page   1
 1                                    Wednesday, 7th October 2009
 2   (Morning session)
 3   (10.00 am)
 4                ARIE JACOB  Zeelenberg (affirmed)
 5   THE CHAIRMAN:  If you would just give us your full names,
 6       please, so that we can have that on the record.
 7   THE WITNESS:  Arie Jacob Zeelenberg, Jacob with a C and
 8       Zeelenberg with four Es.
 9                   Presentation by THE WITNESS
10   THE WITNESS:  I will start with a few remarks.  I would like
11       you to have consideration with my language.  I am not a
12       native English speaker.  I don't know all the
13       sensitivity of all the words I may use and I hope you
14       have some consideration with that.
15   THE CHAIRMAN:  I think your English is probably clearer than
16       some of the English we've heard in this Inquiry so I
17       wouldn't worry about that.
18   THE WITNESS:  For the record all the slides are numbered.  I
19       may refer to it, if I not forget it, but my wife's not
20       here so, please, tell me if I forget something.
21           The other thing is after recent events I ask myself,
22       I wondered why should I do this?  Why should I be in
23       here and why should I don't leave the problems of the
24       Scottish to the Scots?  I was communicating somebody in
25       an English-speaking country and I wondered why is
page   2
 1       everybody in the UK profession so silent about this and
 2       I asked why me and I used a citation of Lincoln who
 3       said:
 4           "To sin by silence when they should protest makes
 5       cowards of men."
 6           He wrote me back -- and I can't say it any better --
 7       he wrote me back and said:
 8           "But the truth is worth the pain, and as I read your
 9       e-mail the quote from Lincoln was already forming in my
10       mind.  We as professionals have to take ownership of the
11       knowledge and skills we are privileged and trusted to
12       process.  In the McKie case it is the science we are
13       protecting from the abuse of human error and ego.
14       Without the truth of what fingerprint science can and
15       cannot achieve then we become redundant and robotic.
16       The pain and discomfort this case and Y7 has brought is
17       something that exists on both sides of the forensic
18       argument but the truth is loyal to only one value,
19       sufficient evidence both factual and logical."
20           I want to be loyal to that too.  That is why I am
21       here.
22           My presentation will be pretty long.  The body of it
23       will be the exercises we were asked to do and some other
24       evidence related to it.  This is the overview of topics
25       and I will commit the ultimate sin of any presenter, I
page   3
 1       will read out what is on the slides but I have to for
 2       the record.
 3   THE CHAIRMAN:  That is helpful, thank you.
 4   THE WITNESS:  The overview of topics is I will start with
 5       some principles of fingerprint identification; then I
 6       will address the placement of Y7; then I will go to the
 7       first phase of the exercise, the analysis of Y7; then I
 8       will do the scrutiny of the SCRO charting, effectively
 9       phase 2; then I will address the presentation of the
10       SCRO to ACPOS in Tulliallan; then other evidence
11       regarding Y7; then I will move into the analysis of QI2,
12       phase 1; then the scrutiny of SCRO.  On the request of
13       Mr Moynihan, I will skip the QD2 involvement and then
14       will have some remarks about identification/elimination
15       in general and then I will wrap up.
16           It will take I think about a day but I think it will
17       be faster than doing it interactively.  I will discuss
18       point by point and you will see how it will go.
19           First, some principles of fingerprint
20       identification.  Many has already said about
21       fingerprints and I think you are already pretty well
22       educated, but I will just highlight some things I think
23       are important.
24           A fingerprint is a two-dimensional image of a
25       three-dimensional surface of the skin.  We can print it
page   4
 1       with ink and paper, with digital scanners nowadays and
 2       then we do refer it as a known print or the reference
 3       print if we are comparing.  Then we talk about
 4       fingerprints, marks, if it is a deposit of the residue
 5       of the sweat or fatty surface we pick up somewhere else.
 6       The fingers will only have pores that will generate
 7       sweat but we may pick up fat by touching our skin and
 8       other things.  So the deposit will have certain elements
 9       of that.
10           If it is on a surface deposited like that we call it
11       a latent because you cannot see it yet so often when we
12       talk about marks we also use the word "latent", even
13       when it is already detected and made visible.  Generally
14       here we use the word mark again.
15           Now, we work from the hypothesis that fingerprints
16       are unique and permanent.  I will not go into the deep
17       of that but there is good evidence that fingerprints are
18       unique.  We prove it about every day.  As we speak
19       billions and billions of fingerprints are compared by
20       computers and the opposite is never proven, so we can
21       take that as an empirical fact.  I come back to that
22       when we speak in the domain of identifying marks because
23       that's another issue.
24           Fingerprints are permanent from before birth until
25       long after death and that is just a matter of empirical
page   5
 1       proof proven by Mr Galton and all others.  Only if the
 2       skin is damaged by scarring or also by damaged by a cut
 3       or another wound and we get scarring but also the wound
 4       after it's healed the scar will also become permanent.
 5           What is a fingerprint that I was just talking about?
 6       I will address it in three levels.  It's already a bit
 7       familiar to you.  The First, Second and Third Level
 8       Detail.
 9           If we look at the First Level Detail, when we talk
10       about the overall flow of the ridges or we call them
11       also patterns, on the first on the left-hand side you
12       see the most common or very uncomplicated pattern that
13       is an arch then to the right of that you see the tented
14       arch; in the middle, up in the middle you see the right
15       slanted loop, that is the ridges come from the right,
16       recurve, go back to the right making a loop and you will
17       always see on the left side then a delta.  You will not
18       see it here but on the left side there will be a delta.
19           On the left side -- left from that, right from that
20       you see the left slanted loop, same principle
21       coming from the left going back to the left and making a
22       loop on the right side there is a delta.  Then we have
23       the double loop.  Then we go down we have the central
24       pocket loop, also in America referred to as the central
25       pocket whorl, central pocket loop to the left and the
page   6
 1       full whorls.  These are full patterns, if we roll
 2       fingerprints but we have also parts of fingerprints that
 3       show some kind of patterns like deltas, spirals and so
 4       on.
 5           The difference in flow of pattern excludes the same
 6       origin.  If the pattern is different it is end of ball
 7       game. If I would have a very faint print of a hand and I
 8       may see just three fingers and I can only see that it
 9       may be three whorls and I know somebody to compare with
10       who has only arches I can simply exclude him.  That's
11       the end of it.  So first level excludes the same origin.
12
13           The Second Level Detail: we can regard a fingerprint
14       as a regular system of parallel ridges and minutiae or
15       dactyloscopic points, as we refer to it in the
16       international meeting, are those points, those events
17       where the regularity of this similar system is disturbed
18       and then we refer to ending ridges and bifurcations.
19           If I point here (indicated) we see here a
20       bifurcation where the ridges are joining or literally if
21       we call it a bifurcation we should say that the ridge is
22       splitting but we generally refer to it as an upwards
23       pointing bifurcation.
24           Here is an upward ending ridge.  (Indicated)  We
25       have a definition of a dactyloscopic point, that is a
page   7
 1       notable event that occurs in a regular flow papillairy
 2       ridges that is subject of analysis.  The event is a
 3       natural biological disturbance to the normal parallel
 4       system of the ridges, eg a ridge stops or starts and is
 5       significant.  What we look at is the direction upwards,
 6       the location and the relation to other minutiae.
 7   THE CHAIRMAN:  I wonder could I just disturb you for a
 8       moment.  It might help the stenographer if we could get
 9       the ... so that she can see the slide more clearly.
10   THE WITNESS:  If we take this minutiae the position of it is
11       firmly established by all the connections to other
12       minutiae and what we do is do ridge count.  If we take
13       this minutiae here (indicated), the position of it is
14       established by the connection to other minutiae and we
15       cannot simply neglect it because if we were to say this
16       minutiae does not exist, this upcoming ridge does not
17       exist then we should have to discard the whole ridge.
18       So you cannot simply rearrange the whole thing by taking
19       one minutiae out.  That's impossible.  They all connect
20       to each other.  So uniqueness or specificity as others
21       address it as established by configuration of similar
22       minutiae type and location.
23           Similarity of a configuration is established through
24       triangulation, so connecting the minutiae to each other
25       based upon direction and ridge count between minutiae.
page   8
 1       I will mention ridge count very often these days.
 2           Second level must be the same.  That is a
 3       non-discrepancy rule and I will come back to that in due
 4       course.
 5           Then I move to look at -- sorry, what we see here is
 6       the triangulation and on the right side, on slide 8, you
 7       see an image where the ridges are distorted.  I did that
 8       artificially on the computer but what you see is that
 9       even under these conditions the corners will be
10       different but the ridge count will still be the same.
11       So if you count the ridges from here and there they will
12       still be the same.  So even under adverse conditions,
13       the main aspects like direction, location and relation,
14       will stay the same.
15           There is one exclusion to that, that is if I were to
16       count ridges on the left-hand and by distortion on the
17       right side the core is moved in between the line then I
18       will have different ridge count so I have to compensate
19       for that.  Typically, if we look here (indicated) the
20       ridge count will stay the same and relative direction of
21       the neighbours will also stay the same.
22           Then I will move to the Third Level Detail, also
23       referred to as ridgeology by David Ashbaugh who invented
24       the word, but it was already said yesterday that also I
25       think in 1916 Mr Locard, a Frenchman, did already write
page   9
 1       about the power of the shapes of the pores and the
 2       ridges that could be used in the comparison of
 3       fingerprints but nowadays it's referred to as
 4       ridgeology.  We could define that as the details of the
 5       ridges such as the pores, the edges and the ridge path
 6       of the individual ridge as far as it is reflected in the
 7       print or the mark.
 8           If we take this on the left side, this is a mark, a
 9       fairly beautiful mark, detected by cyano acrylate
10       fuming, then we see a very crisp detail and if we take
11       this individual ridge (indicated), then it has kind of a
12       river bed shape.  We call that a ridge path of the
13       individual ridge and you can follow that, then the
14       shape, the rhythm of it, is almost the same.
15           We can also look at those little details here like
16       this small incipient ridge (indicated) or if we take
17       this shape over here it's almost like an upside down
18       finger, index finger.
19           If we look it is not exactly the same as with the
20       inked print but you see it is on that location and it's
21       kind of similar and I could use the analogy of an
22       island.  If you have a small island in the sea, the
23       shore will never be the same, the sea will take, the sea
24       will give, but we know the overall shape of the island,
25       we know the overall size and we know the exact location
page  10
 1       relative to the shore and other islands so we will
 2       recognise it but still it will never be completely the
 3       same and that's also the same with Third Level Detail.
 4       Third Level Detail, therefore, is always studied in
 5       conjunction with Second Level.
 6           Third Level Detail can be the same and, if so, adds
 7       significantly to the value of the comparison.  We have
 8       to keep in mind, however, that there are many, many
 9       variables.  There may be distortion.  It is an image of
10       a three-dimensional type of skin, so it's always
11       two-dimensional so there may be variations in the
12       pressure.  The detection technique may contribute to
13       what we detect and what we see.  Ageing or drying out or
14       damage will also damage the Third Level Detail.  So it
15       can be the same and it can contribute to the comparison,
16       but it's not always.
17           At first sight these two fingerprints look similar.
18       If you would look through your eye lashes they look a
19       little bit the same.  There are some points of
20       similarity.  There is this in the left side, we can see
21       the overhanging bifurcation downwards and we see
22       something similar here (indicated) and at first sight
23       this configuration of four points may look the same.
24           However, if we start looking further then this
25       point, the red point, (indicated) the upcoming end of
page  11
 1       the ridge on the left side is not on the left image
 2       where the yellow dot is.  This also counts for this
 3       point, the bifurcation downwards in the right image is
 4       not on the left image where the yellow point is.  Then
 5       we have another one that's not here.  The upcoming
 6       bifurcation is not on the left image and then this
 7       downwards (indicated) is not there and the wider we make
 8       the scope, all those bifurcations and minutiae are
 9       missing.  We pick up one on the right that's not on the
10       left and we pick up one on the left that is not on the
11       right and the wider we make the scope the more
12       differences we will tend to see.
13           Now, if we see this minutiae here (indicated) and
14       it's similar in direction and where it is in location
15       and what we do then is we start counting the ridges so
16       we connect the points, the minutiae, and we count the
17       ridges in between.  On the left side ridge count is 8
18       and on the right side the ridge count is 7, so we know
19       now that it is a distinct discrepancy.  So this is just
20       to confirm that these images are not from the same
21       fingerprint.
22           There's another thing that is to our aid, that is
23       what we call tracing.  This is an ambiguous area in the
24       fingerprint.  We may see an upcoming ending ridge -- I'm
25       now looking at slide number 12 -- but it may be stained
page  12
 1       in that area and I don't know really what I'm looking
 2       at.  What we can do then is trace the ridges and then
 3       count the ridges on both sides and then we can count all
 4       the ridges and there's two on the top side and three on
 5       the downside and what we know for sure now that the
 6       ridge that is in the middle is ending upwards somewhere.
 7       We don't know how it ends, we don't know the exact
 8       location, we don't know the type, we don't know whether
 9       it's a bifurcation or an ending ridge but we know for
10       sure that there is something ending up there.  So we can
11       now demonstrate that there must be an ending ridge and
12       we can demonstrate that there is a divergence of the
13       ridges.
14           This is an artificial example of how we compare
15       fingerprints.  On the left side we could say this is a
16       latent and on the right side we could say this is a
17       comparison print.  If we now look at the noise on the
18       left side, on the contour on the edge of the latent, we
19       can see that there is some noise over here.  We could
20       take that for an ending ridge.  If we would then go to
21       the comparison print on the right side, we see that
22       there is a real ending ridge coming in.  What we could
23       do now is say, "Okay, you see it's in the reference
24       print so that is a real ending ridge and I counted" but
25       you were not sure when you were looking at a latent.
page  13
 1           If we take then this similar type of noise over here
 2       (indicated) and we will also take that for an ending
 3       ridge, then we know it's not in the comparison print.
 4       So we have to be fair then.  If in the analysis we mark
 5       these both as incoming ending ridges, then there will be
 6       a penalty, except if I count this one as an incoming
 7       ending ridge then the other one should be real too.  But
 8       we cannot use the comparison or reference print as a
 9       guide to the mark.
10           If we would take this ridge here (indicated) and we
11       would count it as ending to the left because we see in
12       the reference print that it is an ending ridge, that
13       would be not honest because we cannot know for sure that
14       this line is ending.  Take this line over here or this
15       line over here (indicated), we cannot be sure that it is
16       ending and we cannot take the reference print as a guide
17       to the mark.
18           In our system, we have a penalty for that.  If in
19       the analysis phase we mark these as genuine and we find
20       out later that it is not genuine, then we have to regard
21       it as a discrepancy.
22           On the other hand, if we have an analysis and we see
23       something like this, now the image on the right, and
24       then I would see this incoming ridge and I would argue
25       that this is real because the other ridges around it are
page  14
 1       demonstrating, make way for it, then I can argue that it
 2       is a real, real incoming ridge.  Also the other ridge
 3       from the other side is kind of compensating for it and
 4       my argument would be this is a real incoming ridge.  On
 5       this side, although the ridges don't make that much way
 6       as in the lower example, still the head of the ridge is
 7       so thick and the wide space in front of it is telling me
 8       that it is real and it is ending and that it is ending
 9       there.
10           The same might count for this ridge (indicated)
11       where the ridges are narrowing and where the end of the
12       ridge is really prominent and I can really see it.  If
13       there would be Third Level Detail that would be
14       similar-ish, that would confirm my opinion.
15           (Slide 14)  Let's have a look at this comparison.
16       The core maybe looked somewhat distorted.  We have some
17       upcoming bifurcations here.  If we go upwards to the
18       core we see an upcoming bifurcation to the right, then
19       one ridge we go higher, a bifurcation, another
20       bifurcation to the right, we go up again then we have an
21       ending ridge to the left, and then we have 12 minutiae
22       that are in agreement.
23           If we were to look at this location over here
24       (indicated) we would see that -- and I point to this
25       ridge here (indicated) -- that the ridges are a little
page  15
 1       bit wider and the same effect we see here and we could
 2       also argue that next to this point on the left we see
 3       that this ridge is thickening looking the same, the same
 4       kind of ridge detail.
 5           (Slide 15)  I am sorry that I did deceive you.  This
 6       is not identical.  I apologise if I offended you, but I
 7       wanted you to go to the thought process and I use it for
 8       a purpose because now you have for yourself some kind of
 9       conviction that these were the same and now I have to
10       try, to try harder to convince you of the opposite, but
11       I will show you what is the purpose of this.
12           These were the minutiae I was looking at and if we
13       look closer now, then we see that this ridge here
14       (indicated), if we follow it downwards, it is going
15       underneath this ending ridge and this is just a short
16       ridge.  It is not the same and I should not take it for
17       granted that it is on the other side.
18           These two minutiae are wider apart -- these are
19       wider apart than these two on the left and this one
20       these are between the same ridges and this one is
21       between this ridge and this one is lower (indicated).  I
22       ignored already the core.  I said this may be distorted
23       but it is not the same.
24           If we take the scope wider, then we will -- sorry, I
25       will do a ridge count here, if I count the ridges here
page  16
 1       and I ignore it at first instance there is two ridges in
 2       between and here there are three (indicated).
 3           If I take the scope wider and if I have more detail
 4       and I have the wider image of the latent I can do that.
 5       Then we see an upcoming on the yellow point here an
 6       ending ridge to the left that is not over there
 7       (indicated).  There is another one we cannot find and
 8       there is another one, ending ridge, we cannot see.  So
 9       if we are allowed to take the scope wider, then we can
10       control that it is not identical.
11           There was another warning sign if we look upwards in
12       the tip we see that the lines are curving like an arch
13       on the left side and on the right image they are
14       slanting to the left.  So also the First Level Detail,
15       so the flow of the ridges, could have given it away that
16       it's not the same finger.  This is a look-alike.  This
17       is one of the close calls you may see on CLPEX website.
18       These are the warnings for fingerprint experts that we
19       don't take anything for granted.  We have to check the
20       ridge count and we cannot walk around like a beachcomber
21       picking a point here and picking a point there.  We
22       should take it very seriously.
23           We cannot -- and there is another thing that is a
24       paradox, that is the lower the quality of the mark the
25       larger our tolerances tend to become.  The mark is so
page  17
 1       good that I can know it's not identical but if it
 2       becomes worse, it degrades and degrades it may take away
 3       the chance I see the differences and I should not allow
 4       my tolerances to become bigger.
 5           Now, I will read out some of the IEEGFI report.
 6       This is the Interpol European Expert Group on
 7       Fingerprint Identification.  I refer to it in my
 8       documents already.  It's on the public website of
 9       Interpol.  Mr Mackenzie referred to this working group
10       as being a member of it.  They have issued the report
11       that's endorsed by the Interpol European Working Group
12       in 2000 and I read paragraph 810 on page 26 on this
13       subject -- 25 to 26 -- and it is about tolerances and I
14       read it out.  It also refers to the first report that
15       was endorsed in 2000:
16           "With identifications proven to be mistaken it
17       became clear that the involved experts have ignored the
18       differences.  Evaluation of those comparisons often
19       contain a long list of excuses why the print does not
20       look like how it should disguised as a demonstration of
21       the skill and experience of the expert."
22           I always say -- end of citation -- they have a
23       rucksack full of excuses why the mark doesn't look like
24       it should look.  I continue with the citation:
25           "A difference in appearance between compared
page  18
 1       fingerprints or details of them that is contributed to
 2       normal variations with printing can be tolerated.
 3       Tolerances should be applied consistently and honestly.
 4       Experts should be aware of the paradox that one may be
 5       inclined to accept more differences in bad prints under
 6       the umbrella of distortion than one would accept in
 7       better quality prints.  Distortion not only limits the
 8       perception of the similar but also from the dissimilar.
 9       The rule, therefore, is that tolerances should not vary
10       dependent on the quality of the impression.  Simply put,
11       the paradox is the worse the print the larger the
12       tolerances.  Mistaken identification can often be
13       contributed to this effect.  One starts to assume
14       similarity because one is unable to check it."
15           There is a tendency to just assume that we know the
16       ground truth and then explain all the differences away,
17       claiming to be an expert of distortion.  This is a
18       distinct warning sign.
19           (Slide 16)  I just wrap up about fingerprints.  A
20       fingerprint is an image of a regular system of
21       parallel papillairy ridges.  A characteristic is an
22       event that disturbs this regularity.  Characteristics
23       keep their basic properties even under adverse
24       conditions.  The value of a characteristic is defined by
25       its properties, the location, direction and relations,
page  19
 1       and clarity and ridge detail.  Uniqueness or specificity
 2       is established by a coherent configuration of
 3       characteristics and detail.
 4           Identification is the establishment by an expert of
 5       sufficient coinciding, coherent, characteristics in
 6       sequence -- sequence is the key word -- in combination
 7       with ridge detail and absence of any dissimilar
 8       characteristics.  That is the non-discrepancy rule.
 9       Discrepancy stops the process of identification.  We
10       cannot assume or prove the ownership any more.
11           I will read out the relevant paragraph of the IEEGFI
12       document too.  This is paragraph 851 about
13       dissimilarities and dactyloscopic points of difference:
14           "Compared prints will never look exactly the same so
15       there will always be dissimilarities between prints from
16       the same donor.  If there is only one different
17       dactyloscopic point or discrepancy between points
18       identification is excluded.  The first report that was
19       from 2000 reads:
20           "'Identifications require sufficient coinciding
21       information between two prints.  If features are present
22       in one print and absent in the other and there is no
23       rational explanation based on findings and facts a
24       statement of identification should not be given in
25       principle.'"
page  20
 1           We have SWGFAST.  You may have heard about it, the
 2       Scientific Working Group on Friction Ridge Analysis
 3       Study and Technology, that is a body erected by the FBI
 4       in 1995.  It's a kind of regulatory body that tries to
 5       define best practices, procedures and definitions on a
 6       consensus basis.  They are still around and they have
 7       given standards for conclusions about individualisation
 8       and I read out paragraph 114 -- or I read out all the
 9       conditions.  It's short enough:
10           "Conditions that shall be satisfied with an
11       identification is determined by a competent examiner and
12       applied to a common area in both impressions based on
13       quantity and quality of the friction ridge details and
14       absent any discrepancy and a reproducible conclusion."
15           (Slide 17)  SWGFAST also describes or defines that
16       it is qualitative and quantitative analysis.  We follow
17       the procedure or method of ACE-V, as it is called,
18       analysis of the mark, comparison, evaluation and
19       verification and the key words in that are that it has
20       to be reproducible, absent any discrepancy and I will
21       add to it that it must be demonstrable.  You must be
22       able to define or point out what you say you see and
23       there is a sufficiency rule.  You can only identify it
24       is sufficient coinciding information.  Sufficiency
25       generally now is, in the profession of fingerprints, is
page  21
 1       arrived at in two ways: that is the numeric system and
 2       the non-numeric system as it is addressed.  That is in
 3       the non-numeric system or the holistic approach to the
 4       discretion of the individual expert on a case-by-case
 5       basis or with reference to an empirical standard, often
 6       referred to as the numerical system.
 7           (Slide 18)  Now, I won't fool you again, I promise.
 8       This is an example of a real identified mark.  This
 9       looks pretty easy but if we look into the green box, we
10       will see that the ridges have a kind of a funny flow,
11       not very regular.  They make small curves, there are
12       thinning lines, they make way for each other and we see
13       kind of the same in the green box in the same area in
14       the same or similar location we can see the same events.
15       This is very strong proof that we are looking at the
16       same print.  So this is coming from the same donor.
17           This is not too difficult and we could ask ourselves
18       why is the comparison of Y7 and QI2 more difficult?
19       There are two main reasons for it: it is the lesser
20       quality, they are not very good prints, but the
21       main reason is they are not identical.  We are trying to
22       compare apples and pears and it is very cumbersome.
23           (Slide 19)  I will now move -- you will tell me if
24       it is time for coffee, I think?
25   THE CHAIRMAN:  I think we have a little time now.  It is
page  22
 1       usually 11.30, another hour before we break
 2       unfortunately for you.  Any time you want to stop please
 3       do.
 4   THE WITNESS:  We look now at the placement of Y7.  Why is
 5       the placement of Y7 so important?  First of all, the
 6       dissimilarities and the discrepancies observed are
 7       attributed to double placement and/or twisting.  So we
 8       have to find out what's going on with this mark.  The
 9       other thing is if Y7 is one print from one donor placed
10       on one location then Shirley McKie is excluded by all
11       experts on both sides.  So that is the importance to
12       find out how it is placed.  I must add, however, that
13       the upper and lower part in separation can also be
14       excluded as coming from Shirley McKie's fingerprint but
15       it is nevertheless very important to look at the
16       placement of it.
17           (Slide 21)   I refer to the minutes in Tulliallan
18       where the independent facilitator, the chief scientist,
19       Dr Bramley, who was hired by the organiser of the
20       meeting, Dr Bramley stated -- and it was after some
21       conversation -- that from conversations at lunchtime,
22       that was between Mr Rudrud and myself, Alan Dunbar and
23       Robert Mackenzie, conversations at lunchtime it would
24       appear that there is some agreement.  If the scene mark
25       is considered to be a single finger impression it is
page  23
 1       more likely to be that of a right thumb and not that of
 2       Shirley McKie.
 3           Mr Dunbar stated that he accepted if it was one
 4       piece it was not that of Shirley McKie.  The Mackay
 5       Report reads they did state or, indeed, concede that if
 6       it was proved conclusively to be a single mark then it
 7       could not have been made by Shirley McKie and I think
 8       that is also confirmed lately with the evidence.
 9           (Slide 22)  If we look at the mark and look at our
10       first assessment then we can see there is a lot of
11       background noise.  If we look at around the print we see
12       stains and the stains are also coming back in the print.
13       So there's a lot of background noise that doesn't make
14       the mark much better.  There is obviously distortion in
15       the tip.  The definition of the ridges is not too clear.
16       It's different on the bottom than on the top.  You
17       cannot see them crisp and clear.  There is not much
18       contrast so the clarity is not that good and there is a
19       relatively low number of minutiae or Galton points or
20       dactyloscopic points as you can call them also.
21           There is an idea that it is left slanted because we
22       expect, by the spreading of the ridges here, we expect
23       it to be a delta down here (indicated).  On the right
24       side down we expect a delta.  So we can then assume or
25       hypothesise that it is going to be left slanted but we
page  24
 1       don't know for sure.
 2           It could be a whorl.  It could be a loop.  It could
 3       be a thumb or an index finger.  We don't know.  The only
 4       reason why we talk about the left thumb is because it's
 5       identified to a left thumb by the SCRO but if we look at
 6       the first analysis we cannot know.  We may look at a
 7       loop to the left but you may also look at a whorl.  If
 8       you look in the core, you almost can see that the ridges
 9       here are a little bit curving inside and these are also
10       a little bit recurring.  It just could tell you it also
11       could be a small whorl, something like this (indicated).
12           Another apparent thing is those incoming ridges from
13       the left stopping to the right.  The three ones I just
14       made red I will repeat it, just look for yourself, those
15       incoming ridges are not very contrastful but they are
16       definitely there.
17           (Slide 23)  What I did recently was I went to our
18       office and said, "Give me the last ten ten-print cards
19       that came in", and I just took them randomly and I
20       looked at all the thumbs and the forefingers and I took
21       a sample of them.  There you see a left thumb and on the
22       left thumb you all see three incoming ridges from the
23       right, stopping to the left and we see some left
24       slantedness.  We see one upcoming here (indicated) but
25       the tendency is incoming ridges from the right to the
page  25
 1       left.
 2           On the right thumb, however, we see the opposite.
 3       Incoming ridges from the left to the right and we see
 4       none coming in from the right to the left.  This gives
 5       away right slantedness.  You see the ridges go slight
 6       right down.
 7           The right forefinger, same issue, incoming ridges to
 8       the left and there is one here (indicated) incoming from
 9       the right but the tendency is from the left to the
10       right.
11           Left middle finger, same issue; left forefinger,
12       same tendency; left thumb, only ridges coming in from
13       the right to the left.  We see here (indicated) a whorl.
14       That's a left thumb, same issue, incoming ridges and
15       left slantedness.  This already gives away that what we
16       are looking at in Y7 is not a left thumb but either a
17       right thumb or a right forefinger.  This is a clear
18       warning we are not looking at the right finger.
19           (Slide 24)  We are asking ourselves what are we
20       looking at?  Is it a double placement or a single print
21       or single or multiple fingers.  I must say I was misled
22       by this finger in first instance too.  The properties
23       are indeed misleading because we see colour difference
24       in the upper part and colour difference in the lower
25       part.  So you might think it could be two fingers or two
page  26
 1       placements.
 2           There's also a structure difference.  You see the
 3       tram rail effect here (indicated) because those ridges
 4       here tend to become double, it looks like a tram rail.
 5       The space in between is gone and there is an unnatural
 6       flow here (indicated) in the centre.  I point to the
 7       centre now.  It's like the ridges are bended downwards
 8       and there is a very unnatural contour.  It seems to be
 9       cut off here (indicated).  Later I found out it is
10       caused by a curve in the wood but at first sight it can
11       be misleading.  It looks as if it could be two pieces
12       but on closer examination you see that it's not the
13       case.
14           If now look at the properties and the phenomena
15       relative to placing, then I can see that there is
16       obviously high pressure in the tip and that is given
17       away by the thickness of the lines, they are darker, the
18       tram rail effect as I mentioned it, the furrows, the
19       space between the ridges is smudged and smaller and
20       there is obviously slippage in the centre area and there
21       is no slippage on the contour.  Nowhere around the print
22       will you see any slippage; it's just a print.
23           If we then take into account the circumstances and
24       the location then we know that the fingerprint is found
25       in isolation.  There are no other fingers adjacent to
page  27
 1       it.  It is an uncommon location.  It's not something, a
 2       doorway, that everybody touches or a door of a shop or
 3       something as a place where everybody touches on the same
 4       spot.  It's an uncommon location on the hinges of the
 5       door of a house where not many people lived.  There's a
 6       limited amount of users and donors.  So the chance of an
 7       undiscernible double placing is remote.
 8           Also, the bent skin downwards, this is the skin is
 9       bent downwards, can hardly be explained by having placed
10       this lower piece on its own.  There must be a reason why
11       the skin is bent down and that should be the upper
12       portion.
13           (Slide 26)  We have to keep in mind this is a very,
14       very tiny print.  It's smaller than a 5 pence coin.
15       It's 18 mm and it is really very, very small.  So to
16       place two things on top of that without noticing that
17       would be almost impossible.
18           (Slide 27)   If we now look at the structure of a
19       finger we know this is the bony structure and this is
20       the piece that is printed.  You can almost -- you can
21       try for yourself.  If you push on the tip of your finger
22       you will feel that it is soft and if you push on the
23       straight side you will feel that it's more sturdy and
24       there is a transition area.  If you push hard on the
25       table and you will go round you can almost feel it slip,
page  28
 1       you can also physically feel it.
 2           (Slide 28)  So we now turn the finger and I've put
 3       the wall a bit slanted just for demonstration purposes
 4       and the finger would go down then this has to tip over
 5       on this piece of bone (indicated) (slide 30) and this is
 6       the tip over or friction area and something has to
 7       happen to the skin in that area if you turn it around.
 8           (Slide  31)  I can demonstrate if you made a small
 9       movie of a finger pushed on a piece of glass and then
10       turned downwards.
11           (Slide 32)  Again, we have repeated it over and over
12       again, you put the finger, push it really hard but not
13       too hard and you go downwards and then you get these
14       (slide 33) images.  We did not push very, very hard but
15       what you see is the lines are becoming broader and
16       darker and the skin is bent downwards.  In the middle we
17       have the normal flow of the ridges and the normal print
18       of this finger.  You see the ridges are evenly spread,
19       not darker and in the middle the ridges flow very
20       normally.  If we go to the left image, we see that the
21       ridges are pressed down, there is some slippage.  On the
22       most right print you see already the slippage is more.
23       The ridges are more pushed together and what is also
24       very important that you see here (indicated), that the
25       skin on the left image -- I pointed the arrow -- the
page  29
 1       skin has kind of bulged outwards.  The skin has to stay
 2       somewhere with this pressure and then going down.  You
 3       can almost feel it, if you would do it.
 4           (Slide 34)  This is what we are looking at.  This is
 5       the mark.  This is the pressure area.  This is where the
 6       skin is bent downwards.  This is where the slippage is
 7       and where the lines tend to join and this is the area
 8       where the skin is bulged out and where the ridges have
 9       spread.
10           If I can go back -- I have to find my arrow (slide
11       33) -- you see that the lines are small, hook almost
12       here and the lines have spread more like here
13       (indicated).  So there's more space between the ridges,
14       there is an unnatural hook here and if you look here
15       it's the same effective here (slide 34) the line, the
16       ridges, the skin has kind of bulged outwards.
17           If we place the other image next to it and then we
18       take off the corner because of the groove in the wood,
19       this is pretty much what we are looking at.  If you have
20       an explanation that fits all the facts a hand like a
21       glove then I think you find the right explanation.  In
22       my view there can only be one conclusion: the subsequent
23       placing by same finger/single touch.
24           If I might go up there I would like to give you a
25       small demonstration.
page  30
 1           This is an unnatural place for a left thumb.  As a
 2       child you are taught to not touch the door here because
 3       you lose your fingers.  The other thing is if you would
 4       place your thumb on this height, because it's the tip
 5       that is rolled, it's an unnatural position.  If you
 6       would go down you would roll it more or less lower.  For
 7       a forefinger or an index its much more natural, much
 8       more natural than a thumb and the height, it is for an
 9       average person it is too high.  If I would fantasise
10       about how this happened -- and I cannot help, I do this
11       when I'm at night I'm thinking of this case -- if there
12       was somebody there conscious of not touching anything,
13       "I shouldn't touch it", and he is in the doorway losing
14       his balance, "Oops, I touched it".  That's what I was
15       fantasising about.  You don't want to touch it but you
16       incidentally just shortly, typically you would go lower
17       with your finger, it's just very short touch.  I think
18       that's what we're looking at.  Everything is telling me
19       it's not a thumb.  It's definitely not a left thumb,
20       more likely to be an index finger.
21           If we accept, based on this evidence, there's one
22       single placement, one single finger and we look at the
23       incoming ridges it's not a left thumb, then it's
24       definitely not a left thumb, it's definitely not Shirley
25       McKie's left thumb.  That's given away already by the
page  31
 1       placement and by the position and by the overall
 2       phenomena that we see with Y7.
 3           (Slide 35)  I will now move to the analysis of phase
 4       1 for myself of Y7.
 5           (Slide 36)   This is my chart.  I cannot use this
 6       for this purposes because it's an overall chart and the
 7       quality of the image is not good enough so I took a
 8       digital image and I also consulted other images (slide
 9       37) like the image from Pat Wertheim as it is featured
10       on the onin.com website and I also have an image from
11       Terry Kent, the image we received at the Tulliallan
12       meeting.
13           The first thing I did (slide 38) is -- what we are
14       looking at is the red points are locations where there
15       are minutiae.  The yellow points are the points where on
16       the concurrent place or the other place they are
17       missing.  So there's a red one on the right and it's not
18       on the left image it will be yellow and vice-versa.  So
19       if there's a point on the left side that's not on the
20       right side it will be yellow.
21           It's also striking at first sight that there are
22       more points, yellow points, on the left side than yellow
23       points on the right side.  So there are more points
24       missing from the reference print in the latent than
25       there are points from the latent missing in the
page  32
 1       reference print.  I did already tell you that the mark
 2       had a low number of Galton points and this may be, at
 3       first instance, may be one of the reasons why the SCRO
 4       went wrong.  If you go one way and you don't recur and
 5       don't look in the reference print and you also look in
 6       the reference print for things in the mark but you go
 7       one way and this may be a reason, one of the underlying
 8       reasons, why it went wrong.  It's no excuse but we know
 9       if disasters happen, more often there are more clues,
10       more reasons why disasters happen but I think this is
11       one or could be one.
12           What I did was try to position the mark with
13       reference to the reference print as good as possible.
14       So I took the orientation of some points as leading.
15       These are the blue ridges.  These are the points that I
16       will refer to as the stepping stone, the four or five
17       points that are similar-ish, they are not similar, they
18       are similar-ish -- it may not be a proper word but I
19       don't know of a better one.  I used those blue lines,
20       the orientation of those points, to orientate the finger
21       as proper as possible and this is the height
22       (indicated).  I use these points to put them on the same
23       height.
24           (Slide 39)  This is the total comparison.  Again, I
25       took the numbers from the charts and transferred them to
page  33
 1       here (indicated) and I will now work in -- those points
 2       are all points that are discussed.  So the numbers in
 3       this chart represent points and locations that I will
 4       discuss and should not be regarded as points of
 5       similarity.  I want to be very explicit about that
 6       because it is mistaken before.
 7           (Slide 40)  The structure I will follow is that on
 8       the top you will see the points that I discussed and
 9       marked in the chart presented to the Inquiry team and
10       below I will give you the images, the same images, but
11       uncharted.  So you can check for yourself what you see.
12       When I start, on the left side you will see come in
13       which point I'm going to discuss, otherwise you may get
14       lost or I may get lost, and then the same numbers, the
15       reference numbers of the SCRO, are also appearing in the
16       yellow circles.  So point 3 is also point 3 of SCRO,
17       point 1 is point 15 and point 2 is point 1, but I will
18       discuss these later.
19           I am now going to discuss point 3.  The structure is
20       that you will have to look -- here is point 3 in the
21       upper part and you will have to look then in the lower
22       part and you will see an overlay coming.  There is the
23       overlay (indicated).  I use dots so that you can still
24       see the underlying information and check what I am
25       telling you.  What we see is this is the reference to
page  34
 1       point 3 that these are two ridges that are kind of
 2       symmetrically walking together and then joining in each
 3       other.  If you look in the mark, they are walking
 4       together and they are joining each other and then
 5       there's one ridge continuing.  It is a bifurcation
 6       downwards but pointing upwards.
 7           If we then look in the reference print, we see that
 8       the left leg of it is a straight line and the right leg
 9       of it is kind of bowing or curving to the left.  It's
10       not symmetric.  It took me to check it and I took two
11       other inked images and these are the same points and
12       then you see that here there's a real ending ridge in
13       both the other inked images.  You see that upwards.
14       This is an ending ridge and this is an ending ridge
15       (indicated).  So this is accidentally, if we go
16       downwards, really confirming that this is an ending
17       ridge that is kind of pushed to the left.  The left line
18       is straight, the right line is curving and in the mark
19       on the other hand it is a real symmetric bifurcation.
20       So the ridge detail is already not the same.  This
21       should be the first warning.
22           (Slide 41)  Then I move to point 4.  You know where
23       to look now.  Same thing: we see two even legs forming a
24       bifurcation almost symmetric.  You see here it's a
25       little bit interrupted but I think the main thing is
page  35
 1       symmetric -- is there something wrong?
 2   THE CHAIRMAN:  I am not sure what is affecting the
 3       microphone.  We will just check it anyhow.
 4   THE WITNESS:  If there is a disconnection here, then it will
 5       be the left leg that is disconnected, not the right leg.
 6       Then the right leg is continuing.  But I think it's kind
 7       of symmetric.  Again, if we look at the reference print,
 8       same thing: the left leg is straight and the right leg
 9       is kind of or an ending ridge or a bifurcation when it
10       bends to the left line.
11           Then, again, I consult the inked prints and we see
12       again it is a real ending ridge.  It was already given
13       away by good study of it that it is an ending ridge
14       instead of a bifurcation and this is a real bifurcation
15       or, if not, then the left leg is disconnected and not
16       the right leg.
17           (Slide 42)  These are, so to say, differences on the
18       third level.  People may name it differently, but
19       already warning signs.  I will not call them
20       discrepancies.
21           Point number 5 we have to look in this area here
22       (indicated).  If we look now to the reference print, it
23       is a full bifurcation symmetric, two legs of the
24       bifurcation moving upwards and continuing in one
25       straight line.  If we now move to the mark, we see the
page  36
 1       left line is continuing and the right line is almost
 2       upwards ending.  It's not really connecting.  It's
 3       closer to the right line than to the left line.  It's
 4       not really a bifurcation.  Again, on Third Level Detail
 5       not the same and a warning sign but I will now start
 6       counting the ridges between these two and then we see
 7       that the ridge count between point 4 and 5 is maximum
 8       one.  If I would connect it to the left side to make it
 9       a bifurcation then it will be one ridge count but if I
10       connect it here it will just be none but it's maximum
11       one ridge count and then if I then move to the right we
12       have a ridge count between the minutiae of minimum two.
13       This is a distinct discrepancy we cannot ignore that one
14       and even within the features in the centre of the image
15       that look similar-ish this is a distinct discrepancy
16       which you should notice at first sight.
17           (Slide 43)  Point number 6 we are going to look in
18       this area upwards (indicated).  Then we see here that it
19       is a full bifurcation, left leg is straight and this one
20       is connecting to it and in the mark it is an ending
21       ridge.  There is a white spot next to it so the line is
22       really stopping if it is connecting it may be connecting
23       to the right side but I think it is just an ending ridge
24       and here it is a full bifurcation in the reference
25       print.
page  37
 1           If we start counting the ridges between point 4 and
 2       6 then we see a ridge count and we want to make it fit,
 3       then in the print this has to be regarded as an ending
 4       ridge.  It's now a bifurcation but if we want to make it
 5       fit then we have to make it an ending ridge, then ridge
 6       count 2.5 is one in the print but 2 if regard as an
 7       ending ridge with this minimum of one and the ridge
 8       count in the mark if connected to the right is one and
 9       if connected to the left it is zero, so whatever way we
10       do it there's always a discrepancy.
11           If we move this -- disconnect this one it's still
12       one if we do it like this, if we move this to this one,
13       (indicated) what we should do to make a bifurcation, the
14       ridge count is zero.  If we disconnect it, it is one so
15       it is always a difference, minimum of one.  This again
16       is a distinct discrepancy.  It is the second distinct
17       discrepancy within a small area.
18           (Slide 44)  Point number 7 equals point number 9 of
19       the SCRO.  This is what we could call the banana shape.
20       We can already see if you look at the mark with this
21       kind of sharp, an evenly flowing bifurcation and if we
22       go to the reference print we see if you imagine that you
23       drive your car and this is the road then you would have
24       to make a turn.  It goes downwards, almost.  There is an
25       open space here (indicated), nothing like in the mark,
page  38
 1       and it is higher up and it is much more prominent and
 2       different shape than within the mark.  You can check for
 3       yourself.
 4           If we now consult other images, those of Mr Wertheim
 5       and Mr Kent, then we can see that the area is not too
 6       good but if we see here (indicated) this is the relevant
 7       area, then if this bifurcation exists then underneath
 8       here -- I point here with the arrow -- then this line,
 9       there has to be an ending ridge to this side or this
10       line is continuing here and this line is continuing here
11       (indicated), it's just a mall blur in between but what I
12       want to point out to you is if this is a bifurcation,
13       then there has to be an ending ridge to that side to
14       compensate for the bifurcation.  Not that clear here
15       because that image is very, very high in contrast but
16       also with the image of Terry Kent you may see that if --
17       and I point to the upper one here (indicated) -- if this
18       is a bifurcation then there has to be something that has
19       come back and stopped you.  I will come back to that
20       later.
21           (Slide 45)  Now we move to point number 8, equals
22       point number 14 of SCRO.  Now I have traced the whole
23       area of that and we are looking at this area here
24       (indicated).  This is an ending ridge or it looks a
25       little bit like it is bifurcating but I think it is more
page  39
 1       likely to be an ending ridge.  If we look in the same
 2       area in the mark, then we don't see any event like it
 3       and to confirm that, we can trace the ridges and then
 4       count them.  On this side we have five ridges and
 5       there's five ridges over there, there's four ridges here
 6       and there's five ridges there (indicated).  This is
 7       confirmation that under the way one ridge has ended and
 8       it's not the case in the mark.  You see here (indicated)
 9       the ridges are traced, there's nothing happening here
10       and that's confirmed by the ridge count 5, 5, 4, 5.  So
11       there's one ending to the left.  So the point 8 fails in
12       the mark and is discrepancy number 3.
13           (Slide 46)  Now we move to point 9, 10 and 11 and
14       these are not marked by SCRO.  You see here (indicated)
15       the point in the reference print, another point and then
16       I have traced the whole area and there is the incoming
17       line from the left in the mark ending to the right.  I
18       think this is the one referred to as the Rosetta
19       characteristic.  We see it over here (indicated) and in
20       the reference print we see the two other lines ending to
21       the left.  The two ending lines to the left fail in the
22       mark and the ending ridge to the left in the mark fails
23       in the print.  We can check that by counting the ridges.
24       If we look in the mark, we have seven ridges on the left
25       side and six on the right side confirming that there is
page  40
 1       one ending to the right and I think this is not denied
 2       by anybody so far.  On the right side we count six on
 3       the right side and four on the left side so that
 4       confirms that there are two stopping to the left.  So
 5       these are discrepancies number 4, 5 and 6.
 6           (Slide 47)  Points number 13 and 15 are also not
 7       addressed to by SCRO.  We find them in this area here
 8       (indicated) with the cursor we find number 13 and number
 9       15 over there.  I've again traced the ridges but now I
10       have used two colours not to get confused between the
11       ridges.  This is a reference point number 1.  It is
12       number 15 of SCRO.  I use it as a reference point.  We
13       have the upwards ending ridge and if we take reference
14       point number 1 and we go down one and then follow the
15       ridge, then it ends.  The opposite, if we go to the
16       reference print and take point number 1 then we go up
17       one ridge and follow it and it ends a little bit
18       further.  So it's ending further away and on the other
19       side.  This is a distinct discrepancy.
20           Then if we go up four, there is a distinct upwards
21       ending ridge, other ridges giving way for it and really
22       we cannot ignore it and there is nothing alike in the
23       reference print.  So point 13 in relation to point 1 is
24       wrong and point 15 in the mark is missing in the
25       reference print.  This discrepancy is number 7 and 8.
page  41
 1           (Slide 48)  Point number 12 and number 14 on the
 2       outskirt of the reference print, also not addressed by
 3       SCRO.  If we again trace the ridges in the mark and then
 4       look where they should be, taking the reference print as
 5       a guide, I left some reference points in there and you
 6       can see that there is no ridge ending downwards and no
 7       ridge ending upwards.  There is nothing the like over
 8       there.  So both these two characteristics are missing in
 9       the mark.  These are discrepancies 9 and 10.
10           (Slide 49)  If we now move to point 1 and 2, 16 and
11       19, they are all in the same area and the reference
12       points of SCRO are 15, 1 and 16 and missing.  Then we
13       are talking what is referred to as the deviated break.
14       If I move to the reference print right down then we see
15       it is also referred to, I think, as a handshake.  We use
16       the word "changeover" but the official name, I think, is
17       deviated break.
18           If we now look in the mark, you can see the upwards
19       ending ridge and that is that point over here
20       (indicated) that's point 1.  I can see that there is
21       some kind of an event over there (indicated), an upwards
22       ending ridge, the same as in the reference print, and
23       point 2 there is some kind of event in the mark at the
24       same location it could be similar-ish.
25           Then if we look with the handshake at the line
page  42
 1       coming down in the reference print, there is nothing
 2       alike in the mark.  There is no line coming down and
 3       ending downwards.  If we go down from point 1 in the
 4       reference print we will see a bifurcation, an obvious
 5       bifurcation, and we see nothing alike, if we go down in
 6       the mark there is no bifurcation going downwards.  So
 7       there's two distinct discrepancies.  So we've got now 12
 8       discrepancies.
 9           (Slide 50)  Point 17 and 18 are also not addressed
10       by SCRO.  They are here (indicated).  On the right side
11       there's a distinct formation.  This is a very thick
12       line.  It's a kind of a closed island but the presence
13       of it is confirmed by this ridge on the left side making
14       way for it.  So there's really a bifurcation here and
15       the fact that there is a ridge going downwards is also
16       confirmed by the ending ridge going upwards.
17           Now we see nothing alike in the mark.  We can trace
18       the ridges.  We can see this upwards ending ridge (that
19       was point number 2) but all the other events simply
20       fail.  There is nothing alike in the mark that is in the
21       reference print and these are points 17 and 18 and we
22       don't see them in the mark at all.  We now have a total
23       14 discrepancies.
24           (Slide 51)  My conclusion of my analysis is this: we
25       have a minimum of 14 discrepancies on top of many Third
page  43
 1       Level Detail that is not the same.  The top part is not
 2       even evaluated because the top part was missing in the
 3       reference print.  There are numerous differences in
 4       ridge path and Third Level Detail.  There are pattern
 5       differences.  The flow is inconsistent with the left
 6       thumb.  There are some similar-looking points in the
 7       centre, what I would call the stepping stone where you
 8       could go wrong but then nothing is the same.  The
 9       majority of minutiae are missing from print in the mark
10       rather than vice versa.  There are numerous warning
11       signs that could have told you there's something wrong
12       here.  It is a definitive exclusion that should be
13       recognised in early stages of comparison.
14           (Slide 52)  I would like now to go to the scrutiny
15       of the SCRO charting of Y7.  This is the chart from the
16       SCRO as I received it and I refer the marked points
17       (slide 55) to the digital image to be able to put them
18       in my presentation for better quality.  This is the
19       reference print of the SCRO chart and the points number
20       1, 2, 3, 4, 5, 6, 9, 15 and 16 are already discussed in
21       my analysis.  I will not discuss them again.  But the
22       red circles are my numbers.  So I will only discuss the
23       remaining points that are 7, 8, (slide 56) 10, 11, 12,
24       13, 14 and ultimately 17.  This is part of the SCRO
25       chart phase 1.  I have transferred the numbers to the
page  44
 1       digital image.
 2           (Slide 57)  Point number 7 I take two reference
 3       points here.  The reference point is number 6 and then
 4       point number 7 should be an upcoming bifurcation in this
 5       area (indicated).  There is some grey-ish looking thing
 6       here on the edge that is marked by SCRO as a bifurcation
 7       upwards.  I would never take that for genuine because
 8       there's a lot of other noise but, okay, if we accept
 9       that there is a bifurcation over there, then we have to
10       do the ridge count to point number 6 and then we see
11       that the ridge count is off.  We've got a ridge count of
12       one in the reference print to number 6 and you've got a
13       ridge count of two in the mark.  So even if we accept
14       that to be a similarity, then the ridge count is off.
15           This is what we would call a teased out point.  You
16       take the reference print as a guide to the mark and you
17       say, "Okay, I find it over there", and then if you fail
18       to really check whether it is on the same location you
19       make two mistakes I think because the ridge count is
20       simply off.
21           (Slide 58)  Point number 8 is a point relevant to
22       what I would call the banana.  Here it is (indicated).
23       Now we see in the reference print there is a small dot.
24       SCRO is mentioning a ridge ending to the right and they
25       mark it over there (indicated).  I already discussed
page  45
 1       earlier that the presence of this point over here, we
 2       now call it point 9, this point over here (indicated),
 3       has to be confirmed by an ending ridge to the right and
 4       I think it's, therefore, that SCRO calls it an ending
 5       ridge to the right whilst in the reference print it is
 6       just a dot.  The blob is regarded as an ending ridge in
 7       the mark and in the print it really is a dot.  But if we
 8       start counting the ridges between point 9 and the dot,
 9       then in the mark it is zero and in the reference print
10       it is one.  So even if we take it for granted that it is
11       an ending ridge or a blob, then the ridge count is
12       wrong.  So I call this an additional discrepancy.  That
13       is number 16.
14           (Slide 59)  Then we move over to point 10.  If you
15       look on the left side, we can regard this is the edge of
16       the mark, the existence of the mark.  I will go back and
17       let you check, if you look at the mark then we can see
18       that here (indicated) there's a consistent shift in the
19       grey scale, we could say the mark is stopping over here
20       and I have marked it with my dots.  So this is about the
21       edge of where the mark is stopping.
22           This is a reference point we've used before.  Then
23       this is another reference point, that is the bifurcation
24       that is marked by SCRO over there (indicated).  I've
25       copied it as good as possible.  I copied it from the
page  46
 1       chart over there and put it in there and then I measured
 2       in the reference print where point number 10 should be.
 3       The length of this ridge I've transferred to the mark
 4       and there you see that the minutiae they have marked
 5       should be over there somewhere, if we take it the same
 6       length.
 7           It is not marked over there.  It is not marked over
 8       there (indicated) because then it would be underneath
 9       the stain, it is marked not there but conveniently over
10       there (indicated).  So in an area on the edge of the
11       mark too short, wrong direction, wrong length and I
12       would never mark anything like that.  This is a point
13       teased out.  You cannot simply claim that there is some
14       event, that the ridges here are narrowing and are then
15       forming a bifurcation.  You cannot see the system of
16       ridges and you cannot see the event that is taking
17       place.  And even if we take it there, it is outside the
18       contour of the mark.  It should be here (indicated) and
19       even there it is wrong.  It is a point teased out.  It's
20       simply non-existent.  I will not even call it a
21       discrepancy.
22           (Slide 60)  Points 11, 12 and 13.  I have now traced
23       the whole area.  This again is the reference point, the
24       banana shape, and this is what is marked by SCRO as
25       point number 11 and now I made an enlargement.  This
page  47
 1       small ridge is now marked over there (indicated) as two
 2       minutiae.  I cannot find point 11 because if we follow
 3       the ridges over here, there is no ridge ending.  There
 4       is no bifurcation to that side.  I cannot simply see it.
 5       It should be here (indicated) but it marked over there
 6       somewhere.  We cannot simply say that there is an ending
 7       ridge.  We cannot even see the ridges but then this
 8       small ridge ending is marked as two minutiae, two
 9       concurring minutiae.  Well, the mark simply does not
10       have that quality to see this ridge, let alone the
11       detail that are the ridge endings.  It is simply the
12       mark doesn't have the quality.  These are points that
13       are simply teased out.  In the absence of point 11 in
14       the mark I call a discrepancy.  The others are simply
15       non-existent or non-provable.
16           (Slide 61)  If I briefly move to the Tulliallan
17       presentation and I will come back to that later, we see
18       that points 10, 11, 12, 13 and 17 from SCRO I just
19       discussed are not marked in the Tulliallan presentation.
20       I find it very remarkable because if 45 points in
21       agreement are found but the 10, 11, 12, 13 and 17 the
22       SCRO has used for a court production are not used by
23       Mackenzie in the Tulliallan presentation.  So this
24       confirms my opinion they are not there.
25           (Slide 62)  I will wait for my conclusion about the
page  48
 1       analysis of SCRO until I have discussed the other
 2       evidence relevant to Y7.  So I will postpone my
 3       conclusion on that.
 4   THE CHAIRMAN:  Would this be a convenient time then to take
 5       a short break, if we take 20 minutes?  We will start
 6       again about 11.45.
 7   (11.27 am)
 8                         (A short break)
 9   (11.48 am)
10   THE WITNESS:  We finish the principles of fingerprint
11       identification, the placement of Y7, the analysis of Y7,
12       the scrutiny of SCRO.  I will postpone my conclusion
13       upon my analysis of that and now move over to other
14       evidence relative to Y7.  One of that is the
15       presentation of SCRO to ACPOS and it was also presented
16       to us in Tulliallan.
17           This is the relevant booklet that was handed over to
18       us then during that presentation and that's what where I
19       got the images from.  This presentation, I understand,
20       is delivered for the Board of Directors of SCRO, just to
21       convince them from the right position of SCRO and I
22       think it was also delivered for ACPOS and later on it
23       was a submission to the civil case from Shirley McKie v
24       The Executive.
25           I will not discuss all the points, just address some
page  49
 1       major issues.  It would take much too long to go over
 2       all the points but I will just discuss a few issues of
 3       them.
 4           These points already pointed out again in this
 5       presentation in which no less than 45 points of
 6       similarity are discussed.  These two ending ridges to
 7       the left (indicated) are not discussed within the mark
 8       and I find it pretty obvious and remarkable.  (Slide 64)
 9        This is my earlier slide of my analysis of Y7 where I
10       showed they were nonexistent in the mark.
11           (Slide 66)  I will just move to a detail of page 11
12       of the relevant booklet and show this image of marked
13       points of similarity number 7, 8 and 11, in particular.
14       Now what is marked over here is an ending ridge, an
15       ending ridge upwards and also ending ridge here
16       (indicated).  This is the contour of the mark.  Here is
17       where the mark is ending consistently.  You cannot
18       uphold that there is any system of ridges, parallel
19       ridges, here, let go that you can mark events in that
20       system of ridges.  What really is marked here are
21       grooves in the wood.  We see here the wood grain in the
22       image in the photo and we see it here (indicated) and we
23       see here where the wood grain is causing (slide 67) the
24       print to be chopped off because it is lower and we see
25       here the wood grain continuing and what is done here is
page  50
 1       that (slide 66) these grooves are marked as points of
 2       similarity.  I will not use any kind of adjectives but
 3       this is really very, very poor.  I cannot call this good
 4       practice, not bad practice.  This is ... I have no words
 5       for it to do with that.  I will leave it at that.
 6           (Slide 68)  If we look at the booklet of
 7       August 2000, that was given to us in 2000, and we look
 8       at point number 4, then it did read "core area not clear
 9       on the Internet marks" and then kind of the same
10       presentation was a submission in the civil case but then
11       there was a distinct difference on page 37.  Now it was
12       reading:
13           "The core area not clear on the Internet marks and
14       due to brush mark, six ridge characteristics used in
15       court case productions have been effectively erased."
16           I found that a very strange remark, so I kind of
17       checked that and looked at page 35 and 36 in this
18       booklet (slide 69) where we see minutiae are marked and
19       they are marked straight through the brush mark.  (Slide
20       70)  So the statement that these minutiae were
21       effectively erased is apparently untrue.
22           (Slide 71)  Page 37 is also talking about using the
23       Internet images.  Page 38, I think it is already
24       conceded by Mackenzie yesterday that he used the
25       Internet images.  The problem is that from 2000 onwards
page  51
 1       in the public domain the reproach was that everybody
 2       anybody who would use these images was incompetent,
 3       other words are used.  I will not repeat them but this
 4       allegation was, of course, unfounded and false because
 5       SCRO used these images far before 2000 themselves.
 6       (Slide 72)  Not only that they used them to prove that
 7       they were right and others were wrong.
 8           Another issue in this booklet was the presumed
 9       edgeoscopy features in sequence.  It is already shown by
10       Mackenzie also, I think, in his evidence, that he used
11       this to demonstrate that there were ridgeology features,
12       detailed features that were the same, also proving that
13       we are looking at the mark from Shirley McKie.
14           (Slide 73)  Now if we now look at the reality and
15       just take one piece of it and we take that piece
16       (indicated), can we now uphold that we will draw
17       anything like what is below (indicated).  What is drawn
18       here in this area is nothing alike what is drawn below.
19       It's just fantasy.  There's nothing in there what is a
20       little bit the same in there.  This is not the way you
21       should practice ridgeology in any way.  The level of
22       detail simply fails.  You cannot use that.  Even in the
23       reference print it is just blackened.  I will come back
24       to that later.
25           I transferred the relevant area to the mark and the
page  52
 1       reference print.  We take the reference point as number
 2       15 that is the lower ridge of the handshake, number 1 of
 3       myself and 15 -- 1 of SCRO, I think, or vice versa and
 4       this is the gap we have seen in the ridgeology features.
 5       I may go back.  This is the gap we are talking about,
 6       also referred to yesterday and this is the gap and I
 7       have transferred it over there.  That's the gap we are
 8       talking about (indicated).
 9           If we now look in the adjacent area for similar
10       points it's upward ending ridge, this downward ending
11       and this downward ending ridge and this upward ending
12       ridge (indicated), they are not there in the mark.  Now
13       the Second Level Detail is not the same and then yet
14       Third Level Detail is used to prove that this is the
15       same area.  This is not good practice.  I will read out
16       to you what the procedures are for the holistic
17       identification process as it is mentioned by Scotland
18       Yard and this is a submission to our working group from
19       Scotland Yard.  This is about the holistic process in
20       the UK.  What it says is, I will just read some
21       citations:
22           "It is the configuration of these features relative
23       to one another which determines identification."
24           Then about Third Level Detail:
25           "Whether they are in the same relative position to
page  53
 1       each other in each impression as for Second Level
 2       Detail, where there is sufficient quantity and
 3       qualitative of detail present in each impression to
 4       arrive to a conclusion and whether there are any areas
 5       in disagreement."
 6           So it must be in the same area.  That's one thing.
 7       If we move to what the IEEGFI report says about it and I
 8       read out 773:
 9           "Third Level Detail comprises of all that is seen of
10       the basic properties of a ridge.  The size and shape of
11       pores in the actual ridges.  There is a direct link to
12       quality.  Only good prints can show Third Level Detail.
13       The presence of it can be a very powerful contribution
14       to each individual point and the whole of the print but
15       it's very dependent of clarity.  Quality of Third Level
16       Detail varies not only from print to print but also
17       within one print.  Only a small part may show pores so
18       this type of quality is decided upon on a point by point
19       basis", and then the important thing:
20           "The location of Third Level Detail is decided in
21       relation to Second Level Detail."
22           So it cannot stand on its own, so you have to check
23       it in correlation with Second Level Detail.
24           If we now check the location of the area that is
25       used for the Third Level comparison and we use point
page  54
 1       number 15 and 1 as a reference and then we count the
 2       distance, and you see the orientation is a little bit
 3       funny, but if you then count the ridges, this area is
 4       ambiguous because of the two ending ridges.  I take the
 5       lower count and I use the other point over here
 6       (indicated) and I have a ridge count of 8 over here to
 7       this bifurcation and a ridge count of 9 over there,
 8       (indicated).  So the position is really off.
 9           Then if we look at quality, this is the quality of
10       the reference print used by Mackenzie for his exercise
11       in Third Level Detail and this is the image taken from
12       the Internet and this is the image he used and I wonder
13       where the grey scales are gone.  This image is much
14       better right side, taken from the Internet.  It's much
15       better than what he has used.  So what happened here?
16       All the grey scales are gone, it is just not black and
17       white, it is yellow and white but every detail is lost
18       and here there is a lot of grey scales.  I wonder why
19       this is done.
20           (Slide 76)  This is another issue when it struck me
21       when I was looking at this image that you've kind of a
22       horseshoe-shape over here (indicated).  I point to the
23       upper right corner.  There's a horseshoe-shape and here
24       is a kind of tent shape and this is the image from Pat
25       Wertheim from the Internet and you will see here the
page  55
 1       horseshoe and the tent shape.  This is the image from
 2       the Inquiry in the same position.  You don't see the
 3       same shape.  This is the image of Terry Kent.  You don't
 4       see the same shapes and this is from the Inquiry from a
 5       print, scanned from the print.  This is the negative;
 6       this is the print (indicated).
 7           The only image that shows this kind of detail (the
 8       horseshoe and the tent) is the image from the Internet
 9       from Pat Wertheim.  The conclusion you can derive from
10       this is that the SCRO used this latent to deliver
11       further proof of the identification of the mark and with
12       that they authenticated it but they also condone that
13       the images of the site are obviously superior for their
14       own images.  That's a bit strange.
15           Back to the overview of the presentation.  What
16       struck me was that if we look on the left side we have
17       the points 38 to 45 marked in the mark but not in the
18       reference print.  I find that kind of strange, to say
19       the least.  So I was looking where I could find them
20       back (slide 80) and I could find them back on page 31
21       and 32 of the SCRO presentation of Tulliallan.
22           This is the image from page L of the part of it and
23       now it is rotated.  The image is degraded.  So it is
24       rotated a little to make it in conformity with the other
25       part of the mark and now we find these points over here
page  56
 1       (indicated) in the reference print on the right side.
 2           So this shows that this was done on purpose, with
 3       awareness, and I started to check things and then I
 4       looked in the mark and I put the same points also in the
 5       same reference print.  These are the check points.  The
 6       red dot over here (indicated) in the reference print is
 7       placed over here on the ending ridge and this is the
 8       other control point.  Then I looked at the other points
 9       over here (indicated) and did place the point 43, 44 and
10       41 on the same spot and then I found out that point 26
11       is the same as 43, 27 was the same as 44, and point 28
12       was the same as 41.  That's probably why it's not shown
13       in the first image.
14           I made some ridge counts just to check whether I was
15       looking at the right spot so we take the reference point
16       and we count the ridges to 43.  You see here it is
17       mentioned point number 26 and here it is called point 43
18       on the right side, if you take the other reference point
19       and then we see that all those numbers are used double.
20           I put them back into the mark in the overall
21       picture.  These are point 43, 44 and 42 in the circle
22       and these are the other three points in that circle
23       (indicated).  The consequence of what is done here is
24       that this area over here, this piece of the fingerprint,
25       and that's the presumption of what is told is that this
page  57
 1       portion of the fingerprint is placed here and here
 2       (indicated).  This cannot be true.  That is probably why
 3       it is not demonstrated in this image because it would
 4       immediately show that the three points were used double.
 5           (Slide 83)  These are the numbers of SCRO copied
 6       into our presentation and we used for HMIC and later we
 7       used that to discuss what was discussed in Tulliallan.
 8       What I did is, in the booklet it was -- the points were
 9       segmented in certain segments.  So points number 1 to
10       14, points number 15 to 22, 23 to 32, 33 to 37 and 38 to
11       45.  They also had different colours in the markings up
12       of the chart of Tulliallan.  What I did, I took those
13       clusters, those segments, and put them into our image
14       and then cluster A is point 1 to 22 and I made a circle
15       around it; then circle or area B out of points 38 to 45;
16       then area C is point 23 to 28; area D is 29 to 32; and
17       area E is 33 to 37; and area F is not mentioned.
18           Numbers 23 to 28 and 29 to 30 are split by me
19       because Mackenzie used different colours and because the
20       colours -- these clusters are out of sync.  That's why I
21       did it.  But if we now look at it, we can see that if we
22       compare cluster A and cluster E they are out of sync.
23       They should be in sequence also as clusters if it is
24       from the same finger.  It's obvious that cluster A and B
25       are out of sync and B and C are out of sync and you can
page  58
 1       see now that B and C in the comparison print are
 2       overlapping, demonstrating what I just said that points
 3       are used double.
 4           This is just to check the colours that were used and
 5       the segments that were used by Mackenzie.
 6           (Slide 86)  I was trying to find an analogy of what
 7       is going on here and let us just assume or presume that
 8       we have a situation, that we have a satellite photo of a
 9       certain area of the globe and we wonder what we are
10       looking at, which country it is.  Now we have an aid
11       here, somebody drew some lines and it is obvious it is
12       North America but you may have a satellite photo with
13       clouds and you may not be able to see and you may have
14       this situation (slide 87) where on the left side is the
15       questioned country.  You might go to a scientist or an
16       expert for help and you say, "Okay, I'm looking for this
17       country", and the expert comes back to you and says,
18       "I've found it.  I have a sample database and I found
19       these pieces back, the shapes are a little bit off but
20       we call that distortion.  That happens with clouds and
21       the changes of the coast area but I've found it", and if
22       we accept his conclusion then we are now in Denmark,
23       sir.  (Slide 88).  But this is what is done.  Is this
24       ridiculous?  Yes, it is ridiculous.  But this is exactly
25       the proposition that is made by Mackenzie by doing what
page  59
 1       he was doing with these sequences.
 2           Either way, these are out of sequence and it is not
 3       Shirley McKie's print because sequence is the most
 4       important thing, or they are in sequence and the
 5       consequence of this is, of his proposition, that Shirley
 6       McKie has placed her finger five times, no less than
 7       five times, in this small area.
 8           (Slide 90)  It is this area that it is this big
 9       (indicated).  So if we follow this proof then we have to
10       accept that she has placed her finger five times on this
11       small area without noticing any overlap, any distortion,
12       she must have uplifted it and put it back again five
13       times and then the tip of somebody else because it's not
14       her tip.  That was already conceded.  This is too bad.
15       This is not even bad practice; this is malpractice.  If
16       you use this image to convince somebody that you are
17       right and this is someone's fingerprint, somebody could
18       go to gaol for, this is wrong and you should tell what
19       your proposition is.  You should tell somebody, "This is
20       what I'm showing you and the implication of it is that
21       she has put her finger five times in this area".  You
22       don't need to be a rocket scientist to see that this is
23       impossible.  You don't need even to be a fingerprint
24       expert to know that this is impossible.
25           (Slide 91)  I will now have and intermezzo and look
page  60
 1       at other evidence relative to Y7 and then I will move
 2       over to QI2.
 3           I have tried to find some explanations about (slide
 4       92) Y7, you could ask Y7: what, how and why?  Iain
 5       Evett, with his team it was already discussed and
 6       Mr Williams did a survey in 1996 about the consistency
 7       of marking points in marks, identified marks or
 8       unidentified marks, by experts because he was
 9       investigating the value of a numeric standard.  He
10       received a award of a distinguished forensic scientist
11       in 2000 and he is basically, how do you call it, a
12       statistician but also scientist and he is mostly a
13       statistician.
14           He visited several times in the Netherlands.  We had
15       lengthy discussions.  I think he came over three or four
16       times.  We were not in agreement on every point but we
17       had lengthy discussions about our profession with Alan
18       Scrafton and Bill Hannis who was present, I think
19       Mr Williams came along, but in this report he gives
20       something away he wasn't even looking at and he
21       writes -- and it was already said yesterday but for the
22       notes I will repeat it -- and he mentioned teasing the
23       points out:
24           "Probably because of the 16-point standard, a
25       practice has grown in the UK service which the team did
page  61
 1       not find in the other countries visited.
 2           "A fingerprint expert will generally reach an inner
 3       conviction about the correctness of an identification
 4       long before he has found 16 points.  His or her
 5       subsequent activity will centre on establishing that
 6       features which are clearly visible in the print can also
 7       be seen in the poorer quality mark.  The print is used
 8       as a guide for scrutinising the mark.  This is called,
 9       in some quarters, 'teasing the points out'."
10           Continued:
11           "This contrasts sharply with the practice in
12       Holland, for example, where the expert must decide on
13       all the usable features that are present in the mark
14       before comparing it with the print.  Whereas the UK
15       procedure does not necessarily entail risk, it is
16       questionable from a scientific viewpoint: to search for
17       features in an evidential item of unknown origin based
18       on features observed in one of known origin would not be
19       considered best practice in other areas of forensic
20       science.  An identification based on a small number of
21       clearly demonstrable points in the mark would be
22       preferable to one based on a larger number of unclear
23       points which have only been established after reference
24       to the print of the suspect.
25           "There is a widespread view among UK fingerprint
page  62
 1       experts that, because they work to a higher number of
 2       points than most of the rest of the world, their
 3       standards are, in some sense, higher.  The collaborative
 4       study showed this to be a mistaken view.  If standards
 5       are measured by the avoidance of mistaken
 6       identification, then experts in Holland and Germany are
 7       more cautious than their UK colleagues -- even though
 8       they declare a lower number of points for identity."
 9           Here we may have a find of what was going on, a
10       cause for the mistake.  What he essentially said is that
11       if you have an inner conviction of what the identity of
12       source is then you have a conviction-based system.  You
13       say you think it's him and now you start putting the
14       points in using the reference print as a guide and you
15       explain every difference you see.  So you assume that
16       you know the ground truth and then you explain
17       everything away.  Mr Evett has said already it's not
18       necessarily wrong from a scientific point of view.  I
19       think I disagree with him.  I think it's wrong also from
20       a scientific point of view.
21           He wasn't looking for it.  This was long before the
22       McKie case was going on and we have found kind of the
23       same issue with the Mayfield case where the FBI went
24       wrong.  They also used the reference print as a guide
25       and also found Third Level Detail in coincidence but
page  63
 1       were proven to be wrong afterwards.
 2           (Slide 95)  So I'm looking for what went wrong.  If
 3       we look at the statement of Mr Swann, then on 26th
 4       February 1999 he says in his own statement in
 5       submissions to Justice Committee 1 but I think it's
 6       almost the same to this Inquiry:
 7           "I recall one particular chart of the disputed mark,
 8       which was not too clear, and probably a second or third
 9       generation copy.  In my examination of that chart, I
10       identified 16 ridge characteristics, which were in
11       agreement.  I was satisfied that it was a positive
12       identification."
13           Here we go.  The conviction is there before ACE-V
14       because what happened to ACE-V?  He didn't make an
15       analysis; he didn't make a real comparison because he
16       looked at a second or third generation copy of a chart
17       and he bases a conviction on that.
18           Now, normal people have a problem if they come to a
19       conviction to reverse themselves.  Experts are worse.  I
20       am an expert.  I fear that one.  When we have arrived to
21       a certain conclusion, it is very difficult to reverse
22       yourself and that is why we use ACE-V.  You use your
23       analysis first.  You do your analysis first and then you
24       do your comparison and then you come to a conclusion
25       instead of the other way round because you know that you
page  64
 1       can go wrong.  That's a human weakness.  This is a give
 2       away I think what happened to Mr Swann.
 3           He goes on then, 2nd March 1999, and he said in his
 4       second visit of the second study of the print:
 5           "We entered a very small room, containing exhibits.
 6       The door standard was propped up against the wall.  I
 7       examined exhibit 102, being the door standard, a
 8       fingerprint form in the name of Shirley Jane McKie, an
 9       actual size photograph of the mark on exhibit 102 and
10       other exhibits connected with the case and I am able to
11       produce my sketch and notes.  As a result of that
12       inspection, I confirmed that the mark was the left
13       thumbprint of Shirley McKie with at least 16 ridge
14       characteristics in agreement in both detail and
15       position."
16           Again, no analysis, small room, no adequate lighting
17       and he comes to the same conclusion.  He knows now that
18       the identification or at least the mark is contested.
19           (Slide 96)  In his report of 16th March he tells
20       about the encounter of the Tayside Police.  He said it
21       would be in the autumn of 1999.  I think that is a
22       mistake.  It should have been 2000, about one and a half
23       years later, there is a knock on the door and he writes:
24           "However, at the time of making my statement to the
25       Tayside Police, I had still not resolved one issue,
page  65
 1       which concerned certain characteristics at the top of
 2       the crime scene mark.  Irrespective of that fact, the
 3       position remained that there were in excess of 20 ridge
 4       characteristics in sequence and agreement in the area
 5       below, so as to fully satisfy me as to the
 6       identification.  It was after the meeting with the
 7       Tayside Police that I resolved the issue of the
 8       characteristics at the top of the mark and, in doing so,
 9       realised the severity of distortion that had occurred
10       when the mark had been deposited on the door standard.
11       Thus, it was assessed to be a 66-degree anti-clockwise
12       movement of the tip of the thumb which resulted in ridge
13       characteristics seen at 12.00 on the crime scene mark
14       appearing at 2.00 on the rolled impression, not the
15       plain impression, of Shirley McKie's left thumb.  Latent
16       print distortion is a problem that can easily deceive
17       the inexperienced expert or examiner and cause even the
18       most knowledgeable expert to take a second look."
19           Mr Swann had not even noticed that there were
20       discrepancies in the tip.  He only noticed one and a
21       half years later and he explains the difference one and
22       a half year later.  How can you uphold that there are no
23       discrepancies because that's one of the firm conditions
24       before you can call an identification.  He didn't even
25       see them but still his conviction is leading in
page  66
 1       rationalising everything he sees that is in conflict and
 2       he comes to extreme assertions about clockwise rotation
 3       and other things.
 4           (Slide 97)  The same pitfall I think happened to
 5       Mr Martin Leadbetter.  It is his own account to
 6       Justice 1.  He says:
 7           "In closing I wish to state that I have examined the
 8       charted material produced by Peter Swann and I am in
 9       agreement with his findings.  I have also examined
10       similar charted material produced by Mr John Berry BEM
11       FFS and I again completely agree with the conclusions
12       reached by this examiner."
13           Again, no analysis, looking at charted material,
14       coming to an inner conviction and then agree.
15           These factors may contribute to explaining the
16       outset of the mistakes.  I don't give them as an excuse.
17       It's no argument -- I'm not providing excuse for it but
18       maybe an explanation what happened: if people are
19       convinced of something they cannot move away from it and
20       certainly cannot explain the persistent denial and
21       cover-up after that.
22           (Slide 98)  If we just put things in a matrix and
23       look at the early days of the identification, then we
24       look at the first one who looked at it and then we look
25       at significance, the equipment used, the reflection
page  67
 1       time, the controls, the number of controls and the way
 2       the controls were executed, and the alerts that were
 3       there and we look at the first date, then there is an
 4       expert looking in an elimination mode, so looking at
 5       fingerprints, "Oh, I've found five points.  Oh, Shirley
 6       McKie", pick up the phone and the equipment is not too
 7       good yet, you just use your glasses, there's no
 8       reflection time, there's no control, there are no
 9       warnings yet.
10           But then we have the consultation a day later
11       because someone says, "Are you sure", so the
12       significance has risen.  The equipment is still the
13       same, reflection time has gone up, the control has gone
14       up, the alert has gone up.
15           Then six days later, the 17th, now the significance
16       must have been known and the equipment must have been
17       improved, reflection time has gone up, the control has
18       gone up and if we then move down to the core productions
19       then you know what you're doing, you know what the
20       significance is.  Somebody could go to gaol.
21           The warning signs are there, they are complaining
22       about equipment, so I gave them 4 stars only there, the
23       reflection time had been enough, the controls had been
24       enough, so there is no excuse any more.
25           What is the problem if an expert has come to a
page  68
 1       certain conviction, the train starts running and it
 2       doesn't stop.  There are no mechanisms in place to say,
 3       "Hey, stop".  Some people have called this
 4       institutionalised arrogance.  There may be some truth in
 5       that.  "We never make mistakes."  We see this with
 6       medical people.  The Pavlov reaction is, "We make no
 7       mistakes", and this has probably intertwined with some
 8       culture we find within the police.  The police is, from
 9       old days, kind of a military organisation.  They define
10       the enemy, hook in and they never give up and they fight
11       the enemy until the last days.  This is kind of
12       explaining what happened.  Nobody was wise enough to
13       say, "Hey, let's sit back, let's consult somebody else.
14       What's going on."  The train doesn't stop.
15           (Slide 99)  Let's go back to the submission of
16       Mr Swann.  He is talking about the 66-degree rotation as
17       an explanation of the differences in the tip.  The
18       differences in the lower part are not explained but
19       let's just address the 66-degree rotation.  What we see
20       here is, again, that extreme arguments and exercises are
21       used just to rationalise the position you have taken.
22           (Slide 101)  I have put my image in there because it
23       was better and I want to use and transfer the points
24       into it and I could not use the image supplied by Swann
25       in his chart, so I replaced it.  The ratio may be a
page  69
 1       little bit off but, for the exercise I'm doing, it's not
 2       significant.  This is the chart of Swann and we see that
 3       the reference print is placed relatively high; so I
 4       corrected that for the exercise I'm going to do (Slide
 5       102) because I would like to look at the 66-degree
 6       rotation theory.
 7           Now, I would like to say we have looked at Y7 and
 8       there are no disturbances around it.  That is in
 9       conflict with any rotation theory.  Just keep that in
10       mind.
11           (Slide 103)  The written evidence from Peter Swann
12       to Justice 1 is this:
13           "On 2nd March 1999, I attended at the offices of
14       Levy & McRae.  As a result of that inspection, I
15       confirmed that the mark was the left thumbprint of
16       Shirley McKie with at least 16 ridge characteristics in
17       agreement in both detail and position.  The appearance
18       of the mark had all of the signs of having been made by
19       the natural thumb.  The mark's position, that is to say
20       the way it lay on the exhibit, evidenced the natural
21       manner in which the thumb had made contact.  The ridge
22       structure and presence of sweat pore detail all added to
23       the genuineness of the mark.  There was no sign of any
24       disturbance to either the mark or the area around the
25       mark."
page  70
 1           "No sign of disturbance."  Yet we have to go to the
 2       66-degree rotation theory.
 3           (Slide 104)  This theory addresses the unexplained
 4       discrepancies he was made aware of after the visit of
 5       the Tayside Police.  Now, we use the Rosetta
 6       characteristics as a reference point.  I have marked it
 7       in the mark and I have marked it in the reference print.
 8       I have copied the points that Mr Swann is using in his
 9       chart and in the mark.  First, we perform a ridge count
10       from the banana shape point to the Rosetta
11       characteristic.  We see in the mark it's a ridge count
12       of 10 and in the reference print it's a ridge count of
13       12.
14           Now, this causes some difficulty because is it now
15       the Rosetta characteristic placed two lines further or
16       the mark diminished or is it uplifted and then rotated?
17       I don't know.  Does the Rosetta characteristic belong to
18       the cluster of the minutiae below and/or belong to the
19       cluster of minutiae on the tip?  I don't know.  The
20       ridge count is wrong.
21           (Slide 105)  If we now look at the proposition of
22       the rotation, I made an overlay of this area over there
23       and then I moved it to the mark and I used the Rosetta
24       characteristic as the point for positioning it and then
25       I rotate it 66 degrees.  You can see it is obvious that
page  71
 1       the lines are now crossing (slide 106).
 2           Then I take some reference points, the Rosetta
 3       characteristic and the point over here, and we see the
 4       66-degree rotation, because it should be over there, is
 5       too much.  It should be there (indicated).  So I
 6       corrected it and actually what we see is that, if we
 7       have to follow this proposition, that the rotation is
 8       48 degrees.  In fact, it could be 47 or 49 I'm not that
 9       accurate but something (slide 107) around there.  This
10       would be the right orientation.  Still, there would have
11       to be crossing lines in order to follow the theory.  We
12       should see either way sideways slippage -- we don't
13       see -- that or we should have to see crossing lines.
14       Both are not the case.
15           (Slide 108)  I have taken some images from Alice
16       Maceo.  She has a Bachelor Degree in science of biology
17       of the University of Alaska.  She is the head of the
18       Forensic Lab in Las Vegas.  She does a lot of work on
19       distortion of fingerprints.  She gives classes upon it
20       and I have followed one of her classes in the
21       Netherlands.  I cannot say that I can really repeat all
22       that she had done in the class and I cannot repeat her
23       knowledge, but I am able to use some of her images.  She
24       does a lot of work particularly on distortion.
25           We see on the left is one of her images from the
page  72
 1       training class.  On the left side, we have a standard
 2       image of a loop to the left.  In the middle, we see a
 3       twisted fingerprint on the 25 degrees rotation.  On the
 4       right side, we see 45 degrees of rotation.  You can
 5       imagine now that skin can accommodate for about
 6       25 degrees of rotation but if you go further with
 7       45 degrees, you now see slippage on the sides.  So the
 8       skin can no longer accommodate for it and then the skin
 9       starts slipping and then after that it will print.
10           What I would like to do is now look what is
11       happening to the minutiae if we have this kind of
12       distortion.  (Slide 109)  So I corrected the position of
13       the right image, so the distorted 45-degree image, and I
14       am going to look at what happened at the minutiae.  I
15       take these reference minutiae underneath and this
16       cluster of three as a reference on the upper side.  I
17       draw a line between them, transfer the line to the other
18       side and we see although the rotation of the fingerprint
19       is 45 degrees, the minutiae only rotated about
20       6 degrees -- not 45, only 6.  So that's what can happen
21       with distortion, the friction of the skin.
22           So I don't think we can accept the explanation
23       (slide 110) now already from the 66-degree rotation but
24       still I wasn't satisfied.  I want to look at ridge
25       count.  I may say the point that is marked over here
page  73
 1       (indicated) on the left side in the mark as an ending
 2       ridge I wouldn't find it proper.  I cannot say that a
 3       system of ridges there continuing and I can neither say
 4       or be sure that there is an ending ridge.  I think that
 5       the reference print again is used as a guide to the
 6       mark.
 7           Also the points in the reference print here, this
 8       cluster of three, (indicated) I find them suspicious but
 9       just let us take them for granted and start counting the
10       ridges.  Now, the slippage area over here in the mark is
11       ambiguous.  So if I count ridges over there, I cannot be
12       sure whether I am counting properly.  First of all, if
13       we look over here (indicated), there is clearly an
14       incoming ending ridge.  This ridge makes way for it,
15       curves, this one goes straight and there is an incoming
16       ridge very, very obvious and the ridge should be in that
17       area (indicated) relative to this cluster of three.
18       There is nothing alike over there.  So this cluster that
19       is rotated by Swann, there is one distinct discrepancy
20       that is telling you it is not the same cluster.  So with
21       this already the 66-degree theory is gone.
22           Counting the ridges towards the Rosetta
23       characteristic, it would make it 9 to the most lower
24       point of the cluster (slide 111) and to the left side it
25       would make 7 and this side it would make 9.  This was in
page  74
 1       the marked, the ambiguous area.  I will not be too
 2       critical but I look for the cluster itself and these are
 3       the points marked by Swann, not by me.  A ridge count of
 4       three over there, ridge count of three over there.  Not
 5       too critical.  A ridge count of six over there, ridge
 6       count of six over there.  Ridge count of nine over
 7       there, ridge count of eight over there -- one ridge
 8       count wrong.  (Indicated)  Ridge count of six over
 9       there, ridge count of five over there.  Ridge count of
10       eight over there and a ridge count of just six over
11       there.
12           So the whole cluster is not the same.  There's one
13       point missing, the ridge count is off and the relative
14       position to the Rosetta characteristic is off.
15           If this cluster is not in sequence with the mark of
16       Shirley McKie, then this part of the print is, again,
17       not Shirley McKie's.  We did already know because of the
18       incoming ridges but the 66-degree theory is wrong.
19           (Slide 112)  If the top is not from Shirley McKie,
20       then the evidence of here is in conflict with Mackenzie
21       and it excludes Shirley McKie again as the donor.
22           I move to another topic.  Many allegations are
23       launched against those who opposed the Y7 identification
24       and some of those are founded on material used.  I will
25       not go into all the allegations outside the fingerprint
page  75
 1       domain.  I will not address them.  I am not interested
 2       in them.  I will just contain myself to those in the
 3       fingerprint domain and I will move first to the Internet
 4       images, then I will discuss plain impressions and then
 5       the method of presenting just briefly.
 6           (Slide 113)  The Internet images.  People who use
 7       the Internet images were called charlatans, incompetent,
 8       even by senior police officers were criticised.  What's
 9       the content of that?  Internet is a means to exchange
10       digital images, information.  But digital images
11       nowadays are the standard in fingerprinting.
12       Ultimately, fingerprint systems -- AFR as it's called in
13       the country -- are using digital images.  Nowadays,
14       fingerprints are taken with fingerprint scanners.  There
15       are no ink and paper prints any more.  In a lot of
16       countries they don't even use ink and paper any more.
17       We have moved to the paperless era.  The FBI is still
18       accepting fingerprint cards with ink and paper.  They
19       will scan them and then destroy them and only use the
20       digital images.
21           If we look at photography under lab conditions and
22       we have to detect images we need live images with
23       reflection and all these type of things, you cannot use
24       old fashioned cameras when you need a live image so the
25       pictures that are taken are all digital images.
page  76
 1           The charting machine of SCRO is using digital images
 2       and the Inquiry is using digital images and we see all
 3       the advantages of it.  So digital images and equipment
 4       are very useful in the accurate analyses and comparison
 5       of fingerprints and will or are the standard practice
 6       already.  So that is not the issue.  The issue is
 7       integrity of the images and the authentication.  The
 8       integrity means are we looking at an accurate
 9       representation of the mark in situ.  I think that is
10       proven these days that we are looking at the real thing.
11       There may be variations but nobody is now disputing that
12       we are looking at a picture of this mark.
13           The images of Y7 were taken by Pat Wertheim,
14       authenticated by him, accepted in the court,
15       authenticated by myself and SCRO.  They are used by the
16       SCRO.  There is no issue here.  You can use the Internet
17       images.  They are real.
18           (Slide 114)  The discussion about the brush mark,
19       the brush mark has not altered the mark fundamentally,
20       no minutiae disappeared, no minutiae appeared and it is
21       really impossible to change the configuration of
22       minutiae or the sequence of minutiae in a mark by
23       damaging it.  If that were the case, if that would be
24       possible this would be the end of fingerprinting.  We
25       could not change fingerprints that way.
page  77
 1           Now I accept that using the original material in
 2       court for the chain of evidence, of course, is good
 3       practice but for the analysis, comparison and discussion
 4       about donorship original material is not essential, in
 5       particular not in order to exclude a potential donor.
 6       We use digital images, digital equipment all day to
 7       analyse fingerprints, to make markings up, to make
 8       discussions and they are very, very beneficial.
 9           I must say Mr Leadbetter, Mr Swann and Mr Berry used
10       second or third generation material even from the
11       Daily Mail so what is their approach?
12           (Slide 115)  Mr John Berry in particular but others
13       criticise many of us for using plain impressions.  I
14       think Mr Wertheim was called a charlatan for doing so.
15       I will not repeat any of these allegations but I will
16       read out what Mr Berry said in his statement for
17       Justice 1 and many, many followed him:
18           "Also the fingerprint of Shirley McKie was a plain
19       impression.  I have no knowledge of any fingerprint
20       expert ever examining a scene mark using a plain
21       impression.  The system started in 1901 and, since then,
22       rolled impressions showing 100 per cent of the detail
23       have been used.  Plain impressions show only about 50
24       per cent of the detail -- I might say, grudgingly, that
25       they can show as much as 60 per cent or 70 per cent of
page  78
 1       the detail.  Throughout the world, millions and millions
 2       of computer programmes are used and every one is based
 3       on the use of a rolled impression.  I worked for
 4       37 years in bureaux comparing hundreds of thousands of
 5       marks and always used a rolled impression.  Why did
 6       Wertheim and his associates send a plain impression of
 7       Shirley McKie's thumbprint?  That was my big problem.  I
 8       decided at that point to take an interest in the case."
 9           (Slide 116)  Mackenzie used a plain impression,
10       Berry himself used a plain impression for his first
11       chart he produced in 2000.  Berry used a plain
12       impression for his second chart he used for evidence in
13       2003.  Mr Swann is using a plain impression from the
14       Daily Mail and, again, Mr Swann is using a plain
15       impression from the Daily Mail for the fingerprint
16       Inquiry.  Mr Leadbetter is using a plain impression and
17       the Inquiry is using a plain impression.  I will leave
18       the reproach there.
19           Another issue that was made about presenting: (slide
20       117) Rudrud and I made a presentation to HMIC and we
21       made a report and we wanted also laymen to understand
22       what we were doing and we wanted to discuss the mark and
23       we made a grid in order to discuss relevant areas in the
24       print.  Now it's ridiculed by Mackenzie and others
25       because you never use such a thing to compare
page  79
 1       fingerprints with a grid.  Well, we didn't compare the
 2       fingerprint with the grid; we used the grid to discuss
 3       it.  I'll use words like I look at 2.00 or 7.00 or
 4       12.00.  We could do this much more accurate and if I
 5       read out what we wrote in the report to HMIC:
 6           "In exhibit 1 we show a print of the latent.  In
 7       exhibit 2 we have superimposed the grid with numbers and
 8       capital letters in order to nominate the area discussed.
 9       Within every square there is a clockwise nomination with
10       small letters as shown the K2."
11           As an example:
12           "At the right top and side area diagonal from F1 to
13       I6 the contour may be influenced by a clipping that
14       could be caused by a lower surface in the background, eg
15       a groove in the wood."
16           That is what it is used for; nothing else, not to
17       compare the fingerprint and these allegations I will
18       leave them where they are but as a professional I don't
19       take that reproach.
20           Another issue about Y7: Mr Swann was presenting it
21       to the Justice 1 Committee and he is now repeating it to
22       the Fingerprint Inquiry and he is giving this chart and
23       if we enlarge this, he says:
24           "Point 19 on the crime scene mark is indicating a
25       ridge, not a characteristic, which ends to the right at
page  80
 1       point 20.  On the donor print it refers to a small
 2       independent or island ridge ending to the left.
 3       Point 20 also ends to the right on the crime scene mark
 4       and to the left on the donor print.  Points 23 and 24
 5       are wrongly marked and pointing to non-existent
 6       characteristics."
 7           (Slide 119)  Mr Swann and the SCRO and we used the
 8       same basic configuration, starting with the stepping
 9       stone.  Now Mr Swann says that he is missing that we are
10       pointing to non-existent points.  Here they are
11       (indicated).  He says again:
12         "Point 19 in the crime scene mark is indicating a
13       ridge not a characteristic ... point 23 and 24 are
14       wrongly marked ..."
15           In Justice 1 he goes on:
16           "Can Mr Zeelenberg explain why point 19 on the crime
17       scene mark Y7 is indicating a continuing ridge which
18       ends further to the right at where he has marked
19       point 20 and how these can be related to points 19 and
20       20 on Shirley McKie's left thumb impression, which is a
21       totally different area of ridge structure and pointing
22       in the opposite direction.  Can Mr Zeelenberg explain
23       why point 20 on the crime scene mark Y7 is marked as
24       point 23 on the left thumb impression of Shirley McKie?
25       Can Mr Zeelenberg explain why his points 23 and 24 on
page  81
 1       the crime scene mark Y7 which are indicating
 2       non-existent characteristics and simply continuous ridge
 3       structures be related to his points 23 and 24 on the
 4       left thumb impression of Shirley McKie.  As I have
 5       referred at 2.00 point 20 on Y7 is point 23 on her left
 6       thumb.  Can Mr Zeelenberg explain why on both the crime
 7       scene mark Y7 and Shirley McKie's left thumb, his
 8       point 25 on both, is pointing not to any characteristics
 9       but simply to a continuous ridge?  What is the
10       significance of a continuous ridge in an identification
11       process."
12           Mr Swann:
13           "The answer is quite simple: the marks are not
14       identical.  The numbers used were meant to discuss these
15       points and locations not to identify the print.  This
16       was clearly stated in the report."
17           I make the citation on page 13, paragraph 10.2,
18       heading:
19           "The comparison in order to enable a discussion we
20       have numbered dactyloscopic points and other possible
21       characteristics.  To make the discussion more
22       transparent and the comparison with the SCRO easier, we
23       have maintained their numbers and only added to them.
24       The charted lines and numbers are aids to the discussion
25       of certain areas and events, and are not meant to
page  82
 1       indicate similar points intrinsically."
 2           If he would only have read our report and not only
 3       looked at the charts he would have known.  But the
 4       importance of this (slide 122) is what we have to look
 5       at.  Mr Swann used the same basic configuration as SCRO
 6       and these points should be in sequence but he is now
 7       acknowledging in his testimony now and before that there
 8       are at least four differences, differences he did not
 9       report nor account for in his first report, nor his
10       submissions to Justice 1.
11           Testimony to Justice 1, Peter Swann:
12           "It is very important, however, in my opinion there
13       are no points of dissimilarity in the case of this
14       mark."
15           (Slide 123)  I noticed just recently that Mr Martin
16       Leadbetter had made a submission to the Inquiry.  From
17       his presentation we have a chart number 11 (slide 124)
18       and I must assume that he has charted it to prove this
19       is an identification.  I cannot assume any other thing.
20       So he has marked 14 points in agreement (slide 125) and
21       I have made some of the points on the left side a little
22       bit more red so that they can easily be seen but I have
23       copied them and I have also added to the reference
24       point, let's say, the banana point.
25           If we now take his point 13 on the left side (slide
page  83
 1       126) and here we have the handshake or deviated break on
 2       the right side and made it a little bit better to see
 3       and if we do now a ridge count to point 13 on the left
 4       whether you count this one, this upcoming ending ridge
 5       with or not it's a maximum of six.  If we count on the
 6       right side, we have minimum 11.  This is, number 13, is
 7       an integral point of this whole cluster of similar
 8       points and the ridge count to accepted point of
 9       similarity close to the core is five ridges off minimum.
10           (Slide 127)  If we take point 13 as a guide and we
11       go then downwards in the mark, we see an upcoming
12       bifurcation and if we then go in the reference print,
13       it's not there.
14           We then move to the mark and we have the upcoming
15       bifurcation everybody is using, even SCRO, and we look
16       to point 13, it's not there in the location where it
17       should be.  If we then go up there may be an ending
18       ridge in the same location, also in the mark in the
19       reference print, I accept that, but then we look for the
20       known bifurcation upwards and it's not in the reference
21       print.  This point over there (indicated), the upcoming
22       ending ridge, is not there.  If you look in the area
23       between the core and the cluster we have marked, then we
24       miss this one and this one and this one and this one
25       (indicated).
page  84
 1           Mr Leadbetter is pointing to a totally different
 2       area in the print and marking his points over there.  If
 3       we look at the famous handshake, point 15 of SCRO,
 4       point 1 for myself (slide 130) it is located over there
 5       by Mr Leadbetter.  There is point 15.  We mark it over
 6       there (indicated).
 7           So Mr Leadbetter is positioning the handshake over
 8       there and everybody else marks it over there
 9       (indicated).  There is where we mark it.
10           (Slide 131)  I have upgraded the image a little.
11       The issue is that Mr Leadbetter, in conflict with
12       everybody else, takes a cluster of points and moves them
13       four lines minimum to the side and says here has Shirley
14       McKie -- this part of Shirley McKie's finger has caused
15       this mark.  Now, we have now six deposits of Shirley
16       McKie claimed by SCRO and Leadbetter and Swann, or
17       seven, that the finger of Shirley McKie has now made
18       seven deposits of this mark, in this mark.  I think
19       Shirley McKie deserves a Nobel Prize of impossible
20       physics.  It is impossible.
21           If we follow their assertions, it is really
22       mind-boggling.  This is not possible.  They are in
23       conflict with each other.  Mr Leadbetter is in conflict
24       with Swann and Swann is in conflict with the SCRO and
25       Mr Leadbetter is in conflict ... nothing fits.
page  85
 1           Mr Leadbetter says in his citation at chart 13, next
 2       to it:
 3           "I have also marked the features in an area of the
 4       mark which I consider the least distortion.  It should
 5       be noted that there is an angle of approximately 60 or
 6       70 variation in orientation between the mark Y7 and the
 7       inked left thumbprint of Shirley McKie."
 8           The least distortion?  Weren't we in agreement that
 9       this was the slippage area, where the rotation should
10       have taken place?  60/70-degree variation orientation?
11       So where does Mr Leadbetter now rotate his print?  Not
12       this area (slide 132).  What other area or is the
13       fingerprint uplifted and then deposited again.  But let
14       us look at the rotation he is claiming and let us use
15       his own points to look what is happening.  If we then
16       look at the presumed rotation and we check it, then this
17       is 60 to 70 degrees we have to maybe correct a little
18       bit because his positioning is marked a little bit
19       different to everybody else, but it is no way it is
20       65 degrees.  It's not even -- it may be 6 or 15 degrees
21       maximum.  But this is the slippage area and I wouldn't
22       know Mr Leadbetter is now making the rotation.  Is it a
23       rotation?  Is it lifted up again?  I don't know but I
24       see no crossing lines, I see no slippage and I don't
25       know how he comes to 60 or 70 degrees.  I think it's
page  86
 1       just fantasy.
 2           (Slide 133)  One remark in between.  I will stay
 3       away from allegations and personal things but I know
 4       Mr Leadbetter and Mr Swann are icons of fingerprinting
 5       in the UK.  I've always respected them.  But I also know
 6       a lot of honest, competent practitioners in the UK and I
 7       leave it to the fingerprint world in the UK to study
 8       this material and take appropriate action.  This cannot
 9       decide upon the image of fingerprinting in the UK.  This
10       is as bad as it is.
11           (Slide 134)  I was going back to my conclusions on
12       the analysis of Y7.  This mistake is of an unprecedented
13       magnitude.  There are 17-plus discrepancies.  It is
14       highly unlikely for an expert of reasonable competence
15       not to detect the discrepancies between Y7 and the
16       reference print.  It is impossible -- that's my firm
17       opinion -- it's impossible for any expert and, indeed,
18       layman, you have checked it for yourselves, you have
19       seen it for yourself, and indeed any layman not to
20       recognise these discrepancies when you are alerted when
21       someone is pointing it out to you.  The evidence
22       presented to defend the identification reveals
23       inconsistencies, bad practice, and I hesitate to say it
24       but I can only call it malpractice, awareness and
25       concealment.
page  87
 1           The reasons for the development of the case after
 2       the initial maybe possible honest mistake cannot be
 3       contributed to the fingerprint nor fingerprinting and
 4       have to be sought outside the domain.
 5           You have to think of culture, power, psychology.  I
 6       don't know.  I don't know what the causes are for
 7       persisting the denial.  There are two options within the
 8       domain.  If we reject an honest mistake, then it's
 9       either incompetence or fraud.
10           (Slide 135)  We are not talking about an
11       identification.  The process of elimination comes before
12       identification and if you notify so many discrepancies
13       in the early stages then we are not talking about a
14       disputed identification, we are talking about an
15       indisputable exclusion.  Asbury was excluded, Marion
16       Ross was excluded, others were excluded, nine fingers of
17       Shirley McKie were excluded very briefly, very quickly.
18       So we cannot uphold that we cannot exclude Y7.  So there
19       is no dispute about an identification.  It is an
20       exclusion, with 17 discrepancies: period.  The SPSA as
21       it is now cannot deny that this is an exclusion.  They
22       cannot remain sitting on the fence and saying it's
23       undisputed or it is disputed.  It is an exclusion:
24       period.  This is what the science of fingerprinting
25       requires.
page  88
 1           (Slide 136)  Because it was an indisputable
 2       elimination that was not acknowledged as a mistake it
 3       deteriorated to an undefendable identification.  This
 4       has led to extreme assertions, misrepresentations, acts
 5       varying from bad practice to malpractice, et cetera,
 6       both within the fingerprint domain and outside it.  I
 7       have got my share of it.  We have heard several things
 8       yesterday.
 9           The phone rang in 2000 when HMIC asked me to assist
10       in it.  I did it.  I didn't solicit for it.  I was
11       invited.  History will prove me right but that's not the
12       point.  Will this Inquiry prove Scotland right and pave
13       the way for justice.
14           Another example of the extreme assertions and
15       misrepresentations is the collaborative study that was
16       used or misused by Mr Leadbetter and others.  I will not
17       go into detail.  I have given an account for what
18       happened with the collaborative study in my witness
19       statement to the detail, if people want to read it.  It
20       is there.  There's some offence to the practitioners in
21       Holland that are incompetent.  If people want to read
22       it, they can read my account there but I just give one
23       example of why we were different in our opinion than
24       maybe others.  First of all, it was a discussion we had
25       with Evett, it was another discussion then how are we
page  89
 1       pointing the points.  Our discussion was how do we
 2       practice fingerprints in the proper way and I think has
 3       Evett has already given an account that we were doing it
 4       differently but not, per se, wrongly.
 5           (Slide 137)  If I go to the collaborative study and
 6       I take just one example that is marked B.  Many people
 7       have identified it.  You cannot take the collaborative
 8       just as competency exercise because you were just given
 9       A to compare with mark A and if you could count to
10       23 then you were competent.  This is just a monkey test.
11       I can give it to anybody else.  If you were capable of
12       counting 24 points then you were competent.  It is not a
13       competency test and not meant to be that way because
14       another issue is you don't know the ground truth because
15       these are actual cases.
16           But let me take this one as an example.  You study
17       it and if you look at it you can see an ending ridge
18       coming upwards here (indicated) and ending ridge coming
19       upwards on the right side in the reference print.  Then
20       we see an ending ridge to the right somewhat higher and
21       I will mark it on the reference print over there.
22           You know by now I am going to count ridges.  We
23       count eight over there and seven over there.  So we
24       found a discrepancy and this discrepancy is confirmed by
25       another cluster of minutiae that we find here
page  90
 1       (indicated).  It's a kind of a change over there with an
 2       intervening ridge.  We see this change over and now we
 3       find, in the mark, we find an ending ridge to the left
 4       we're not finding in the reference print.  So we find
 5       two discrepancies in the mark or between the mark and
 6       the comparison print.
 7           If we find this we can either say the identification
 8       process stops or we can say, "Okay, I have an
 9       explanation.  There's probably a double placement here".
10       There is some argument for that because we see here an
11       image of a contour, not a real good connection of the
12       ridges but then I would have to ignore the whole area.
13       If the whole area is not part of the whole mark, then we
14       have to ignore the whole area.  So then we would
15       discount this area.  That's what we did.  Then if we
16       look at this incoming ridge here (indicated), you see
17       here an upcoming bifurcation, it's not there.  Another
18       dissimilarity.
19           I can explain that by saying there is a double tap
20       here of another incoming mark and there's some argument
21       for that because you can see here these lines a little
22       bit different of structure than the lines here but then
23       again I would have to ignore that area.  That would be
24       just sensible or reasonable.
25           If I then move to the core of the mark, then if I
page  91
 1       look at the formation or the shape of the core, then I
 2       see kind of a more or less pointing shoulder with an
 3       appendix that is going straight up and in the reference
 4       print it looks different.  It's more a round shoulder
 5       and an appendix that is coming down but it's shorter.
 6           Also, next to it is a bifurcation.  It's much more
 7       pointing in the mark and it is more round in the
 8       reference print and the same counts for the bifurcation
 9       here (indicated).  It's lower and it looks different.
10       So I have no confirmation whatsoever on Third Level
11       Detail that I'm looking at the right print.  It may be
12       the right print; I don't know.
13           The fact that we did not identify this one and not
14       found enough points you cannot hold it against us.  We
15       have very good reasons not to do that.  But if you are
16       just looking at numbers saying, "Oh, the Dutch are
17       incompetent", so be it.  I accept it.
18           (Slide 139)  Shooting the messenger does not change
19       the message.  Who is right about Y7 and who is wrong
20       cannot be decided upon assumed or pretended competency
21       but should be decided upon by the evidence presented and
22       demonstrated.  For me, the litmus test of competency is
23       Y7 and QI2.
24   THE CHAIRMAN:  Would this be a convenient time to take a
25       further break and we will sit again at 1.50?
page  92
 1   (1.03 pm)
 2                      (Luncheon Adjournment)
 3   (1.55 pm)
 4                      ARIE JACOB ZEELENBERG
 5                     PRESENTATION (continued)
 6   THE CHAIRMAN:  We are going to move now to QI2.
 7   THE WITNESS:  (Slide 140)  The analysis of QI2, phase 1,
 8       that's my analysis of the mark.  This is the chart I
 9       presented to the Inquiry, the same structure.  I will
10       (slide 142) take the image and copy the points in there
11       to digitise it just not to have quality loss.  (Slide
12       143)  This is the mark in situ and on the left side
13       there is an image of the tin in colour and these
14       colours, in a way, explain some of the difficulties with
15       the print.  They have tried to photograph it to take out
16       the background but you can still see this sharp V-shape
17       is still coming back in the mark.  But we have to take
18       the mark as it is and if we then look at the overall
19       quality of the mark (slide 144), the overall assessment
20       is there are many fragments all over the place inside
21       QI2, and outside it.  The V-shape I already mentioned
22       is, in general, I can think it's a poor quality print.
23           There's the coherence of the points in itself and
24       the segments is questionable.  The determination has
25       caused that, whether it's a single print or what the
page  93
 1       source of the print may be or whether multiple prints is
 2       of course fairly difficult.  There are stains in there.
 3       We see two vertical swipes.  I will point them out here.
 4       On the right side from the core (indicated), you can see
 5       them starting here and the swipe going downwards and on
 6       the left side we see kind of the same thing start a
 7       little bit lower and we see a swipe over there, left, on
 8       the right side of the core.
 9           There's very few reliable Galton points at first
10       sight.  You could use the mark to exclude but in our
11       system and I think many would not find the mark eligible
12       or suitable for identification.
13           The finger: as I said, we cannot determine what
14       finger it is coming from.
15           If we look at the overall shape, I have a habit of
16       trying to follow the line starting in the core and we
17       can see if we follow the lines here, there is a tendency
18       to rotate out so that the rotation is clockwise -- so to
19       the right -- and typically this will tell you that it is
20       most likely coming from a left hand.  If it's turning to
21       the left, it's most likely to come from a right hand.
22           If we try to do the same in the latent, then here it
23       is obviously ambiguous but if we look at the larger
24       picture, if we look here (indicated) we see that the
25       lines are going down and out and below the left delta
page  94
 1       and if we look here there is a tendency that we see that
 2       the lines go down and out.  Not 100 per cent but there
 3       is some idea that this is rotating differently than the
 4       mark.
 5           (Slide 145)  Same structure as with Y7.  On top we
 6       have the charted fingerprints and below we have the
 7       clean ones.  I tried to keep them as clean as possible
 8       for your own observations and we start with point 1 and
 9       2 at the same numbers as SCRO.
10           You see down where I'm looking at and I'll leave
11       point 1 for what it is -- sorry, if I move back, if we
12       see here (indicated) this line on the right side is more
13       prominent than on the left side, where we see it is an
14       eyelet.  I will move back to discussion of the core
15       later.  But if we now look at a bifurcation in the mark,
16       left leg and right leg are kind of symmetric.  You could
17       say they are almost even.  But if we move to the
18       reference print and I move now downwards to be able to
19       see it better, we see that the left leg is considerably
20       thinner, I think it is half as thin as the right leg.
21       So a distinct difference in appearance of the legs of
22       the bifurcation.
23           The core I said I would discuss later in phase 2.
24           (Slide 146)  If we move over to point 3, which is
25       also point 3 of SCRO, then the occasion we have to look
page  95
 1       at is here (indicated) so we see here in the reference
 2       print an upcoming ending ridge.  If we move over to the
 3       corresponding area in the mark, I have traced the
 4       ridges, there is no such event.  It is there in the
 5       reference print and it should be there in the mark and I
 6       cannot see it.  It is not there.  If we measure from the
 7       reference point 2 downwards, then we can see that it
 8       should be somewhere over there, but it isn't.  This is
 9       the first distinct discrepancy.
10           If we now move to point 4, we see the area, it's
11       down left of the core and I'll mark the area in the
12       reference print where the overlay should appear
13       (indicated).
14           (Slide 147)  What we see in the reference print is
15       the pointing bifurcation with some small lines, a little
16       bit smaller than the adjacent lines, and here in the
17       mark they are pretty fat, very prominent, a little bit
18       different outlook but we could say that the point is
19       over there and we could use it as a reference point.
20           The ridge path of the original ridges is different,
21       not really the same, but just let us just accept that it
22       is there and use the location as a reference point.
23           (Slide 148)  I must say that in an earlier
24       evaluation of QI2 I used another point as a reference
25       point, more to the left, and therefore I had different
page  96
 1       conclusion but I will explain that later as well.
 2           We move now to point number 7 which is not discussed
 3       by SCRO.  We find it over there (indicated), we find
 4       it -- we should see it over here.  This is the point we
 5       just discussed (slide 148) and there is another line
 6       coming in and definitely something is stopping over
 7       there which we don't see in the reference print.  The
 8       whole area is moving upwards and thinning and there is
 9       another event which I cannot see in the reference print
10       and this is a distinct discrepancy.
11           (Slide 149)  Point number 6 which is the same as
12       SCRO number 13 -- I will discuss point number 5 later.
13       Point number 6 we see over here (indicated).  We have to
14       look in the reference print in this area.  We can now
15       see the overlay and, again, then we expect point 6 to be
16       there but if we take a ridge count then it is four in
17       the mark and three in the reference print.  Even if we
18       assume that it's the same point, then the ridge count of
19       the reference print number 4 is off and I said in an
20       earlier assessment that I used another point and I
21       accept 4 is the right point to be but then -- sorry,
22       number 6 is the right point and then number 4 is off but
23       which way do you go?  But in between the ridge count in
24       one is minimum four and the other one it is three so
25       there's a distinct discrepancy.
page  97
 1           (Slide 150)  We then move to point 5 and I must say
 2       this is one of the better points in the mark.  This
 3       upcoming ridge here (indicated) is pretty clear.  The
 4       line on the left side is pretty clear; the line on the
 5       right side is a little bit more ambiguous but you can
 6       trace it and it's absolutely without discussion that
 7       there is an upwards ending ridge -- one of the better
 8       points in the mark.  If this event is not true, almost
 9       nothing would be true.  Here is the overlay.  We use 6
10       and 13 of SCRO as the reference point and we should find
11       it there and it isn't.  So this is, again, a distinct
12       discrepancy number 4.
13           (Slide 151)  Then we move to point 8 and 9.  We can
14       find them over there (indicated) in the reference print,
15       look in that area and I have no issue with point 9.  It
16       could be similar and in the same location almost, so I
17       will leave it where it is but now we start looking for
18       point number 8.  If we look over here and we follow that
19       ridge here and we follow the yellow ridge here
20       (indicated), there is at least one extra line stopping.
21       There could even be another one but at least one over
22       there is stopping and there is no such event in the
23       reference print which is another distinct discrepancy.
24       You may even assume, as I said, another upwards ending
25       ridge but there is at least one distinct discrepancy
page  98
 1       making it number 5.
 2           (Slide 152)  Point number 10 and 11.  These are not
 3       marked by SCRO, which I find very peculiar.  If we
 4       follow this ridge over here (indicated), then you see
 5       that it bends downwards.  This is one of the thicker
 6       ridges and there is another line coming in, this kind of
 7       a confirmation of the event of the curving ridge, this
 8       line is coming in.  There's little dispute about that, I
 9       think.
10           From the other side, we see that it is an incoming
11       ridge.  It may be a little shady but you see it coming
12       in and it has to end over there.  The area around it is
13       really white and there is a lot of space between the
14       adjacent ridges so there is a ridge coming in from the
15       left.
16           If we take on the right side here -- I will use the
17       overlay -- then here is the overlay for yourself to
18       check.  This is the point that could be there,
19       although -- and this is the missing one.  If we look at
20       the Third Level Detail, this is a ridge that is really
21       thinning out very, very long.  It almost becomes
22       incipient, you have used the word before, but I will
23       concede that it could be there but then at least point
24       number 11 is failing and is a distinct discrepancy
25       number 6.
page  99
 1           (Slide 153)  The conclusions of my first analysis is
 2       that overall it is a poor quality, there's lots of
 3       distortion and poor coherence that will say the
 4       configuration and sequence of points is poor.  There's a
 5       minimum of 6 discrepancies found in 11 locations
 6       discussed and the core is not yet extensively evaluated.
 7       I will do that later.  The mark is fit for elimination
 8       but due to defined core area and some certain points you
 9       can say, "Okay, there is the core area.  I must find
10       things over there so I can exclude certain donors".  We
11       know that certain donors are excluded like Mr Grey and
12       all other people that are discussed are excluded as
13       possible donors.  So the mark is fit for doing that.
14           There are numerous differences in ridge path and
15       Third Level Detail.  I have shown them to you.  There
16       may be a pattern difference.  I cannot be certain
17       100 per cent but there is a kind of a warning sign.  In
18       this case, I have no stepping stone.  I cannot say,
19       "This is similar.  I go from there".  There is nothing
20       for me to say, "I start over here", and I find it
21       alarming.
22           There are numerous warning signs and it is a
23       definitive exclusion that should be recognised in the
24       early stages of the comparison.
25           (Slide 154)  I will now move to the scrutiny of the
page 100
 1       SCRO charting, phase 2.  Again, this is the charting.
 2       (slide 156)  I transferred it to the digital image for a
 3       better quality.  Here are the points that are
 4       transferred.  (Slide 157)  This is my first chart and
 5       the points over here on the charted points from SCRO.
 6       As I said, I will do the core in the end but let's now
 7       move to point number 3.  It is marked over there.
 8       (indicated)
 9           Now what we see here, as I discussed before, we have
10       a vertical swipe here in the mark and this is a kind of
11       a white area with very low or virtually no definition
12       but it is definitely affected by the vertical swipe.  We
13       have to look for point 3 over there and it has to be on
14       the white location.  I cannot see a system of ridges,
15       let go that I can see that there is an ending ridge in
16       there.
17           It is inserted in the mark in a distorted area of
18       very poor quality.  I will not count it as a discrepancy
19       because I cannot prove it's there.
20           (Slide 158)  Then we look at point number 4.  We see
21       the area below.  We now use point 3 as a reference point
22       in the reference print and the location of it where it
23       should be in the mark as a reference and then we have to
24       look for point number 4 in that area (indicated).  I
25       simply can't find it.  I simply can't see it and my
page 101
 1       point number 3, I call it now number X, should also be
 2       in that area and I find that very strange that this one
 3       is marked and the other one is not marked, whilst the
 4       area there is a little bit better definition but we have
 5       to look for the point over there.  So point 4, like
 6       point 3 in the reference print, is inserted in the mark
 7       in a distorted area of very poor quality and an adjacent
 8       point that fails in mark, point X in the area of better
 9       quality, is ignored.  Ian Evett called this teasing the
10       points out and the more general term is reverse
11       engineering.  The IEEGFI report has some remarks about
12       that and I will read out the relevant quotation or the
13       relevant paragraph of this report, that is about
14       circular reasoning:
15           "Circular reasoning is a scientific fallacy and has
16       to be strictly avoided.  A scientific method follows a
17       step-by-step procedure with the aim of an unbiased
18       objective conclusion.  With circular reasoning a
19       prejudgment is more or less apparent steering the
20       process to the wished and preset outcome.  In other
21       words, an assumption is the basis for a conclusion and
22       then this conclusion proves the assumption to be right.
23       This is a scientific fallacy."
24           The first report said:
25           "The pitfall is that the premature assumption of
page 102
 1       donorship leads to transplantation of data from the
 2       original into the blur of the latent.  It is circular
 3       reasoning like this print comes from the donor.  Prints
 4       are unique, thus all the data must be the same and
 5       subsequently all differences are not real.  Circular
 6       reasoning is often hard to detect.  Examples in
 7       fingerprinting are, 'If I ignored this difference it is
 8       a perfect match, so the difference cannot be real and
 9       must be caused by distortion".  Different dactyloscopic
10       points between prints from the same source don't exist.
11       This is reverse engineering, reverse talking.
12       Generally, of course, it is true that fingerprints from
13       the same finger cannot have real differences so as a
14       factual statement for the base of fingerprint is true
15       but it's a fallacy if used during the methodical
16       forensic procedure of proof."
17           This is from the IEEGFI report from Austin in 2000
18       and, I think that is what we see here, Ian Evett also
19       referred to it.
20           Point number 5 and number 6 I take together.  (Slide
21       159)  In the reference print it is a short ridge between
22       two lines and we see it over here.  I made it green.
23           In the latent I see two yellow lines that are
24       parallel and in between I see the green which is short
25       piece of ridge and continuing over there with a gap in
page 103
 1       between.  The point is that, very strangely, the SCRO
 2       has marked the ending ridge, the upper piece of the
 3       ending ridge over there.  Now, it's against the stain so
 4       you cannot know that the ridge is ending there.  You
 5       know there is a stain but you cannot be certain that the
 6       ridge is ending over there.
 7           The lower one is not taken to the right but
 8       downwards on a continuous ridge.  You could ask yourself
 9       why is that done that way?  This one is between the
10       lines and this one, these two, go outside the flow of
11       the ridges.  I think there is a reason for it and I will
12       go (slide 160) to the next point, that's point number 9.
13       We find it over there (indicated).  That's kind of the
14       broken ridge and we see it over there.  Now point 9 is
15       marked in the latent on the gap between the ridges and
16       this is it in the reference print (indicated).  In the
17       reference print we see the adjacent line in between the
18       short ridge and point number 9.  Just to make it kind of
19       the same, if the small ridge was marked here and there,
20       then the existence of point 9 was wrong and the location
21       of point 9 would be wrong.  So I think that's the reason
22       why they marked it downwards just to make it fit.
23           As I said, if point 6 was marked in the proper
24       direction, point 9 would become non-existent and if it
25       is done this way, you have virtually no ridge count.  I
page 104
 1       even have not counted it as a discrepancy, point 7 and
 2       point 8.
 3           (Slide 161)  These are two upcoming ending ridges.
 4       This is the area where these two upcoming ridges are
 5       marked and this is a blow-up of it.  The dominant
 6       direction of information is horizontal, like this
 7       (indicated).  The vertical is non-detectable.  They have
 8       now assumed there is a system of vertical lines that can
 9       also detect ending ridges in that same area.  This is
10       where they mark the points.  I would like to invite you
11       to look for yourself whether you can see upcoming ending
12       ridges.  (Pause)
13           Now, if you don't see them, there's nothing wrong
14       with your eyes.  If the kitchen fitter tells you that
15       your kitchen is not skewed but that you have to go to
16       Specsavers, don't believe him.  Don't believe a
17       fingerprint expert that he can see what you can't see.
18       If he cannot demonstrate it, it is not there.  You may
19       struggle to interpret the things, to validate things and
20       to compare them but if you can't see them, they're not
21       there.  A general requirement is demonstrability.  You
22       should be able to demonstrate what you say you see.  If
23       you cannot see it, it's not there.
24           (Slide 162)  Point number 11.  We use 12 as a
25       reference point.  This is in the area where there's a
page 105
 1       vertical swipe downwards here (indicated) and the ridge
 2       should be there.  I cannot confirm it.  It's on the edge
 3       of where the information stops to be reliable or even
 4       visible.  This is what I would see as a line.  As well I
 5       can say here it stops where I can see, still it is
 6       marked, I wouldn't do it.  This is where it should be.
 7       This is again a point that is teased out using the
 8       reference print as a guide to the mark.
 9           (Slide 163)  Point number 12 is discussed in my
10       analysis as phase 1 as number 4 and I will use it as a
11       reference point for the rest of the points on the left
12       side of the mark.
13           I have got some support, I think.
14   THE CHAIRMAN:  There is a concert next door, I'm afraid.
15   THE WITNESS:  (Slide 164)  Point number 13.  I've made the
16       overlay over there.  There's a definite point in the
17       reference print.  I use this as a reference.  It should
18       be there.  I can't see it.  If you just said that if the
19       other point over here, which is much more prominent,
20       it's not used by SCRO and a non-existent point or at
21       least a very ambiguous point in the mark is charted as
22       similarity and I think this is bad practice.  You cannot
23       ignore this one and mark one over here just because you
24       need it just because this is the reference print.  If we
25       compare it with this information, if we look at the
page 106
 1       upcoming ridge here, and we see that in this area
 2       point 7 and 8 were marked and point number 3 was marked
 3       and this point is ignored, I find it very suspicious to
 4       say the least.
 5           (Slide 165)  Point number 14.  It's an upcoming
 6       ending ridge.  It could be there but it's ambiguous in
 7       the reference print.  It could also be on the other line
 8       because there's a small crease over there.  It could be
 9       there.  I have no issue with it.
10           However, if we locate its position against point 13
11       and we start counting the ridges, if we go down in the
12       reference print and the count to point 13 is about 1, 2
13       and down here there is 1, 2, 3, 4-something.  So even
14       then the location of point 14 in reference to the other
15       points is off.  Whatever choice you make about where you
16       locate the points here, the ridge count never fits.
17           (Slide 166)  Point 15 is marked in an area next to
18       the swipe, the downward swipe.  If we would look in that
19       area, then here you see the swipe and look in the same
20       area we could also mark other points of ridges that stop
21       in that area or cannot be seen further in that area
22       because there is a swipe.  If you look at the blue dots,
23       you could suppose and say okay, there's all ending
24       ridges but because the swipe area is over there.  All
25       these are not marked as points of similarity but this
page 107
 1       one conveniently is because it is in the reference
 2       print.
 3           This is a distorted area of low quality.  Point 15
 4       is assigned to a questionable event of arbitrary origin,
 5       similar events are ignored.  This is inconsistent
 6       assessment of information based upon what is desired and
 7       what is not.  This is bad practice.
 8           (Slide 167)  If we now move to the core area point
 9       number 1, 10 and 16 -- and this is the last one -- if
10       you look at what the shapes are in the core, then you
11       can see over here -- I will move back -- there's an
12       elongated kind of a thick ridge.  This may be not part
13       of it but it is typically long.  I see it as this --
14       this is not a presumed ridgeology, just my assessment of
15       what I see -- and it is as long as this here and in the
16       reference print it's kind of a small comma, triangular.
17       It is different.
18           If I then move to what I see in the latent and I go
19       back or I see here is a kind of an eyelet -- in English
20       we call it an eyelet, it is called that way in the
21       IEEGFI, in German they call it an auge.  That is also to
22       say an eye.  In Spanish they call it a circle.  Somebody
23       commented that the word "eyelet" is never used.  I just
24       ask attention that there are many other languages than
25       English that also use fingerprints.  So we call it an
page 108
 1       eyelet and we see it like this.  The left leg is
 2       considerably thinner than the right one.
 3           Go back again, an eyelet, this one is thin, this one
 4       is prominent and cannot be ignored and the line is
 5       continuing.  This is how I see it.  In the reference
 6       print what we see is it is a thick line on the left, on
 7       the contrary, and a small hook downwards that is
 8       supposed to flow downwards in the latent.
 9           Again, moving back this is the very thick line that
10       is interrupted because if we look at the Porelon pads
11       but this is the thick line on the left and this one is a
12       hook downwards that is supposed to run through, which
13       isn't.
14           Then we move upwards and if we look in the reference
15       print we see this line which is pretty prominent and we
16       see it here forming a kind of an island or shape or
17       eyelet or whatever you call it and then moving onwards
18       much more thinner.  We see nothing alike in the latent
19       in the mark.  It is just a line that is evenly spread
20       and also the other ridges are not demonstrating that
21       there is some kind of event like that.  If we look at
22       the mark, the reference print, we see there's a lot more
23       room for it.  There's even also incipient ridges a lot
24       of space to allow for this island to happen if you could
25       say it like that.  So definitely the whole area looks
page 109
 1       different and I see that there is definitely a
 2       discrepancy over there.
 3           (Slide 168)  That concludes my analysis of QI2 and
 4       the scrutiny of the SCRO markings.
 5           My conclusions are this: the mark is of such poor
 6       quality that the suitability for identification is
 7       questionable beforehand.  The mistaken identification is
 8       of a considerable magnitude.  You cannot miss it.  There
 9       are about three points that are similar by type and
10       location only.  There are at least six discrepancies.
11       All other points are either debatable or considered
12       non-existent of 16 discussed in total.  It is highly
13       unlikely for an expert of reasonable competence not to
14       notice the discrepancies between QI2 and the reference
15       print.  It is impossible for any expert and, indeed,
16       layman not to recognise these discrepancies when alerted
17       to them.
18           The absence of quantity and the distortions open
19       avenue's for guided interpretation, however there is no
20       stepping stone that can explain an honest mistake at the
21       outset.  The evidence presented to defend the
22       identification reveals inconsistencies, awareness by
23       avoided discrepancies and bad practice and if bad
24       practice is executed obviously with awareness it becomes
25       malpractice.
page 110
 1           (Slide 169)  On the request of Mr Moynihan, I was
 2       going to skip the QD2 involvement so I will go very
 3       quick ...
 4           (Slide 176)  The next ones a few remarks and then I
 5       will wrap up about identification and elimination.
 6           (Slide 177)  I will show you a computer screen of
 7       our old AFIS systems, presently replaced, but we see on
 8       the left side a mark.  This is a mark that is searched
 9       in our computer.  It works this way, that we take a
10       cluster of reliable minutiae in the latent, we mark them
11       up and then these clusters in sequence is compared to
12       the database.  The AFIS or AFR system will bring back
13       fingerprints with similar clusters, similar
14       relationships.  You could say that what we are looking
15       at on that basis are the kind of lookalikes, the most
16       similar prints, and the system will generate a candidate
17       list of most likely persons and the best lookalike or
18       the best one will be on top.  This is typical for any
19       system.
20           This is a random latent search and I asked them to
21       film how they go through the candidate list.  These are
22       the candidates, here you see the list and I ask your
23       patience for looking at it for a short while.
24           (Slide 178)  Here is the latent.  This is the movie.
25       You see the hand of the latent expert on the left side
page 111
 1       and he is clicking.  This is an exclusion, an exclusion,
 2       an exclusion, another exclusion, another exclusion.  Now
 3       he enlarges it because he just wants to see -- he's not
 4       looking at the forest, he is now looking at some trees
 5       so he goes into the minute detail in order to be able to
 6       exclude this possible donor.
 7           You see, they are all loops to the left of a certain
 8       size.  Again, he is enlarging it just to make sure what
 9       he's looking at.  Now he is done.  He has excluded all
10       of them.
11           (Slide 179)  He has excluded 30 candidates in two
12       minutes.  I don't think he did time it but it was 2
13       minutes.  That is 4 seconds per exclusion.  That is how
14       fast it can go.  These are prints that are selected by
15       similarity of configuration.  If somebody says that it
16       takes him 90 seconds to exclude it, you may excuse him
17       of being slow.  That's just all I wanted to show to you.
18           (Slide 180)  This is how it works.  The process of
19       identification starts with elimination.  You take a
20       cluster in your head and you go over a ten-print sheet
21       and you say no, no, no, no, no because you look for
22       differences.  The process if you go to -- and only if
23       your exclusion fails you start making comparison.  If I
24       make a search in AFIS and if I see a similar one and say
25       "Hey, this one could be the one" and I hook on to some
page 112
 1       cluster of points, then the process of comparison will
 2       start and then we will use ACE-V.
 3           The process of ACE-V, yes, that may take hours.
 4       When we have a difficult print and it is a questioned
 5       one, it will be executed by three independent experts.
 6       They will make annotations, they will write things down,
 7       they will make markings-up and then we will have
 8       discussion amongst them in the lab and that can indeed
 9       take hours.  But this is not a matter of a disputed
10       identification but of indisputable exclusions.  To me,
11       the discussion about standards in relation to Y7 and QI2
12       is futile because there is no standard that changes a
13       non-identical print into an identification.
14           (Slide 181)  If this mistake is not acknowledged by
15       the profession in Scotland, then my worry is that there
16       will never be mistakes that you can prove in the future.
17       These are two really bad ones and they are acknowledged
18       by the profession.  Then I stop.  The most worrying
19       aspect is that the wrong lessons are learnt.  There are
20       people here in Scotland that have learned nothing from
21       it.  This will inevitably have long-term negative
22       effects on fingerprints as an instrument in fact-finding
23       and assisting justice and the truth.  To me, this
24       fingerprint should be debated in the fingerprint
25       profession in Scotland.  Only the science should rule
page 113
 1       the science, nothing else.  No police officer in charge,
 2       no management: the science should rule the science.
 3       That's why I'm here.  We should discuss this print.  We
 4       cannot avoid it.  We should learn lessons from how the
 5       FBI handled the Mayfield case.  They had an in-depth
 6       investigation, they acknowledged the mistake, they
 7       learned from it, they changed their procedures and that
 8       is how it should be done and that's why I'm here.
 9           Thanks for your attention.
10   THE CHAIRMAN:  Thank you.
11                     Examined by MR MOYNIHAN
12   Q.  Mr Zeelenberg, I want to go back to some more elementary
13       questions and to ask you to comment, please, and then
14       give some assistance.  If you don't understand anything
15       that I say, please just indicate because it will be my
16       fault without question.
17           First of all, if I start at the most elementary
18       level, would I be correct in understanding that the
19       conclusion reached by a Fingerprint Examiner in relation
20       to an individual fingerprint is to be properly regarded
21       as a matter of personal opinion, albeit informed by
22       skill -- personal opinion rather than an object of fact?
23   A.  Yes, that's true.  We call it an informed opinion but it
24       also depends of course on the level of information we
25       are looking at.  If I might explain, if we have a
page 114
 1       ten-print search and we look at ten fingers and we
 2       compare them and if we then say that this is the donor
 3       of this fingerprint, it cannot be disputed.  I could say
 4       my mother-in-law could do it because it's an
 5       overwhelming amount of information that you cannot
 6       ignore it.  Whether this makes it scientific I don't
 7       know, but it makes it certain.
 8           The point with fingerprint identification with marks
 9       is the question is not is it reliable -- the basic
10       question is how much is enough to come to this
11       conclusion but it is an informed opinion about the
12       origin or the attribution of source who has deposited
13       the fingerprint.  Yes, that's true.
14   Q.  If I just follow your answer on, you mentioned that with
15       the SWIGFAST paper, there was a question of the print
16       requiring or the details requiring to be demonstrable.
17       I think you have covered that already in what you have
18       said today at page 101 of the transcript.
19           What I also wanted to discuss with you was the
20       question of permanence.
21   A.  Yes.
22   Q.  The features, barring a scar on the finger, the
23       characteristics that form the fingerprint are understood
24       to be constant through life, yes?
25   A.  Yes.
page 115
 1   Q.  SWIGFAST speaks of not just demonstrability but also
 2       that the characteristics must be reproducible.  What
 3       does that actually mean?
 4   A.  Here's a misunderstanding.  The specific word
 5       "demonstrability" is what I added in.  I think I did try
 6       to say that.  But of course it is a general requirement
 7       if you go to the further procedures, yes, you must be
 8       able to demonstrate it.
 9           Reproducibility is more looking at the fact that if
10       you repeat the whole process, the outcome should be the
11       same and that's more faces towards ACE-V and
12       verification.  So if it is verified, then it's also
13       reproducible.
14   Q.  I am grateful to you because that assists.
15       Reproducibility has got nothing to do with the
16       individual characteristics?
17   A.  No.
18   Q.  It rather is the conclusion of the officers?
19   A.  Yes.
20   Q.  I will come to that later.  Forget reproducibility,
21       though, and carry on with the particular theme that I
22       wanted to ask you about.  On any one human finger, the
23       features are constant over time barring injury.
24       However, does the discipline of fingerprinting have to
25       allow for the fact that when a comparison has been made,
page 116
 1       what the officer is looking at is in fact an
 2       illustration or an impression -- if that's the correct
 3       word -- an impression of the fingerprint?  By
 4       "impression", what I mean is it is not literally a
 5       constant image of what's on the finger but rather can be
 6       the subject of very subtle and indeed wide-ranging
 7       variations?
 8   A.  That is definitely true.  It's kind of confusing.  When
 9       you talk the science of fingerprints and when we say
10       that its uniqueness in permanence, in effect we're
11       talking about the finger itself.  You are right in
12       saying that when we looked at fingerprints we are
13       looking at one type of representation of it.  That's
14       true and these representations may vary even between --
15       we have seen it already -- rolled impressions and plain
16       impressions.  I showed you two inked impressions in
17       which there were ending ridges and in the one other one
18       it was a bifurcation seemingly.  So there is always some
19       time of variation.
20           But what does not change is the sequence of the
21       events if you take them together.  If you look at the
22       clusters and you look at the relationships and the
23       in-between ridge counts, that will always be the same.
24   Q.  I happen to have an example which I have used earlier
25       that I think makes this point for me.  If I could bring
page 117
 1       up for you, please, an image that was used on the first
 2       day of this particular chapter of evidence, FI2209.01
 3       and 02.  I have brought up two images of two inked
 4       impressions of Shirley McKie.
 5   A.  Yes.
 6   Q.  Do you recognise the areas that I have highlighted with
 7       the red circles?
 8   A.  Yes, I do, yes.
 9   Q.  At least for me, because I think in terms of SCRO
10       numbering, these are the areas that SCRO number as 4.
11   A.  Yes.
12   Q.  The lower of the two related bifurcations, the lower one
13       being 4 leading up to the higher one 5.
14   A.  Yes.
15   Q.  On the right is an inked plain impression with a detail
16       and on the left is a rolled impression.  I understand
17       these to be the same corresponding areas.
18   A.  Yes.
19   Q.  Do you accept that?
20   A.  Yes.
21   Q.  To some extent, you have already covered this in your
22       own evidence, but when you look at the inked plain
23       impression on the right, what would your own preferred
24       interpretation be of the characteristic that is in the
25       red circle?
page 118
 1   A.  Without looking at the left one, I would prefer to say
 2       or to assume that it is an ending ridge but I must add
 3       you are never certain.  But if I look at the ridges
 4       adjacent to it, they seem to follow their course
 5       straight and the other one is kind of more stopping, the
 6       lines above it are not really moving towards each other
 7       that much.  My best guess would be that it is an ending
 8       ridge, but we are never certain.
 9   Q.  Accordingly, if the image on the right is construed as
10       an ending ridge, you would therefore, I think I would
11       understand, see the rolled impression in entirely the
12       same way, as an ending ridge?
13   A.  Yes, probably.  I must say that the deposit on the right
14       side there seems to be a lot of inking just on that
15       spot.  There may be some extra pressure.  So this can
16       make an ending ridge to appear connecting -- this can
17       also happen by dirt or by ink -- appear to be
18       connecting, but actually it could also be in the finger
19       that it is connected but just a little bit lower.  So
20       not printing in that effect.
21           There are also people that are able to see that if
22       you roll from left to right or from right to left that
23       sometimes the ending ridge is moved to the other side
24       and if you do it the other way round, it moves to the
25       opposite side.  I'm not that able but there's people
page 119
 1       that seem to be able to look at it that way.
 2   Q.  If I understand correctly, you were reading from for us
 3       I think it's the SWIGFAST document which speaks in terms
 4       of an "event"?
 5   A.  Yes.
 6   Q.  An "event" could be a collective term that would cover a
 7       variety of characteristics?
 8   A.  Exactly.  I think that we have done the same in the
 9       Interpol Working Group.  I was an advocate of that
10       because we have been talking way too long about
11       bifurcation, ending ridges, deviated breaks, do we call
12       it a short ridge, do we call it whatever.  I'm advocate
13       of saying it is an event.  A disturbance of the regular
14       is an event and all the others are formations of one or
15       two or three events.
16   Q.  What I wanted to ask you to comment on, because this is
17       just there as an illustration of a point, I might
18       myself, as a completely lay person, simply choose to
19       avoid this challenge and debate about whether this
20       feature is a ridge ending or a bifurcation, avoid it by
21       simply calling it an event?
22   A.  True.
23   Q.  I could then have all experts on the range of opinion
24       just agree there is an event there.  Are there occasions
25       when it is necessary for the expert to actually commit
page 120
 1       himself to the precise nature of the event in that
 2       location?
 3   A.  Basically for checking the sequence that is not so much
 4       of importance.  As long as if I say this is not an
 5       ending ridge but a bifurcation on that side, if I'm
 6       checking the relationships I must keep in mind that I
 7       first said it has to be on this side and that I not
 8       change it when I am counting on the other side.  So then
 9       I have to be consistent in that.
10           However, if we are studying and analysing latents,
11       it is beneficial to look at a detail as I already showed
12       you saying, "Listen, this is a real bifurcation I'm
13       quite certain of it, the event looks like this and I
14       would like to see it back in the reference print as I
15       determined it".  That gives confidence.
16           Those small things that was also talked about, the
17       hawk-eye, the Rosetta characteristic that makes a
18       strange bend downwards.  I would see that, I would note
19       that in my analysis and I will find it very nice if I
20       look in the reference print, "Hey, this gives
21       confirmation.  This really looks like what I'm looking
22       for" and if it's not, it should be kind of a warning
23       sign, you cannot ignore everything.  But the basic
24       thing, yes, you are right the basic fabric is the
25       locations of the Second Level Detail.
page 121
 1   Q.  If I can move from what we have just been talking about
 2       just now to another example, I will take you to the SCRO
 3       charting of Y7 to follow this particular point if you
 4       allow me just a moment.
 5   THE CHAIRMAN:  Would it be helpful if we took the short
 6       break now?
 7   MR MOYNIHAN:  We can, sir.
 8   THE CHAIRMAN:  I think it might be while you are finding it.
 9       We will resume at 3.00.
10   (2.50 pm)
11                         (A short break)
12   (3.00 pm)
13   MR MOYNIHAN:  I was trying to take a temperature in the
14       room.  I have been trying to check if I don't ask some
15       questions this afternoon if I still have a realistic
16       expectation of finishing Mr Zeelenberg tomorrow
17       afternoon so that I don't detain him any longer.  I have
18       a rough impression that that is perhaps achievable.  I
19       am conscious that Mr Zeelenberg has had a very powerful
20       stint and it would also probably assist me if we just
21       simply --
22   THE CHAIRMAN:  It's a lot to absorb and a lot for you to
23       demonstrate in a day.  I am obviously very anxious, as I
24       am sure you are, that we finish you tomorrow.  But if I
25       have reasonable assurance that tomorrow will allow
page 122
 1       Mr Zeelenberg to be free, then I am quite content to
 2       stop today at this point.
 3           Mr Holmes, it is likely that you will take longer
 4       than Mr Smith.
 5   MR SMITH:  Sir, I can say at this stage I have no questions
 6       for Mr Zeelenberg.  It may be that some of the questions
 7       raised by Mr Holmes we will --
 8   THE CHAIRMAN:  What about Miss Grahame?  Are you
 9       anticipating time?
10   MISS GRAHAME:  We had intimated a line of cross to my
11       learned friend Mr Moynihan last week and I understand he
12       is adopting that as part of his examination.  So I
13       wouldn't anticipate having any questions.
14   THE CHAIRMAN:  Certainly between Mr Holmes and Mr Moynihan
15       then as to whether they both can be reasonably certain
16       that any questions will not take longer than tomorrow.
17       It is a bit unfair to ask you at this stage.
18   MR HOLMES:  Given the discussions that I have had with
19       Mr Moynihan already, subject to anything my clients
20       might raise with me flowing from Mr Zeelenberg's
21       presentation this morning, I think we are reasonably
22       confident we can finish tomorrow.
23   THE CHAIRMAN:  It is important that the way that the Inquiry
24       is proceeding is that those that you represent will have
25       an opportunity to give their response and that is
page 123
 1       really, I think, the best way of working and that is the
 2       way we will be.  The fact that there isn't long
 3       cross-examination should not be interpreted in any way
 4       that any of the Core Participants accept the evidence.
 5       I understand that but I think it is important that it
 6       should be more widely understood that perhaps silence
 7       should not be taken as acceptance.
 8           On that point then, we will rise.  Thank you very
 9       much for the time you have given us today and we will
10       sit again at 10.00 tomorrow morning.
11   (3.05 pm)
12        (Adjourned until 10.00 am the following morning)
13
14
15
16
17
18
19
20
21
22
23
24
25