CHAIRPERSON: Good afternoon. I would like to just thank you very much for making yourself available for this hearing.

PROF VANEZIS: My pleasure.

CHAIRPERSON: Prof Vanezis, this is a hearing of various amnesty applications, which have been made to the Amnesty Committee of the Truth and Reconciliation Commission and the subject matter of the amnesty applications, is the death of Ms Ntombi Kubheka, which took place during or about May 1987.

I would just like to briefly introduce the people to you before we commence. The Panel consists of the three of us who you can see. On my left is Mr Ilan Lax, he is an Attorney. On my right is Ms Francis Bosman, she is an Advocate, and I am Selwyn Miller, I am a Judge. Appearing for the applicants are Mr Visser, I don't know if you can see him, with his Attorney, Mr Wagener, Mr van der Merwe and Mr Hugo and Mr Nel. I think he is behind us here.

Appearing for the victim's family is Mr Wills and appearing for an implicated person is Mr Samuel and the Evidence Leader for the Commission is Ms Thabethe. Those are the people who would be asking questions.

Professor, if we could get right into it straight away, unless there is anything that you wish to say before we start?

PROF VANEZIS: I don't think so, no.

CHAIRPERSON: Professor, would you prefer to take the oath or make an affirmation?

PROF VANEZIS: I will take the oath.

PETER VANEZIS: (sworn states)

CHAIRPERSON: Thank you very much. Professor, you have been called as an expert witness who has carried out a crania facial identification on a skull. Could you start off by please briefly tell us of your qualifications and your experience in this field, to set up the basis for being called an expert?

PROF VANEZIS: Yes. First of all I qualified as a Doctor in 1972 and I have been engaged in the field of forensic pathology since 1974 and during that time I have dealt with many situations where I had to assess skeletal remains.

During that time, in the 1980's, I developed an interest in facial reconstruction and developed a technique using computerised methodology to do this, in the late 1980's, which since then have been developing and have been improving ever since. I have also been engaged in other areas in relation to facial identification since the mid-1980's, such as the use of superimposition techniques for examining photographs and overlaying them onto images of skulls, to assess whether they belong to the same person.

And indeed I have been carrying out research in a number of areas related to facial identification and also the Director of the Human Identification Centre at Glasgow University which was established a few years ago now, initially being in London and then moved up to Glasgow, and where we carried on working in that area, we have been working there ever since.

CHAIRPERSON: Yes thank you. I see from your report and other documents that we have, that you have written a number of articles in the field as well?

PROF VANEZIS: Yes, I have written a number of articles and even papers at conferences, both oral and poster and have given advice in previous court cases in this area, which have been accepted by Courts both in the United Kingdom and abroad.

CHAIRPERSON: Yes, and your academic qualifications are reflected at the base of your report, your short report that you submitted, is that correct

PROF VANEZIS: That is correct, yes.

CHAIRPERSON: Yes. Professor, we have seen the video and we have read your short report, and we have also today been handed an article by, I have forgotten the name, it is Shahrom et al, on facial reconstruction. I wonder if you could tell us about what you have done in this matter, regarding the skull that is the subject of your video.

Then, once you have explained to us exactly what you have done, how you arrived at your conclusions and perhaps if you can reiterate them to us, then I will ask various legal representatives if they wish to put any questions to you.

PROF VANEZIS: Yes. The process that was carried out, was really in two stages. We received the skull from, sent to us by Dr Naidoo. There has been an anthropological assessment of the skull, for example that we were dealing with an African female, and basically we were then asked to carry out a reconstruction based on that basis, which we did.

The technique we used, was to place the skull onto a platform, acquire the information using a laser scanner by beaming a leisure scanner onto the skull to pick up all the co-ordinates of the skull, which are then transmitted and produced a three dimensional picture within the computer. This was then used as our scaffolding, shall we say, for us to place a face on top of this skull.

We use a face of someone of approximately the same type of age group initially, and it acts as a facial mask and that facial mask then moulds onto the underlying skull. Obviously it doesn't just kind of fit over like a glove, you have to put landmarks on the skull and put landmarks on the face, which correspond to each other, to be able to produce this. Also taking into account ...(indistinct) data that is available for that particular racial grouping.

So that was all carried out and we produced a facial reconstruction on that basis. That was really to see whether the shape of the face would be in keeping with the photographs that we received. It was ... (break in recording) ... we used a technique called video superimposition, which has been well established since the late 1970s, and used effectively in numerous court cases throughout the world, this technique.

What we do is, we capture the image of the photographs of the person in question, so that they are then recorded and frozen and placed onto a TV monitor. These are then, remain static as images, and then whilst we have them on the monitor, we then use a quality broadcast camera, that is a camera which is really the highest quality that we have available, so in order to do this, we employ the Media Services Department of our University, that does lots of professional work.

The quality broadcast camera is then, then takes an image of the skull, which is placed on a universal tripod, this is a tripod which is designed to enable us to move the skull, in order to take the skull in any direction we wish, so that we can tilt the skull forward, we can tilt it backwards, sideways and so on.

However, we need to compare landmarks on the skull directly with the photographs, so that we can align these images properly. The way we do this is we make markings on the photographs, we do this before they are captured, we make markings on the photographs, so that we can align for tilt and rotation, and we do similar markings on the skull as well.

And then, whilst having the photograph of the person on the monitor, on the TV monitor, we can also see an image of the skull come onto the TV monitor at the same time, which we can then move around to get it into the same orientation exactly as the other image, and we do this by using the lines that I have mentioned, and the landmarks.

Once we are satisfied with that, then we can do a direct comparison to see whether there is a contour match with the outlines of both images, match or whether they don't match. We found this to be a particularly powerful technique, because it also allows us to exclude quite quickly and to give negative identifications. However, I have to say that one of the difficulties, if the person has a reasonably normal face, is that if they match in every respect, we still cannot say a hundred percent that it is that person, but we can ...(indistinct) very strong evidence, that we believe that there is an excellent match, as with the case here.

Obviously if we have more than one photograph, we are then in a position to improve the situation slightly and confirm our contour match, say in one view, as we did in this case as well. We used two views and we were able to get a good match with both views. And so really, that is the extent of my work, I then produced a video recording for your purposes, so that it could demonstrate what we did.

CHAIRPERSON: Thank you very much, Professor. At this stage, I will ask Mr Visser, who is appearing for some of the applicants, whether he has any questions that he would like to put. Mr Visser?

CROSS-EXAMINATION BY MR VISSER: Thank you Chairperson. Hello Professor. Professor, yes, I received a photocopy of the extract from Computer Aided Facial Reconstruction by Shahrom, and I will certainly want to ask you a question or two about that.

First of all can you hear me clearly enough?

PROF VANEZIS: Yes, I can, yes, thank you.

MR VISSER: Professor, firstly, there are other experts in the field as well, I just wanted to know whether you know some of them, I am only going to mention two to you. Have you heard of a Dr Susan Loth?

PROF VANEZIS: Loth, Susan Loth, from, she used to be in Miami, I believe?

MR VISSER: Correct?

PROF VANEZIS: Is that the one, yes.

MR VISSER: Yes. And Mr, I am not sure whether he is a Professor or a Doctor, I am not sure about my pronunciation here, Mechmud Jassar Iskan?

PROF VANEZIS: Iskan is also from Miami, or was in Miami.

MR VISSER: And do you consider them to be experts in the field of Forensic Pathology?

PROF VANEZIS: Not Forensic Pathology, but Forensic Anthropology.

MR VISSER: Alright. Will you just tell us what the difference is, please, very briefly.

PROF VANEZIS: The difference is that a Forensic Pathologist is a Medical Doctor who then specialises or begins his career, or her career, as a Pathologist, in other words the person who diagnoses diseases, carries out autopsies from the point of view of determining the cause of death, manner of death, looking into the circumstances, so that they can carry out and ...(indistinct) of thing, and then they may well specialise in Forensic Pathology, which is more directly involved with criminal cases and so on, medical/legal cases.

A Forensic Anthropologist is a person who is involved more in looking to the identification process of a person, and also of course there are different kinds of Forensic Anthropologists. The one that most Forensic Anthropologists consider themselves, is a sort of sort of major in osteology-type work, so they rely on looking at bones for identification. They, because of this, they are in certain circumstances, because of the nature of their work, also quite proficient in being able to assess whether someone has been killed in a certain way, but that really is in the first instance in the ...(indistinct) of the Forensic Pathologist, they have to make ...(indistinct) assessment rather than ...(indistinct)

MR VISSER: Yes, thank you, Professor. Now Professor, you obviously would have read widely on your subject, are you acquainted with a book called Introduction to Forensic Sciences, Second Edition, edited by William G. Eckert(?), have you come across this work? I am not sure whether you can see where I am holding it up?

PROF VANEZIS: I know of the books that Bill Eckert has written in the past, and I know Bill Eckert personally as well. This is one of many he has written, I am not sure whether I have read that one myself, but I have certainly heard about him, yes.

MR VISSER: Yes. I will tell you why I refer you to this particular work, is because it contains inter alia a paper by Iskan and Loth, which deals with identification and that is what we are dealing with here.

I am going to point out, if you will allow me, that there are certain strong comparisons between the article which you have sent us a photocopy of, and what they have to say and if you will allow me, I would just like to refer to certain points.

The first thing is a question of age in this particular matter that we are dealing with, and as far as age is concerned, I would like to differentiate between the age at the time of death and the age of the skeleton since death. What I want to ask you Professor is, did you run any tests or try to determine the age since death - let's start first of all, with the age of the deceased when she died. Did you make any assessment of that aspect?

PROF VANEZIS: No, what we did was, we were instructed to actually carry out facial reconstruction. We took it on face value, the information that we received regarding the anthropology.


PROF VANEZIS: We did not make any further assessments.

MR VISSER: Alright, and that age that was provided to you was 35 years old, if I am not mistaken?


MR VISSER: Yes, alright. If the actual age was - well, may I ask you the question this way, normally speaking, what would the age variation be to effect a noticeable change in your enquiry into identification and reconstruction of the skull or is that a difficult question?

PROF VANEZIS: Well, let's put it this way, when you do a facial reconstruction, you can only approximate to the appearance of a person in any case. So there will be certain features, like details of the eyes and things like that, that you will not necessarily be able to get correct. So as far as fine features are concerned, it will be difficult. It is not that age sensitive a subject, so if we are dealing with say, somebody who is between 30 and 40, or even 30 and 50, I do not find any difficulty in being able to put a face onto someone within that age group. But obviously when it is a bigger discrepancy of about 20 years, and depending at what time of their life that is, then obviously there is a difference, we do make some changes.


PROF VANEZIS: But I must stress that all we did here was we just did the reconstruction to assess the shape of the face in relation to the skull, we did not rely on that to actually do our identification. Identification was done where we used superimposition.

MR VISSER: Alright. Yes. You see in your report under the heading, the second heading "Procedure Carried Out", the first numbered paragraph states -

"... the skull was examined both to confirm the autopsy findings and the anthropological data provided."

and that is what made me ask you the question, whether you went into that kind of detail.

PROF VANEZIS: No, we didn't go into - well, obviously what we do is we have a look at the skull and if there are any obvious discrepancies between what we are provided with and what we have, then we will examine it further, but it seems to me as approximately stated to us by the Forensic Pathologists who sent the details in.

MR VISSER: Yes, speaking of obvious differences, this skull which you projected on the video which we obtained from you, Professor, would that be the skull that you received from Dr Naidoo?

PROF VANEZIS: That is correct, yes.

MR VISSER: And can you, have you looked at that video lately again, before today?

PROF VANEZIS: I have seen it in the last month, I don't know, I can't say I have seen it in the last week or so, no.

MR VISSER: Can you perhaps recall the condition of the top part of the skull, any fractures or wounds that you can remember, what it looked like?

PROF VANEZIS: Yes, I mean there appeared to me to be an obvious gunshot entry wound.

MR VISSER: Well, on the video it appears to be a complete fracture of the skull? Would that be incorrect?

PROF VANEZIS: I would have to remind myself by looking at the video again.

MR VISSER: Yes, alright. If you are at a disadvantage, we can deal with this in another way, but I wanted to give you the opportunity if you could contribute at all to this, because what we know is a skull was found with an orbital bullet wound at the top of the skull and what we see on your video is a complete fracture of the skull on the top, which seems to us as laymen, completely different from what was found. But if you cannot give any opinion or make any comment on that, we can deal with that in another manner.

Could you just excuse us a moment? Alright, I am just told Professor, that we can show the video simultaneously and you will probably be able to see it. I would have done so if we weren't time restricted, as we are, and if I couldn't do it in any other way, but I think I can. In any event, if we cannot, we will come back to you on another day and take up some more of your time.

In paragraph 2, have you got your report in front of you, Professor?

PROF VANEZIS: I am afraid, I don't.


PROF VANEZIS: I don't have it with me, I am afraid.

MR VISSER: Oh my, that may be a bit of a problem, then I will have to read it to you. In paragraph 2, at the first page of your report you say -

"... it was then prepared for image capture, for reconstruction by replacing teeth in sockets and articulating the mandible."

What I want to ask you about that is that we have been given to understand that not all the teeth were present when the skull was exhumed and certainly some teeth were removed for DNA testing, presumably before they were sent to you. Can you at all recall what the position was as far as the teeth were concerned, particularly the anterior teeth?

PROF VANEZIS: No, I can't. All I know is that some were removed, and whatever we had, we replaced by reconstruction.

MR VISSER: Alright.

PROF VANEZIS: As far as we were concerned, what you have to appreciate what we were doing, we weren't doing an odontological examination, we were basically doing a facial reconstruction for identification here.

MR VISSER: Yes. Coming to that, well, just before we come to that, you say also in paragraph 2 -

"... standard methodology was then used to reconstruct a face onto the skull, using a 3D computer."

When you speak about standard methodology, would that be the methodology referred to in the article of Shahrom, that you sent to us?

PROF VANEZIS: That describes our technique that we used, that article, as you can see, was obviously a little while ago and it combines reconstruction and superimposition as well, but that is the technique, yes and it is a technique that we have been using ever since.

MR VISSER: Can you give us the date of which this work was published, perhaps? Would you have ...

PROF VANEZIS: It should be there on your, it should be there.

MR VISSER: Well, unfortunately it isn't.

PROF VANEZIS: I think it is 1996, I think it should say.

MR VISSER: Alright. We can possibly check that here. You see, coming to the question of the facial reconstruction, please stop me if I am wrong, but it seems to me that what you sent is a skull as well as a photograph or more, more than one photograph?

PROF VANEZIS: That is t, that is t.

MR VISSER: You must forgive me Professor, but I find that intriguing, because in terms of evidence in a court of law, I would suggest to you that a facial reconstruction would have value if you did the reconstruction and compared it later to a photograph and said "well, it compares" or "it does not compare", but here you seem to have had the photograph at the same time you were doing the reconstruction?

PROF VANEZIS: We were sent the photographs at the same time, we didn't look at the photographs, we did the skull blind, first of all, to produced the reconstruction and then looked at the photographs afterwards.

MR VISSER: Oh, I see. And ...

PROF VANEZIS: You must appreciate, what you have to appreciate is that the work is not carried out just by myself alone, I have a research assistant with me who actually does the reconstruction at my instructions, and then afterwards, we then look at the photograph after the reconstruction had been carried out.

MR VISSER: And, on that point, who was your assistant here?

PROF VANEZIS: It was my wife, Maria Vanezis, who carried it out.

MR VISSER: What this article of Mr Shahrom says at page 198 is that in marking exact locations, peg marks, sometimes requires subjective judgement that is required. Do you go along with that?

PROF VANEZIS: We are talking about, are we talking about clay model facial reconstruction here, or computerised reconstruction?

MR VISSER: No, no, we are talking about facial reconstruction first of all?

PROF VANEZIS: Yes, I mean there is - what you have to appreciate is that the landmarks on the skull are chosen by a person who is trained in knowing where landmarks are, in anatomy and so they can actually place the landmarks correctly, but, and so there is a degree of skill, that is used by the person carrying out the work.


PROF VANEZIS: From that point of view, it is subjective.


PROF VANEZIS: And then the landmarks on the face were also put again in by the person again, on the face, again that is subjective.

MR VISSER: Who would have done that in this case?

PROF VANEZIS: That would have been my wife.

MR VISSER: And any subjective judgement would have been exercised by her, in doing so?

PROF VANEZIS: ...(indistinct), well, what you have to appreciate is that the reconstruction is like any technician, the reconstruction is carried out under my instruction, the landmark placement by someone who is trained, is fairly straightforward. Although it is a subjective judgement, you are only allowed a certain degree of limitation and freedom to make that judgement, In other words you have to sort of follow the rules and place the landmarks where the skull dictates. So for example if you can see clearly the point such as the end of the nose, the size of the ...(indistinct) and so on in the skull, they are straightforward to place, but occasionally you need to use your subjective, your own assessment to do this, and yes, she obviously had this discretion to do this, yes.

MR VISSER: Yes. You just mentioned that it is rather straightforward, but I gathered from page 198 of the article which you sent us, right at the top in the right-hand column, it says:

"A further drawback is that the error of the positioning peg marks on the images, may be as high as two to three millimetres in all directions from the true location."

So, it does not appear from a reading of this, that it is all that straightforward, it takes some skill?

PROF VANEZIS: Well, what you have to appreciate is that I have sent you a paper to give some guidance on our technique that we use, what you've got to appreciate is what you have there is based on the methodology. The principle is the same, but the software and the technology is going back to 1990 and slightly earlier. We have got newer software which enables us to be much more accurate with our landmark placements and the reason why I sent you that paper is because that gives you reconstruction and it gives you superimposition together. But we do have a more recent paper which I told Dr Naidoo about, which should be made available to you, it was published at the beginning of this year in Forensic Science International, which would give - if you really want to go into that detail regarding the reconstruction, then that is the paper you should have a look at, that is the most up to date.


PROF VANEZIS: This is a problem with publishing papers in the past, you know, you don't stand still, you develop things you see.

CHAIRPERSON: Sorry Mr Visser, if I could just intervene. Professor, is that article that you are referring to the one that you've made reference to in the footnote of your report? An article by yourself, Forensic Science International, 108, pages 81 to 95?

PROF VANEZIS: That's come out I think, in the February 2000, that one. That is the one I think.

CHAIRPERSON: That is the one you have made reference to now?


CHAIRPERSON: Thank you. Sorry, Mr Visser.

MR VISSER: Thank you Mr Chairperson. Just allow me a moment. I am sorry, if I did ask you this question before, because I am following your line of evidence, with my questions, please forgive me, but did you make an assessment of the age of the skull since death?

MR VISSER: You didn't.

PROF VANEZIS: No, no, all I did was, I looked at the skull, saw that it was an adult female and I did not make any ...(indistinct) assessment, no.

MR VISSER: Alright. Could we possibly agree on categories of certainty as far as identification is concerned. Would you agree with four possible categories, the one is possible, the second indeterminate, the third is a positive identification and the fourth possible category may be called perhaps rule-outs? Would you agree broadly speaking that we can talk of those four categories as far as identification is concerned?

PROF VANEZIS: Rule-out, positive, indeterminate and the other one you said was possible?

MR VISSER: Possible, yes.

PROF VANEZIS: I mean that is obviously the big, we have a lot of flexibility ... but broadly speaking, yes.

MR VISSER: Yes, well are you going to concentrate in fact, on the possible. Because let me immediately put it to you that in terms of the article, looking at page 200 and I said to you before that Iskan and Loth appear to be in conformity with this article and what I was referring to was at page 200 where this article says -

"... however, we must stress that the degree of medical certainty of this method (and this refers to a video superimposition) cannot be considered adequate for positive identification. It is used in combination with other corroborative evidence when confirmatory evidence, such as fingerprinting, dental records and DNA profiling are not available."

So what it says, if I understand the English, is that it is a good, it is a useful method of providing assistance for identification where other foolproof, if I may call it that, information does not exist. Would you agree with that?

PROF VANEZIS: That is quite correct, I agree, yes.

MR VISSER: Now, the degree it says -

"... the degree of medical certainty of this method increases when more anatomical landmarks on the ante-mortem photograph match those on the skull, especially when the dentition landmarks are also included."

PROF VANEZIS: That is correct.

MR VISSER: Those were not included in this particular case?

PROF VANEZIS: That is t.

MR VISSER: Yes. The value of superimposition has been challenged on the basis that the alignment and enlargement factors are too variable and you refer to Defaur - do I pronounce that correctly, or the article does, emphasise strongly that -

"... this method cannot be used for positive identification, since the magnification and angulation of the original picture, are unknown. It has more value in exclusion, than in positive identification ...",

would you agree with that last statement?

PROF VANEZIS: Certainly you cannot positively identify, well, in one or two circumstances you can, but mostly you cannot.


PROF VANEZIS: I think one has to be also cautious on the exclusion side as well, again you can mostly exclude, but again bearing in mind there's sometimes difficulties with photographs and you may not be able to exclude a hundred percent, again.


PROF VANEZIS: But generally speaking it is slightly better for exclusion than for, you know, giving you a positive ID, ...(indistinct).

MR VISSER: Yes, what you have just said is in fact confirmed by Dr Loth, to whom I incidentally spoke before, absolutely confirmed.

It goes on, and you must forgive me for reading this because this is crucial in our submission.

PROF VANEZIS: Of course.


"... a misidentification by photographic superimposition of skull and anti-mortem photograph has been reported ..."

you say - well, they say one, well let's leave that aside and we go on to read -

"... the error was confirmed after positive identification using fingerprints ...",

so what he is saying here is there is on record a misidentification. Be that as it may, the article goes on to say -

"... this illustrates the danger of over-estimating the capability of this method. In conclusion, despite some limitations in obtaining a positive identification, the Computer Aided Facial Reconstruction and Video Superimposition methods are very useful in enhancing the process of identification, especially in the case of a severely decomposed or mutilated body and skeleton remains."

That would meet with your approval as well?


MR VISSER: Alright. You see Professor, why I read that to you is because it appeared to us when we received your report and when we received this article, that you were aligning yourself with the article and yet, if you look at your results and conclusions at the last page thereof, you say -

"... I am satisfied that there is an excellent match between the photographs examined, and the skull in question, and I am of the view therefore that it is (and I stress this) highly likely that the skull is part of the remains of Ntombi Kubheka."

PROF VANEZIS: That is right.

MR VISSER: Now, that is not the same as saying it is a possible?

PROF VANEZIS: It is not the same as saying it is a positive.

MR VISSER: That is true, that is true. But are you then saying it is indeterminate?

PROF VANEZIS: Not at all, I am saying that in my experience and bear in mind also, I have now another seven or eight years experience in this work from when that article was written in, well that was published in 1996, it was written a lot earlier than that and based on software that we used in the earlier stage, and I spent a lot of time looking very carefully at the skull and the photograph in different positions, we took a great deal of care, and this is a very carefully considered opinion and as you can see, I stress that it is not a positive identification, but nevertheless it is still an excellent match.


PROF VANEZIS: And it is on that basis that I say "very likely".

MR VISSER: Likely, alright. Now, coming to the category of positive identification, I noticed in the article that the author there said that, or seemed to suggest that video superimposition cannot ever be a positive identification and if that is so, certainly Dr Loth disagrees with him.

PROF VANEZIS: I have never said that.

MR VISSER: No, no, I am not saying you did, I am saying that the article seems to suggest that and I read that to you just now, but if that is the suggestion, can we be in agreement that there could be in fact, such a positive identification that it could be beyond reasonable doubt, in fact?

PROF VANEZIS: In this case you mean?

MR VISSER: No, no, in general.

PROF VANEZIS: In any case? Well, as the article says where you have a number of landmarks, more landmarks that you can compare and also if you have teeth marks as well, then you can make an assessment where you can actually do a positive identification. This has been done on teeth marks, on a skull photo superimposition, and we have done it ourselves with a case that was presented in Italy three years ago. But the problem remains that in virtually all other cases, unless you have some quite extraordinary bony abnormality of the skull, which you can match up, then I would have to say that you cannot positively identify.

MR VISSER: Yes, well you have just come very close to what I am going to put to you straight away, because in the Forensic Science Second Edition, which I mentioned to you earlier, there is a paper by Iskan & Loth and what they say is this, and I would like to read it to you, and forgive me, it is a little long, but I believe it is relevant. It starts off by saying -

"... numerous prospective matches survive initial screening, but most of these will wind up in the "indeterminate" category. This is due to the fact that a large number of very similar features are shared by the members of any given age, sex, race group or nationality and this cannot be deemed diagnostic identity."

I then skip two sentences and I read the following -

"... if no idiosyncratic characteristics or factors of individualisation can be isolated and matched, the comparison can only be considered indeterminate or inconclusive."

So what they are saying is you start off with a possible, but if you cannot find individualistic features in comparison with either what you know from evidence of this person, anti-mortem or from the photograph, in your comparison with the skull that you are examining, it can at best be termed indeterminate.

PROF VANEZIS: Well, I think it is perhaps an emphasis situation here, I am saying that with, when I look at the skull and photograph, I look at as many landmarks as I can on both and we do quite close contour match, and we use two different photographs. My view is that you can say more than indeterminate, my view is that if you get an excellent contour match, that says a great deal about the matching process, but I would agree to the extent that it is inconclusive when we are saying whether or not we are looking at positive identification.


PROF VANEZIS: It cannot conclude it is a positive identification, no, I agree with that entirely.

MR VISSER: So, can we agree then that you and Dr Loth and the rest agree that in the absence of an individualistic discernible similarity, there cannot be a positive identification, am I putting that correctly?


MR VISSER: Thank you. How many landmarks do you use when coming to conclusions in regard to identification from radiography of the skull?

PROF VANEZIS: We don't use radiography of the skull.

MR VISSER: Alright. How many landmarks do you use on the skull of Ms Kubheka here, in this case?

PROF VANEZIS: It varies. I think in this case we used 37, but sometimes we use 40, it depends on whether or not we feel it applicable. We have used less. As far as the superimposition is concerned, what I stress is important here, is contour matching. There are landmarks at certain points, yes, but it is important that you match the profiles, the contours of the actual image.

MR VISSER: And how many landmarks would you use for contour matching?

PROF VANEZIS: Well, you don't use landmarks for contour matching, as such, you are using contours.

MR VISSER: Right well, can I just refer you to a work which is called Forensic Analysis of the Skull by Iskan and Helmer, you have heard of Helmer?

PROF VANEZIS: Richard Helmer, yes, he's with us today in South Africa, yes - sorry, in Washington, I beg your pardon.

MR VISSER: Well, if he is here, I would be surprised to hear. I was looking up everybody I could find.

At page 180 we find the following, it says under "Conclusions - Discussions and Conclusions",

"... this chapter presents an analysis of the relationship between the landmarks on the face (which you have been talking about) and corresponding locations on radiography of the skull."

and he says this, he says -

"... four groups of standards, using a total of 58 examining targets have been established for identifying a skull by superimposition. These include eight targets for the examining lines, nine for contour curve (I think which is what you have referred to), 13 for soft tissue thickness, and 28 for the relevant relationships of the landmarks."

Would that be an over-kill or would you say that would be what is necessary to do?

PROF VANEZIS: I am reasonably familiar with Iskan and Helmer's book, because I reviewed it for one of the journals a few years ago, there are a number of different articles which state a number of different opinions in that book, as you would appreciate once you peruse it, and there is quite a lot of repetition as well, which sometimes is a good thing in a book like that. But I have to say at the end of the day, the technique that we use, we're quite happy with and we're fairly rigorous in what we carry out, and because - we don't use radiography, because radiography, I believe is not relevant to these particular images, but if you use radiography, then there are different issues involved.

On would need to have a much fuller discussion on that area if that is what you point to, but I would need to have the book in front of me as well to refer to it.

MR VISSER: Yes, I was suggesting that we go to Glasgow, but that suggestion fell on deaf ears, so we will have to make do with what we've got, Professor.

PROF VANEZIS: Okay, no problem.

MR VISSER: In any event, at 180 the authors also say this, they say -

"... if all visible targets are completely superimposed, using our system of standards, it can be considered a positive identification."

That is the 58, they say if they are completely superimposed, then it is positive.

"... on the other hand, a match is ruled out if two or more targets cannot be superimposed."

and then it says ...

"... finally, when only one target does not fit, the possibility of identification exists."

So coming back to the categories, he says if there is one that does not fit, it is still a possible, when two don't fit, it is out.

Would you go along with that in principle?

PROF VANEZIS: This is entirely arbitrary, this is work that's been carried out by some Chinese workers in Tailing city, Lan and others, and it is based on Chinese standards and looking at radiography in Chinese subjects in all the Chinese over a period of time, and basing it on their work. We in fact tried to repeat their works and we found that we couldn't. Personally I think quite frankly, what they are suggesting is perhaps too rigorous and your word "over-kill", I think is quite appropriate in its use. I think one has to use one's judgement with these sort of cases, and that is precisely what we did. We are obviously aware of all the different methods involved, but at the end of the day, we use a method which we happy with and we stick by it. And that is how we do it.

MR VISSER: Well, it is my job to ...(intervention)

PROF VANEZIS: ...(indistinct) different. I'm just saying, if an expert sometimes uses slightly different methods for, you know - if you talk to, if you actually talk to the Chinese researchers, they will tell you that they get positive identification every time, in most cases. ...(indistinct) which cannot be the case. I certainly have some doubts.

MR VISSER: Yes, well, thankfully you didn't come to a positive identification, otherwise we would have been here for a few days. But what you are saying, we already have enough trouble with each other because you said it is highly likely.

PROF VANEZIS: That is right.

MR VISSER: What I am going to suggest to you now is what follows from Iskan and Loth which I started reading to you, is the following. It says -

"... even if there appears (and I am talking about superimposition all the time) even if there appears to be a strong probability, classification must be in this category."

I am sorry, I am told I have left out a sentence, may I start over again. It says -

"... even if there appears to be a strong probability of a match, without a unique feature to set that individual apart, the final classification must be in this category, that being of indeterminate or inconclusive."

PROF VANEZIS: Well I mean, that is their view, my view is ...(indistinct) positive identification, but I don't agree with them with the concept of indeterminate. Indeterminate means you just don't know, and I don't think we are in that situation here.

MR VISSER: Yes. But it is something stronger than possible?

PROF VANEZIS: Well again, if we define the word possible, I mean you propose the four different categories and you said possible and I questioned you about the flexibility of possible, I would say from my point of view, if we're looking at the degree, or should we say of a probability of it being the same person, then to my mind it is pretty high, but not positive.

MR VISSER: Not positive. Yes. Would you allow me a moment to see whether there is anything else that I have to take up with you now.

How many photographs did you receive of Ntombi Kubheka?

PROF VANEZIS: I know I used two, I think I received, I sent them all back actually, it was four I think, I recall four, they may be repeats of the same view, but not as clear, but certainly I know I used two, two ...(indistinct) were used.

MR VISSER: And would they have been sent back to Dr Naidoo?

PROF VANEZIS: That is correct, yes, sent back to him, yes.

MR VISSER: On the photograph which you used on the video, would it be correct to say that the outlines on that photograph were rather fuzzy?

PROF VANEZIS: Let's put it this way, they weren't crisp as you would expect, you know the best photograph photographic quality images, but they were adequate.


PROF VANEZIS: If you say they were fuzzy, the degree of fuzziness, I mean obviously they could be improved.

MR VISSER: Yes, I am going to argue, Professor, and I am putting this to you so that you can make comment on it if you wish, I am going to argue that the eyes on the superimposition of the photograph on the skull of vice versa, is not a match.

PROF VANEZIS: Well, I disagree. If I examine the superimposition and notice that the eyes were not a match, then I would have said straight away that in my view, this is not the same person. So ...

MR VISSER: Yes. So only on the eyes not lining up, you would have been prepared to say "this is a rule-out"?

PROF VANEZIS: Well, one of the difficulties as you appreciate with all of this, is getting the orientation correct and getting everything aligned in the proper position, and if my view I didn't think that the eyes matched, then I would have said it couldn't have been the same person.

MR VISSER: Yes. We looked very carefully at this, and I tell you what we did, we marked on the video screen, on the TV screen, we marked the horizontal and the vertical lines and to me as a layman, I must tell you, on the left to right scan, there is no way that eye fits into the socket. You don't agree with that?

PROF VANEZIS: Well, no, I don't, no.

MR VISSER: Alright. And when one looks at the right-hand jaw, the picture that we have has a position of the person looking to her left, and on the scan, both ways, I would suggest to you that where one expects the jaw bone to be, it just isn't when you run the scan both ways.

PROF VANEZIS: Well, all I can tell you merely is that I did very carefully look at this and it was my view that it was, you have to take into account of course the degree of skin thickness in an area, the distance from the edge of the skin to the actual bone itself, so these are all factors that we took into account.

MR VISSER: You know, we might be compelled to take up the offer of looking at the video. I'm just wondering how much time is that going to take, is that going to be difficult?

MR LAX: You can stop it and talk to him.

MR VISSER: Perhaps, if you don't mind, could we do that, Professor, thank you.

PROF VANEZIS: No problem.

MR VISSER: If you could set it up.

MR LAX: It is done, it is all ready to roll.

MR VISSER: Well, let's roll it.

CHAIRPERSON: Do you know which part to get to.

MR VISSER: I am just interested in the scans.

TECHNICIAN: Professor, can you see the video?

PROF VANEZIS: Yes, I can, thank you, yes.

MR VISSER: It's so short, perhaps we can just run through it and we can stop it at the appropriate ... That would be your wife, would it be, Professor?

PROF VANEZIS: That is correct, yes.

MR VISSER: Thank you. Could we just go a few frames back? Thank you Chairperson, yes, I just looked down for a moment, and I missed it. That is the one. Now, Professor, how would you describe the top of that skull as far as the fracture is concerned?

PROF VANEZIS: Okay, well, that doesn't actually show a fracture, what that actually shows is lots of captured data by the laser scanner, which it does this in virtually every image that we produce, the top of the head has an area where the ...(indistinct) on all our scans. The reason for that is because the laser light basically glances off the top of the head and you don't get a proper image capture in this area, but because we are dealing with faces, this is an area which is of, basically of little importance to us and we're concentrating further down on the face. This isn't in fact a fracture, this is in fact just an artefact really.

MR VISSER: Yes. What happened to the skull after you had completed your examination, was that sent back to Dr Naidoo as well?

PROF VANEZIS: Yes, it was, yes, a courier picked it up and sent it back.

MR VISSER: Yes. Now, I want to ask you and I am not sure whether you can see me, whether you used this photograph in order to run your tests?

PROF VANEZIS: Sorry which photograph is that, on the reconstruction?

MR VISSER: I am told we must just wait, he is going to zoom in on the photograph. I don't know where the camera is.

MR LAX: Here it is, it is on you. Come in a bit more. That is it.

MR VISSER: Would that be a photograph which you used?

PROF VANEZIS: I am not sure I used that one, I would have to check the photographs that I was sent exactly, but I would have to see the video superimposition.

MR VISSER: Can I show you another one?

PROF VANEZIS: I certainly used one from the side.

MR VISSER: Can you possibly see this one?

PROF VANEZIS: I used that one, yes.

MR VISSER: Did you use this one?

PROF VANEZIS: Yes, I can see that. That's more like it, yes.

MR VISSER: Yes, and are you saying that according to your opinion, the reconstruction which we saw on the video is similar to the photograph, to the extent that it is highly likely to be the same person, or are you basing that on the video superimposi-tion?

PROF VANEZIS: I never said it was the reconstruction, I am basing my views on the video superimposition, on the photographs I received of the skull, the reconstruction was just to show the shape of the face was similar, bearing in mind it conforms to the overall general structure of the skull, but that is as far as that went.

MR VISSER: Alright. When you say in your "Results and Conclusions" that -

"... the reconstructed facial image reflected within acceptable limits, the photographs of the deceased Ntombi Kubheka ..."

are you again referring to your video superimposition or are you now referring to the reconstructed skull?

PROF VANEZIS: Yes, what I am saying there is that the general shape, I am not talking about individual small features, I am talking about the general shape of the face, because from reconstruction it is very difficult to get things exactly, shapes of noses, ears and ...(indistinct), that sort of thing, but certainly the general shape was within acceptable limits, were similar.

MR VISSER: What are these acceptable limits ...?

PROF VANEZIS: General shape.

MR VISSER: What are the acceptable limits, Professor?

PROF VANEZIS: Well, we bear in mind that we cannot judge exactly what the eyes looked like, we cannot judge exactly what the nose looked like, you cannot judge exactly what the ears looked like, we cannot judge exactly what the lips looked like, but the contours, the shape of the jaw, the cheekbone structure, forehead structure, that sort of thing, the general shape of the face, there was nothing dissimilar about it. In other words it looked as if it could have come from the same source as the photograph. In other words the skull of the reconstruction, and the photographs, the skull ...(indistinct) the photograph was the same person or the same source.

MR VISSER: So the shorter answer, you considered it to be a possible?


MR VISSER: Thank you. If we could then perhaps look at the video again, thank you.

I am sorry Sir, just while we are waiting, you told us about the laser beam, are you saying that we see on the video isn't in fact a hole in the head, to put it that way, is that what you are saying?

PROF VANEZIS: That is quite correct, it is not a real hole in the head, it is ...

MR VISSER: It is a shadow?

PROF VANEZIS: ... it's the fact that the laser cannot pick up that part of the head.

MR VISSER: Thank you.

PROF VANEZIS: It is not a hole in the head, no.

MR VISSER: Right, thank you.

MR LAX: Just stop there.

MR VISSER: Could you just stop there?

MR LAX: Sorry Professor, it is Lax on the record, I am one of the Panel Members. Are those the sorts of marks that you were talking about, the landmarks, the little dots?

PROF VANEZIS: Yes, these are what we would call face marks or landmarks, yes, that are placed on the face, correct.

MR LAX: Thank you.

MR VISSER: Just leave it to run from now, thank you. Would that be the revolving to get the angulation right, which you explained earlier to us, Professor?

PROF VANEZIS: No, not at all. The angulation that I explained to you was superimposition work and really had nothing to do with this reconstructive face. All I am doing here is, I am just demonstrating the reconstructed face in different positions.

MR VISSER: Alright. That is a photograph which you had in your possession and sent back to Dr Naidoo, is it?

PROF VANEZIS: That is correct, yes.

MR VISSER: Thank you. Professor, I am advised that one must now look, when the scanning goes on, from left to right, to the right socket of the skull. The first one lined up. Did you notice the second eye?


MR VISSER: Just watch it again. That is the eye already, now we come to the socket, that one lines up perfectly. The right eye lines up, the left eye doesn't, don't you agree with that?

PROF VANEZIS: My view was, when I did this and was able to see it with obviously better quality images than I can see it now, I was happy that they lined up.

MR VISSER: Let's see how we go on. Perhaps you can now also look at the right-hand jaw. Do you say that that lines up?

PROF VANEZIS: Yes, in my view it does, yes.

MR VISSER: Professor, lastly, I didn't direct your attention to it, but there also seems some doubt that the mouth actually lines up with the skull on that last superimposition, but I take it you will say that when you considered the scan, you thought it did?

PROF VANEZIS: You're perfectly correct, yes.

MR VISSER: We are looking at the jaw particularly now, and that is what we find. Again do you say that that is a perfect line-up, a perfect match?

PROF VANEZIS: I am saying is that it certainly is an excellent match, yes, and I am quite happy with the position we placed the skull in and the matching, yes.

MR VISSER: Yes. There is a lot to look out for, but of course the eye, you will accept that as far as the eye is concerned, that remains a problem with us and I am not going to every time refer to it, but you can perhaps look at the eye again, apart from the jaw.

That superimposition it would appear, you did on a different photograph, one of the person looking to its left, would that be correct, not a flush-on photograph?

PROF VANEZIS: That is correct, yes.

MR VISSER: Would that have been the one that I showed you just a moment or two ago?

PROF VANEZIS: Yes, I think it was the second one you showed me.

MR VISSER: Yes, thank you. Perhaps to save time, may I ask you, Professor, on the hypothesis that the eyes did not line up, I believe you have already stated you would have ruled it out as a match?


MR VISSER: So much the more if the jaw does not present a match, and perhaps the position of the mouth, then so much the more it would be ruled out as a non-fit?

PROF VANEZIS: Well, that is correct, the more areas that you find matches, the more doubt creeps into a situation, yes.

MR VISSER: Yes, and I have asked you this before, but just to make it absolutely clear, there is no individualising phenomenon in the skull compared to the photograph that you could identify to say "this belongs to this person and no one else in the world"?

PROF VANEZIS: That is perfectly correct.

MR VISSER: Lastly, I just want to refer to Forensic Sciences again, Loth and Iskan, they say at page 359 under "Skull to Photo Superimposition", they say this -

"... this method can be attempted if photographs or X-rays are available of individuals that answer the description derived from a skeleton with the skull and mandible intact."

I then go on a sentence later -

"... if the bony landmarks of the skull align with the size, configuration and replacement of the features in the photograph, a match becomes possible, but cannot be considered conclusive. If however, superimposition reveals an obvious disparity, such as the position of the orbits or the length of the nose or size of the chin, that individual can be eliminated as being the victim."

That is basically I think what we have come to an agreement on, already?

PROF VANEZIS: That is correct, yes.

MR VISSER: Yes, Professor, you might welcome the fact that I don't believe I have any further questions right now, the problem which we have is that we have not been able to discuss the evidence that you have given this morning or this afternoon, here in South Africa, with Dr Loth for her comments. I don't believe that there seems to be any disparity between your evidence and hers as I've pointed out before, but if it does happen again, and I would apologise if it does, that we need to talk to you again, I might have to ask for that to be arranged, and you can blame me for taking up your time, but I hope that it won't be necessary.

PROF VANEZIS: It is quite all right, I am very happy to cooperate in any way.

MR VISSER: Yes, and thank you for that. Could you just perhaps bear with me a moment to make absolutely sure that I didn't leave out something material? Chairperson, that is as far as we can possibly take it at the moment.


CHAIRPERSON: Thank you Mr Visser. Do any of the other legal representatives for the applicants, wish to put any questions, Mr van der Merwe, Mr Hugo, Mr Nel? None? No further questions from the applicants.




CHAIRPERSON: Mr Wills, who appears for the victim's family, will now put questions, Professor.

CROSS-EXAMINATION BY MR WILLS: Thank you Mr Chairperson. Professor, John Wills, I am appearing for the victim's family in this matter.

Just, your wife, Maria, could you place on record her qualifications, Sir?

PROF VANEZIS: Yes certainly, she has a, she's a Radiographer by profession, she has ...(indistinct) started her career. She does diagnostic radiography and therefore she was ...(indistinct) to using a laser scanner and the health and safety implications involved in that. She then got a first-class degree in Health Studies, which included ...(indistinct) Anatomy, Physiology and so on. She has been working with facial reconstruction work now since approximately four years. She's actually carrying out a Doctorate in looking at the validity and the psychology behind recognising faces from facial reconstructions, where she's done a ...(indistinct) at Glasgow University.

MR WILLS: Yes thank you. The software programme that you use, I understand it is largely your own creation?

PROF VANEZIS: The software programme was, the intellectual part - I am not a computer expert, don't get me wrong, I basically, I want a problem solved and what I do is I go along to my computer people and we put together what is an acceptable package. They write the software to our specifications basically, and this was an improvement on the much earlier software that was available. So it is really ...(indistinct) software.

MR WILLS: How current is the software that you used, when was it finalised?

PROF VANEZIS: This software was produced and started to be fully used about two years ago, and improvements are being made to it all the time.

MR WILLS: Thank you. Now I see that one of the problems that you have in this field is to determine the correct angulation of the skull to the photograph.


MR WILLS: The article that we have been provided earlier, indicates that in order to minimise this problem, it is helpful to have an anti-mortem photograph with the face looking directly at the lens, would you agree with that?

PROF VANEZIS: That is right, yes, yes.

MR WILLS: Now if I can just show you a photograph, and I believe this is one of the photographs that you used, it is a copy of one you have sent back to Dr Naidoo, did you use this photograph?

PROF VANEZIS: Yes, I believe I did, yes. Or I think I did.

MR WILLS: As I understand this, this is a photograph of the deceased looking directly at the camera?

PROF VANEZIS: Yes, what you have to understand is that my definition of what I would regard as directly inside the camera, it may be slightly different to a lay person's understanding of it, because we have different terminologies, such as the Frankfurt plain, in other words kind of standardised the view of the photograph and so there may be slight variations from this. I mean, even the frontal photographs, because of that, there still remains a slight problem in alignment.

MR WILLS: But the photographs you used, you are happy were good enough for the purposes of this?

PROF VANEZIS: Absolutely yes, indeed, yes.

MR WILLS: Just finally, there is obviously points of convergence in the superimposition that you performed, and my understanding is that you didn't come across any point of divergence, is that correct?

PROF VANEZIS: That is correct.

MR WILLS: Yes. Thank you Professor, I have no further questions.


CHAIRPERSON: Thank you. Mr Visser has asked if he can ask a question arising out of that, but you can go ahead and ask, Mr Visser.


Professor, it is actually something I forgot and not arising from my learned friend, Mr Wills' questions. May I ask you, can you give us some assistance as far as DNA testing is concerned, because I've only got one question in that regard. Perhaps I should ask you the question, and you can tell us whether you can help us or not.

PROF VANEZIS: Sorry, could you just repeat that, you didn't come over very clearly there.

MR VISSER: I am sorry. I am not entirely certain whether it is part of your field of expertise to express an opinion on DNA testing, but I have one question and if I could pose it to you and perhaps you could help us, and perhaps you cannot, if you'd just say so.

PROF VANEZIS: If you put it to me, I will yes.

MR VISSER: Would it be true or false to say that DNA testing where no previous record exists, can assist in making a positive, I am sorry, cannot assist in making a positive identification of a victim, in other words, no record, DNA testing can't give you a positive identification?

PROF VANEZIS: That is probably correct, what you need is, when you are doing a DNA, when you have a DNA profile that you, say got from bone, ...(indistinct) or DNA STR, that sort of thing, you need to compare it with something and if there is no comparative record, then that DNA profile just remains on file, with no positive identification made.

MR VISSER: And in fact it would be the same as far as comparison of abnormalities of teeth are concerned, you would also need a previous record?

PROF VANEZIS: Well, of course, yes. You have to attach what was there before life to what is there after death.

MR VISSER: Yes, thank you. Just the last point here is, as far as DNA testing is concerned, if one did not have a record of the deceased, and you for example took a sample of a living close relative, I am informed that the only purpose that you could achieve would be an exclusion factor, in other words, to be able to say "this relative is not a relative of the deceased", but not vice versa to say "this relative is in fact a relative of the deceased"? Would that be correct?

PROF VANEZIS: Now you are getting to the point where it becomes difficult for me.


PROF VANEZIS: ...(indistinct)

MR VISSER: All right, it is not that important, thank you very much, Professor.


CHAIRPERSON: Thank you. Mr Samuel, do you have any questions you would like to put?

MR SAMUEL: No questions.


CHAIRPERSON: Ms Thabethe, do you have any questions you would like to put to the Professor?

EXAMINATION BY MS THABETHE: Yes Mr Chair, thank you.

Professor Vanezis, in this case of Ntombi Kubheka where a DNA test has been unsuccessful, we have failed to get a dental ...

PROF VANEZIS: I cannot hear you. I am very sorry, I cannot hear a thing, you have to speak up, I'm afraid.

MS THABETHE: Okay, I will start again. I am saying in this situation, or in this case of Ntombi Kubheka where we have failed or where we have failed to get a successful match in the DNA test, the same applies with dental and fingerprints experts, would you say it is reasonable for us to rely on your expert evidence on the facial reconstruction?

PROF VANEZIS: From the point of view that you have no other available evidence at this time, you can make some kind of judgement as to the likelihood of the identity but you cannot positively, a hundred percent, identify just from my technique alone.

MS THABETHE: Thank you. You have also indicated in your report that it is highly likely that this is Ntombi Kubheka, what I want to find out is what made you to reach a conclusion that it is highly likely as opposed to a positive identification?

PROF VANEZIS: I have done many cases of superimposition and from my experience of matching people and looking at exclusions and so on, it is my judgement, my opinion that there will be, because there was an excellent match of the superimposition to the skull of the photograph, it was my view that it would put it into the bracket of being "highly likely". And looking at two views as well, not just one picture, but looking at two views of a same face, one slightly different, one at a different angle.

MS THABETHE: Just to further that question, what would it have taken you to reach a decision that it is a positive match? What would you have required?

PROF VANEZIS: Yes, the only thing I would be - if I was to say it was a positive match, what I would have needed would have been some highly significant abnormalities, say on the teeth, a chip on the tooth or something like that, with the face smiling and one able to see the two images of the tooth on the skull and a tooth in the mouth. Or a very, very close mouth formation, say even from an old fracture of the skull, which may have reflected itself in the photograph. That's the kind of thing, but there wasn't anything individualising like that, to be positive.

MS THABETHE: Thank you. What would you say to an argument that any African female person's photograph would have been taken and the likelihood would have been a match? What would your response be to an argument like that?

PROF VANEZIS: Can you say that again, I want to get this right, I didn't quite hear the beginning.

MS THABETHE: Okay, what would be your response to an argument that you could have taken any photograph of a female African and it could have given you, the likelihood is that it could have given you the same match? What is the possibility or the probabilities of that?

PROF VANEZIS: Well, I have looked at a lot of faces and a lot of skulls, we're all individuals, if we look at our contours on our skulls and our faces, there are slight differences which we can see. In my view, if the person is of the same race, there is still a lot of variation between people of the same race, which one can pick up. We are all individuals and all I would say to that is, that I would still say that this was highly probable it was of this person, based on what I have done and based on the contour matching that I have carried out.

I have done, I also looked at skulls of other races as well, I look for example at South Americans from Chile and the same thing applies there. And then other groups, I've looked at Sri Lankans and again the same thing applies there. So everybody as individual within that racial group, the same standards apply. Obviously if you have, say a European and comparing it with an African skull, there are things you will pick up on the skull anyway, regardless of the facial features, which will tell you they are different. So I have no problem with looking at images from people from the same race, and drawing conclusions like the one that I have.

MS THABETHE: What about images from people who are related, would there still be distinct features between people who are related?

PROF VANEZIS: Sometimes there are some slight features which are similar, but there is enough information to show enough dissimilarity between them, so that you can pick up.

MS THABETHE: Professor Vanezis, my learned friend, Adv Visser has referred to Dr Loth, I just wanted to find out, in our own opinion, or if it is a fact you can say so that it is a fact, as an Anthropologist, would Dr Loth have an experience in giving expert evidence on facial reconstruction?

PROF VANEZIS: I am sorry, I didn't catch that very well at all, all I heard was facial reconstruction, I am sorry.

MS THABETHE: Would Dr Loth who is an Anthropologist, have experience in giving expert evidence on facial reconstruction?

PROF VANEZIS: Sorry who? You said who was ...(indistinct) the person. Sorry, I didn't catch that one.

MS THABETHE: Dr Loth, Dr Loth?

PROF VANEZIS: Mr who? Loth?




PROF VANEZIS: Right, okay. If she what, sorry, I haven't ...

MR LAX: The question that is being put to you - it's Lax here on the Panel, Professor, does Dr Loth have sufficient experience as a Forensic Anthropologist to be able to do these kinds of analyses, facial reconstructions and superimpositions, and to be an expert ...(intervention)

PROF VANEZIS: I am not sure, I cannot answer that question.

MR LAX: ... and to be an expert witness?

PROF VANEZIS: I cannot answer that question, I am not sure, all I know is she has published in the field, but I really don't know.

MR LAX: Thank you.

MS THABETHE: Maybe just to put you at the corner a bit, would you say in terms of experience, who is more experienced in this field between you and her?

PROF VANEZIS: I am probably the wrong person to answer that question, if you're saying what I mean, I think that is really for the Court to decide, not for me to say.

MS THABETHE: Okay, I won't push it any further. Just the last question, we are going to argue that the skull that you made a facial reconstruction from, is the skull belonging to Ntombi Kubheka and on the basis of your finding, that it is highly likely that the skull is part of Ntombi Kubheka's remains, can one argue that it is highly probable that the skull is Ntombi Kubheka, based on your findings?

PROF VANEZIS: I would like to leave that with the phraseology that I have used, that it is highly likely, and it is for you and others to put other evidence together, to corroborate what I have said.


MS THABETHE: Thank you, Professor Vanezis. Thank you Mr Chair.


CHAIRPERSON: Thank you. Mr Lax, do you have any questions you would like to put to the Professor?

MR LAX: Thank you Chairperson. Just one, two aspects. The first is, if one reads these other articles that were referred to, you have indicated they were published in 1996 and probably written a little bit earlier than that, some of the problem areas that are raised in these articles are areas with regard, for example, to the sufficiency of the image, the data that is used as the basis, the images that are used as the basis for the data to create reconstructions, in other words as I understand it, one takes numerous photographs of various people, of different features and then you agglomerate those features and separate out individual aspects of them and then use those to reconstruct. Have I explained it or described it reasonably accurately?

PROF VANEZIS: You have probably complicated it a little bit, to be honest with you. If I can just simplify it a little more, all you are doing is you are just shining a light onto someone's skull and that picks up all of the information around and puts it into a 3D image into the computer. It is as simple as that.

MR LAX: Alright. The bottom line though is, has the amount of data that forms the basis of your, that you use as the basis for coming to these comparisons, has that obviously increased since the time that these articles were written?

PROF VANEZIS: Well, what we've got to do here is to separate the facial reconstruction that we carried out, which was really just a guide to show us that we were on the right line, and from the photographs that we had, and the skull, to produce the superimposition position, yes, it has improved. The technology has improved, because we have good quality broadcast images that we can put on screens and we can manipulate these images, magnify them much more accurately and check them more accurately, and when we magnify images, we can see better detail when we are matching. So it has improved from that point of view. Perhaps also - I hate to say it, but after doing more and more cases, as we go on looking at these and we have accumulated more, then we get the experience. That obviously counts for a lot, seeing a lot of these cases, seeing where we fall down and where we can improve things, and from that point of view, we are therefore more confident in the work we carry out, than we were, say a number of years ago.

As we pointed out in the one misidentification that was done, that was only on one photograph that was done, and it wasn't standardised, to the extent that this one was, where the line is drawn and the points marked out, and so on. So there are big leaps from then.

MR LAX: Yes, you have inadvertently moved onto my next question, which was, if we were to provide you with more images, more photographs for example, better photographs, would the degree of certainty improve, or lack of certainty, as the case may be?

PROF VANEZIS: Well, let's put it this way, I come from the standpoint that based on these two photographs, I think it is highly likely. If we look at it statistically, if you provide me with more photographs of the same person, it would then be quite possible for me to improve my position on that, based on what I have already said. However, if - there may also be a very remote possibility that if you provide me with more photographs, that it may make it less, slightly less shall we say, likely. It all depends. I mean I really - if I see more photographs, it gives one a much more accurate appreciation and we get more and more towards a positive or even possibly exclusion, but in this case, I believe we will get more towards a positive, I mean that is my own personal view. But I would, if I had more photographs now, probably I would say it would be difficult for me personally to carry out this test again, I would get another expert to do it, because I have already given an opinion on this case based on the two photographs, and I mean you know as well as I do, the difficulties in getting a witness to repeat something on which they already have an opinion on.

MR LAX: No, fine, thank you Chairperson. Thank you Professor.


ADV BOSMAN: Just one question.

Professor Vanezis, you describe the match between the photograph and the skull as an excellent match, was there anything lacking which prevented you from describing it as a perfect match?

PROF VANEZIS: There was nothing lacking, I had enough information in my view from these photographs to describe it in the manner that I described it.

ADV BOSMAN: Well, why could you not describe it as a "perfect match" then?

PROF VANEZIS: As a perfect match?


PROF VANEZIS: Well because nothing is perfect in this world, there are, nothing is black and white, if you will excuse the phrase, there are different shades in-between and we can only give our opinions to the best of our ability and that is what I tried to do.

ADV BOSMAN: I regard that as a rather philosophical answer to my question, but thank you. Thank you.


CHAIRPERSON: Thank you. Is there anybody who would like to put any further questions before we conclude?

FURTHER CROSS-EXAMINATION BY MR VISSER: Chairperson, there is one very brief question, and perhaps I should have asked you this before Professor, and I am sorry that you have to listen to me all the time.

What is your experience - you spoke, or the article that you sent, spoke about a data bank, computer data bank with images that you can project on skulls which makes it more accurate and makes life a lot easier and makes it much quicker, on that basis, what is the database on, for example on South African black people that would exist in America, or in Glasgow, Scotland for that matter?

PROF VANEZIS: Well, there aren't that many to be honest with you, in Glasgow, Scotland, that is true to say, but nevertheless we do have a small number representative of black people of different age groups, but I have to say once again that what we are looking at here are broad similarities as far as skin thickness types and fitting onto a particular skull to show actually just the face shape, nothing else with the reconstruction, that is all we're doing with the reconstruction.

MR VISSER: Yes, what you are saying to me is, I am also making it more complicated now, what you really did was a superimposition on a photograph and that was it?

PROF VANEZIS: That really, the reconstruction was in fact just a guide at the beginning, the true meat of the work was the superimposition work on the skull and photograph.

MR VISSER: Yes. And do I understand you correctly to say that race, etc, wouldn't make that much difference to that part of the work, because you would have a photograph and you would superimpose it?

PROF VANEZIS: Well, I mean you wouldn't put a white face say on a black skull and vice versa, I mean that would be ridiculous, but the point is, if you have broad racial groups and you use a data bank which reflects that, that is fine if you are using it as a guide. If you are obviously going to take it much further than that, and I am talking here of police work and using publicity to try and get somebody recognised, etc, then obviously you have to use different criteria, and then we would go into identikit images and so on, so it becomes more complicated.


PROF VANEZIS: Plus also if you notice, with this reconstruction the eyes are closed, because that's, the laser beam does not affect the person's eye when we collect the data. So as I say, it was just used as a guide and it was a superimposition which mattered in this case.

MR VISSER: Thank you Professor.


CHAIRPERSON: Professor, thank you very much indeed. Once again, I would just like to express our gratitude in you giving up your valuable time and testifying in this matter, on a point that is very important to the whole hearing. Thank you very much indeed.

PROF VANEZIS: It's been a privilege.


I would also like to thank certain persons for hosting and organising this video conference at the United States Consulate's Office, where we are sitting in Durban here. I would like to thank Craig Kiehl the Consulate-General and also Jerry Williams and Devor Govinsami for their assistance. From the FBI in Washington, Mike Tuyster, and in Johannesburg, Peggy Morgan and from Giant Video Screens, Johnny Dolnick, Ian Zulch and Cecil and from the South African Department of Foreign Affairs, Mr Ronnie Mamoepe and Dumisa Ntshinga and from the TRC, our own Paddy Prior and of course Debby Quinn who has done so much work.

Thank you very much indeed, I think it has been - and Dr Naidoo as well for the work that he did in sending the skull and receiving it back again and supplying certain details that were relied upon by the Professor.

It has been a first time for me, and I am sure for most of us here, and I have found it an exhilarating experience and I think that it is something that will become more and more commonplace in the future. Thank you very much indeed, Professor.

PROF VANEZIS: My pleasure.