that that probably
 6       would be the case but until the work is done it's
 7       difficult to be precise.
 8   MR MOYNIHAN:  I think we may well benefit by being fortified
 9       by caffeine before we move to the next stage, which is
10       probability analysis.
11   THE CHAIRMAN:  Just while I remember to ask you is there any
12       possibility that the characteristics, the same
13       characteristics, might appear nearer the core than would
14       further out, if you see what I mean, the similarities
15       might appear?
16   A.  There are -- yes, there are some research papers that
17       look at the distribution of different characteristic
18       types within different areas.  So you do get a higher
19       likelihood for some bifurcations in some areas rather
20       than others.  The simple answer is probably.  One thing
21       with the model we have it doesn't make any assumptions
22       of that type.
23   THE CHAIRMAN:  I was just wondering whether not only the
24       rarity of the characteristic but the place where it is
25       found in the print has a bearing on it?
page  55
 1   A.  Yes, as I say, there's limited research around the area
 2       but I'm sure that is the case, certainly in terms of if
 3       you look at the delta-type formations you have more
 4       characteristics around the delta formation than in other
 5       parts of the print.  So the type, I am sure, probably is
 6       relevant to those areas as well.
 7   THE CHAIRMAN:  What we normally do is take a short break now
 8       and we will sit again at 11.25.
 9   (11.04 am)
10                         (A short break)
11   (11.28 am)
12   MR MOYNIHAN:  Mr Chamberlain, if I just reverse back into
13       some of the questions we were talking about earlier just
14       to cover some more of the detail.  I will ask you
15       something about verification in a moment and what you
16       are currently working on.
17           If I step back to the beginning, we have talked
18       about the volume of work that you are doing and how your
19       practices may relate to the volume of work.  I don't
20       want to breach commercial confidentialities but the type
21       of agencies that you work for, do you do work for police
22       forces?
23   A.  Yes, we do, for police forces, some Government
24       investigative agencies.
25   Q.  Any other agencies?
page  56
 1   A.  No, I think that covers ...
 2   Q.  Do you ever get involved in defence work; in other
 3       words, are you ever approached by solicitors acting for
 4       accused persons?
 5   A.  It has occurred in the past very rarely.  Most of our
 6       work in defence comes through the IPCC(sic), the
 7       Independent Police Complaints Authority, and a couple of
 8       other review commissions but direct from defence,
 9       rarely.
10   Q.  So in other words, you are essentially servicing the
11       public sector?
12   A.  Essentially, yes.
13   Q.  The second thing is I had asked you about note-taking
14       and I am quite conscious that perhaps I may be looking
15       at note-taking practices today.
16           We are looking at events in 1997 (in other words,
17       more than ten years ago).  When you were working for
18       police forces was there a practice then of taking notes?
19   A.  No, there wasn't.
20   Q.  Even today, would you regard your own practices as being
21       the exception rather than the rule for authorities in
22       England?
23   A.  Within England and Wales, from my knowledge, yes, I
24       believe we are the exception.
25   Q.  However, you are also the exception in that you are the
page  57
 1       one fingerprint agency that is accredited to ISO17025?
 2   A.  That's correct, yes.
 3   Q.  It least it is your perception that note-taking was an
 4       essential feature of complying with that standard?
 5   A.  Yes, as an auditor for 17025 I don't think you can meet
 6       the standard unless you are making contemporaneous
 7       notes.
 8   Q.  The final point that I said I would reverse back into is
 9       verification.  We had some discussion about the
10       processes that you currently apply and, in particular,
11       what independence means.
12           I understand from you that you are actually carrying
13       out work within your organisation in at least
14       considering the potential for a greater degree of
15       independence?
16   A.  Yes.  We're looking at a technological solution to this,
17       a computer-based comparison system.  It will actually do
18       two things.  First, in the analysis phase it will not
19       allow the examiner to see the known print until the
20       analysis has been completed and the note-taking will
21       become an audit trail of the actions undertaken on the
22       computer.
23           Secondly, in terms of verification, once you have
24       made a match or you need a verification and you push the
25       button, the computer will decide where that verification
page  58
 1       is taking place and, at least in theory, that
 2       verification could take place in an entirely different
 3       agency or part of the world.  It doesn't have to be
 4       someone within your own department.
 5   Q.  If it is within your own department, would you still
 6       envisage, because you are a small community, that one
 7       examiner would almost inevitably know that somebody else
 8       has been involved before him, even with a computer-based
 9       system?
10   A.  Yes.  If it was restricted within that department, yes,
11       clearly.
12   Q.  The technology you are talking about, is that still very
13       much under development?
14   A.  Yes, there are a number of technical issues to solve so
15       it's still very much in development.  I wouldn't imagine
16       it will be available for a couple of years.
17   Q.  What I want to do before we move on to probability is
18       looking at the conclusion in relation to current
19       fingerprint practice.  In the article, if I go back to
20       the article, it is at page 78 it begins and then I will
21       quickly move from page 78 to page 80, at page 78 you
22       begin with the context and you are looking at
23       fingerprint evidence in court and you're looking at a
24       number of unreported cases in relation to this.  That is
25       the context.
page  59
 1           If I then turn you, please, without diminishing the
 2       conversation that has taken place preceding this, turn
 3       to page 80 you might write in the second full paragraph
 4       on the page -- probably, I should start on page 79 at
 5       the foot.  You are mentioning some American case law.
 6       It is page 79, four lines up from the foot you say:
 7           "Later in Mitchell the Court of Appeals for the
 8       third circuit, the Appeal Court gave, while accepting
 9       fingerprint evidence under Daubert, a fair assessment of
10       the field: on balance the probative benefits outweigh
11       the risks, but the field lacks clear standards and the
12       debate is marred by the ill-defined concept of the
13       criteria for sufficiency."
14   A.  That's correct.
15   Q.  If I understand what you mean by "the debate being
16       marred by an ill-defined concept of the criteria for
17       sufficiency", is that going back to the conversation I
18       had with you earlier about the intuitive judgment about
19       whether there is, on a non-numerical basis, sufficient
20       observed characteristics consistent with a unique
21       identity?
22   A.  Absolutely right, it's back to that issue of personal
23       standards, personal thresholds.
24   Q.  You then go on and say:
25           "The scientific status of identification evidence
page  60
 1       and in particular fingerprint evidence still receives
 2       critical attention from scholars and commentators."
 3           You mention a number.
 4           "Simon Cole in particular published a series of
 5       papers pointing out some critical weaknesses in latent
 6       fingerprint identification", and you refer to his series
 7       of articles.
 8           I then turn to what I wanted to ask you about with
 9       that background:
10           "The present situation is that the UK courts accept
11       the non-numerical standards and for the most part do not
12       challenge the identification."
13           So if I just stop there, that is your personal
14       experience, is that fingerprint evidence remains largely
15       undisputed in the judicial process?
16   A.  That is certainly my experience.
17   Q.  You say:
18           "The reasons for this may be twofold.  First, the
19       general perception among the public (including
20       judiciary) is that fingerprint evidence is irrefutable
21       and safe."
22   A.  Yes, very much so.
23   Q.  Then:
24           "This view is consistently reinforced by the media."
25   A.  Yes, I think you can again take a rather simplistic
page  61
 1       example, if you look at the TV programme CSI,
 2       fingerprint comparison there is carried out by a
 3       computer and hardly plays a part in the process at all.
 4       It's accepted that it's scientifically sound and there
 5       is no variance in the decisions.
 6   Q.  I suppose CSI has actually made your discipline much
 7       more attractive, so I understand.
 8   A.  Apparently so, yes.
 9   Q.  Then you say:
10           "Second, there may be a lack of adequate defence
11       expertise.  The defence, therefore, tend to try to
12       devalue the evidence through claims to legitimate
13       access, attacks on the chain of evidence or occasionally
14       discrediting the examiner."
15           So, in effect, collateral challenges to fingerprint
16       evidence.
17   A.  Yes, I think in the past talking to barristers there was
18       very much a view that the identification could not be
19       attacked, that it would always be essentially correct
20       and, therefore, the only areas to attack were the
21       examiner themselves and the chain of custody.
22   Q.  Then you go on to say:
23           "In fact, there are a number of other lines of
24       questioning that may be appropriate (especially in the
25       aftermath of the Mayfield case): for example, the lack
page  62
 1       of a demonstrable process, issues relating to ongoing
 2       competence of the examiner and lack of contemporaneous
 3       notes and detailed records of the conclusion and its
 4       basis."
 5   A.  Yes and, again, that's very much looking at the American
 6       justice system at the moment, where they are making and
 7       have done for the last 10 to 15 years, numbers of
 8       attacks, for example, specifically around the existence
 9       of notes taken at the time of the comparison have been
10       issues raised in open court for debate.
11   Q.  Again, those may be regarded as process or collateral
12       issues.  On the basis of the work that you have been
13       doing, the thinking you have been doing over the years
14       that feed into this paper, do you believe that in
15       particular with a non-numerical standard that there
16       ought to be scope, in fact, if there were defence
17       expertise for a head-to-head dispute in relation to the
18       conclusion of fingerprint evidence rather than just
19       collateral process issues?
20   A.  I think there's certainly scope for them.  Some of what
21       we discussed before about the issue of events versus
22       actually saying it's a bifurcation/ridge ending I
23       believe does give scope for challenging an
24       identification.
25   Q.  So, in other words, there would be scope, if I use a
page  63
 1       perhaps more technical expression, there would be scope
 2       for dispute in relation to the interpretation of
 3       particular events perhaps -- perhaps -- giving rise to a
 4       dispute as to the conclusion reached?
 5   A.  Yes, I think that's exactly right.
 6   Q.  For the reasons that you have canvassed in that
 7       paragraph, it just simply doesn't occur today?
 8   A.  Correct.
 9   Q.  You then go on to look at the manner in which matters
10       are presented in court.  When you were looking at how it
11       was presented in court are you primarily writing, or
12       with Professor Champod, your own experience of matters
13       in England?
14   A.  Yes, that's correct.
15   Q.  Do you have any personal experience of practice in
16       Scotland?
17   A.  None in Scotland, no.  I do have some knowledge of
18       practices in the United States and some other countries
19       but strangely enough not Scotland.
20   Q.  It is just so that when we are reading this you are
21       tending to think of those jurisdictions and not ours.
22       Nonetheless we will speak of it from its parallels.  You
23       say:
24           "Examiners supply statements in accordance with the
25       prevailing rules.  However, in general, these statements
page  64
 1       do little more than offer the opinion of the examiner
 2       and provide rather limited detail on how the conclusion
 3       has been drawn."
 4           So if I tell you I understand Scottish practice both
 5       in 1997 and today is to present a very, very brief
 6       report consisting of not much more than the conclusion
 7       of the examiner, that would in fact be consistent with
 8       the jurisdictions that you are familiar with?
 9   A.  With my experience of them, yes.
10   Q.  What you then go on to say is:
11           "The practice of charting the mark and print to
12       demonstrate the features in agreement is no longer
13       common ..."
14           That is your experience of other jurisdictions?
15   A.  Yes, it is.  Even in the United States where that was,
16       up until very recently, a very common practice it seems
17       to have fallen into disuse.  Certainly in the United
18       Kingdom I don't know personally of anyone who has
19       charted a mark for a long period of time.
20   Q.  In fact, that, if I tell you, seems to be the experience
21       in Scotland and this week we also had Mr Logan from
22       Northern Ireland and that seems to be his experience as
23       well.
24           You then say though, and this is the conclusion of
25       this that is written by you and Professor Champod:
page  65
 1           "... thus the transparency of the examination
 2       process may be questioned."
 3           Do you have difficulties about the transparency if
 4       there is no note-taking and no charting being produced
 5       in court?
 6   A.  Well, I think this stems from perhaps
 7       Professor Champod's and my view that as forensic
 8       scientists we should be providing all the information to
 9       the court, so showing transparently exactly what we have
10       done, how we have arrived at our decision.  Note-taking
11       is part of that but if we refer to the statements,
12       statements from the FSS, for example, provide perhaps
13       not the most robust but at least some information about
14       how we arrived at our decision rather than just the
15       decision itself.
16   Q.  That leads naturally to the question, we have talked
17       about note-taking and about the degree to which you
18       would note specific characteristics and you said at the
19       end of the comparison stage you would tend to capture
20       images of mark and print with the features of
21       correspondence highlighted in some way.
22   A.  That's absolutely right, yes.
23   Q.  In the reports that FSS produces for the various
24       agencies that it works for, will there be a definitive
25       description given of the characteristics that are
page  66
 1       observed in mark and print that are in sequence and
 2       agreement?
 3   A.  At the present moment in time, no we don't.  We provide,
 4       as I say, this explanation of the process.  We declare
 5       the existence of what we call these feature maps but we
 6       do not provide much more than the conclusion at the end
 7       of the statement.
 8           Again, unfortunately, that is rather driven by
 9       customer requirement rather than by necessarily my view
10       of how we should be producing that.
11   Q.  When you say customer requirement, your customers it
12       would sound are more interested in your conclusions than
13       in the reasoning that's preceded that?
14   A.  I think that's absolutely correct, yes.
15   Q.  Then you say:
16           "There is a common practice within the fingerprint
17       community to document factually what has been done
18       without documenting why a given conclusion has been
19       drawn."
20   A.  Yes that's right.  It tends to be lots of documentation
21       around chain of evidence, for example, and techniques
22       that have been used in detection but less documentation
23       supporting the formulation of the findings.
24   Q.  Again, inhabiting the practical world you do and you
25       have described, you would accept that even your own
page  67
 1       practices may not match up to what one might take out of
 2       some of the features of this article or chapter as the
 3       ideal?
 4   A.  Absolutely.  We are a way from perhaps the standard that
 5       I'm writing about there.
 6   Q.  Just stopping there, before we go on to look now next at
 7       probability, are there key features of even what you
 8       practise that you regard as matters that should be given
 9       more careful consideration, from the consumer
10       perspective?
11   A.  Yes, it's difficult but I think the fundamental process
12       that I'm involved in in providing evidence or
13       intelligence to my customers is based on the findings
14       that I record.  So I would like my customers perhaps to
15       be more interested and more supportive of the need for
16       me to document what I do, absolutely.
17   Q.  We will look at the future, which looks perhaps at the
18       more headline areas and doesn't look at documentation.
19           You look in the first paragraph at developments in
20       technologies to recover fingerprints.  We need not take
21       up time with that.  You are also expecting AFIS
22       technology to continue to improve and then the point
23       that I did want to discuss with you was the final
24       paragraph on page 80.  You say:
25           "... probably the most important development will be
page  68
 1       the design of statistical models to evaluate low
 2       features matches."
 3   A.  (Nodded)
 4   Q.  That takes us into the realm of probabilities?
 5   A.  It does indeed.
 6   Q.  Before we get too far into the realms of probabilities,
 7       first of all, if the dispute in a particular case
 8       relates to what we have just discussed previously,
 9       namely, the interpretation of particular characteristics
10       as a ridge ending or a bifurcation, that is one;
11       secondly, the presence of coincident characteristics, in
12       other words, whether there were 16 points in sequence
13       and agreement or relatively fewer than that; probability
14       analysis is irrelevant to that?
15   A.  Yes.  If the issue is whether or not there is a
16       corresponding configuration for whatever reason then
17       probabilistic analysis is not going to assist in making
18       that decision.
19   Q.  Then looking at probabilistics and probability analysis
20       do I therefore take it you begin with a conventional
21       fingerprint analysis of mark and print to form a
22       conclusion about the presence in both of a number of
23       points that are in agreement?
24   A.  Yes.  The ACE-V process remains identical to as it is
25       today.  It doesn't change that process at all.
page  69
 1       Essentially, what we're doing is once a corresponding
 2       configuration of minutiae is found is providing a
 3       mathematical measure of the weight of that evidence as
 4       an additional piece of information that's used in the
 5       evaluation stage.
 6   Q.  As I have said, perhaps it helps because your article is
 7       speaking of statistical models to evaluate low features
 8       matches.  Are you envisaging that probability analysis
 9       would be used in every case or only in cases where there
10       are relatively few minutiae?
11   A.  I guess within those people in the community talking
12       about this that is the debate, whether you use
13       probabilities across all matches or just low feature
14       matches.
15           What our research and other research has shown is
16       that when you get in excess of 12 minutiae likelihood
17       ratios become very large numbers which appears to
18       indicate that genuine configurations of 12 minutiae plus
19       are quite safe and very specific.  So, therefore, at the
20       moment my view is that these models are aimed at those
21       low features, between 3 and 12.
22   Q.  In what you have said you have used, to some extent,
23       a vague, if you will forgive me for saying, quite safe,
24       reasonably specific, you are not claiming from a
25       scientific point of view using the word "claim"
page  70
 1       appropriately.  The proposition is not there is
 2       100 per cent certainty above that level, it's just that
 3       there is a relative degree of security in the findings?
 4   A.  Yes, the words are chosen carefully.  It's this concept
 5       of difficulty with individualisation as provable, which
 6       it's not so probabilistic models will indicate there's a
 7       very slight chance that it could come from some other
 8       individual but they will not support an
 9       individualisation or identification.
10   Q.  So they will not support a unique identification but
11       they will show -- and you have said experience is from
12       12 points and upwards, there is a low chance of a
13       coincident observation of these features in two
14       different individuals?
15   A.  One way of putting it is to say the likelihood of that
16       mark coming from a different individual is so
17       diminishingly small that I can reasonably dismiss that
18       as a proposition.
19   Q.  Therefore, I take it from that that, as you say, the
20       balance of the debate tends to be whether probabilistics
21       or probability analysis is a useful tool in the criminal
22       justice system when one is encountering lower levels
23       than 12?
24   A.  That's I think where the debate should be had, yes.
25   Q.  Again, based on what we discussed before the break, do I
page  71
 1       take it that part of the thinking that underpins this,
 2       the example I used earlier was eight characteristics,
 3       you might have eight genuinely identical characteristics
 4       in two different sets of matching pairs and your studies
 5       would suggest that the likelihood ratios associated with
 6       those matching pairs can be different.
 7   A.  Correct.
 8   Q.  So the likelihood is not a feature of the number of the
 9       number of characteristics that are common, it's rather
10       peculiar to the precise configuration?
11   A.  That's correct, yes.
12   Q.  If I understand, equally, what you have written in the
13       article, by looking at low frequency of minutiae you are
14       looking contemplating going very low down, down as far
15       as, for example, three minutiae might be discriminating
16       to an acceptable level for the criminal justice system?
17   A.  Yes, from a mathematical statistical point of view we
18       can generate likelihood ratios on three minutiae which
19       appear to be useful evidence or intelligence for the
20       court process.  I think it's worth saying that, to a
21       certain extent, the acceptability of that within the
22       criminal justice system is something that will be
23       decided in the fullness of time.
24   Q.  So, in other words, the whole model is one that remains
25       under examination rather than having been verified and
page  72
 1       accredited?
 2   A.  The current situation is we're approaching the end of
 3       its scientific validation.  The next stage will be to
 4       move to acceptance into judicial systems.  That's not
 5       necessarily an easy process to go through.
 6   Q.  If I can then just look at it with those qualifications
 7       in mind, first of all, it is based on a statistical
 8       model; is that correct?
 9   A.  That's correct, yes.
10   Q.  It is a computer system based on a statistical model.  I
11       understand that the author of the statistical model is
12       Professor Champod?
13   A.  This particular model in the FSS is actually Dr Cedric
14       Neumann who worked with Professor Champod.
15       Professor Champod provided a lot of the early momentum
16       to this project when he was an employee of the FSS.
17   Q.  So if I wanted to ask about the detail of the statistics
18       Professor Champod would be the better person?
19   A.  Absolutely.  He would be the person to ask, yes.
20   Q.  With that qualification I will be relatively superficial
21       in my conversation with you.
22           Is the statistical base something different from the
23       national fingerprint register or catalogue?
24   A.  Yes.  The model works on a database which is essentially
25       of minutiae configurations and generates a statistic
page  73
 1       based on that database.  So it's not attempting to say
 2       how likely a particular configuration is within the
 3       national database, it is providing a pure statistical
 4       measure of the likelihood of that
 5       configuration appearing.
 6   Q.  Do you use a particular restrictive population for your
 7       database?
 8   A.  Yes, the database at the moment is based on
 9       configurations which have been extracted from the
10       fingerprints of Swiss prisoners.
11   Q.  Therefore, do you recognise at least the possibility of
12       limitations in that?
13   A.  It's one of the interesting things.  We would, at the
14       moment, say fingerprints are the same regardless of race
15       and nationality but undoubtedly at some point in the
16       future someone is now going to test that assumption.
17   Q.  So far as, again, the work that we would understand that
18       you are having to undertake, is the statistical
19       reliability of the database something that's having to
20       be considered?
21   A.  Indeed it is and publications will be produced that show
22       the theoretical basis for why that is a suitable
23       database to use.
24   Q.  Again, because of the commercial application we'll not
25       go too far in relation to that.
page  74
 1           So it's based on a data population of fingerprints;
 2       do I understand?
 3   A.  Yes.
 4   Q.  What it is trying to do, is it, to answer the Chairman's
 5       question, it is trying to look across a database of the
 6       prevalence of particular combinations, for example, in
 7       proximity to the core or the delta.  That's the type of
 8       thing it is trying to do?
 9   A.  Absolutely, yes.
10   Q.  It results in something called a likelihood ratio and I
11       understand what that is trying to do is to tell me what
12       the chances that a print, for example, is mine compared
13       with the chance that it might be the print of somebody
14       else completely unrelated to me?
15   A.  Yes.  As with all these things, it's very easy to slip
16       into transposing conditionals and saying things which,
17       strictly speaking, are incorrect but the easiest way to
18       think about it is it's a measure of the weight of
19       evidence, the likelihood that that particular mark came
20       from a particular individual is a very simplistic way of
21       looking at it.
22   Q.  Is the object then in looking in particular at
23       relatively low minutiae, for example three, to enable a
24       Fingerprint Officer ultimately, if this is all tested,
25       to be able to answer the type of question I was posing
page  75
 1       to you earlier, namely, it is not a claim this is
 2       indicative of a unique match but is nonetheless
 3       expressing the chance, in completely lay terms, that
 4       that fingerprint may indeed have been left by the
 5       suspect or an incriminee or someone of that sort?
 6   A.  Yes.  We see it as a tool for even the higher numbers of
 7       minutiae to assist the examiner but, essentially, it
 8       does open up that possibility of taking marks with very
 9       low numbers of minutiae, three minutiae, and providing
10       some evidence, intelligence, some information to a
11       court.  So, yes, one of the -- it's not a driver for the
12       project but one of the benefits of the project could be
13       that there are more marks out there which can be used
14       within the judicial process.
15   Q.  Because it is helpful to express it in that way, what is
16       the driver for the project?
17   A.  Well, the project actually established with
18       Professor Champod before we were a commercial
19       organisation and was, essentially, an academic
20       experiment to look at configurations of minutiae and how
21       specific they were.  It's now developed into an attempt
22       to provide an objective tool that can be used by
23       Fingerprint Examiners.
24           I should add that at the moment we're very -- this
25       is very much first generation.  We're only looking at
page  76
 1       bifurcations and ridge endings and no other feature sets
 2       so models of this type will develop over time to become,
 3       probably, a fundamental tool in the arsenal of
 4       Fingerprint Examiners.
 5   Q.  Do you see that as a tool that would be used in every
 6       case or do you, for the reasons we have discussed, see
 7       it as being essentially one where we are looking at
 8       below 12 characteristics?
 9   A.  I think this particular tool, and probably for the
10       foreseeable future, we're looking at below 12
11       characteristics but as the feature sets are added and
12       the tools mature it's possible they will become used
13       right across fingerprint comparison.
14   Q.  I started with the proposition that the probability
15       analysis won't assist if the dispute is a dispute among
16       practitioners about the observable characteristics, it
17       simply converts what the examiner records as having seen
18       into a number of likelihood.
19   A.  Yes.
20   Q.  So far as that is concerned then, as you envisage
21       operating the system, how would you expect the examiner
22       to be inputting the data?  Would he be looking simply at
23       confirmed characteristics or would he be looking at
24       debatable characteristics as well?
25   A.  I think initially the process would be to look at those
page  77
 1       characteristics in which they have a high confidence,
 2       those confirmed characteristics, to measure a likelihood
 3       ratio around those.
 4           If there were characteristics of less confidence
 5       then to add those in in subsequent runs of the
 6       likelihood ratio is a logical thing to do.  It's
 7       possible that one of those less clear characteristics
 8       will not affect the likelihood ratio of the ones that
 9       have already run or it could turn out to be very
10       significant and that is part of that comparative process
11       to establish what could be used to base the findings.
12   Q.  So, in other words, you could envisage two ranges of
13       likelihood: one on the more certain characteristics; one
14       on the more debatable or including the more debatable
15       characteristics?
16   A.  I think it's quite possible in any examination you would
17       generate several different likelihood ratios and several
18       different sets of minutiae and then the evaluation stage
19       would be to look at those and to base your findings on
20       what you thought that was telling you.
21   Q.  Something that we may hear a little bit more about next
22       week, the danger in all of this, in particular, as you
23       say, because of the intuitive feel that from 12
24       characteristics and above the likelihood of a match is
25       high, is that a probability analysis might give a
page  78
 1       misleading impression of the reliability of the work
 2       because, for example, someone who is observing a high
 3       incidence of points of similarity may inappropriately
 4       discount a difference and the probability analysis will
 5       not reflect that.  Is that fair?
 6   A.  That is fair and I think we are back round to this issue
 7       of transparency and to the judicial process actually
 8       challenging those decisions.  Certainly, I would
 9       envisage that reporting a likelihood ratio
10       probabilistically is going to include probably some
11       charts and certainly far more detail about what has
12       actually been done in order that that may be challenged.
13   Q.  In a sense, the very fact that it's being presented with
14       a degree of mathematical precision means that it is all
15       the more necessary for the assumptions that have been
16       made to be clearly stated and to be capable of being
17       audited?
18   A.  That's correct, yes.  The production system that we
19       would envisage would audit trail every input that's put
20       into a particular mark so there would be clear
21       demonstration of what the examiner has done, how they
22       have placed particular minutiae, how they may have moved
23       the placement of particular minutiae and how that's
24       affected the likelihood ratio.
25   Q.  Because this involves you working, as you say, not the
page  79
 1       driver to begin but what is seen as a collateral benefit
 2       can involve you working with low minutiae fingerprints,
 3       is it envisaged that this will result in scenes of
 4       crimes officer, in fact, detecting and ingathering a
 5       greater volume of marks from crime scenes?
 6   A.  We certainly saw that as a possibility.  There is some
 7       work being undertaken at the moment to try and establish
 8       whether that is the case and try to quantify how many
 9       additional marks there are out there to be captured.  I
10       think logically I know that there are but quite what
11       that impact will be is difficult to assess at this time.
12   Q.  As you say, because as a scientist you work both on the
13       disclosure of the marks and their identification, you
14       are in a position to say from personal experience to
15       what extent additional volumes of mark may, in fact, be
16       disclosable or do you yourself find this is something
17       that just needs further research?
18   A.  I think it might be difficult to base it on my -- from
19       my own experience are there additional marks to recover?
20       Yes, there certainly are.  Maybe 6 to 10 per cent but
21       whether that is a reliable figure to place on crime
22       scene examination and other types of item that are
23       examined I couldn't say.
24   Q.  As you say, research is being undertaken in relation to
25       that?
page  80
 1   A.  Yes.
 2   Q.  Obviously from a police force point of view, let alone a
 3       prosecution point of view, there might be a degree of
 4       trepidation that fingerprint bureaux are going to be
 5       swamped by an increased volume of fingerprints of
 6       doubtful value.  You are careful to say that you
 7       envisage probability as being a tool.  Do you envisage
 8       that it will be one that will be deployed
 9       proportionately by the investigators to decide whether
10       they indeed do want to engage with low value minutiae in
11       particular cases?
12   A.  Absolutely, I think this is part of perhaps something
13       which has not really been undertaken very well by
14       fingerprint bureaux around the world, is that, yes, you
15       have a tool, yes, you have the ability to look at these
16       low feature marks, whether you do, whether you recover
17       them, whether you look at them, is based on the value to
18       that particular investigation.  So I think each
19       department will draw up its policies for how that is
20       used.  It's not illogical for a department to say, "We
21       will not do that because of the volume of our work
22       because of the type of work we do we will not report
23       probabilistically we will always report very, very clear
24       marks in every case".  That, as an administrative
25       decision, is valid.
page  81
 1   Q.  Ultimately, from the point of view of moving beyond the
 2       police and even the prosecution, from the courts'
 3       perspective, if I understand it correctly, at least what
 4       you are aiming to do with all the uncertainties that
 5       surround reliability of observation characteristics you
 6       are at least aiming to give a quantifiable measure for
 7       the output of the observations of the examiner?
 8   A.  Yes.  It's introducing some objectivity.  It's something
 9       external to the examiner and should be giving the same
10       result regardless of which examiner inputs the
11       information so, if you like, it's providing a baseline
12       measure of the weight of evidence in particular cases.
13   Q.  But looking at the totality, therefore, of the
14       discipline of fingerprints you are giving an objective
15       output, however, to subjective inputs; is that correct?
16   A.  To an extent, yes.
17   Q.  Do you see, as a practitioner involved in various
18       committees that you are, the need for standardisation as
19       to the inputs?
20   A.  Yes.
21   Q.  Do you have any thoughts yourself about the appropriate
22       direction for that to proceed or is it too early to say?
23   A.  I think it's too early.  I think it's a piece of work
24       that naturally springs out of what we are doing.  As you
25       say, we have looked already at the input from
page  82
 1       Fingerprint Examiners into the model and we know there's
 2       a variation, as there always will be to a certain
 3       extent, but I think there is quite a bit of work that
 4       can be done to try and reduce that variation in terms
 5       of, for example, if you're marking a bifurcation, a very
 6       clear and defined definition criteria for marking that
 7       is something will be required.
 8   Q.  But do you see in looking at this, in a sense, of the
 9       chicken and the egg, that what you are doing is giving
10       the aura of scientific quantification and, therefore,
11       accuracy to inputs that are, in fact, subjective and
12       inherently variable, that perhaps the appropriate place
13       to start would be properly gauging the inputs before you
14       translate those into a mathematical expression?
15   A.  That is true.  In fact, to do that would be to have some
16       technology that located the minutiae in the first place
17       and, therefore, did it objectively.  Such technology
18       does exist but is very unreliable.  I think if we look
19       forward into the future, will there be systems that
20       replace that Fingerprint Examiner in terms of locating
21       minutiae?  I think yes.
22   Q.  If I am looking at it today, what is technically
23       available, if I understand what you are saying, is that
24       because there is the variability of the inputs the
25       controls for that go back to having proper procedures in
page  83
 1       place, structured around the ACE-V methodology, proper
 2       procedures to have a quality assurance as to the inputs?
 3   A.  Absolutely.  I think the other thing is that the model
 4       that we have is laptop-based so if there was a
 5       disagreement about where a particular minutiae has been
 6       placed by a particular examiner it's quite possible to
 7       move that and demonstrate actually in a court what
 8       effect that would have on the likelihood ratio if it was
 9       placed in a slightly different position.  So it does
10       provide an absolute tool for, not just providing the
11       likelihood ratios and assisting examination, but
12       actually demonstrating the examination as well.
13   Q.  If I just conclude in this way in trying to understand
14       the inputs.  We have obviously heard of the Evett
15       & Williams that studied the observation of comparison of
16       particular individuals, you have obviously done some
17       work of that sort in your own exercise.  Are you aware
18       whether there is a regular enough process within bureaux
19       of studying the extent to which identifications and
20       comparison work are susceptible to variability,
21       dependent on the observations of the individuals?
22   A.  I'm not aware of study in that area, no.
23   Q.  Is that something that merits being studied?
24   A.  I would think so, yes.
25   Q.  Why?
page  84
 1   A.  I think, as a general rule, what has happened over the
 2       years in this sort of identification side of
 3       fingerprints is a lack of study into different areas and
 4       I think any study that looks at the variation between
 5       examiners is going to be useful in terms of informing
 6       how we go forward.
 7   Q.  So you do yourself see it, as a practitioner in the
 8       field, that there has been a lack of experimentation and
 9       study in the past?
10   A.  Undoubtedly, yes.
11   MR MOYNIHAN:  I have no further questions.  Thank you very
12       much.
13   THE CHAIRMAN:  Perhaps, Mr Holmes, I might ask you first.
14   MR HOLMES:  Yes, sir, there are four areas I would like to
15       cover briefly.  The first relates to the usefulness of
16       note-taking; the second relates to dealing with
17       discrepancies; the third relates to the expression of
18       the examiner's opinion in court; and the fourth is a
19       single question relating to confirmation bias.
20   THE CHAIRMAN:  Yes, very good.
21                   Cross-examined by MR HOLMES
22   Q.  Can I ask, first of all, you have spoken a bit about
23       note-taking.  Your own figures record only two disputed
24       identifications in the last 15 years or so; is that
25       correct?
page  85
 1   A.  Yes, I don't think that's a precise figure, that isn't
 2       recorded but that is about right, yes.
 3   Q.  We have also heard evidence from Mr Logan of the PSNI
 4       that in a relatively busy bureau the taking of
 5       comprehensive notes is not practical.
 6           What would your comment be to that?
 7   A.  I think, as I said before, I think, yes, it's a
 8       challenge but we're in an issue of whether what should
 9       be done is dictated by resources or what should be done
10       dictates the resources.
11   Q.  Will there not always be a question of resources where
12       we're dealing with a service that's publicly funded?
13   A.  I think that's absolutely true, yes.
14   Q.  If those resources are to be applied to avoiding a
15       situation which is fairly rare, is that a good use of
16       those resources?
17   A.  My personal opinion is that note-taking is a central
18       part of the process.  If you look at the note-taking
19       that we are doing it is below what I would recommend and
20       part of that is resource issues and other constraints.
21       I think it's a difficult one to give a simple solution
22       to.
23   Q.  Whatever your view of it is, is it practical to use this
24       as a solution?
25   A.  I believe that, practically, more extensive note-taking
page  86
 1       could be made within fingerprint bureaux of which I'm
 2       aware.
 3   Q.  I would also like to speak about dealing with
 4       discrepancies.  I have you noted as saying that you have
 5       been able to explain in most cases where a discrepancy
 6       occurred between the known print and the crime scene
 7       print.
 8           Can you have an identification with an unexplained
 9       discrepancy so long as you have a sufficient volume of
10       ridge detail in sequence and agreement?
11   A.  Yes, I think that's possible.  Again, it's talking
12       hypothetically but if you had a particular mark with a
13       lot of correspondence with the known print and you had
14       one discrepancy for which you could not provide a
15       reasonable understanding -- I'd say it does depend a lot
16       on what that discrepancy is, it's possible.
17   Q.  When you are providing an explanation about a
18       discrepancy, how much detail would you be expected to go
19       into?  You are, after all, only providing a theory of
20       something you can't possibly prove, aren't you?
21   A.  Yes.  The explanation of discrepancy is effectively
22       opinion but I would be basing that on some observation
23       I've made.  The most easy one to understand is if there
24       are two prints overlayed which has created a discrepancy
25       or a double tap, I would be attempting to show why I
page  87
 1       believe that was a double tap.
 2   Q.  You said or you made reference there to your opinion.
 3       You are, when you are giving evidence, only giving your
 4       opinion -- is that correct -- I think you said earlier
 5       on, rather than something based on fundamentally
 6       scientific principle?
 7   A.  I think a fingerprint identification is an opinion, yes.
 8   Q.  Indeed, you have identified a sort of third option
 9       that's given to an examiner when you are giving what you
10       have called the comfortable option of saying that you
11       simply cannot come to a conclusion on the basis of the
12       detail that's before you.  Is that right?
13   A.  I think that was the phrase that was used, yes.
14   Q.  Is there, therefore, room within any fingerprint
15       organisation for examiners to hold different opinions on
16       the same mark?
17   A.  Yes.  I think it's perfectly feasible, bearing in mind
18       the subjective nature of the process, on a particular
19       mark, it's possible for competent examiners to formulate
20       differing opinions.
21   Q.  In relation to the expression of opinion in court, is it
22       ultimately the intention that a Fingerprint Expert will,
23       in future, give evidence as to a likelihood of
24       identification based on something like a percentage?
25   A.  It's too early to say.  I think Professor Champod and
page  88
 1       myself what we're saying is that we believe that
 2       qualified reporting of a match is possible.  So I can
 3       envisage a time when a Fingerprint Examiner will say
 4       there is a strong likelihood or a weak likelihood.
 5       Exactly how that is articulated, whether it uses
 6       numbers, whether it uses verbal scales, whether it's
 7       actually explained graphically I think is too early to
 8       tell.
 9   Q.  Finally one question in relation to confirmation bias.
10       You said earlier on that the decision on whether to do a
11       probability analysis should be based on the value of the
12       mark to the investigation.  Does knowledge of the value
13       of a mark not carry with it the risk of influencing the
14       examiner's conclusions?
15   A.  Yes, that's a possibility.  I think this is again one of
16       those big discussions that is sweeping through a lot of
17       areas of forensic science, not just fingerprints.  If
18       you work totally in the abstract with the mark and the
19       print, then you are going to examine (for example, in a
20       murder scene) every mark that's recovered.  If you have
21       some knowledge or someone within the process has some
22       knowledge to say, "Well, actually you can ignore marks 1
23       to 100, I'm only interested in 102", then you can reduce
24       those work loads and be more effective.  So I think the
25       decision-making about which mark should be examined
page  89
 1       doesn't have to be the Fingerprint Examiner's.  It could
 2       be another party so therefore the Fingerprint Examiner
 3       doesn't know necessarily why that decision is being
 4       made.
 5   Q.  Are you speaking about something like investigators
 6       relaying to Fingerprint Examiners that they would like a
 7       particular fingerprint looked at before the others or
 8       they would like it looked at as a matter of priority
 9       rather than the Fingerprint Examiner themselves
10       knowing the value of that particular mark to the case.
11   A.  I think that resolves that confirmational bias issue,
12       yes.
13   MISS GALBRAITH:  Sir, thank you.  There are four points I
14       would like to raise with this witness, largely arising
15       out of the article that has become available today.
16       Firstly, I do appreciate this is to an extent going over
17       ground already covered both by Counsel for the Inquiry
18       and Mr Holmes, but I would like some clarification as to
19       Mr Chamberlain's position on unexplained differences,
20       particularly with reference to comments in the article.
21   THE CHAIRMAN:  Yes.
22   MISS GALBRAITH:  Secondly, to clarify a comment regarding
23       the training and experience of individual officers;
24       thirdly, to ask about cultural reasons behind errors
25       (again, it is a matter that is raised within the
page  90
 1       article; and, lastly, to cover the comment that has been
 2       made about lack of challenge of fingerprint
 3       identification being perhaps due to a lack of defence
 4       expertise.
 5   THE CHAIRMAN:  Yes.  You may ask about those matters.
 6                 Cross-examined by MISS GALBRAITH
 7   Q.  Mr Chamberlain, you just have been asked some more
 8       questions regarding the issue of unexplained differences
 9       within prints and this is an issue that has been of
10       considerable interest to the Inquiry.  I wonder if I can
11       just ask you for some clarification as to your position
12       on that.
13           Can I perhaps ask you to look at page 69 of the
14       article that we have been looking at.  If we go to the
15       second paragraph down at the end of that paragraph, what
16       is written there is:
17           "In fact as soon as a difference is observed between
18       the mark and the print that cannot be reconciled in the
19       light of tolerances defined during the analysis stage,
20       then an exclusion conclusion will be reached.  Logically
21       the absence of any irreconcilable discrepancy is a
22       prerequisite for identification."
23           My understanding of that is that there is a basic
24       principle of fingerprint identification which is that
25       obviously fingerprints are unique and that would lead on
page  91
 1       to a principle that a conclusion of identity cannot be
 2       made where there is an unexplained difference.
 3           What you seem to be saying today doesn't appear to
 4       be reconciled --
 5   A.  Okay, if I can clarify.  Perhaps I need to clarify this
 6       slightly.  It depends what we mean by "discrepancy".  If
 7       we had a bifurcation appearing in the print that was not
 8       in the mark or in the mark that's not in the print,
 9       clearly that is something that can't be reconciled.  If
10       it is a flow of the ridge or some other artefact of the
11       ridge, pore position as such, these things possibly are
12       less of discrepancy, in a sense.  Though a mechanism for
13       why that discrepancy occurred might not be obvious, if
14       there is a substantial amount of other information
15       available, it's possible that you could determine that
16       they are from the same source even though you haven't
17       got the absolute mechanism for why that discrepancy has
18       occurred.  So there is a level of discrepancy
19       essentially I think is the clarification I'm putting in
20       place.  Certainly if there is a mismatch, a difference,
21       in the actual minutiae themselves, then that would be
22       irreconcilable.
23   Q.  There seems to be something of a sliding scale from the
24       way you are describing it.  There may be a discrepancy
25       that's not a discrepancy?
page  92
 1   A.  Not at all.  I think there are just different levels of
 2       discrepancy.  If I explain again, if there were a
 3       bifurcation appearing in the latent or the print that is
 4       not in the known, then that discrepancy could not be
 5       reconciled.  It is quite clearly a discrepancy.
 6           The type of discrepancies that I was talking about
 7       are more around the flow of the ridge, perhaps the
 8       appearance of a particular ridge.  So it's not a sliding
 9       scale, but there are different discrepancies that can
10       occur.
11   Q.  Perhaps as a lay person you can help me with
12       understanding how is anyone to know what the sliding
13       scale of differences is?  What is an important
14       difference?  What is a reconcilable difference?  Is
15       there any sort of accepted way of approaching the issues
16       of unexplained differences?
17   A.  As I said, the starting principle is to actually be able
18       to articulate why there is a discrepancy there (be it
19       double tap, be it movement) within the mark.  The
20       situation I was being given to talk about was where you
21       have an awful lot of agreement between the mark and the
22       print, lots of features in agreement, and one
23       slight discrepancy.  In that situation, I would expect
24       to articulate to the court that that discrepancy is
25       there and I cannot explain why that discrepancy takes
page  93
 1       place, but my decision is based on the fact that I have
 2       found substantial other agreement.
 3   Q.  I notice there you have used the word a "slight"
 4       discrepancy.  Again, it sounds from the way you're
 5       explaining it like there is, for example, if there was
 6       to be ridges missing or a bifurcation going the wrong
 7       way that wouldn't be something you would consider as a
 8       slight discrepancy?
 9   A.  An unfortunate turn of phrase.  The discrepancy is a
10       discrepancy.  It is whether or not you can explain that
11       discrepancy .
12   Q.  So the explanation is the key aspect.  You gave the
13       example to Mr Holmes earlier of there being a double
14       tap.  Now, if there was a double tap, would you expect
15       to see some sort of evidence of crossing ridges or
16       movement or something that would allow you to reach the
17       conclusion that there may have been double tap?
18   A.  Yes, I would expect to see that.
19   Q.  So, again, that perhaps allows for an explanation of any
20       difference rather than there just being simply nothing
21       to explain it?
22   A.  Yes.
23   Q.  Can I ask you is there any science that you are aware of
24       or research that has been done in relation to an
25       identity being reached but there being any kind of
page  94
 1       unexplained difference?  Have you ever known it to
 2       happen?
 3   A.  Not in my personal experience, no.
 4   Q.  If I can move on from that, please, to ask you about a
 5       comment you made in relation to the training and the
 6       relevance of training and expertise of individual
 7       officers.  If we read on further down that page, in the
 8       next paragraph you say:
 9           "In the absence of dissimilarities, the examiner
10       will weigh the corresponding features in reference to
11       the standards of identification.  In a nutshell, the
12       individualisation will be reached when the examiner
13       observes a level of agreement (across the three levels
14       of legible features) that exceeds the highest level of
15       correspondence he observed through his/her training and
16       experience in comparisons involving non-matching
17       entities."
18           Now, the way that that reads would appear to suggest
19       that the examination is dependent on the level of
20       training and expertise of the individual examiner.  Is
21       that correct?
22   A.  I think, yes, the examination is very firmly based on
23       the experience of the examiner -- training to a certain
24       extent but more the training provides the ability and
25       the mechanics to carry out that comparison.
page  95
 1   Q.  You have spoken a little bit about the subjective nature
 2       of the examination.  Would this mean that a different
 3       conclusion could be reached by somebody with many years'
 4       of experience and training rather than a novice?
 5   A.  Well, I think there probably is a threshold.  I don't
 6       think your ability as an examiner increases just because
 7       of the number of years, it's actually the work that you
 8       undertake.  But could a more experienced examiner
 9       perhaps formulate an identification that a novice could
10       not?  It's possible.
11   Q.  If we were to be looking at any kind of objective
12       standard, are you aware of there being any national or
13       international accepted standard or guideline in terms of
14       training or experience that should be expected of
15       Fingerprint Officers?
16   A.  Within England and Wales, there is a national training
17       scheme for examiners which I'm not familiar with the
18       scheme as it stands at the moment but presumably has a
19       certain level of assessment to be reached.  Across the
20       world there is a huge variation in how Fingerprint
21       Examiners are actually trained.
22   Q.  Would you agree that even within the UK there's a huge
23       variation between bureaux and individual examiners as to
24       levels of training and expertise?
25   A.  There's clearly a difference in the expertise.  In
page  96
 1       theory, the training is all the same because it's
 2       through national training schools but I suspect there is
 3       some variation.
 4   Q.  In terms of the research that you have done and are
 5       aware of, are you aware of there being any published
 6       accounts of error rates within identifications?
 7   A.  I'm not familiar with any per se, I don't think.  I'm
 8       aware that there are some publications that emanate, I
 9       think, from the FBI around error rates but I'm not
10       cognisant with those.
11   Q.  Sorry, I am just reading your answer there.  You are
12       aware of some publications that emanate from the FBI?
13   A.  I believe, yes.
14   Q.  But nothing that's been done within the UK or anything
15       that you are aware of?
16   A.  Not to my knowledge.
17   Q.  Do you feel that the lack of that information or that
18       information would have been valuable in terms of your
19       work on the probability aspect of fingerprint analysis?
20   A.  I'm not sure to my area of work within that which is to
21       take the system and enter it into casework, no.  I'm
22       sure that the researchers are more aware of those papers
23       than I am.
24   Q.  Can I ask you to look further on in the article, please,
25       to page 80.  This is a part you referred to earlier, the
page  97
 1       second complete paragraph there, and what is being
 2       considered here is the reason for a lack of challenge of
 3       fingerprint identification.  From the third sentence,
 4       you say:
 5           "First, the general perception amongst the public
 6       (including judiciary) is that fingerprint evidence is
 7       irrefutable and safe.  This view is consistently
 8       reinforced by the media.  Second, there may be a lack of
 9       adequate defence expertise."
10           Can I just clarify when you say defence expertise, I
11       take it you're not referring to defence Fingerprint
12       Experts but rather the solicitors and defence team that
13       may be involved?
14   A.  I think we were referring to the actual defence experts
15       but actually it's probably both.
16   Q.  In relation to the legal team side of it, if I can put
17       it like that, would you accept that in the entirety of a
18       criminal prosecution there may be other defences that a
19       particular accused may be using so that there isn't a
20       denial of being somewhere or touching something they
21       shouldn't have been which may mean that a challenge to
22       the fingerprint identification may not be required, may
23       not be an appropriate use of resource or time?
24   A.  Absolutely.  I recognise a fingerprint examination in
25       any case is a part of a larger picture and is not
page  98
 1       necessarily the most probative piece of evidence.
 2   Q.  In terms of a lack of expertise, I think, as you fairly
 3       said yourself, it would be any defence Fingerprint
 4       Experts that would have the expertise to advise the
 5       legal team as to any possible challenge?
 6   A.  Yes.
 7   Q.  Just bear with me for a second, Mr Chamberlain.  (Pause)
 8           I have entirely forgotten something I said I would
 9       ask you about.  If we could go back in the article to
10       page 74, this isn't a part that has been put to you
11       previously.  The last paragraph on that page really
12       deals with possible reasons as to why there may be
13       errors in identification.  What you say there is:
14           "Research aimed at identifying the potential bias to
15       which examiners may be susceptible has made significant
16       recent progress."
17           There is reference there to Dror and Charlton:
18           "This empirical research highlights the existence
19       and the need of a full awareness of the impact of
20       contextual information on the final decision-making of
21       Fingerprint Examiners.  The studies by Dror et al are
22       mainly focused on the final decision arising from the
23       comparison.  We believe that a strict adherence to the
24       ACE-V protocol with a distinct analysis and comparison
25       stage is critical to mitigate the risk.  Clearly there
page  99
 1       are dangers here that may be more pronounced given the
 2       close working relationship between the examiners and the
 3       police investigators."
 4           Is that something that you yourself have found to be
 5       of concern where there can be seen to be a close
 6       relationship?
 7   A.  I'm not so sure "concern" is the right word.  I'm not
 8       able to put forward any examples where I think that's
 9       caused an issue but I think, from the personal
10       perspective, it's logical that the two are separated.
11   Q.  I think just in the last sentence you go on to mention
12       the importance of a clear and documented analysis which
13       would minimise the risk of any bias creeping in but you
14       say:
15           "It's not universally seen in the UK and it is
16       unlikely that marks are regularly analysed in this
17       fashion."
18   A.  Yes.  I think, as was mentioned before, the ACE-V
19       process, the different segments tend to be blurred and
20       run together and I think certainly in the UK my
21       experience is that you do not see a clearly documented
22       analysis as a separate stage.
23   Q.  In relation to the point I made previously about the
24       perhaps lack of defence expertise and defence experts,
25       just for your comment if I can put to you what the
page 100
 1       exercise the Inquiry is undertaking is on the basis of
 2       an inquisitorial system and I just wondered from your
 3       position whether there may be use in having a report
 4       provided more for the court's benefit rather than one
 5       for prosecution, one for defence.
 6   A.  I think, again as a personal view, the inquisitorial
 7       system has a lot of merits.
 8   Q.  Can I finally just going back briefly to the unexplained
 9       differences point, you have described your probability
10       procedure, and from my understanding of the way you have
11       explained it, that programme is really based on the
12       ACE-V process properly carried out and that the analysis
13       of the mark remains crucial and the quality of the
14       analysis will be key in the ultimate probability that
15       arises out of that work.
16   A.  Yes.
17   Q.  With that in mind, how would any unexplained difference
18       be treated if the probability tool were to be used?
19       Would that factor into the model at all?
20   A.  In a way it doesn't.  The model measures the
21       correspondences between a mark and a print.  If you were
22       to put in -- in a sense, you can't put in a disagreement
23       because you can only mark it on one or other.  So the
24       probability does not necessarily help that but, again,
25       to reiterate, it comes back to what that discrepancy is.
page 101
 1       If the discrepancy is something like an additional
 2       bifurcation or ridge ending, you probably wouldn't even
 3       get to -- well, you wouldn't get to the probability
 4       model because you already will have excluded it.  If
 5       it's something around a ridge flow, our model will mot
 6       measure a ridge flow.  So, again, it's not within the
 7       scope that measurement.
 8   MISS GALBRAITH:  Thank you very much, Mr Chamberlain.
 9   MISS GRAHAME:  There is one matter I would just like to
10       clarify if I may regarding the independence of an
11       expert.
12   THE CHAIRMAN:  Yes.
13                  Cross-examined by MISS GRAHAME
14   Q.  If you take it from me in Scotland that experts are
15       required to be independent and the purpose of an expert
16       is to assist the court, whether that be a judge sitting
17       alone or a jury, in deciding the factual issues in the
18       case.  The expert is there to give advice on perhaps the
19       scientific matter or their area of expertise rather than
20       being there as advocate for one particular side.
21           It was suggested to you today that perhaps the
22       inquisitorial system made that different.  In your
23       experience in England, when you give evidence in court
24       are you there to assist the judge or the jury or are you
25       there to give evidence on behalf of one party to the
page 102
 1       case?
 2   A.  I'm quite clear as a forensic scientist that my role is
 3       to assist the court.  So I tend to give my evidence with
 4       no bias in either direction.
 5   Q.  Does it make any difference if you are dealing with a
 6       situation which is a hearing in an inquisitorial or an
 7       adversarial system?
 8   A.  No.
 9   Q.  It is possible no doubt in England, in the jurisdiction
10       in which you mainly work, or perhaps in other
11       jurisdictions, for individual barristers or solicitors
12       to challenge an expert, an individual expert, on the
13       basis that perhaps, because he only prepares reports for
14       one side, that he is in some way biased or not
15       independent such as he only ever prepares cases for the
16       defence or for the prosecutor or in the public sector.
17   A.  Yes.
18   Q.  Is it fair to say that, leaving aside that type of
19       challenge against an individual expert on that basis, it
20       doesn't necessarily mean that the comparison of
21       fingerprinting as a science is not some sort of
22       independent scientific discipline?
23   A.  No -- if I understand you correctly.
24   MISS GRAHAME:  Thank you very much.
25   THE CHAIRMAN:  Ms Jones?
page 103
 1   MS JONES:  Two quick points, sir, if I may, one on blind
 2       verification and one on workload.
 3   THE CHAIRMAN:  Yes, very good.
 4                    Cross-examined by MS JONES
 5   Q.  Mr Chamberlain, hopefully I will not keep you very long.
 6           You gave evidence earlier about the workload of FSS
 7       and you said, I think, you have got six fingerprints
 8       experts; is that right?
 9   A.  That's correct, yes.
10   Q.  What I wasn't quite clear on was you said they work on
11       between 12 and 30 cases.  Is that at any one time or a
12       year?
13   A.  That's over the course of a year.
14   Q.  Thank you.
15           The other point I wanted to ask you about was in
16       your statement you mentioned blind verification and I
17       think you say it's impractical at present but you say
18       that you are looking at a technological solution to deal
19       with that.
20           I wondered if you were able to tell us what you
21       meant by that?
22   A.  I did mention this is the idea of computerising the
23       comparison process, part of which would be the
24       verification.  So on the computer system, once I've
25       declared that I have a match, I will push a button for
page 104
 1       verification and that verification will then be sent to
 2       another examiner to undertake, who -- I don't know who
 3       that examiner is and that examiner may not even be
 4       within my organisation.
 5   Q.  Would that be within your organisation, within the six
 6       people who are in the organisation, do you mean?
 7   A.  Well, you would say naturally it would be within the
 8       same organisation but potentially that verification
 9       could be sent to anybody with that system so you could
10       envisage actually having verifications or some
11       verifications being undertaken by different
12       organisations.
13   MS JONES:  Thank you.
14   THE CHAIRMAN:  I am not sure if anybody else's LiveNote has
15       stopped but I am sure we can cope.
16                    Re-examined by MR MOYNIHAN
17   Q.  Mr Chamberlain, I just wanted to return to some of the
18       questions you were asked by Mr Holmes, who was the first
19       one who asked you some questions and it may simply have
20       been that I was imprecise when I asked you some
21       questions.
22           It is back to this question of asking a Fingerprint
23       Officer something in a comfortable formulation.  What
24       Mr Holmes put to you was that I had suggested that I
25       would be asking in court whether a Fingerprint Officer
page 105
 1       had reached a conclusion or not.  Plainly if I, as a
 2       lawyer, since I have a bias in a case, I'm trying to
 3       prove one proposition or another, if I were simply to
 4       ask you have you reached no conclusion it would be very
 5       comfortable because you would be at home with your pipe
 6       and slippers if you had reached no conclusion, I assume,
 7       in un-adversarial situation; is that fair?
 8   A.  That's fair.
 9   Q.  However, what I was, in fact, intending to discuss with
10       you was accepting, as you previously said, that
11       universally the training of Fingerprint Officers is in
12       terms of 100 per cent or nothing I, as a lawyer, might
13       for one side or the other -- this is neutral -- might
14       try to get some evidence out of a Fingerprint Officer
15       where he is less than 100 per cent certain.  That's the
16       scenario.  It's a question of the way I formulate the
17       question.  If I ask you have you reached a conclusion as
18       a traditional Fingerprint Officer I would understand
19       that you would say no?
20   A.  That's right.  I think I understand what you are saying,
21       yes.
22   Q.  However, if I asked a differently formulated question,
23       have you been unable to exclude either the suspect or
24       perhaps more importantly an incriminee, then I might get
25       a different answer, I might get an answer, "No, I've
page 106
 1       been unable to exclude"?
 2   A.  That's right, yes.
 3   Q.  You have had experience of giving that answer, in other
 4       words giving evidence that you have been unable to
 5       exclude?
 6   A.  Yes, I certainly have given evidence.
 7   Q.  Then simply this is, I suppose, a reflection of advocacy
 8       training, if I try to chance my arm and go further and
 9       convert that negative answer, "I have been unable to
10       exclude", into a positive, ie to infer a match and ask
11       you, therefore, the chances that the mark and the print
12       may come from a suspect or an incriminee, at present you
13       would simply have to give an intuitive answer to that,
14       if any?
15   A.  Yes.  It's as the case I described.  If I could give
16       more than just unable to exclude I already would have
17       done in any case and I would have given some opinion as
18       to how likely it came from a particular individual.  So
19       hopefully it wouldn't happen in court.  You would
20       already be aware I could go further than just unable to
21       exclude.
22   Q.  So the extent to which you would be able to give a
23       confident answer would simply be, "I am unable to
24       exclude"?
25   A.  Yes.
page 107
 1   Q.  Then one would need to arrive at it by different
 2       inferential ways about how small the population is that
 3       may be under consideration?
 4   A.  Yes.
 5   Q.  But perhaps more positively, if I wanted a positive
 6       answer, that's the realms where the probability analysis
 7       that remains under consideration really does enter the
 8       field?
 9   A.  Absolutely.  In a sense, it removes that unable to
10       exclude and replaces it with some measure of the
11       likelihood that it came from a particular individual.
12   Q.  So it removes what I would call the negative, unable to
13       exclude, and puts it into some quantified positive
14       likelihood of match?
15   A.  Yes.
16   MR MOYNIHAN:  Thank you very much.
17   THE CHAIRMAN:  Just one or two matters if you could clear up
18       for me.  This Handbook of Forensic Science, your
19       article, roughly what date is it?  Is it recent or --
20   MR MOYNIHAN:  Sir --
21   THE CHAIRMAN:  Did you mention it earlier?
22   MR MOYNIHAN:  No, sir, I apologise.  When I took the
23       photocopy from the book and we then sent the book back
24       to Strathclyde, I forgot to take the editing date.  I
25       think it's relatively recent.  I think it's within the
page 108
 1       last year.
 2   A.  It was published, I think, the early part of this year,
 3       May or such, but it was written actually a couple of
 4       years before.
 5   THE CHAIRMAN:  The reason I was asking was at page 74 you
 6       speak of:
 7           "At the time of writing no uniform competence or
 8       proficiency scheme is employed throughout the UK.  Plans
 9       exist for its introduction."
10           I was wondering has that happened as yet?
11   A.  To my knowledge and I'm a little disconnected from the
12       police, sir, to my knowledge, their plan still exists
13       there is no ...
14   THE CHAIRMAN:  The plans exist.
15           Another matter I wanted to ask you about: we have
16       had an example or maybe examples in the course of the
17       Inquiry, where an image taken from the same negative has
18       appeared different to examiners and that, I'm sure, is
19       something of which you would have had experience that it
20       all depends on the way that an image is developed?
21   A.  Yes.
22   THE CHAIRMAN:  Is it ever possible for the examiner to
23       actually look at the fingerprint on whatever surface?  I
24       mean, I'm assuming it's something that's portable and --
25   A.  Well, yes, it is.  I mean, in fact, because we process
page 109
 1       the items as well as carry out the comparison of the
 2       marks recovered the actual fingerprint on the item is
 3       available to us in most of our cases.  The only time
 4       it's not available is if, as you say, it's a
 5       non-portable item or is attached to a building.
 6           Where you have a fingerprint lift, for example,
 7       essentially, that is the original mark, in some respects
 8       so, yes, it is possible.
 9   THE CHAIRMAN:  It just seemed to me it might be one should
10       go back to the original because anything else is a
11       reproduction of it which may depend on the way it has
12       been reproduced.
13   A.  Within our process, obviously with a fingerprint lift,
14       which are generally imaged to produce a photograph for
15       comparison, I would always go back to the fingerprint
16       lift.  In my own casework I have seen the original so I
17       have the ability to look at the photograph and the
18       original mark to see that they a true representation, so
19       I do have that ability in those cases.
20   THE CHAIRMAN:  You referred at one stage in your evidence to
21       sometimes you and your colleagues would use a piece of
22       plain paper but you do have a form that you can use for
23       note-taking.
24   A.  We do have forms for note-taking, yes.
25   THE CHAIRMAN:  That is something you have devised yourselves
page 110
 1       for the purpose?
 2   A.  Yes.  They are very simple and they are devised for that
 3       process.  For me the main issue is the information is
 4       recorded so, to a certain extent, our procedures allow
 5       variation away from that form if it is a more
 6       appropriate way of documenting the information.
 7   THE CHAIRMAN:  I think I do know the answer to the next
 8       question but you said, I hope I quote you correctly, but
 9       something to the effect that fingerprint evidence is
10       part of the larger picture and not necessarily the most
11       probative piece of evidence, but I take it that doesn't
12       mean if the only evidence was a fingerprint that
13       wouldn't be or may not be sufficient?
14   A.  Oh no, there are cases where fingerprints is the only,
15       physical evidence, in a case then clearly, therefore, it
16       is the most probative.
17   THE CHAIRMAN:  If the Fingerprint Expert is certain then
18       that is the evidence.  It is up to the court and the
19       jury whether it is accepted or not.
20           Thank you very much.  I think those were all the
21       questions I wanted to ask you.  Thank you for your
22       assistance and for travelling up today.
23           That, contrary to my estimate that we had to start
24       at 9.30, now means we have finished this part of the
25       evidence earlier than expected.
page 111
 1           We can release you from having to stay in the
 2       witness box.
 3           The only other question is whether it is an
 4       appropriate time to deal with the question about the
 5       closing statements.
 6   MR MOYNIHAN:  Sir, just before we do, I do apologise, the
 7       estimate for starting at 9.30 this morning was simply
 8       due to the fact that I had to finish Mr Chamberlain
 9       today and I had no means of knowing how long it would
10       take.  So I do apologise to all concerned for that.
11   THE CHAIRMAN:  For some people it means early journeys.
12   MR MOYNIHAN:  Sir, before I forget I have also been asked to
13       relay a message to the public.  It looks as if there is
14       a fair chance we will have Mr Wertheim's evidence by
15       video link on Monday afternoon and we may be, as I
16       understand the message I have received, convening in
17       Edinburgh to have that evidence taken by video link.
18           What we can say is that if it does transpire that we
19       are doing that we will place a message on the website to
20       disclose that we are giving the evidence, the time and
21       the place and if there is an ability for members of the
22       public to attend that will be disclosed as well.
23   THE CHAIRMAN:  However, any evidence given will appear on
24       the website.
25   MR MOYNIHAN:  Yes, sir, any evidence given will ultimately
page 112
 1       appear on the website.  It's just simply there may be,
 2       at relatively short notice, a session of the Inquiry in
 3       a venue as yet to be determined in Edinburgh on Monday
 4       afternoon and we will try to give notice to the public,
 5       if at all possible, of that fact.
 6   THE CHAIRMAN:  I assume that is because we haven't the
 7       facilities here?
 8   MR MOYNIHAN:  That is correct.
 9   THE CHAIRMAN:  I think if it avoids Mr Wertheim coming from
10       the United States yet again, it is reasonable to do it
11       in that way.
12   MR MOYNIHAN:  Sir, in relation to the submissions the piece
13       of information is that Mr MacPherson is not here from
14       Simpson & Marwick.  We understand he might take about
15       half-an-hour, if that assists.
16           There is however one other matter that we're trying
17       to resolve with him which means that the exact duration
18       of his information for a closing statement is
19       unresolved.  Subject to that it won't unduly impact on
20       others, I don't think, and therefore if you wish to have
21       a discussion today about the duration and manner or
22       order for closing statements then that would be
23       possible.
24   THE CHAIRMAN:  I am quite content if it is arranged between
25       the Core Participants as to the order in which they want
page 113
 1       to go.  It is not going to influence me in the least
 2       which order I hear people in, if I may say, so it is
 3       more for your own convenience.
 4           My initial feeling is that it would be reasonable if
 5       Turcan Connell and Digby Brown came towards the end but
 6       that is merely because they have had the heaviest burden
 7       in a way and that they should have longer to prepare
 8       their submissions but if that is agreeable to everyone
 9       then we will do it in that order and, as I say, it's not
10       going to make any difference to me who has the last
11       word, as it sometimes used to be said in trials.  That
12       doesn't arise.
13           So I would be quite happy to leave it to you to
14       decide between yourselves and if you can't decide, then
15       I will decide for you but I hope that won't be
16       necessary.
17           That would conclude today's --
18   MR MOYNIHAN:  That concludes today, sir, and if I take
19       responsibility for what I am next about to say: again,
20       we can have a conversation about timing but Mr Pugh is
21       coming up from London with two of his colleagues.  They
22       will be here as a panel of three to give evidence,
23       primarily Mr Pugh but if there are any technical
24       questions he is bringing colleagues with him.  He is
25       only available on Tuesday and, therefore, with
page 114
 1       considerable apologies to all concerned, I would prefer
 2       to err on the side of starting at 9.30 even if that
 3       turns out to be unnecessary simply because we don't have
 4       the opportunity to overspill with Mr Pugh.
 5   THE CHAIRMAN:  Well, since it's the last week of the public
 6       part of the Inquiry I am sure everyone will tolerate one
 7       more day at 9.30.
 8   MR MOYNIHAN:  Sir, if I can confess again, the same
 9       difficulty may arise with Professor Champod.
10   THE CHAIRMAN:  All right, two more days.
11           Thank you very much.  So we will sit at 9.30 on
12       Tuesday but the Inquiry will officially take place, we
13       believe, on Monday subject to the notification that will
14       be given.
15   MR MOYNIHAN:  May take place.  The arrangements have to be
16       finalised -- may.
17   THE CHAIRMAN:  Thank you.
18   (1.00 pm)
19     (Adjourned to a date and time to be fixed and published)
20
21
22
23
24
25