1 Monday, 27 June 2005

2 [Open session]

3 [The accused entered court]

4 --- Upon commencing at 2.18 p.m.

5 JUDGE PARKER: Mr. Topolski.

6 MR. TOPOLSKI: May it please Your Honours. Events have occurred

7 in the lives of some of us since we last met. In those in relation to

8 this case, may I indicate two matters before Professor Wagenaar joins us.

9 Your Honours, may I make it clear something which I think is already

10 understood, that he is being called, as it were, on behalf on all three

11 defendants. And the second matter which perhaps has not yet come to the

12 attention of the Chamber, and certainly for my part it will considerably

13 shorten matters. Your Honours have I think Professor Wagenaar's report.

14 JUDGE PARKER: Yes.

15 MR. TOPOLSKI: Your Honours will of course have read it and of

16 course have appreciated that it is in three parts. It is only part 1

17 upon which we, the Defence, rely, that is to say that which deals with

18 the Rules. Those two other parts that deal, if you like, with how those

19 rules may be interpreted and/or applied as far as this case is concerned

20 certainly seem to us who defend to be more appropriately and properly the

21 subject of submissions in due course rather than evidence from Professor

22 Wagenaar.

23 JUDGE PARKER: So does that mean you only seek to rely on one

24 part of the report?

25 MR. TOPOLSKI: Yes. It did not seem to us to be necessary to

1 provide a supplemental amended report, because of course that's entirely

2 a matter for Mr. Whiting, the ambit upon those matters which he

3 cross-examines and that entirely of course is for him. But certainly,

4 that which I propose to elicit in chief and that which we propose to rely

5 upon are those matters that are contained in part 1.

6 JUDGE PARKER: Yes.

7 MR. TOPOLSKI: Your Honour, with those words in mind, with your

8 leave, may I call Professor Wagenaar?

9 JUDGE PARKER: Perhaps before you do there are some lurking 92

10 bis statements.

11 MR. TOPOLSKI: Yes.

12 JUDGE PARKER: And having reached the point of a draft decision

13 only to find that the Prosecution response seemed to be conditional upon

14 some agreed facts, the matter's been held in abeyance. And I wonder if

15 really the two things are connected. Perhaps I will look at the other

16 side --

17 MR. TOPOLSKI: Would you please --

18 JUDGE PARKER: Yes. I can see that your mind is not at this

19 point. Yes.

20 Mr. Whiting.

21 MR. WHITING: Your Honour, if I could just briefly address the

22 point about the expert. It will be my intention only to question the

23 expert on part 1 of the report, and that will of course shorten

24 dramatically my cross-examination. If parts 2 and 3 had been introduced

25 and relied upon in any way, I would have had substantial

1 cross-examination in relation to those parts.

2 Now, with respect to the 92 bis motion filed on behalf of Fatmir

3 Limaj with regard to three statements. The agreed facts are connected to

4 the statements and I can actually -- I have them here and would be happy

5 to put them on the record and that would clear the way, as it were, for

6 the Court's decision on the matter. And it will be apparent why the

7 facts -- why they're conditional. If these facts had not been agreed,

8 the Prosecution would have asked to cross-examination the witnesses.

9 JUDGE PARKER: Very well.

10 Now, Mr. Khan, is this a course that you're content with?

11 MR. KHAN: Your Honour, it is. There is nothing controversial at

12 all either in the 92 bis witness statements or in what my learned friend

13 I anticipate is going to read out.

14 JUDGE PARKER: Thank you.

15 Mr. Whiting.

16 MR. WHITING: Thank you, Your Honour. There are three statements

17 which pertain to the three different 92 bis statements. The first one

18 pertains to the RTK statement and the agreed statement is as follows --

19 JUDGE PARKER: Sorry, may I ask. Are these in a written form

20 that we can follow?

21 MR. WHITING: They're in a written form, but I only have one

22 copy. So they're in a written form but you can't follow them because I

23 only have one copy.

24 JUDGE PARKER: Proceed orally and you'll be able to let the court

25 officer have a written form at the break. Is that right?

1 MR. WHITING: That's -- yes, I can do that and it's very short.

2 JUDGE PARKER: Thank you.

3 MR. WHITING: The first statement is that:

4 "Other political figures and former members of the KLA from

5 Kosovo also appeared regularly on RTK from September 1999 onwards,

6 including Hashim Thaqi, Jakup Krasniqi, Sylejman Selimi,

7 Agim Ceku, Ramush Haradinaj, Rexhep Selimi and others. RTK also

8 ran many stories about the war that included footage from the war and

9 interviews with former KLA fighters."

10 The second one is in reference to the newspaper 92 bis statement.

11 "In addition to Epoka e re, other newspapers published in Kosovo

12 are Koha Ditore and Bujku. Other political figures and former members of

13 the KLA from Kosovo also appeared regularly in Epoka e re from the

14 beginning of 2000 onwards including Hashim Thaqi, Jakup Krasniqi,

15 Sylejman Selimi, Agim Ceku, Ramush Haradinaj, Rexhep Selimi and others.

16 Epoka e re also had stories about the war that included photographs of

17 former KLA fighters."

18 And then finally with reference to the last statement from

19 Dukagjini Printing Company:

20 "Dukagjini Printing Company also provided printing services for

21 other political figures in Kosovo, including Hashim Thaqi, Ramush

22 Haradinaj, and others."

23 JUDGE PARKER: So those are the facts which are agreed, we

24 understand, with the parties.

25 MR. WHITING: That's correct, Your Honour.

1 JUDGE PARKER: And are put forward on that basis. Is that so,

2 Mr. Khan?

3 MR. KHAN: It is, Your Honour. There is one very brief matter

4 arising out of that.

5 JUDGE PARKER: Can it be clear whether all accused agree with

6 those facts.

7 MR. GUY-SMITH: Yes.

8 MR. TOPOLSKI: Yes.

9 JUDGE PARKER: Thank you.

10 Sorry, Mr. Khan.

11 MR. KHAN: Your Honour, not at all. The Rules make provision

12 under 92 bis for this procedure. Your Honour, we're totally in your

13 hands. There doesn't appear -- well, there's no dispute between the

14 parties as to the form of these statements. They are, as I said, very

15 uncontroversial.

16 JUDGE PARKER: You're going to try and ask the Chamber to invent

17 a rule that would allow you to avoid Rule 92 bis.

18 MR. KHAN: No. Your Honour, of course Rule 92 bis is ex

19 specialis, but of course it's subservient to 89 (C) and it appears to me

20 in relation to costs and expedition there's no dispute between the

21 parties. Matters can be, for example, read into evidence as my learned

22 friend has done now. If you're minded to proceed with the 92 bis

23 procedure, of course that can be done and, in fact, a formal request has

24 been given last week to registry for the formalities to be started. But,

25 Your Honour, if you were minded to admit these statements under 89(C), it

1 might save the precious time of the registry as well as certain expenses

2 to the Tribunal. But, Your Honours, it's a matter totally for you.

3 JUDGE PARKER: We're well aware, Mr. Khan, that in circumstances

4 there is in essence agreement the Rules can be somewhat cumbersome.

5 Nevertheless, the Appeals Chamber tends to require us not to ignore Rule

6 92 bis when we look at Rule 89. And in those circumstances, I'm afraid,

7 our course must be to say that you should follow the procedures required

8 by Rule 92 bis.

9 MR. KHAN: And, Your Honour, that will be done. I'm grateful.

10 JUDGE PARKER: Unless of course you can get agreed facts. If you

11 can manage that. It's a bit hard to convert a statement into agreed

12 facts.

13 MR. KHAN: Your Honour, perhaps we'll leave it as it was.

14 JUDGE PARKER: As foreshadowed, some other 92 bis --

15 MR. GUY-SMITH: They're fast approaching. However, there is

16 another matter that I would -- I'd like to attend to. I've been reminded

17 that it should be taken care of which is if we could have the medical

18 reports that were discussed in our last session which have been admitted

19 be under seal.

20 JUDGE PARKER: Exhibit DB6?

21 MR. GUY-SMITH: Exhibit DB6.

22 JUDGE PARKER: Should be received under seal.

23 MR. GUY-SMITH: Thank you.

24 THE REGISTRAR: DB6 would be under seal, Your Honours.

25 JUDGE PARKER: Thank you.

1 I think then we had better have the witness and Mr. Topolski can

2 get underway. He's champing at the bit.

3 MR. TOPOLSKI: I'm completing my housework. May I inquire of

4 Your Honours whether attached to Dr. Wagenaar's report you have the

5 document headed the Police and Criminal Evidence Act 1984, Code D: Code

6 of Practice For the Identification of Persons By Police Officers.

7 [The witness entered court]

8 MR. TOPOLSKI: I see Her Honour nodding.

9 JUDGE PARKER: I have two out of three.

10 MR. TOPOLSKI: Your Honour, may I just indicate it emerged in

11 consultation with the professor this morning that he and I were looking

12 at a different format of this document. If it turns out be the case,

13 that is to say it differs from the one that you have, I now have spares

14 here which would mean we all have the same. Would Your Honours indicate,

15 should the need arise, if you're not following the pagination.

16 JUDGE PARKER: Thank you.

17 Good afternoon, Professor. Would you be kind enough to read

18 aloud the affirmation that is on the card.

19 THE WITNESS: I solemnly declare that I will speak the truth, the

20 whole truth, and nothing but the truth.

21 JUDGE PARKER: Thank you. Please sit down.

22 Mr. Topolski.

23 MR. TOPOLSKI: Thank you, Your Honour.

24 JUDGE PARKER: While that is happening, I forget to indicate, Mr.

25 Khan, that the three Rule 92 bis statements will now be received.

1 MR. KHAN: I'm grateful.

2 MR. TOPOLSKI: Does that mean that Your Honours' view of the

3 professor is obscured?

4 JUDGE PARKER: We're blessed with a clear view of the professor

5 and of you, Mr. Topolski.

6 MR. TOPOLSKI: Life can't get much better than this.

7 WITNESS: WILLIAM ALBERT WAGENAAR

8 Examined by Mr. Topolski:

9 Q. Professor, as you know, I appear together with Mr. Powles for

10 Isak Musliu but you are called on behalf, as it were, of all three

11 defendants. We're grateful for your attendance and assistance. May I,

12 first of all, establish with you one or two preliminary matters before we

13 get to your curriculum vitae. Before I do anything else at all, if I

14 appear to you to speak more slowly than normal, it is on purpose. You'll

15 appreciate that there are simultaneous translations going on. Would you

16 be good enough to wait for my questions before answering them and I will

17 try and wait for your answers to finish before the next question so the

18 interpreter can keep up with us.

19 Professor, is it right that you were first instructed in this

20 matter on the 15th of April of this year?

21 A. Correct.

22 Q. We may need you to sit a little closer to the microphone. Thank

23 you so much.

24 Included in those instructions, Professor, were you sent what you

25 were to describe as case files, materials in relation to this matter,

1 which are set out I think at Annex A of your subsequent report?

2 A. Yes, that's correct.

3 Q. You were sent those I think, for the record, on the 18th of April

4 of 2005.

5 Is it right that you reported in writing and in detail on the

6 22nd of May of 2005?

7 A. Yes.

8 Q. Is it right that subsequent to that you had a number of

9 discussions and meetings with members of the Defence teams?

10 A. Yes, that's correct.

11 Q. The last meeting you had, apart from today, with myself and Mr.

12 Guy-Smith and Mr. Powles was a joint meeting between the Defence and Mr.

13 Whiting, who I think was the only representative of the -- oh, no, there

14 was another. Mr. Nicholls was there, I beg your pardon --

15 A. Somebody else was there as well.

16 Q. Messrs. Whiting and Nicholls on behalf of the OTP, a joint

17 meeting with Defence counsel, and that took place last week on the 24th

18 of June?

19 A. That's correct.

20 Q. Professor, it's a long CV. May we seek to summarise it together.

21 Can I ask you first to deal with your own education qualifications.

22 Would you like me to remind you of them?

23 A. No. Well, if you have a question about it.

24 Q. No. Well, I can lead it from you I think. 1953 to 1960 you were

25 at the Gymnasium in Utrecht?

1 A. That's correct.

2 Q. 1960 to 1965 at the University of Utrecht where you took a

3 doctorate in experimental psychology.

4 A. That's correct.

5 Q. 1962 you took a doctorate in social sciences at Leiden

6 University.

7 A. Yes.

8 Q. And you are by trade and profession an experimental psychologist?

9 A. That's correct, yes.

10 Q. You have been since the early 1960s been employed in a number of

11 places and a number of institutions.

12 A. Yes.

13 Q. Including here in the Netherlands, of course, but in the early

14 1970s, you were visiting professor at Pennsylvania State University in

15 the United States, where you were the recipient of a Fulbright

16 scholarship.

17 A. Yes.

18 Q. You headed a psychology department at the Institute for

19 Perception in Soesterberg, have I --

20 A. Soesterberg.

21 Q. In the Netherlands, thank you, mid-1970s to mid-1980s. In 1982

22 to 1985, and I'm sure it is not a typing error, it is said that you were

23 an extraordinary professor of experimental psychology

24 A. That is what it is called. You would now call it a part-time

25 professor. I like the other title better.

1 Q. I'm terribly tempted to say something but I won't. 1985 until

2 the present day I think you enjoy the seat of professorship of

3 experimental psychology at Leiden University.

4 A. Yes, that's correct.

5 Q. You have connections with a university in Belgium.

6 A. Yes, I had a special professorship at the University of Leuven.

7 Q. You're an permanent fellow of the Netherlands Institute for

8 Advanced Studies in Wassenaar?

9 A. That's correct.

10 Q. Dean of University College, Utrecht?

11 A. I have been that, yes.

12 Q. Been that.

13 A. Yes.

14 Q. You are an honorary university professor at Utrecht, and a

15 professor of psychology and law at the Faculty of Law at Leiden

16 University from 2004 until now.

17 A. Yes.

18 Q. You've held a number of administrative positions, Professor,

19 which are all set out in your curriculum vitae; I don't go into them. As

20 far as honours are concerned, as we have established you received

21 Fulbright scholarship in the early 1970s, other honours that you have had

22 conferred upon you, including membership of the Royal Dutch Academy of

23 Sciences and an overseas fellow of Churchill College, Cambridge, and not

24 least, an honorary citizenship of Crete, are all honours that you have

25 received through the 1980s and through the 1990s. Is that right?

1 A. Yes.

2 Q. You are the author and co-author of at least ten books on the

3 subject of which you are to speak today. Is that right?

4 A. Yes.

5 Q. A large number, perhaps in excess of 150 articles, papers written

6 on the topic.

7 A. On various topics.

8 Q. Various topics?

9 A. Not all on this topic.

10 Q. You have in the Dutch jurisdiction, have you not, given evidence

11 in a large number of trials?

12 A. That's correct.

13 Q. And perhaps, if I may so, more relevantly for our purposes you

14 have in fact appeared as a witness twice in this Tribunal.

15 A. Yes, that's correct.

16 Q. First -- in its very first case of Tadic. As you understand it,

17 Professor, who called you as a witness in Tadic?

18 A. I was first contacted by the Prosecution in Tadic and I worked

19 with Mr. Alan Tieger at that time. And then the Defence also asked some

20 questions and it was agreed I would answer questions of both parties

21 simultaneously.

22 Q. Subsequent to that you gave evidence in the case of Kupreskic,

23 did you not?

24 A. That's correct.

25 Q. There you were called as a Defence witness?

1 A. Yes.

2 Q. Professor, the report that we all have in front of us is split

3 into three parts. And as you are aware, we only rely upon, and I say

4 that of course with the utmost respects to parts 2 and 3, we only rely on

5 part 1. Do you follow?

6 A. Yes.

7 Q. And it's only upon part 1 that I want to ask you a few questions

8 and I apprehend Mr. Whiting will have some questions for you about it as

9 well.

10 May I deal with it this way, Professor, because we have all seen

11 the report and the Tribunal has been good enough to indicate its redit.

12 So it's going to profit no one to go through this line by line. It may

13 be of assistance, and I'll stop if it isn't, if we as it were by way of

14 headline only see if you and I can agree upon a headline notation, as it

15 were, for each of the rules, because that's going to be the focus of the

16 questions I and I think and Mr. Whiting will ask you.

17 Annexed to the report is a document called the code of practice

18 for the identification of persons by police officers. Is that correct?

19 The English PACE zone 3/1:51 -

20 A. Yes.

21 Q. -- codes.

22 Can I deal first with aims, Professor, aims of the rules. If one

23 were to look at the code of practice, PACE codes, if one were to look at

24 my page 128, the very first page.

25 MR. TOPOLSKI: Your Honours, may I pause to see whether our

1 pagination does correspond?

2 THE WITNESS: It might be 124 in --

3 JUDGE PARKER: 124 is the copy we have.

4 MR. TOPOLSKI: Well, then, can I hand up three clean sets,

5 please.

6 Does Mr. Whiting have it at 124 or 128? It's Code 1. I'm sorry

7 it's not been stapled together. I should say the contents of course are

8 exactly the same, it's just a different format. Same. Well, I'm going

9 to do it by code paragraph number, not by page number. And I hope it

10 will be followable, as it were.

11 Q. Professor, Code 1 is of -- D is the introduction. Code 1.2 reads

12 as follows, does it not: "Identification of witnesses arises, for

13 example, if the offender is seen committing the crime and the witness is

14 given an opportunity to identify the suspect in a video identification,

15 identification parade, or similar procedure."

16 These are the words I want to focus on: "The procedures are

17 designed to test the witness's ability to identify the person they saw on

18 a previous occasion and provide safeguards against mistaken

19 identification."

20 They are the aims of the code. Does that aim differ in any

21 significant way to the aims of your rules?

22 A. No, they are exactly the same aims.

23 Q. Having dealt with the aim, therefore, can I move through the

24 rules themselves. Again, Professor, attempting if we can, as it were, a

25 shorthand description of each. And if there's anything that you briefly

1 wish to add to the shorthand notation, please do, but bear in mind I know

2 Mr. Whiting is going to be asking you questions about them.

3 Rule 1: "The perpetrator must be unknown" is the first rule in

4 summary form, is it not?

5 A. Yes. That is to say at the time of the crime the witness is

6 should have seen the perpetrator not further known to him or her.

7 Q. Rule 2, insofar as it is possible, of course, there should be no

8 prior sightings?

9 A. Not in between the crime and the test. Right.

10 Q. Rule 3, again if one were to have a byline for this rule, could

11 it be that the suspect should fit the description?

12 A. Yes, that's correct.

13 Q. Rule 4, that when we are dealing, as we are here, with photo

14 line-ups, insofar as it is possible, the foils should match the suspect?

15 A. Yes, and it's preferable that that's also tested before the

16 photos are actually shown to witnesses.

17 Q. Tested upon whom and by whom?

18 A. So that -- as I described, that could be tested by a set of mock

19 witnesses who are provided with a description of the crime and who come

20 from the same population or ethnic group as the real witnesses. And if

21 they are able, although they've never seen the criminal at the scene of

22 the crime, if they are able to identify the one suspect in that line-up,

23 then the suspect is said to be biased against that suspect.

24 Q. Rule 5, and here I want to cross-refer it, if I can, to something

25 in the codes, the PACE codes. Rule 5 says, in essence, does it not, that

1 witnesses should be properly instructed. Do you agree?

2 A. Yes. And there are some designs for standard instructions that

3 can be used.

4 Q. Can we see where that - if the Tribunal does, it's a matter for

5 them, of course - would have a mirror in the codes. Could I invite you

6 and Their Honours to go to Annex E of the codes, which you might either

7 find is on page 168 or 172, depending upon which bundle you're looking

8 at. It's Annex E entitled "showing photographs."

9 A. 168 in my copy.

10 Q. Looking at paragraph 5, coincidentally and putting it beside your

11 rule 5, does paragraph 5 of the annex to the code say this: "When the

12 witness is shown the photographs, they shall be told the photograph of

13 the person they saw may or may not be amongst them and if they cannot

14 make a positive identification, they should say so. The witness should

15 also be told they should not make a decision until they have viewed at

16 least 12 photographs. The witness shall not be prompted or guided in any

17 way but shall be left to make any selection without help"?

18 Would you say, Professor, that that rule, that annex code mirrors

19 your Rule 5?

20 A. Yeah, although there are a few small differences. One is the

21 number 12 --

22 Q. Oh yes.

23 A. -- which of course follows from another section in the code where

24 it is specified that 12 is the proper number.

25 Q. Oh, yes.

1 A. And the other thing is that in my description of the rules, a few

2 aspects are added, like it should be absolutely clear to the witness that

3 they may point at only one person. They cannot claim to recognise anyone

4 if they recognise two or three people in that line-up. So it should be

5 very clear that they can point at either no one or one person; there's no

6 other possibility. My experience is that it helps to specify that in the

7 rules.

8 Q. Rule 6, although it goes entirely without saying, that the test

9 you would suggest should include no suggestions and, in fact, should be

10 conducted by someone independent. Can you define for Their Honours,

11 please, who an independent person should be, in your view, to maximise

12 the efficacy of this procedure, who that independent person should be.

13 A. In fact, I did not use the word "independent" but I specified

14 that the person who administers the test should not him or herself know

15 who in the line-up is the suspect so that there is no way consciously or

16 unconsciously to give away, to pass that information to the witness.

17 Q. For example, in the United Kingdom, as I understand it, this

18 procedure would be conducted by a police officer who has no connection

19 whatsoever to the investigation, that -- such a person would qualify.

20 A. Yes. It would be very desirable to have such a person, but the

21 essential thing is that this person doesn't know who in that line-up is

22 the suspect. It might even happen that the police officer not connected

23 to the case for one reason or the other has become aware of the identity

24 of the suspect in that line-up. That should not happen; that is the

25 essential part of it.

1 Q. Rule 7 could be summarised, could it not, Professor, do you

2 agree, as "all attempts by a witness should be reported and recorded"?

3 A. Yes, but maybe it's not so clear that this applies to two sets of

4 identifications. One set are identifications concerning the same

5 suspects, but it also applies to identifications that were made by the

6 same witnesses, maybe for other suspects. Because if witnesses were

7 either successful or unsuccessful at identifying other people, that would

8 be highly relevant information.

9 Q. Rule 8 perhaps could be summarised by two words,

10 "non-contamination." Would you agree?

11 A. Yes.

12 Q. Rule 9 really deals with, as you describe it, various modes of

13 investigation, if you like, in this respect. I wonder if one turned to

14 find a mirror of this we might find it at paragraph 3.4 of the codes,

15 which -- which is under the heading "cases when the suspect is known and

16 available."

17 A. Yes.

18 Q. Really, what Rule 9 appears to be doing, Professor, is to set

19 out, as it were, steps that may be available to an investigator in terms

20 of identification evidence. Would that be right?

21 A. Yes. There is some sort of graded scale, one might say, an

22 ordered scale of more- or less-desirable procedures.

23 Q. If one looks at the English code in 3.4, in the graded scale we

24 do not see photo line-ups mentioned at all. Where, in your opinion, do

25 such line-ups rank in the scale of effective and fair safeguarded

1 procedures, Professor? Where do they rank?

2 A. Well, to begin with, there are two types of photographic

3 identification procedures. One is in -- with -- one is that -- a

4 procedure that uses photographs of the entire persons, from top to toe,

5 often with some sort of a background with lines indicating the height of

6 the people so that you can see whether they are tiny or big or tall or

7 small. The second procedure is a procedure that uses faces, photographs

8 of faces only, which tells you nothing about, say, the size of those

9 people, or the posture of those people, or the build of those people, so

10 that's limited to faces only. Obviously if the witnesses saw height and

11 build and also describe height and build, it might be a good thing to use

12 photographs that also show this so that the probability of correctly

13 identifying or correctly not identifying such a person would be bigger.

14 Q. Rule 10 simply states "no dock IDs".

15 A. Well, you can do that for administrative purposes, but it cannot

16 replace a proper line-up procedure.

17 Q. I just want to ask you one matter arising out of the next

18 paragraph, that having completed a summary of the rules themselves, and

19 then I have one only matter only to ask you about, Professor. The

20 question you pose in the report between paragraphs 20 and 21 after Rule

21 10 on page 7 is what if the rules are violated. Now, of course you

22 appreciate whether this Tribunal adopts any of these rules, first of all,

23 and/or whether it finds, as a matter of evidence, any of those rules it

24 adopts have been breached, it's entirely a matter for it.

25 But do you have, Professor, an understanding of the way that the

1 English courts deal with breaches of the codes? What do you understand

2 to be the test that the English judge would apply when he or she is

3 looking at the admissibility of identification evidence and considering

4 whether there has been or was a breach of codes of the practice?

5 A. Although I'm not an expert in law and definitely not in English

6 law, my understanding is that -- which of course is also sensible, that

7 when you talk about violation of the rules, you do not talk about small

8 deviation from the desirable rules, but you talk about serious and

9 significant breaches of the rules which -- which to my mind implies that

10 the probability of a false positive, a false positive identification is

11 substantially increased, of course at the risk of the accused.

12 Q. Yes. A final question: Are you aware, first of all, of any

13 criminal or quasi-criminal jurisdiction in the world that has adopted all

14 or most of these ten rules, Professor?

15 A. Well, you have already referred to the PACE rules which of course

16 in Europe is a brilliant example of qualification in an early stage even

17 of such rules. In the Dutch jurisdiction, as I have noted in I think

18 footnote number 1 of my report --

19 Q. Page 4, footnote 1.

20 A. The Minister of Justice has issued a decree that there are also

21 rules for the Dutch jurisdiction which are laid down by the -- what's

22 called the Recherche Advies Commissie, the criminal investigation

23 advisory board, of which I was a member at the time, and in the handbook

24 written by Mr. Von Amelsvoort for the police academy which is used by all

25 police officers who perform such tests in our country. The decree of the

1 minister in 2002 has its essential intention that the courts should test

2 actual procedures being used in trials against these rules. And since

3 there is not much difference between the rules from 1992 and the ones

4 published by Mr. Von Amelsvoort in his textbook, the courts could choose

5 either one.

6 Q. That's the Dutch national experience. Are you aware of any

7 connection, as it were, closer to home, closer to this Tribunal and its

8 investigating wing, as it were, as far as these rules are concerned?

9 A. Well, I've not seen rules for identification procedures, say,

10 decreed by this Tribunal.

11 Q. No.

12 A. But I've noticed a few things that I can give to you for what

13 they're worth. One is that in the trial of Mr. Kupreskic, in the ruling

14 of the Tribunal they explicitly underscored some of the rules that I

15 presented then because they were particularly relevant in that case, and

16 indeed decided not to use a certain piece of evidence that was brought

17 forward then.

18 The other thing that I can tell you is that - and of course

19 you're all aware of this - that Mr. Von Amelsvoort -- that the handbook

20 of Mr. Amelsvoort was translated by this Tribunal in English and then

21 subsequently used as a textbook for the training of the investigators of

22 this Tribunal. I think this happened in 1999. So I don't know exactly

23 what officially that means, but at least it should mean that at that time

24 it was thought -- it was decided that this set of rules would be very

25 useful to the Tribunal.

1 Q. Which year was this, Professor, as you understood it?

2 A. I think it was 1999 -- 1998 or 1999, but I think the Office of

3 the Prosecutor can tell you more about it because I think they were more

4 involved with it than I am. So they should absolutely be able to provide

5 you with that translation and the details of this training.

6 Q. And I'd just like to be clear precisely what your evidence is

7 about it, please. Bearing in mind that we are dealing with a large

8 organisation here.

9 A. Yeah.

10 Q. Which particular organisation within the ICTY do you say that

11 these guidelines, these rules, were provided?

12 A. I think I should not guess.

13 Q. No, you shouldn't.

14 A. I talked to Mr. Von Amelsvoort who explained me exactly -- the

15 reason I talked to him was to ask him whether there would be an English

16 translation of his book to provide to the Tribunal.

17 Q. Have you seen --

18 A. And he answered they already have it.

19 Q. Right.

20 A. That was when he explained that it was used in a training course

21 for investigators - that's all he said and that's all I can tell you

22 about it - investigators of this Tribunal.

23 Q. Have you seen this document before?

24 A. No, I've not seen this document. I must also say that right now

25 this book has its -- is in its fifth edition. The second edition was

1 translated for this Tribunal but the differences are slight.

2 Q. Can I show you just for identifications purposes at the moment

3 only a document, Professor, and ask you whether you've seen it before.

4 MR. TOPOLSKI: I'm very happy for Mr. Whiting to have a copy

5 while I'm doing this. One for the witness, one for Mr. Whiting for the

6 moment.

7 Q. Professor, just look at this document for a moment, will you,

8 which is headed "guidelines For Multiple Photo Identification Procedure"

9 and appears to have the name "Amelsvoort" at the bottom and a date in

10 1997. Have you seen this before?

11 A. No, I've not seen this, but this seems to me, because I know the

12 Dutch text, a translation of part of his handbook.

13 Q. And it's your understanding, is it, that this or something like

14 this is the material you've just been talking about that went, as it

15 were, from Amelsvoort to the Office of the Prosecutor sometime in the

16 late 1990s? Is that the position?

17 A. Well, I don't want to guess and I think there's no reason for

18 me --

19 Q. Well, that's why I'm doing it carefully and, I hope, fairly,

20 Professor. If you can't say, then we can't take that any further. I

21 don't know whether Mr. Whiting may or may not be able to help us.

22 A. I can't say that this was the document used for training in this

23 court. I can only say that this, to my understanding, is a translation

24 of a section in Mr. Von Amelsvoort's handbook and it might well be it

25 would have been a very useful basis to teach a course on, but I can't

1 know whether this was the document used in that training.

2 Q. Very well. In those circumstances I take it no further and I

3 have no more questions for you. Thank you. Would you wait there?

4 JUDGE PARKER: Mr. Guy-Smith.

5 MR. GUY-SMITH: Based on the position the Defence has taken, no

6 questions at this time.

7 JUDGE PARKER: Mr. Khan.

8 MR. KHAN: No questions.

9 JUDGE PARKER: Thank you.

10 Mr. Whiting.

11 MR. WHITING: Thank you, Your Honour.

12 Cross-examined by Whiting:

13 Q. Good afternoon, Professor Wagenaar.

14 A. Good afternoon.

15 Q. As you know, my name is Alex Whiting and I am one of the members

16 of the Prosecution team and, as Mr. Topolski said, we met, did we not,

17 last Friday at your home. Is that correct?

18 A. That's correct.

19 Q. Now, I just want to repeat what Mr. Topolski said. We have to

20 speak slowly. We also have to pause after each one of us speaks for the

21 benefit of the interpreters. So after I ask my question, if you can

22 pause before responding and I will try to pause before asking my next

23 question. Do you understand?

24 A. Yes.

25 Q. I also want to make it clear that I'm only going to ask you

1

2

3

4

5

6

7

8

9

10

11

12 Blank page inserted to ensure pagination corresponds between the French and

13 English transcripts.

14

15

16

17

18

19

20

21

22

23

24

25

1 questions about part 1 of the report. Do you understand that?

2 A. Yes.

3 Q. Finally, I would encourage you to focus on the questions and be

4 as specific and brief as possible in your responses. With some luck, we

5 can complete your testimony today. Do you understand that?

6 A. Sure, yeah.

7 Q. Now, of course if you need to respond at length to give a proper

8 answer, of course you must do that.

9 A. I will give you a warning then first.

10 Q. Professor, I want to start with the role of an expert in a case,

11 and I want to ask you if you agree with this statement. And just so

12 there's no mystery, it's actually from your own book, from the Ivan book

13 from page 157. "There is considerable truth in the argument that the

14 expert does not know which witness will tell the truth, but it is not the

15 expert's task to make such assessments. The expert can only present

16 information about variables that are possibly relevant. The actual

17 assessment of witness reliability is in the domain of the court."

18 Do you still agree with those words that you wrote?

19 A. Absolutely.

20 Q. Now, I'd like to agree on some things just in the -- if we start

21 to discuss some of these rules, I think many of the terms are clear in

22 your report, but so that -- because sometimes there's variance in use of

23 language in this area. The first is estimator variables and system

24 variables. Now, these are words, terms, that you have used yourself in

25 the articles. Correct?

1 A. [No verbal response].

2 Q. You're nodding your head, but that doesn't get recorded on the

3 transcript.

4 A. Yes.

5 Q. Now, estimator variables are variables that are -- that in a

6 witness identification that are -- occur at the time the witness sees the

7 suspect, for example variables such as lighting, distance, exposure time,

8 stress, things like that. Is that correct?

9 A. Yes.

10 Q. And system variables are variables that are within the

11 identification procedure; that is, some of the rules that you have been

12 talking about, how many photos are shown, how you arrange the photos, who

13 administers the test, and so forth. Is that correct?

14 A. That's correct.

15 Q. And those terms, were they first used by Gary Wells?

16 A. They were introduced by Gary Wells, that's right.

17 Q. And Gary Wells, he is an expert who is similar to yourself, who

18 works in the United States. Is that correct? When I say "similar to

19 yourself" I mean --

20 A. I don't work in the United States.

21 Q. No, no, he works in the United States.

22 A. That's right.

23 Q. When he say "similar to yourself," he works in the same area you

24 do and in fact he takes many of the same positions you do on these

25 issues. Is that correct?

1 A. I know him as a researcher. I have no way of knowing whether he

2 also studies cases or testifies in cases, actual cases.

3 Q. Well, have you read his articles, some of his articles?

4 A. Yeah, sure.

5 Q. And have you seen in his articles that he discusses many of the

6 same topics in this area?

7 A. Absolutely.

8 Q. And he takes many of the same positions that you do on these

9 topics?

10 A. Yes, correct.

11 Q. And you yourself, you've written about both estimator variables

12 and system variables. Correct?

13 A. Yes, that's true.

14 Q. And while we have used two different terms for these topics,

15 there is a relationship between the two; that is, that the system

16 variables are created in some sense because of the estimator variables,

17 that is -- let me try and put it this way: It is because of concerns

18 about witnesses' ability to identify because of things like distance and

19 time and stress that you create procedures such as the system variables.

20 Do you understand what I'm saying?

21 A. Yes. Yes, but now I need a slightly longer answer to that,

22 because if you suggest that if the conditions of the first perception or

23 the conditions for remembering the person that you saw are ideal, that in

24 that case you do not need a special precaution with the system variables,

25 then I would not agree. It's not so that if the estimator variables are

1 perfect you may sort of fiddle with the system variables. If that's the

2 intention of your question, I would disagree. I would say -- I would

3 make a slightly different distinction than you do. Estimator variables

4 are not under the control of the investigators; they happen at the scene

5 of the crime or because of other conditions not under your control. The

6 system variables are totally under control of the investigators, so that

7 is their main responsibility. And rule-giving or even legislation,

8 therefore, in the first place is applied to the system variables.

9 Q. Okay. Just to press this point a little bit -- just a little bit

10 further, do you understand that in some jurisdictions such as in the

11 United Kingdom and in the United States that if a court has some concern

12 about the system variables, that is the rules that were followed, what

13 often is done is there's a look at the estimator variables to see, that

14 is what the conditions of the identification were. Do you -- are you

15 aware of that?

16 A. Well, in as far as it's at the discretion of the Court to accept

17 or not accept identification evidence, not simply dictated by the fact

18 whether or not rules were followed but just on the merits of the entire

19 -- of the entire - how would I say that? - situation behind those

20 investigations, in as far as that is at the discretion of the court, of

21 course the court should also weigh the estimator variables to make such a

22 decision.

23 Q. Okay. If I could continue with the -- with just some terms so

24 there's no confusion as we continue. A target-present line-up is a

25 line-up where the suspect is present, correct? In a study. That's when

1 the suspect -- when the suspect is present, you have a target-present

2 line-up?

3 A. That's correct.

4 Q. If the suspect is absent you call it a --

5 A. A target-absent line-up.

6 Q. Now, a foil is a photograph of a person who is not the suspect.

7 It's sometimes called a filler or another term, but the term you use is

8 foil. Correct?

9 A. Yeah, but there's one more descriptive characteristic which is

10 the foil is absolutely innocent.

11 Q. Understood.

12 A. If the witness points at the foil, we know that the witness is

13 wrong.

14 Q. A -- in a study, a hit, what's called a hit is an identification

15 of the suspect.

16 A. A hit is a positive, correct identification in the target-present

17 line-up.

18 Q. A false alarm is an identification of someone who is not a

19 suspect, that is an identification of a foil.

20 A. It is an identification of a foil in a target-present line-up or

21 an identification of a foil in a target-absent line-up.

22 Q. I want to move on to ask you about an estimator variable which is

23 exposure time. Now, would you agree with me that most studies look at --

24 when they test witnesses, whether they be laboratory studies or

25 real-world studies, generally they test exposure time of a matter of

1 seconds, at most minutes. Is that correct? When you're describing most

2 studies that are done in this area.

3 A. Yeah, most studies reflect an actual situation, which is that

4 most trials in which identification tests are used, that means in which

5 the witness originally was confronted with an unfamiliar person. In most

6 of these situations, the encounter in fact did not last much longer than

7 a few seconds until a few minutes, and that is why also most studies

8 reflect that situation.

9 Q. Looking at your -- the study that you did with others in 2003 on

10 light and distance - and I appreciate that this was a study of

11 recognitions as opposed to identifications --

12 A. Yeah.

13 Q. -- you described at one point in that study, and I have it

14 available if you need to refer to it, that the faces in the study were

15 viewed for a relatively long time."

16 Do you recall that you used that description?

17 A. Yeah, but now I must warn you that maybe there is an

18 understanding -- a misunderstanding here. That study that you cite is on

19 recognising familiar people, which means the first encounter -- say, one

20 of these familiar people might be Bill Clinton, right? Now, the

21 encounter with Bill Clinton might have been for hours on television,

22 newspapers, whatever, only the test is described in this publication.

23 But your earlier questions were about the length of the original

24 confrontation. So you switched to another topic, which is all right with

25 me, but I should at least --

1 Q. I tried to be clear about that. I didn't want to be misleading

2 about that. I tried to be clear about that by saying that this was

3 recognitions as opposed to identifications --

4 A. Yeah --

5 Q. But I appreciate your further --

6 A. But the confrontation was during the test, not at the scene of

7 the crime.

8 Q. I understand. In any event, in that case the exposure in the

9 second round was for 12 seconds in your study. Correct?

10 A. That's say is the recognition test, which is -- in terms of

11 identification tests the time that you spend at viewing each of the

12 picture -- each of the pictures is in terms of so many seconds, not

13 hours.

14 Q. It was 12 seconds in your study. Correct?

15 A. Yes.

16 Q. We can move on to some examples of identification studies that

17 were done. Are you familiar with a study that was done by Memon Hope and

18 Bull in 2003?

19 A. I think that's one that you submitted to me, right?

20 Q. I did provide it to you, that's correct.

21 A. Let me see, because I'd like -- if you want me to comment on it

22 I'd like to have it in front of me.

23 Q. I have it available.

24 A. Yeah, I think I have it.

25 Q. It's --

1 A. Here. That's 2003 in the British Journal of Psychology. Right?

2 Q. That's correct.

3 A. Yeah, I have it in front of me.

4 MR. WHITING: I have copies. It may be of some assistance to

5 have this provided. The Defence and the witness have it, but it may aid

6 the Court to have copies.

7 MR. GUY-SMITH: Just to make sure I have the right document, is

8 that the one "Exposure Duration Effects on Eyewitness Accuracy and

9 Confidence"?

10 MR. WHITING: Yes.

11 Q. Now, Professor, this study reviewed previous studies that had

12 been done on exposure time and found that the -- and just to be clear,

13 this is with respect to identifications now, which is, as correctly

14 stated, what I originally asked you about. This study reviewed the prior

15 studies that had been done and found that the exposure time that had been

16 examined in those prior studies ranged from - and this is at page 4 -

17 ranged from 6 to 8 seconds to just over 12 minutes. That was the range

18 of the prior studies. Is that right?

19 A. Yes.

20 Q. And in their own study what they did is they compared an exposure

21 time of 12 seconds - this is at page 7 - 12 seconds to an exposure time

22 of 45 seconds and made a comparison. And is it right that even in that

23 -- in that comparison between the 12 seconds and the 45 seconds, it was

24 found that there was a dramatic increase in correct identifications? I

25 think --

1 A. Where do you find that conclusion? Maybe you can help me.

2 Q. Well, I found it at page 17.

3 A. 17? Because I seem to remember from this paper that there was a

4 effect on confidence.

5 Q. Well, there is also that. But if you look at page 17 in the

6 second paragraph it says, I think it's the third -- fourth sentence it

7 says: "There was a dramatic increase in hits from 32 per cent in the

8 short exposure to 90 per cent with a longer exposure to the face."

9 A. On page 17? Not in my copy, but --

10 MR. WHITING: Is it titled "Discussion" at the top of... It's

11 page 18, I apologise. I have a different pagination.

12 THE WITNESS: Okay. I see what you mean.

13 MR. WHITING:

14 Q. And I take it that that conclusion does not particularly come as

15 a surprise, that with longer exposure time there would be an increase in

16 accuracy of identification.

17 A. Yeah, I'm -- I agree with what you say about what they found.

18 But there can be a confusion with respect to the words "identification."

19 I'm not sure if they call it identification, and the danger is that you

20 would apply such numbers on identification tests, as they are done in,

21 say, criminal trials or criminal investigations. This is a -- what I

22 would call a recognition study, so such high numbers as 90 per cent

23 correct you should not, without further ado, I think, translate to a

24 situation in a criminal investigation and draw the conclusion that after

25 45 seconds of exposure witnesses can identify 90 per cent of the suspects

1 shown to them. With that understanding --

2 Q. Well --

3 A. I'm not saying that --

4 Q. I'm a little unclear because this is an identification study, not

5 a recognition study; that is, that the participants were shown a video of

6 a simulated robbery and then they were then shown photo spreads and asked

7 to identify the suspect.

8 A. Yeah.

9 Q. And that then they were -- the two groups were compared. One saw

10 -- had an exposure time to the suspect in the video of 12 seconds, one

11 had an exposure time of 45 seconds, and the rates of hits, which we

12 defined earlier, were compared. Is that your understanding of this

13 study?

14 A. Yeah, yeah, you're absolutely right. But since you introduced

15 the difference between estimated variables and system variables, I can

16 easily point out to you that this study is on the influence of an

17 estimator variable.

18 Q. Correct. But it's identification, not recognition. Is that

19 right?

20 A. Yeah, we use the term -- I know the Tribunal uses identification

21 and recognition in another manner as we in our --

22 Q. I'm using --

23 A. That's slightly confusing.

24 Q. Well, I'm using the terms as you used them in your report.

25 A. I used identification --

1 Q. And recognition. You used those terms in your report. And this

2 is an study of identification.

3 A. I was trying to point out to you that this was a study on

4 estimator variable with the system variables kept perfect.

5 Q. Understood. Understood.

6 A. Right.

7 Q. I'm trying to focus on the estimator variable at the moment.

8 A. Which does not say very much about identifications in criminal

9 investigations, where there might also be problems with the system

10 variables.

11 Q. Well, that's another subject which we may talk about later. But

12 I'm trying to focus -- trying to isolate at the moment the estimator

13 variables.

14 A. You're absolutely right and the point I think confirms I think

15 what I have said in my report, that if you use optimal procedures,

16 identification can be highly reliable.

17 Q. Well, the point is -- that's not the point I was trying to

18 address -- trying to ask you about with the study. The point I'm trying

19 to ask you about with this study is the effect of increased exposure time

20 on accuracy of identifications. And it's fair to say that all things

21 being equal, an increase in exposure time is going to lead to -- is --

22 correlates with increased accuracy in identification. Is that a

23 statement you would agree with?

24 A. Yeah, there's no doubt about it.

25 Q. Okay.

1 A. If you would have asked what's the minimum exposure that you do

2 need I would say 300 milliseconds. Above that it becomes quite good.

3 Q. Nonetheless, this study found a dramatic difference between 12

4 seconds to 45 seconds. Correct? Yes?

5 A. Yes.

6 Q. Now, in your book the -- about the Ivan Demjanjuk trial, you

7 talked about a study -- it's actually called -- what is fairly called a

8 meta analysis done by Shapiro and Penrod --

9 A. Yes.

10 Q. -- in 1986, and I take it since you have cited that in your book

11 that's something that you're familiar with. Correct?

12 A. Yes, I am.

13 Q. And this -- what this study did is it analysed a number -- a

14 significant number -- I think it's -- is it 190 studies that were done,

15 and did an analysis of all the work -- of all of those studies and kind

16 of put it together. Is that correct?

17 A. Yes, that's what a meta analysis does.

18 Q. And is it correct that that study found a -- by looking at those

19 studies -- and to be clear, I think only 8 of the 190 dealt with the

20 issue of exposure time --

21 A. Yes.

22 Q. But looking at those eight studies that the Shapiro and Penrod

23 studies found a correlation, what they said was an expected linear

24 relationship between exposure duration and hit rates. Is that also

25 accurate?

1 A. Yes.

2 Q. Okay.

3 A. Yes.

4 Q. And I take it that you are not aware of any study that studied

5 exposure times in excess of 12 minutes or 15 minutes.

6 A. Depends on what you call a study. Of course there are cases

7 described where people were exposed for days or years to other people and

8 still misidentified them. But that's --

9 Q. Well, let's -- so --

10 A. So what do we call a study --

11 Q. Let's talk, for instance, about laboratory studies.

12 A. Because in my report about the Demjanjuk case, also the Porta

13 Debalus [phoen] case, where clearly his manservant who seen him for every

14 day for a year identified the wrong person. So we're talking about a

15 statistical relationship between exposure time and identification, which

16 does not mean that exposure time is a guarantee for correct

17 identification. But that's not what you mean.

18 Q. No. And this business is all about statistics, isn't it,

19 Professor? I mean, that's -- there -- all of these issues are about

20 statistics. Isn't that correct?

21 A. Sure, but if you mean by statistics averages, I can't agree

22 because statistics also deals with the outliers [sic].

23 Q. But my point is there are no guarantees -- there are very few

24 guarantees anywhere in this business. Is that correct?

25 A. Yeah.

1 Q. This is all about percentages and estimating and evaluating in

2 terms of percentages.

3 A. Yeah, some sort of statistics can be characterised like that and

4 some sort of statistics deal with the range of things, the outliers in

5 the worst cases.

6 Q. Now, just to get back to my question, to focus on the laboratory

7 studies -- because it's fair to say that the laboratory studies comprise

8 the overwhelming majority of studies done in this field, perhaps even

9 over 90 per cent of studies done in this field are simulated laboratory

10 studies. Looking at these simulated laboratory studies, you're not aware

11 of a study which uses an exposure time of more than 12 or 15 minutes, are

12 you?

13 A. No.

14 Q. Are you aware of the survey of experts in your field that was

15 done by Kassin, Tubb, Hosch, Haas and Memon in 2001?

16 A. That's about the number of statements, the number of -- the

17 percentage of experts agreeing or disagreeing with certain statements?

18 Q. Yes.

19 A. Okay, let me... Yes, I'm familiar with that research.

20 Q. Did you yourself participate --

21 A. Wait a minute.

22 Q. Okay. I'm sorry.

23 A. You give me such a pile. I mislaid it --

24 Q. I can provide it.

25 A. But let's see whether we can deal with it. Oh, yeah, I got it --

1 no, no.

2 Q. It's just been provided to you.

3 A. Yeah, okay. I got it. Yes.

4 Q. Did you yourself participate in this survey?

5 A. No, no. I think these are all American experts. As you may see

6 -- well, I mean Memon is from Aberdeen but he did this when she was --

7 anyway -- no, I did not participate in this study.

8 Q. Actually, if you look at -- and I hope it's the same pagination,

9 page 3, look under the method in the second paragraph under the method

10 under the experts you'll see that there were -- 53 per cent of the

11 respondents were employed in the United States and the others came from

12 the United Kingdom, Canada, Germany, Australia, the Netherlands. I

13 thought perhaps that you were included?

14 A. Yeah. Well, let's not rely on my memory for this.

15 Q. Oh, dear.

16 Now, this was a -- as we said, this was a survey of experts that

17 was done in 2001, and it was to see whether there was agreement about

18 certain propositions in this field. Is that a fair description of what

19 this was?

20 A. Yes.

21 Q. And would you say -- would you agree that there was pretty

22 widespread support for the notion that there is a correlation between

23 exposure time and accuracy in identification? It's negatively expressed

24 in this, that is the shorter duration, the less accurate the

25 identification. But nonetheless, there's an agreement upwards of 85 per

1 cent that agree with that proposition -- with that correlation?

2 A. Yeah, I would say 81 per cent, but anyway. There is a problem

3 with that research, as you are aware of, of course, because it very much

4 depends on what exactly the question is that was asked to them. For

5 instance, to give you an example, if I ask you: Do you think it would

6 make a difference whether exposure time was half a second or one minute?

7 Then all the experts would absolutely say yes, and I can testify about

8 that. If you would ask: Would it make a difference whether exposure

9 time would have been one minute or one and a half minutes? Then probably

10 most experts would say, no, I don't think it makes much difference and I

11 would not testify about it. So the question is: What exactly was the

12 proposition? Were there any numbers mentioned?

13 Q. Well, if you -- I'll draw your attention --

14 A. Or was it just generally exposure time.

15 Q. If I could draw your attention to page 15. Were you familiar

16 with this survey before it was provided to you last week?

17 A. Yes, yes.

18 Q. So page 15, if you look at table 1, number 6, it appears that the

19 proposition that was put --

20 A. Yeah.

21 Q. -- was: The less time an eye witness has to observe an event,

22 the less well he or she will remember it.

23 A. That illustrates my point. That's a very vague statement and the

24 question is how the various experts would have interpreted this

25 particular sentence, whether they would have said -- this question is

1 about half a second versus one minute or is it about one minute versus

2 one and a half minutes. And --

3 But I don't want to be -- I don't want to be difficult. I think

4 exposure time is one of the important, if that's your point, one of the

5 important estimator variables and of course exposure time is important

6 for a court to know. And a short exposure time, a very short exposure

7 time, makes it doubtful whether a witness had sufficient opportunity to

8 see the criminal. And a long exposure time, combined with other

9 favourable conditions like lighting, not wearing a mask and showing your

10 face, and so on, is a very positive predictor of the ability to recognise

11 a person.

12 Q. Okay. Let me --

13 A. If that's what you want to know, maybe we could go faster.

14 Q. Let me before the break to put one further refinement on that,

15 and that is in the first study that we looked at -- you can put that

16 aside, I'm finished with that -- but the one that was done by Memon,

17 Hope, and Bull, and I'm just going to put -- it's at -- it's my page 19

18 but it's probably your page 20. I think is it not that one on top there?

19 Professor, it's being provided to you by the usher.

20 A. Okay.

21 Q. If I could draw your attention to page 20, I just want to see

22 if --

23 A. 20?

24 Q. Page 20.

25 A. Yes.

1 Q. I want -- the authors of this study are analysing the results and

2 they say at the top of the page there, they say: "The opportunity to

3 view the culprit in a variety of poses and expressions may be more

4 important than length of exposure. Changes in pose, expression, and

5 details of appearance can significantly influence identification

6 accuracy."

7 Now, that is a -- that is a proposition that the authors of this

8 study suggest in relation to their results. Is that something that you

9 would agree with, that proposition -- that suggestion made by the

10 authors?

11 A. I think you should follow the discussion of this paper. See, one

12 problem with exposure time is that you consistently find effects of

13 exposure time, not always the same size of effect, and you do not find it

14 in 100 per cent of these studies. And the problem they're discussing

15 here is that if exposure time is just exposure to a stationary image that

16 doesn't change, then it's probably not doing much. The effect of

17 exposure time might be that a longer exposure time gives you, with moving

18 persons, the opportunity to see that person from more angles. That's

19 what they surmise here, that's the hypothesis. Now what do you ask me,

20 whether I agree that that's a good hypothesis or whether I already have

21 the answer to that hypothesis?

22 Q. I'll put the question to you again and more precisely. Based on

23 your experience and your work, do you -- have you found that any support

24 for this hypothesis?

25 A. No. In my own work I've never varied the number of angles in

1 which you've seen the culprit, no.

2 Q. So based on your own work you cannot offer anything about that?

3 A. I can confirm that it's a sensible hypothesis.

4 Q. So it makes sense to you, that is it seems that it would be

5 correct.

6 A. It makes sense that if you talk about real-life conditions and

7 you've seen a criminal for minutes or even half an hour or an hour, then

8 the fact that you've probably seen the criminal from many different

9 angles will lead to a much richer storage of images in your memory.

10 Q. Thank you, Professor.

11 MR. WHITING: I think perhaps a convenient time.

12 JUDGE PARKER: Thank you, Mr. Whiting.

13 We must break now to enable the tapes to be changed. We will

14 resume at 5 minutes past 4.00.

15 --- Recess taken at 3.44 p.m.

16 --- On resuming at 4.09 p.m.

17 JUDGE PARKER: Yes, Mr. Whiting.

18 MR. WHITING: Thank you, Your Honour.

19 Q. Professor Wagenaar, it's been suggested to me that we have

20 violated at least one of the rules of the Tribunal, which is we are not

21 pausing after question and answer and we are instead interrupting each

22 other. So we have to try to be more disciplined.

23 Could I turn, please, to paragraph 22 of your report. Do you

24 have the report in front of you?

25 A. I do. Paragraph 22?

1 Q. Paragraph 22.

2 A. Yes.

3 Q. Now, in that paragraph you talk about certain cases where DNA has

4 shown mistaken identification, and I want to draw your attention to the

5 penultimate sentence on the page where you say: "In the survey published

6 by the American Psychological Association Monitor, 1999, it is reported

7 that 36 out of 40 wrongful convictions involved erroneous eyewitness

8 identifications."

9 Do you see that sentence?

10 A. Yes.

11 Q. Then the following sentence says: "Important shortcomings in the

12 procedures were," and you list four different shortcomings.

13 A. Yes.

14 Q. Are those -- are you suggesting in this paragraph that those four

15 shortcomings occurred in these 36 of 40 cases? Is that how I'm to

16 understand what you've written here? Or are you talking generally about

17 shortcomings and procedures?

18 A. I have this paper in front of me from the APA Monitor.

19 Q. Yes.

20 A. Mm-hmm.

21 Q. Maybe I could provide that to -- I have copies of that and if I

22 could provide that to -- I think the Defence has, but if I could provide

23 it to the Court. You're referring to the paper that you footnote in

24 footnote 5 of your report. Correct?

25 A. Yes.

1 Q. And this article by Scott Sleek is the source for what you've

2 written in this article. Correct? You nodded your head yes, but if you

3 could respond verbally.

4 A. Yeah, yeah, what the paper -- the paper discusses several

5 things --

6 Q. Well, Professor, you're going -- you're getting a little ahead of

7 me. If we could take it one step at a time because you and I know about

8 this but maybe not everybody. This paper -- this article, this

9 three-page article is the source you cite for what you say here.

10 Correct?

11 A. Yes.

12 Q. It is the only source you relied on for this information about

13 this study. Correct?

14 A. Yes.

15 Q. Okay. Now, if we could look at the source, at the first page it

16 says in the fifth paragraph it talks about the study which reviewed 40

17 convictions and found that 36 involved eyewitness -- eyewitnesses

18 identifying the wrong person. And that's what you say in your report,

19 right?

20 A. Yes.

21 Q. Okay. Now, if we could turn, please, to the second page -- and

22 as you say -- before we turn to the second page, this article is not just

23 about that study, it's also about other experiments and studies that have

24 been done in this field. Correct?

25 A. I was just going to say that when you stopped me.

1 Q. As I said, we have to take it one step at a time. So that's

2 correct. This article is about other things, in addition -- not just

3 about the study of 40 convictions, but about other matters. Correct?

4 A. Well, I would like to explain in my own words --

5 Q. Please --

6 A. -- what it is about. The article says: "In response to this

7 number, the outcome of the DNA research, in the response, a group of

8 psychologists, prosecutors, police, and Defence attorneys are prepared

9 within the next several months to release a set of guidelines."

10 To my understanding, then it says on page 2 at the bottom: "So

11 far the group," so in response to these cases, which also includes

12 psychologist Ronald Fisher have developed several recommendations. And

13 then they made the following recommendations and then you get to the list

14 of which I mentioned some of the issues in my report on the top of page 8

15 of my report.

16 Q. Okay.

17 A. So that's the connection between these recommendations, right,

18 and the DNA cases.

19 Q. Okay.

20 A. Which is that a study -- a study group studied these cases and

21 concluded that the following recommendations then are needed.

22 Q. The -- again, it's important to take this one step at a time.

23 Not to belabour the point but to make it clear. The article on which you

24 relied talks about other studies. For example, if you look at page 2,

25 it's under polluted procedures it says: "Wells and other psychologists

1 have offered the fallibility of eyewitness identification." Correct?

2 A. Yes.

3 Q. Gary Wells is one of the people on this committee that's being

4 talked about in the article. Correct?

5 A. Do we know that? I know that Ron Fisher was.

6 Q. Well, if you look at page 1 it says at the bottom: "Psychologist

7 Gary Wells, a member of the Department of Justice panel on eyewitness

8 evidence. This is the panel that has been appointed by attorney general

9 Janet Reno at the time."

10 A. Wait a minute. My page 1 says Gary was a member of the

11 Department of Justice panel on eyewitness evidence. It doesn't say that

12 he belongs to the study group instituted by attorney general Janet Reno.

13 Q. You're a aware that that's what it is, that Janet Reno appointed

14 this group, the Justice Department panel on eye witness evidence, which

15 then issued the recommendations?

16 A. I can't tell whether they're different or the same. You may be

17 correct, but I don't read it.

18 Q. Okay. If you look at page 2, it says that Reno appointed the

19 working group in 1997.

20 A. Yes.

21 Q. So a considerable time before this article was written.

22 A. Yes, two years.

23 Q. Now, let's look at -- it says that recommendations that are going

24 to be offered are these four things at the top of page 3. The officer

25 working with the witness doesn't know which person is the subject, the

1 eye witness understands that the person being sought for the crime might

2 or might not be in the line-up, and these are the four things that you

3 have written at the top of your page 8. Correct?

4 A. Yes.

5 Q. Now, this article does not say that these were matters which were

6 found in those -- in that study in those 36 cases, does it? Does it say

7 that anywhere, Professor?

8 A. Well, it says that the working group in response to studying

9 these cases came with the following recommendations. So logically I

10 assume --

11 Q. Where does it say that?

12 A. On page 1 it says: "In response a group of psychologists, and so

13 on, are prepared within the next several months to release a set of

14 guidelines."

15 Q. Uh-uh.

16 A. And it's in response to these 40 convictions. So my supposition

17 is that their work is based on those 40 cases and not on something else

18 because if they were based on something else like the general literature

19 or so, then I can only say there's nothing new in their recommendations

20 because we've known these recommendations long before. The only new

21 thing to me is that the study group of attorney general Janet Reno has

22 found that especially these recommendations were important in the 40

23 cases. I may -- you may read it differently, but otherwise I can see no

24 logical connection between the way the article starts with 40 false

25 convictions reversed by DNA evidence and the recommendations. The only

1

2

3

4

5

6

7

8

9

10

11

12 Blank page inserted to ensure pagination corresponds between the French and

13 English transcripts.

14

15

16

17

18

19

20

21

22

23

24

25

1 connection can be that the study group instituted because of these 40

2 cases --

3 Q. That's --

4 A. -- came to the conclusion that these were important factors in

5 these cases.

6 Q. That's the inference that you've drawn from this article?

7 A. That's the inference, otherwise the whole paper makes no sense to

8 me. Because just saying that these recommendations are important is not

9 worth having a study group for because we've known that long ago, like

10 the first one, the officer working with the witness doesn't know which

11 person is the suspect. I've written that in 1988 already. We don't need

12 a working group instituted by the attorney general to come to that

13 conclusion.

14 Q. Well, isn't it possible, Professor, isn't another fair reading of

15 this article that this group was brought together because of cases such

16 as these 36 out of 40, but that the recommendations that it's making are

17 drawn from lots of sources, not just these 36 out of 40 cases, but from

18 lots of sources, studies done and lots of information in the field.

19 Isn't that also a fair reading and a sensible reading of this article?

20 A. Yeah, we can speculate about other interpretations of this paper.

21 I think if you want to know, you might well contact one of the authors,

22 one of the participants of this group, but the way I read it is that

23 these recommendations were inspired by looking at these 40 cases. That's

24 the logic of this paper.

25 MR. WHITING: Your Honour, could this be marked as an exhibit,

1 please?

2 JUDGE PARKER: Do you mean that to be confined to that one

3 document? This is the third you've put to the Professor.

4 MR. WHITING: Yes.

5 JUDGE PARKER: Very well. This will be received.

6 [Trial Chamber and registrar confer]

7 MR. WHITING:

8 Q. Now, Professor, there is a -- at the end of this article, it says

9 -- at the very end it says -- it refers you --

10 THE REGISTRAR: Your Honours, the document "Many Eyewitnesses

11 Identified the Wrong Person" would be given Prosecution Exhibit 253.

12 MR. WHITING: Thank you.

13 Q. Professor, drawing your attention to the end of the document,

14 there's in italicised type, there's a reference to -- for further

15 information to the paper on the topic from the December 1998 issue of Law

16 and Human Behaviour. Have you read that article?

17 A. I think I need the specifics. I read Law and Human Behaviour;

18 it's the leading journal, in your profession.

19 Q. Well, I'll show you the article, maybe you'll recall it. It's

20 entitled "Eyewitness Identification Procedures For Recommendations and

21 Line-ups" --

22 A. Yeah, it's in my collection.

23 Q. I don't want you to have to hunt around for it. I'll provide

24 them to you.

25 A. I'm grateful.

1 Q. Now, I provided this article to you last week as one of the

2 articles I might ask you about.

3 A. Right.

4 Q. Had you read it before I provided it to you?

5 A. Yes, I did.

6 Q. Okay. Now, I just want to draw your attention to page 3. And do

7 you see there's a table there, table 1, and it's a sample of 40 cases in

8 which DNA evidence exonerated persons wrongfully convicted. And this is

9 in fact a list of the 40 cases that are referred to in that study.

10 A. Yes.

11 Q. Now, you'll see that in the very right-hand column it says how

12 many identifications were in the case and if there was any other

13 evidence. So, for example, for the first one there's a witness ID, third

14 one it says: "Five witness IDs."

15 A. Yes.

16 Q. Now, would you -- I'm not going to go through the -- a laborious

17 exercise of counting here, but there are 36, as has been reported, 36 of

18 the 40 cases involved witness identification of some sort and 30 of those

19 36 involved a single witness identification. Would you accept that

20 point?

21 A. Yes.

22 Q. Okay. And the dates also of the convictions are on this chart

23 and I'll suggest that they range - you can check it - but they range from

24 1978 to 1991. Is that -- do you accept that?

25 A. Yes.

1 Q. Okay. And these cases, as are described here, are all rapes,

2 murders, violent crimes of that sort. Sexual assaults. Is that correct?

3 A. Yes.

4 Q. Okay. You can put that aside. We may come back to it at another

5 point. Now, if -- can I turn, please, to paragraph 3 of your report.

6 And I'm certainly not going to go over your qualifications, except I want

7 to ask you about this point and it's at page 2, and it's something that

8 we talked about when we met on Friday. It says: "Since 1985 I have

9 served as an expert on questions of perception and memory in an average

10 of 50 trials a year."

11 A. That's correct.

12 Q. Now, when we talked on Friday - and correct me if I'm wrong - but

13 I think you explained to me that ten of those 50 involved trademark

14 litigation.

15 A. That's still on perception of memory.

16 Q. I understand.

17 A. Okay.

18 Q. Of the 40 there were a number of cases that actually did not go

19 to trial, where you consulted before charges were brought. Is that

20 correct? And charges in some of those cases were never brought.

21 A. Yes.

22 Q. That's included in the 40. Is that right?

23 A. I was consulted by the Office of the Prosecutor, yeah.

24 Q. And that's included in your 40?

25 A. 50.

1 Q. Well, I've gone down to -- I'm talking now about criminal trials?

2 A. Yeah.

3 Q. In fact, you told me, as I understand, that you're on a committee

4 that reviews all recovered memory cases in Holland and what -- before

5 charges are brought?

6 A. Recovered memory cases.

7 Q. Yes, that's what I think I said. Yes, all recovered memory

8 cases. And in some of those cases, charges are not brought. Is that

9 correct?

10 A. That's correct.

11 Q. Okay. Now, in fact, in -- you provided us a list. In fact, in

12 2005 you were involved in -- so far you've been involved in four trials

13 involving criminal -- four criminal trials involving identification. Is

14 that correct?

15 A. Not counting this one.

16 Q. Not -- this is number five. In 2004 there were six, and in 2003

17 there were also six. Is that correct?

18 A. That's correct.

19 Q. Okay. So that's the number in those last three years, that's the

20 number of trials -- criminal trials where you've been involved in

21 identification issues. Correct?

22 A. Yes.

23 Q. And on top of that in those years there may be a few more

24 trademark trials, but certainly no more than ten each year.

25 A. Yes.

1 Q. Okay. Now, would it also -- with respect to the Tadic trial in

2 which you testified, would it surprise you to learn that in the judgement

3 you're described in paragraph 547 of the judgement as being a Defence

4 expert. Were you aware of that?

5 A. I'm not least not responsible for it.

6 Q. That wasn't the question.

7 A. My position was very clear. I acted as an expert both from the

8 Defence and for the Prosecution. In fact I was first contacted by the

9 Prosecution and gracefully the Prosecution agreed that I would also

10 answer questions by the Defence.

11 Q. And --

12 A. And how I am described by others I'm really not so much concerned

13 with.

14 Q. Are you aware that you were called to testify at the trial by the

15 Defence? Are you -- were you aware of that?

16 A. I probably would have been aware of it. I think it was in 1997,

17 wasn't it? I don't really remember that.

18 Q. Okay. Fair enough. I think it's 1996 --

19 A. I won't disagree with you on that. You're the expert on legal

20 matters.

21 Q. Well, I ask the question only because you suggested that you were

22 an expert for both sides, and I just wanted to be clear on that point.

23 That's the only reason I've put the question to you.

24 A. I was. I wrote my first report to the Prosecution, not to the

25 Defence.

1 Q. I understand. Now, turning to paragraphs 6 and 7 and 8 and 10 of

2 your report --

3 A. Can you repeat where you're going? Excuse me.

4 Q. Yes. 6, 7, 8, and 10. I'm going to try and --

5 A. Paragraph or page?

6 Q. Paragraphs.

7 A. Yeah, okay.

8 Q. I'm going to go by paragraphs and I'm going to try and deal with

9 those paragraphs as a group. They pertain really to Rule 1. You make a

10 distinction, as you say, between recognition cases and identification

11 cases.

12 A. That's your terminology. That's the terminology of this

13 Tribunal, which is quite confusing to the profession because we don't

14 call it that way, anyway.

15 Q. Whatever the terminology, the concepts represented by the

16 terminology are certainly familiar ones in your profession. Correct?

17 A. No, because we mean different things by identification and

18 recognition than the Tribunal means. But I'm very happy to adopt the

19 terminology the way it's used here.

20 Q. Okay. Well, let's use the terminology you've used in your report

21 because that's all that's going into evidence in this case and that's

22 what we have to talk about. Now, a recognition case is a case where the

23 witness has seen the suspect before the crime occurs.

24 A. Yeah, that's not the way I've used it in my report. For

25 instance, on the top of page 4 you see that I say: "The recognition of

1 an unfamiliar person," and that would be wrong according to your

2 terminology because an unfamiliar person, according to you, cannot be

3 recognised. You see, we use recognition as a term for psychological

4 process, and identification as a term for a legal procedure.

5 Q. Well, would it be --

6 A. You see?

7 Q. Would it be easier -- would it be easier if we used recognition

8 of familiar person and recognition of an unfamiliar person?

9 A. Yeah, but then we are a little bit at odds with the terminology

10 in my report. As long as you're aware of that, it's okay.

11 Q. Well, I'm trying -- I'm trying to make this clear, so I'm trying

12 to use the terminology in your report. Okay. So let's just say that --

13 let's say that a recognition case is a case where you knew the person

14 beforehand --

15 A. Okay.

16 Q. Or you'd seen the person beforehand and an identification case is

17 a case where you never seen the person before the crime is committed?

18 A. Okay. That's the Tribunal's terminology.

19 Q. Okay?

20 JUDGE PARKER: Please understand that is Mr. Whiting's

21 terminology.

22 THE WITNESS: Yeah.

23 JUDGE PARKER: We have no view yet; we are learning.

24 THE WITNESS: Yeah, Your Honour, of course you are true and I'm

25 simply -- I'm definitely not going to contradict you. But I've had a

1 discussion with a different Chamber of the Tribunal before, so my

2 impression is that you're very different -- the various Chambers are very

3 consistent in using this terminology, that's why I said it's the

4 Tribunal's terminology. But of course you're right, it's an inference on

5 my side. It's the terminology that I've only heard within these walls,

6 let me say that. But let's use the terminology because that might ease

7 matters a lot.

8 JUDGE PARKER: Thank you. Now, before you do, Mr. Topolski.

9 MR. TOPOLSKI: I wonder if I could stir the pot a little further

10 and invite caution, particularly on Mr. Whiting's part. At line 20 he's

11 asking for a definition of a recognition case "where you knew the person

12 beforehand." Answer: "Okay." "Or you'd seen the person beforehand as

13 an identification case."

14 So can we bear in mind that we need to stick to those and with

15 those. Recognition is knowing the person, identification is seeing the

16 person. Otherwise confusion will reign, I suspect.

17 MR. WHITING: Well, I -- actually I think that's further confused

18 the issue.

19 The -- let's talk about -- a recognition is if you have ever seen

20 the person before the crime is committed, whether you know them or you've

21 just seen them, we'll call that a recognition case. Is that -- and I

22 draw your attention to your first rule, which is in paragraph 10, where

23 you say: "There should not even be a single occasion at which the

24 witness might have seen the perpetrator before he encountered him at the

25 scene of the crime."

1 So that -- one category case is when the witness has seen the

2 perpetrator before the commission of the crime.

3 A. Yes.

4 Q. Okay.

5 A. That's what I call familiar people.

6 Q. You call what?

7 A. Familiar people.

8 Q. Okay. In that case, in those cases, it's your position that you

9 do not use an identification procedure. Correct?

10 A. No, not entirely. What I say is a multi-person line-up test in

11 that case cannot prove the fact that the witness saw the suspect at the

12 scene of the crime. I'm not talking about all identification procedures.

13 Q. Let's talk about multi-person identification procedures, whether

14 they be a photo spread, a line-up or some other procedure.

15 A. Right.

16 Q. Multi-person procedure. Your position is then that there is no

17 reason to conduct such a procedure if the witness has seen the suspect

18 before the commission of the crime.

19 A. No --

20 Q. Is that fair?

21 A. No, not really. I didn't talk about reason, I think. What I

22 said is if you conduct a multi-person line-up in a situation where the

23 witness might know the suspect from another occasion before the crime,

24 then this multi-person identification test does not prove, cannot prove,

25 that the witness saw the suspect at the scene of the crime.

1 Q. And then --

2 A. I'm not saying there's no reason to do it. I don't know what

3 sort of reasons an investigator might have. I'm only saying such a

4 multi-person line-up does not prove that particular point.

5 Q. Isn't that the point of a multi-person identification test, in

6 your view?

7 A. That's why I would do it, but don't ask me about reasons

8 investigators could have.

9 Q. But is there any other reason to conduct a multi-person

10 identification test? Is there anything else that is accomplished, in

11 your view, when the witness has seen the suspect before the commission of

12 the crime? Does it accomplish anything, in your view?

13 A. Well, it's a quite laborious way to establish a few simple facts.

14 For instance, I give you the example. If I claim that I saw in the

15 middle of the night that my neighbour smashed my car window and stole my

16 radio, then obviously placing my neighbour in a multi-person line-up with

17 five or six unknown people would not constitute a difficult task for me.

18 Of course I can point out who my neighbour is. It might help, that test,

19 also it is somewhat laborious to make sure which neighbour I mean, right?

20 Because I have many neighbours. So it's -- that assertance [sic] that

21 it's this particular neighbour and they didn't arrest the wrong

22 neighbour. It still doesn't prove that I saw him in the middle of the

23 night.

24 Q. And to do that verification that you're talking about, the right

25 neighbour, you don't need a multi-person test?

1 A. I'm saying it would be a quite laborious way of doing it.

2 Q. You could just as easily and with the same result show one

3 picture. Correct?

4 A. Absolutely.

5 Q. You said absolutely?

6 A. Yeah, but you asked me if there would be any other reason to

7 conduct such a test and I said some investigators may do it in that way;

8 I don't know of their reasons. I only say that such a test does not

9 prove that I met the suspect at the scene of the crime. That's the only

10 thing I'm saying.

11 Q. And -- and I side from --

12 A. It's a matter of logic. It's only logic.

13 Q. And I'm just trying to pursue the logic here. Then therefore, as

14 a matter of logic there is no reason that you can think of to show -- to

15 use a multi-person identification test when the witness has seen the

16 suspect before the commission of the crime?

17 A. Yeah -- well, I don't know exactly why you're asking it. My

18 experience is that in criminal trials I'm often confronted exactly with

19 multi-person identification tests in a situation where the witness knew

20 the suspect already before the crime. Apparently all these investigators

21 saw reasons to do that and I cannot even start to guess what their

22 reasons were. I do not think those tests prove their point.

23 Q. The only reason I'm asking you about this is it's your first

24 rule. Your first rule says that there should not be a single occasion at

25 which the witness might have seen the perpetrator.

1 A. Yeah.

2 Q. So that is your rule, and I'm not talking about what other

3 investigators might think.

4 A. No.

5 Q. I'm trying to get your view on matters.

6 A. I think we can easily agree on that rule because I think it is a

7 very wise rule.

8 Q. So in a familiar-person case, when the witness has seen the

9 suspect before the commission of the crime, then really the only issue is

10 whether the witness had the ability, sufficient ability, to actually see

11 the person during the commission of the crime and whether the witness is

12 being honest.

13 A. Right. In a familiar-person case, the main question is estimator

14 variables, if we come back to that terminology. Was a sufficiently

15 favourable condition for the witness to see what he claims he saw? A

16 multi-person identification test does not prove that point any further,

17 so what you have to do is analyse the viewing conditions. And that's

18 where it stops. If you want to believe that witness, you believe the

19 witness but you cannot prove the truth of what the witness says by a

20 further multi-person identification test.

21 Q. With respect to viewing conditions, that is a subject that you

22 have written about in your -- I believe it's your 2004 article, which is

23 entitled: "Familiar-face Recognition As a Function of Distance and

24 Illumination, a Practical Tool For Use in the Courtroom."

25 A. Yes, and I talked about two estimator variables there, which is

1 distance and illumination conditions.

2 Q. And with respect to those two estimator variables -- and you cite

3 this article in your report.

4 A. Yes.

5 Q. With respect to those two estimator variables, is it fair to say

6 that you -- after doing your study, you come to the conclusion that

7 recognitions of familiar faces that occur at a distance of less than 12

8 metres and at an illumination of 30 lux, l-u-x, which is translated in

9 the article as being a badly illuminated room, that those two positions

10 with those two estimator variables or better are reliable?

11 A. I'm not sure whether I used the word reliable. I think I defined

12 the quality of such perceptions in terms of diagnostic values and whether

13 it's reliable or not, that's a rather subjective judgement. But what I

14 said is that given a certain criterion value of diagnostic values for

15 which I usually use 15, then in those conditions that you mention such

16 diagnostic values can be obtained. That's what I said.

17 Q. Well, if I could -- if I could put --

18 A. You may call it reliable.

19 Q. Well, let --

20 A. I think I was slightly more exact.

21 Q. Well, why don't we look at the article; I have copies of it.

22 Can I draw your attention to page 95. In that paragraph that

23 starts: "Since the total diagnostic value," then it says four or five

24 sentences into that paragraph it says: "Area 1 is defined by distances

25 of no more than 12 metres and illumination values of 30 lux or more.

1 Recognitions under these conditions can be said to be very reliable."

2 A. Yes, if diagnostic values are over 15.

3 Q. Yeah?

4 A. It's part of the same sentence.

5 Q. Well, let's look at the last full paragraph on that page. "We

6 concluded that recognitions in area 1, distance of no more than 12 metres

7 and illumination level of 30 lux are reliable and recognitions in area 2

8 defined therein are questionable."

9 A. Yes, that's what it says, but of course you should keep in your

10 mind that I'm really talking about diagnostic values.

11 MR. WHITING: Your Honour, could this document be given an

12 exhibit number, please?

13 JUDGE PARKER: Yes, it will be received.

14 THE REGISTRAR: Your Honours, the document "Familiar-Face

15 Identification As a Function of Distance and Illumination, a Practical

16 Tool For Use in the Courtroom" will be given Prosecution Exhibit P254.

17 MR. WHITING:

18 Q. Now, just to the clear on this area of familiar-person cases, I

19 just want to put an example to you and tell me -- ask you some questions

20 to it. Let's say that a witness knows -- has a neighbour and knows the

21 neighbour for a period of, let's say, a month and sees the neighbour

22 every day, sees the neighbour in various conditions, has some

23 conversations with the neighbour, does not know the neighbour's name. If

24 the neighbour later committed a crime against the witness, under your

25 rule number 1 you would not use a multi-person identification procedure.

1 A. No. To me that would be a familiar person.

2 Q. Okay. All that you could do is have the person recount the

3 circumstances of the crime and then you would have the person come and

4 you'd have -- the court would have to test that person's honesty,

5 correct, evaluate that person's honesty?

6 A. No. The court then should look at what we have agreed to call

7 the estimator variables, which is distance, illumination, amount of

8 stress, and the number of things that you've listed.

9 Q. Okay. And of course evaluate whether the Court believes that the

10 witness is telling the truth. That's of course --

11 A. That's the ultimate question, I would say.

12 Q. Okay. And so there is no multi-person identification procedure

13 that can be done to assist in the Court's evaluation of that scenario.

14 Correct?

15 A. It could be done, but it wouldn't prove the point.

16 Q. It wouldn't prove anything, according to you.

17 A. No.

18 Q. Okay. Let's move to --

19 A. Apart from the facts whether it's the right neighbour or the

20 wrong neighbour.

21 Q. Right. Which you've already told us could be done with a single

22 photograph.

23 A. Yes.

24 Q. Let's move to rule 2, paragraph 11. This is talking about an

25 identification procedure in an unfamiliar-person case. And in this --

1 under Rule 2, if the witness sees the suspect after the commission of the

2 crime, then a multi-person identification procedure also does not have

3 any value, in your view.

4 A. Any logical value, because if then you test this person, the

5 effect would be that you confront the person with one face or one picture

6 of a person that looks familiar for at least one different reason and

7 other people they've never seen before. So when the witness is able to

8 indicate this one person, you don't know whether this is because the

9 witness saw this person at the scene of the crime or because the witness

10 saw this person at another occasion and looks familiar for that reason.

11 Q. As a matter of logic then, the -- an unfamiliar-person case where

12 the witness sees the suspect after the commission of the crime becomes

13 logically like a familiar-person case. That is, there's no logical

14 reason to conduct a multi-person identification test, in your view?

15 A. Yes.

16 Q. Okay. Now, let's move to Rule 3, paragraph 13. You say in

17 paragraph 13 that: "The suspect and all the others," meaning the foils?

18 A. Yes.

19 Q. "Must fit the description given by the witness at an earlier

20 occasion. If this description is in severe disagreement with the outer

21 appearance of the suspect, it makes no sense to test whether the witness

22 has an accurate memory of this suspect; we know he does not."

23 Okay. What I want to get clear on is what exactly this

24 description requirement means. Would you agree that when you're talking

25 about this description requirement, what you are really focussing on is

1 very broad categories, such as man/woman, white/black, very tall/very

2 short. Is that correct?

3 A. Yes.

4 Q. Okay.

5 A. White/dark, long beard/no beard.

6 Q. Okay. My next question was going to be that if there are

7 discrepancies with regard to facial hair, height, weight, build, this

8 would not prevent you from doing the test, would it?

9 A. Well, depends. I said here: "The same holds for gross

10 deviations with respect to age, posture, hair, facial air, specific

11 marks."

12 So with respect to facial hair if the witness says this person

13 has a very conspicuous long beard and in fact you have a suspect of who

14 it's known that he never had a beard and was, say, clean-shaven every

15 day, then there is a -- I would say a gross deviation. If the discussion

16 is about, if you do not shave for three weeks, do you describe that as

17 unshaven or having a beard, right? Then you get in a grey area where it

18 is not settled what you should. But I'm not talking about such an issue;

19 I'm talking about gross deviations.

20 Q. What if the suspect is described as having a beard of some kind?

21 Would it be a violation of this rule, in your view, to show photographs

22 -- a photo spread with -- where all of the persons, including the

23 suspect, did not have a beard?

24 A. Well, as I've said in my report, if you do that, you show the

25 suspect without a beard, clean-shaven, and others, foils, without a

1 beard, you transmit some new information to the witness. You tell the

2 witness in that way, you described a beard, but actually the person you

3 saw -- we believe the person you saw has no beard. And I'm always

4 hesitant to present -- through the testing of someone's memory to present

5 new information to that person's memory. Because what if next time, for

6 instance after he's seen -- he's participated in this line-up procedure

7 and he now testifies in this court, what if the next time he would argue,

8 Well, in fact now I remember he had no beard. Did he change this

9 testimony on the basis of re-finding his original memory or did he adapt

10 his memory to something that he has learned in the line-up procedure?

11 And it's my understanding that line-up procedures are not for giving

12 witnesses additional information, they are for testing their memories.

13 Q. Would it, in your view, invalidate the test? Put aside the issue

14 of transmitting information. Would it, in your view, invalidate the

15 test?

16 A. Well, you see, in some situations a test is weakened. You have

17 weaker and stronger tests, and they can be so weak that in the end you

18 would say, Well, this is not a valid test anymore. I would say that

19 would weaken the test.

20 Q. Can --

21 A. Ideally, in a test - and I think there's no doubt about it, say,

22 in the various publications, even the publications that you have given me

23 - that the people shown in line-up tests must meet, must conform to the

24 descriptions given by the witness as much as is possible. And if it's

25 not possible, yeah, then you can only weaken the test, if you still want

1 to run it. But ideally that would not happen.

2 Q. I'm going to show you the report that you wrote future Tadic case

3 or one of the reports because you wrote several reports.

4 A. Okay.

5 Q. This is --

6 MR. WHITING: This became Exhibit D91 in the Tadic case.

7 Q. It's a letter from you to Defence counsel dated 10 October 1996.

8 And you recognise this letter, Professor Wagenaar?

9 A. Yes, I do.

10 Q. This is a letter that you wrote?

11 A. Yes.

12 Q. Can I draw your attention, please, to paragraph 13. It's a

13 little long, but I think it's worth reading.

14 "One topic that recurred during the hearings of witnesses was Mr.

15 Tadic's facial hair growth. It appeared that Mr. Tadic was sometimes

16 clean-shaven, sometimes badly-shaven, and sometimes bearded. For the

17 subjects who knew him well, this can hardly have been an obstacle to

18 recognition, either at the scene of the crime or at later recognition

19 tests because they were familiar with his various appearances. For the

20 four subjects in group 2," and I'll just interrupt here. Those four

21 subjects were unfamiliar persons. Correct?

22 A. Yes.

23 Q. "For the four subjects in group 2, there may have been a problem

24 because the photograph used in the test shows Mr. Tadic only in one

25 condition, more or less clean-shaven. If the person they saw commit a

1 crime was bearded, they may have had a problem recognising him in the

2 photographic line-up, but that is exactly the reason why the line-up is

3 organised as a difficult test. If the beard problem formed a real

4 obstacle, the witnesses would either not have indicated any picture or

5 the wrong picture. The fact that all four witnesses identified the

6 picture of Mr. Tadic indicates that the beard problem was not an obstacle

7 to identification. The only exception is when the perpetrator was a

8 look-alike of Mr. Tadic was this is not specifically related to the

9 problem of facial hair. As stated before, multi-person identity parades,

10 no matter how well they are organised, cannot exclude the possibility of

11 a look-alike who is the true perpetrator. The beard problem is related

12 to a more general problem of identity parades. The accused in the foils

13 should conform to an earlier description of the perpetrator given by the

14 witnesses."

15 That's your rule. Correct?

16 A. [No verbal response].

17 Q. You're nodding.

18 A. Yes.

19 Q. "I did not obtain a description of all four witnesses in group 2,

20 nor am I certain that they were systematically collected. The general

21 rule is that an identity parade should not be held when a witness's

22 description is decisively different from the accused's appearance, such

23 as when a small suspect is described as a remarkably tall person, a black

24 suspect as a white person, a male suspect as a woman, a long-haired

25 person as a bald person. The inclusion or omission of a beard and a

1 moustache in this case is not such a decisive discrepancy."

2 Do you still hold the view that you expressed in this -- section

3 13 of this report?

4 A. Yes, and I don't see a conflict with what we discussed before.

5 Q. Okay.

6 A. I've said a gross discrepancy would be a person with a remarkably

7 long beard who's now without a beard. I've said a smaller problem --

8 where you only have a problem is where someone is badly shaven or has a

9 -- a beard of two weeks or so and is now shown without a beard, that's

10 only a problem, that's not the basic obstacle to running the test. But

11 ideally, of course, you would present in the test the person who looks

12 like the description. My criticism in this point 13 of this letter, and

13 I hope you have noticed it, is not the way the pictures were presented

14 but the fact that the earlier descriptions were absent. It makes it very

15 difficult to decide where the pictures as used in the test were

16 appropriate or not, and that's where my -- my judgement stops. If I

17 don't have these descriptions, I can't comment on it anymore. So that's

18 the point of point -- of item -- of section 13 is the fact that these

19 prior descriptions were not provided.

20 Q. Now, you -- on this point in your testimony in the Tadic trial,

21 didn't you say that if a photo -- a multi-person identification test, say

22 a photo spread, is shown to a witness who described the suspect as having

23 a beard and in the photo spread the person does not have a beard, then

24 the test, if all of the persons do not have a beard, then the test will

25 show whether the witness's perception was good enough or not, whether

1 they are able to see through the beard.

2 A. What I say here is that if the beard problem would have been an

3 obstacle in the sense that witnesses might have recognised him with the

4 beard but not without, then in fact since all four witnesses identified

5 the picture of Mr. Tadic, you can't say the beard problem was an obstacle

6 here. Right? But mind you, the quality of a test is not defined by the

7 number of positive identifications only, it's also defined by the number

8 of missed identifications. You don't want to set up a test in such a way

9 that the likelihood that witnesses will not recognise a guilty person,

10 that that likelihood is minimised.

11 Q. Well, Professor Wagenaar --

12 A. And if you picture -- if you picture the suspect in a way that it

13 becomes very hard to recognise him, you might increase the probability

14 of, say, a missed identification. That's not what I recommend.

15 Q. That -- okay. My only -- what I wanted to focus on because I

16 suggest that you shifted to another topic. What I want to focus on is

17 whether there is a problem in -- if the suspect is described as having a

18 beard, whether there is a problem in showing -- using a photo spread

19 where the suspect and the foils do not have a beard. My understanding is

20 that if the perception is good enough at the time of the encounter, then

21 a witness may be able to recognise the suspect without a beard, and the

22 photo test, the line-up, has the -- the identification procedure has the

23 potential to show that. Is that correct?

24 A. Yes. But the diagnostic value is still reduced. You're saying a

25 person might recognise the suspect in that case. Of course he might do

1 that. What I'm saying is if it's possible to compose a line-up test with

2 the suspect portrayed just exactly as he is described by the witness,

3 that's to be preferred.

4 Q. Let's see how you express this in Tadic, and I'm going to provide

5 you with the transcript of the first day of your testimony in Tadic.

6 It's from the 24th of October, 1996. And if you could draw -- I'd like

7 to draw your attention to page 5769 of the transcript. And specifically

8 line 23 you say: "We have to distinguish between these two groups," and

9 the two groups you're talking about is the familiar-person group and the

10 unfamiliar-person group. Correct?

11 A. Yes.

12 Q. And you go on to say: "The group who knew him already before

13 with facial hair might not have had a problem because that is the face

14 they were used to. Those who did not know him and saw him for the first

15 time there and then with the beard, initially there is no problem. They

16 do not recognise anything at that time, because that is their first

17 perception of that person. Only later there is a problem, there might be

18 a problem, because if the test presents someone without the facial hair,

19 then they may not recognise him. But again, the test solves that

20 problem; that is the beauty of it. If the perception was not good enough

21 initially, they will not recognise him; they will fail the test. And we

22 know it -- then we know it. If they recognised him in the test,

23 apparently the perception was good enough; that is what the test proves."

24 A. Yes.

25 Q. Do you still hold that view?

1

2

3

4

5

6

7

8

9

10

11

12 Blank page inserted to ensure pagination corresponds between the French and

13 English transcripts.

14

15

16

17

18

19

20

21

22

23

24

25

1 A. Yes, but I would like to ask your attention for something you did

2 not quote, which is on the same issue on the top of page -- the same

3 page, 5769: "That reasoning holds only if the test is well performed,"

4 you see. You're describing a problem that originated in the estimated

5 variables. Now, if you show a person that looks different in the test

6 than the way he was perceived before at the scene of the crime but then

7 you run a suggestive test, then the problem that is created by this

8 discrepancy is easily overcome; that's what I say here. So the test will

9 show whether this problem, facial hair or not facial hair, will lead to a

10 nonrecognition. The beauty of that test exists only if the test is

11 performed properly.

12 Q. I understand. I appreciate that point.

13 A. That is the essential point, because if you have a suggestive

14 test it really doesn't matter what they saw originally or who you show in

15 the line-up. If the test is suggestive enough, you can always manage to

16 let the witness point at one particular picture.

17 Q. Well, a suggestive test is a problem whether you have the beard

18 issue or not. Correct? It's not only a problem if you have this beard

19 issue; it's always a problem.

20 A. It's always a problem.

21 Q. Okay.

22 A. But in the case where there is a discrepancy between what they

23 saw originally and what's presented to them in the test, if you want to

24 resolve that problem, then the quality of the test really becomes

25 critical. That becomes the decisive point. You see?

1 Q. Okay. Staying focussed on this issue of description -- we've

2 talked about beards. I want to talk a little bit about other

3 descriptions, height, weight, build, things like that. Is it fair to say

4 -- would you agree with me that people can know other people very well

5 and be very capable of identifying their face and yet be incapable of

6 providing an accurate description?

7 A. Yes, and there's even some research on that, of course. I

8 myself, I think it was published in Dutch, always use what I call the

9 rule of five. Five years, 5 kilos, 5 centimetres can always be -- you

10 can always make a mistake of 5 kilos, five years, 5 centimetres. There's

11 nothing to worry about. When it becomes substantially more, the question

12 becomes whether the aspect of length, of age, of posture, of weight

13 should not also be a part of the test. The problem, of course if you

14 show only faces and there is a discrepancy of, say, 10 kilos in the

15 weight description, you confront the situation -- you confront the

16 witness with the situation where the witness cannot discover himself that

17 something is wrong with the weight, and this cannot be the person.

18 Q. Professor, you're jumping ahead there. We're going to talk about

19 that later on. I'm just trying to focus on the description, not on the

20 type of multi-person identification test that is conducted.

21 You also have done an experiment, did you not, at your university

22 where you -- you did an experiment with a class of a colleague of yours,

23 a professor who had a -- quite a remarkable moustache, as you described

24 it. And after a ten-week course where the students sat for two hours a

25 week, you asked at the end for the students to describe the professor and

1 you were struck by the fact that quite a few students missed the

2 moustache.

3 A. Yes.

4 Q. Are you also aware that studies have been done which have shown

5 that there is no relationship between a witness's ability to accurately

6 describe somebody and their ability to identify a photograph of that

7 person?

8 A. Sure, yeah.

9 Q. Okay.

10 A. But, I must say, this is not about gross differences.

11 Q. Understood.

12 A. The argument is not that people who describe a person as a black

13 person would later be reliable witnesses in identifying a white person.

14 Q. Right. Okay.

15 A. There must be something wrong there.

16 Q. Aside from those gross cases, very tall, very short, white/black,

17 man/woman, you would agree that there is no relationship between a

18 witness's ability to accurately describe a person and that person's

19 ability to identify a person?

20 A. If you talk about minute details, within the rule of five limits,

21 then there is no relationship anymore. With the gross descriptions, of

22 course there is a relationship.

23 Q. Okay. You appreciate that there's quite a bit of distance

24 between what you've described as the gross differences and the rule of

25 five; there's a big middle area?

1 A. Sure, yeah.

2 Q. Okay. And in that middle area you would also not be surprised,

3 would you, to find individuals who may not accurately be able to describe

4 somebody but who could very well identify that person. Correct?

5 A. Well, let me -- I know you think I'm jumping ahead, but the -- to

6 me the meaning of this, if there are gross discrepancies between the

7 description and our suspect, you do not even start to run a line-up test

8 because you wouldn't know how to compose the line-up test. If it's

9 within the rule-of-five type of discrepancies, there's no problem

10 whatsoever. You can go ahead with constructing your line-up test, and in

11 between there you have to think that it's quite possible that in between

12 those two extreme conditions you think, Oh, okay, let the test tell us

13 the truth. But that all depends on the quality of the test. Because in

14 those situations if then subsequently you do not run the test properly,

15 then you might go wrong. So there is all the more reason if there is

16 some discrepancy in this, what we have called the grey area, to be very

17 careful that you run the proper test so that you won't be misled in the

18 end.

19 Q. I appreciate that answer. Are you also aware of studies which

20 suggest that asking a witness to describe somebody verbally could later

21 actually interfere with their ability to identify that person in a

22 multi-person identification test?

23 A. Yeah, that's called an overshadowing effect.

24 Q. You're aware of that --

25 A. Sure.

1 Q. That some support has been shown for that?

2 A. Jonathan Schooler ran such experiments. Absolutely.

3 Q. Would you agree that if there is a discrepancy of the description

4 of the -- provided by the witness and the suspect, that the foils should

5 match the suspect and not the description provided by the witness. Would

6 you agree with that?

7 A. Well, that depends on what sort of discrepancy we are talking

8 about. See, there is always the requirement that the suspect should not

9 stand out in the line-up from the foils. So clearly, if the suspect does

10 not meet the descriptions and all the foils do, then the suspect would

11 stand out; it would be an improper trial. So the logic would be that at

12 least all the foils should look very similar to the suspect. If that

13 means that in the end no one fits the prior description, you're running a

14 rather funny test, especially when we're talking about gross

15 discrepancies.

16 I've seen a test, just to give you an example of how gross it can

17 get, where a person describes a white culprit, and then he's confronted

18 with a line-up of black people with the question of who of these black

19 people is the white person you described? That doesn't make sense to me,

20 and I think we'll easily agree on that. And the question is: How far on

21 the road to such a strange confrontation can you go? When does it become

22 meaningless? And that's a matter of judgement. Ideally, that problem

23 should not arise. Ideally, the suspect should fit the prior descriptions

24 and you have no problem.

25 Q. But just to be clear, if there are minor discrepancies, then the

1 most important thing is to match the foils to the suspect.

2 A. Sure.

3 Q. Okay. Now, let's move to paragraph 14, Rule 4, and this is

4 matching of the suspect and the foils. We've already talked about this

5 to a large degree. You say that it is necessary to do the and Test.

6 Right?

7 A. Yes, that's what the report said -- what my report says.

8 Q. Now, that is not something that is required by the PACE rules

9 which you have attached to your report, is it?

10 A. No.

11 Q. Now, it -- we had talked before about this commission that was

12 appointed by Janet Reno to come up with guidelines for the Department of

13 Justice in the United States. Are you -- have you seen the guidelines

14 that that committee came up with? Have you seen --

15 A. Yes, we've just looked at them. Yes.

16 Q. Before I provided them to you last week, had you seen them?

17 A. Yes.

18 Q. Okay. Are you aware that the Department of Justice in the United

19 States guidelines also do not require that the Doob and Kirshenbaum Test

20 be done. Okay?

21 A. Yes.

22 Q. Now, you also say that the foils should belong to the same ethnic

23 group as the suspect. Correct?

24 A. Yes.

25 Q. Okay. And that is also, as I understand it, not a requirement of

1 the PACE rules or of the Department of Justice rules. Right?

2 A. No, it's not even in the Dutch rules.

3 Q. Okay.

4 A. And as I told you, I become aware of that problem only in the

5 Tadic trial. And subsequently when doing -- I did some research, as I

6 told you, on the effect of different ethnic groups in the former

7 Yugoslavia. And right now I'm running various experiments on other

8 ethnic groups that might end up in the same line-up, unnoticed by those

9 responsible for the line-up because they're not sensitive to this

10 problem.

11 Q. Now --

12 A. And I think it's a real -- it's a real problem that especially

13 emerges with investigators, prosecutors, Defence counsels, and judges all

14 coming from the same ethnic group and not being sensitive to something

15 which the witnesses may be sensitive who come from a different ethnic

16 group. It's a real problem. You're quite right that it has not been

17 realised until recently, so it did not end up in the formal rule. And

18 what I say in point 14, because you made a quick shift there in citing

19 me, I say: "It is necessary to establish that the test is not biased

20 against the suspect."

21 That is clear, and then I go on saying: "This is done in the

22 so-called Doob and Kirshenbaum Test."

23 I do not say it is necessary to do it through that test, there

24 are many other ways, but it is at least necessary to make sure that the

25 test is not biased. In the Dutch rules, translated for you as I

1 mentioned before, in the Dutch rule it says not that you run a formal

2 Doob and Kirshenbaum Test, but that you have two test observers who judge

3 the fairness of the line-up, and that's the practice in the Netherlands

4 that every line-up is tested by two independent test observers who are

5 not familiar with the case. And I think that's a good procedure.

6 Q. Professor --

7 A. I still think the Doob and Kirshenbaum Test is an excellent

8 manner of doing it.

9 Q. With respect to the ethnic identity of the foils, you said:

10 "It's a real problem."

11 Wouldn't it be more accurate to say it's maybe a real problem,

12 that it's a potential problem?

13 A. Well, I provided you with a list of cases in which I acted as

14 expert on identification in the last three years. If you read the names

15 of the suspects in those cases, you may realise why I said that because

16 many of these names indicate people who came from different ethnic

17 groups. And much of my testimony involved the problem of an unfair

18 composition of line-ups, consisting of people from different ethnic

19 groups. So it's -- that's why I call it a real problem. It's a problem

20 in reality.

21 Q. But it's a problem -- it's not a problem which has yet been

22 demonstrated in studies, has it?

23 A. Well, I demonstrated it for the people inhabiting the former

24 Yugoslavia, together with my students Shara Lochun. I think that paper

25 has also been copied for you because you asked for it. I demonstrated

1 that there is a very real problem, and that as far as we could trace it,

2 because it wasn't known, that the people composing the line-up of Tadic

3 in fact had come from different ethnic groups.

4 Now, what you will see, and that is I think clearly established

5 in the PACE rules, is that the identity of the foils in line-ups should

6 also be recorded so that we know who they are and if a discussion starts

7 on are they from different ethnic backgrounds, that at least that

8 question can be solved.

9 Q. Well --

10 A. That's in the PACE rules and that is of course with this problem

11 in mind.

12 Q. Let's -- let's look at the study that you provided regarding the

13 former Yugoslavia. I have copies of it. And while this study's being

14 distributed, it's fair to say that this is the only study that's been

15 done on the former Yugoslavia, is that correct, that you're aware of?

16 A. The only study that I'm aware of, yeah.

17 Q. This study has not been published, has it?

18 A. It's a dissertation, and dissertations are, in principle,

19 available.

20 Q. But not published in any journal?

21 A. No.

22 Q. Okay. Now, this was a study of possible stereotypes that people

23 may or may not hold about people from the former Yugoslavia, number one;

24 and number two, about different ethnic groups within the former

25 Yugoslavia. Is that right?

1 A. Yes.

2 Q. The study -- this did not study whether people from the former

3 Yugoslavia or from outside the former Yugoslavia could actually

4 distinguish by appearance between different ethnic groups within the

5 former Yugoslavia.

6 A. Could you kindly repeat that?

7 Q. Okay. This study that -- you said -- this study focussed on

8 stereotypes; it did not study whether people could actually distinguish

9 in reality the ethnic identity of a -- an individual from the former

10 Yugoslavia. Correct? It did not study that, did it?

11 A. No.

12 Q. Okay.

13 A. But that's also not what you do in a Doob and Kirshenbaum Test.

14 Q. Okay. I'm just trying to get clear on what was studied and what

15 was not studied. Okay. Now, this study had 26 people from the former

16 Yugoslavia and 26 people from Holland participated in the study.

17 Correct?

18 A. As subjects. Not their faces.

19 Q. They were the ones going through the tests. Correct?

20 A. Yes.

21 Q. Okay. Of the 26 people from the former Yugoslavia, seven

22 identified themselves as Serbs, one as Hungarian, and 18 said they could

23 not describe their ethnic origin. Correct?

24 A. Yes.

25 Q. So apparently no Albanians participated in the study?

1 A. Apparently --

2 Q. Apparently no Albanians participated in the study. Is that

3 correct?

4 A. They didn't tell us they were Albanians.

5 Q. Nobody himself or herself as an Albanian?

6 A. No, which doesn't exclude that in fact they were.

7 Q. The 52 subjects were asked to perform three tasks. The first was

8 to look at a group of photos and say whether or not they thought the

9 people in the photos were from the former Yugoslavia. They were not

10 permitted to say, I cannot judge, I cannot tell. They were forced to

11 choose. Correct?

12 A. Yes.

13 Q. If they thought that a person was from the former Yugoslavia,

14 then they were asked to identify the ethnicity of the person, even if

15 they had to guess. Correct?

16 A. Yes.

17 Q. Again, they were not allowed to say, I can't tell. They simply

18 had to guess the -- take a guess at the ethnicity. Correct? Yes?

19 A. Yes.

20 Q. They were then asked to provide descriptions based on a list of

21 characteristics of their stereotypical image of the different ethnic

22 groups in the former Yugoslavia. Correct?

23 A. Yes.

24 Q. So again, the study was not whether somebody could tell if

25 somebody was Albanian, Serb, or Croat. Right?

1 A. No, and of course we couldn't do that because the identity of

2 these people is not known.

3 Q. Okay. The people you were working -- the photographs you were

4 working with, the identity was not known.

5 A. These are people from the Tadic line-up.

6 Q. Right.

7 A. And the commission was to study the possible misleading effects

8 in that line-up.

9 Q. Okay.

10 A. So it's given that we don't know who these people are.

11 Q. Now --

12 A. Right?

13 Q. Isn't it true that among the 26 people who participated in the

14 study from the former Yugoslavia, it was found that their stereotypical

15 images of Serbs and Albanians were in fact quite similar?

16 A. Yes.

17 Q. Okay.

18 A. It's also true that they had quite distinct stereotyped images of

19 what typical people from those groups look like, and the whole point of

20 this exercise is that if just by accident, just by accident, if the

21 suspect is known to have come from a certain ethnic group, if just by

22 accident the suspect appears to be one, the only one, or one of the few

23 people shown to a witness who conformed to this stereotyped image, he

24 will attract the attention. And in fact, a number of effective foils in

25 the line-up becomes much smaller than you think it is. That's the danger

1 of running an identification test with people who have come from

2 different ethnic groups. They might conform or not conform to some

3 stereotyped images and the only way to find out whether witnesses will

4 recognise these stereotypes is run something like a Doob and Kirshenbaum

5 Test; that's the excellent way of doing it. And in that way you can get

6 a line-up that's neutral with respect to ethnic stereotypes and the test

7 becomes more reliable. That's the only point.

8 I'm not saying that people are able to recognise different ethnic

9 groups; I'm saying that if only they have some stereotyped idea, then it

10 could go wrong. It might be easier for you to understand. Let's assume

11 we're looking for an Italian who committed a crime and now we place a

12 person from the Netherlands who is not Italian but who's short and dark

13 between others who are short and dark. The question is whether witnesses

14 might now conclude, this must be the Italian we're looking for. You see?

15 It's only the stereotype.

16 Q. But, Professor Wagenaar, in that example you would be able to

17 look at the photo spread and see distinctions between the suspect and the

18 foils or some of the foils. For example, you would see that the suspect

19 was short and dark and the foils were Dutch-looking which is perhaps tall

20 and blond.

21 A. What I'm saying is that there are some ethnic groups that look

22 different to people who are familiar with those ethnic groups but not

23 familiar to us. So we cannot easily judge whether such a line-up just by

24 looking at the pictures is a fair line-up or not. You can only ask

25 witnesses who have experience with this group.

1 To give you an example again, you might make a line-up with all

2 sorts of Chinese people. Now, would a Chinese witness be sensitive to

3 the difference between Chinese from the north or from the south or from

4 Taiwan or even from Mongolia or maybe even a single Korean among them or

5 Japanese? It's possible that we, all coming from the West, would not see

6 there is something wrong in the line-up, whereas witnesses coming from

7 that area would immediately see it. That's why I'm saying you need some

8 sort of pre-test before you actually show a line-up to witnesses, you

9 need a pre-test to make sure that this pre-test is not biased and that to

10 the actual witnesses in fact some of the foils are not acceptable choices

11 because they do not fit the description.

12 Q. Just to take two points from this. Number one, you might expect

13 that a Chinese person who could distinguish between the north and the

14 south Chinese, for example, might be able to -- even though you would not

15 perceive it initially, that witness or somebody else in that witness's

16 shoes could point to differences that might be minor or overlooked by

17 somebody who is not sensitive to them but which would still be visible.

18 Correct?

19 A. I don't know exactly what you mean. What I mean to say is that a

20 line-up test should not be biased in this manner.

21 Q. But what I'm trying to determine is whether these are invisible

22 differences or they are visible differences which are not -- that the

23 witness -- that the observer might not be sensitive to. Do you

24 understand the distinction?

25 A. It's quite possible that the witness, although he's sensitive to

1 it, might not be able to express what the difference is --

2 Q. And if you did --

3 A. And unconsciously see that there is only one person who really

4 comes from the group from which the actual culprit came. It's not

5 certain at all that the witness that would perceive this would also be

6 able to tell you this, hey, this line-up is no good, there is only one

7 candidate in it. It's possible that the witness might not realise that,

8 but through this suggestive composition be led to the one picture, pay

9 all the attention to this picture, and then seem to recognise this

10 person.

11 MR. WHITING: Perhaps this is a convenient time, Your Honour.

12 JUDGE PARKER: Thank you, Mr. Whiting.

13 We must adjourn for the tapes again and resume at 6.00.

14 --- Recess taken at 5.40 p.m.

15 --- On resuming at 6.04 p.m.

16 JUDGE PARKER: Mr. Whiting.

17 MR. WHITING: Thank you, Your Honour.

18 Your Honour, could the last study that was discussed, "Ethnic

19 Bias and Identification Procedures" be given a number, please?

20 JUDGE PARKER: Yes.

21 THE REGISTRAR: That will be Prosecution Exhibit P255, Your

22 Honours.

23 MR. WHITING:

24 Q. Professor Wagenaar, moving on to Rule 5, paragraph 15, you state

25 that: "Witnesses must be properly instructed," and specifically you say:

1 "The instruction should explain to the witness that the wanted person may

2 not be in the line-up."

3 Correct?

4 A. Yes.

5 Q. Okay. Now, the -- the study that is most often cited in support

6 of that rule is a study that was done by Malpass and Devine in 1981. Is

7 that correct?

8 A. Yes, that's one of the sources but not the only source.

9 Q. That's a study that, in fact, you have cited. Correct? Not in

10 this report but in other places.

11 A. Yes.

12 Q. Now, in that study, that study compared a group of people who

13 were told that the perpetrator was in the line-up with a group of people

14 who were told that the perpetrator may not be in the line-up. Correct?

15 That's how that study was done.

16 A. Yes.

17 Q. Okay. Now, the danger of not telling a witness that the

18 perpetrator may not be in the line-up is that the perpetrator -- is that

19 the witness will assume the perpetrator is there and if the witness does

20 not recognise the perpetrator will guess. Isn't that the --

21 A. Yes --

22 Q. -- danger of not following that rule?

23 A. Yes, that is a danger.

24 Q. Now, if you have a fair line-up, those guesses will be evident.

25 Correct? It will be evident they are guesses.

1 A. If you have a fair line-up procedure.

2 Q. Yes, a fair line-up procedure and a fair line-up --

3 A. You said a fair line-up.

4 Q. Okay.

5 A. The line-up is just the composition of the photographs. The

6 procedure is what the investigator does.

7 Q. You need both, a fair line-up and a fair line-up procedure.

8 A. And a fair procedure. That's why I -- yeah.

9 Q. If you have those two things, then the guesses will be evident.

10 Correct? They will show up as guesses because they will be random.

11 A. But if you combine -- just to finish your sentence. If you

12 combine say an increased tendency to guess with something suggestive in

13 the line-up or a suggestive behaviour on the part of the investigator,

14 then the effect will be disastrous. And I also mention that I think in

15 my report, it's the combination, the combination of weaknesses in the

16 procedure that may lead to the wrong results; it's never just one

17 problem, it's some defaults in the procedure combined with other

18 defaults. And the wrong instruction is a very good candidate for

19 combining with other factors.

20 Q. Okay. That's very helpful.

21 Let me -- let me read to you a sentence from your -- or a couple

22 of sentences from your book "Identifying Ivan," which I have a copy of

23 here if you need to refer to it. On page 50 -- and I think it

24 illustrates this point that you've just been making. On page 50 you

25 write: "In the previous paragraphs, I have demonstrated that the false

1 identification of a suspect in a target-absent line-up is enhanced by the

2 tendency towards making positive responses, but as explained in chapter

3 1, there is a second line of defence. Even when witnesses respond to

4 target-absent line-ups, there is no special reason why they should point

5 to the suspect instead of a foil."

6 A. There ought not to be.

7 Q. Correct. There ought not to be?

8 A. In a fair line-up, that is so.

9 Q. Okay. Now, you also say in paragraph 15 that the question: Do

10 you recognise any of these people? Is wrong. Now, isn't the only

11 problem with that question that if somebody says, for example, in

12 response to that question: I recognise the person in photograph 2, that

13 you then have to ask follow-up questions such as: How do you recognise

14 that person or from where do you recognise that person?

15 A. No, I don't think so. This particular sentence is preceded by

16 another sentence where it's explained that if the witness is confronted

17 with a line-up in which there's only one suspect, but if that witness has

18 described a number of perpetrators it should be made clear in the

19 instruction to the witness for which of those perpetrators he is tested.

20 So for instance, if he describes one as, say, the person who beat me,

21 there were various people, there's one person who beat me up and he

22 looked like this, then in the identification test it should be said, this

23 test is testing the question whether in this line-up you recognise the

24 person who you described as the one who beat you up. And then you give

25 him the warnings that he may not be there and so on. So it should be

1 clear in advance if the witnesses described various culprits for whom

2 specifically they are tested now, otherwise -- and it's just a matter of

3 I would say optimising the outcome of the test. Otherwise, you're making

4 it very difficult for the witness and the likelihood that he doesn't know

5 for whom to look makes the witness so uncertain that in the end he is not

6 pointing at anyone.

7 And the point of an identification test is not to let the witness

8 fail. In fact the point of an identification test is of course to

9 collect evidence against a possible really guilty person. So you should

10 not design procedures in such a way that make it unlikely that the

11 culprit will ever be recognised, so therefore you should be specific and

12 tell them there's one person you described like this, do you recognise

13 him, can you point him out, in this line-up or can't you? And that's why

14 I add the question: Do you recognise any of these people? Is wrong. It

15 should be: Do you recognise any of these people as the person you

16 described as. It should be specific.

17 Q. And the danger of the more general question: Do you recognise

18 any of these people, is that the witness may not be clear on what the

19 task is and may fail, therefore, to recognise anybody?

20 A. Yeah, or give you very confusing evidence. Like -- I'll give you

21 an example which comes from the IRA murders in Roermond here which caused

22 quite -- two Australian tourists were murdered here. There they were --

23 various people described by witnesses and they were confronted with

24 several line-ups, but they were not told for which described perpetrator

25 each of the line-ups was. So a witness said in line-up one, yeah, there

1 I recognise, say, the tall gunman. And in the second he said, there I

2 also recognised the tall gunman, which -- and it's quite possible in one

3 of them he recognised them. But now you end up with a logical

4 impossibility, evidence that cannot be used, just because the

5 instructions were unclear and the witness had to guess what he should do.

6 That's not helping anyone.

7 Q. Some of those problems of course could be clarified by follow-up

8 questions. Correct? Is that right?

9 A. Quite often, no. Say, an identification is something you do and

10 is a -- once you have said, I recognise that person, and then later on it

11 must be turned back because he was looking for, say, the wrong person,

12 you cannot turn it back. So you better get it right straight from the

13 beginning and give them the proper instructions.

14 Q. Now, you also say in this paragraph that: "A positive response

15 should only be made in case of an immediate and certain recognition."

16 Now, in your book "Identifying Ivan," you say, do you not, that it is

17 firmly established, and this is your view, there's perhaps some

18 controversy about it, but in your view "it is firmly established that

19 confidence and identification accuracy are not closely related. Is that

20 correct?

21 A. Yes.

22 Q. You also say, at page 89, that "uncertainty itself is not a

23 predictor of low accuracy."

24 A. Yes.

25 Q. So why then have a rule where only a certain response is

1 acceptable?

2 A. Because witnesses, if they're not certain, say phrases like:

3 This person looks familiar. I think I recognise this person but I don't

4 know where I've seen him. Or This person I recognise for 70 per cent.

5 And uncertain recognitions just ask for such statements, and it's very

6 difficult to know what the court should do with such statements.

7 So it is far better to tell the witness, Just point -- only point

8 at someone if you're certain and your evidence is: I recognise this

9 person with 100 per cent certainty; then we know what we're talking

10 about. And I'm sure that with the uncertain recognitions, some of them

11 might be correct. But if they are -- if these sentences are added to it,

12 we don't know anymore what they mean. And in my experience, a

13 recognition with full certainty may occur or will occur more often than

14 you think, so you not exclude large numbers of recognitions that are now

15 not counted anymore.

16 Q. But if --

17 A. So there is no disadvantage in saying: Just respond if you're

18 certain and only then, and don't tell us: I recognise him for 70 per

19 cent.

20 Q. But, Professor, if confidence and accuracy are not correlated, as

21 you say they are not, then isn't it -- then isn't it true that what you

22 said about less-confident responses that witnesses might give could also

23 be said about confident responses, that is that you can't be certain

24 about what it means. Just because a witness says: I'm 100 per cent

25 certain, how do you know any better what that means than I'm 80 per cent

1 certain if there's no correlation between confidence and accuracy?

2 A. No, I didn't say that a certain recognition is -- if the witness

3 says: I'm 100 per cent certain that I recognise him, that the witness is

4 100 per cent of the time correct; I didn't say that. What I said is that

5 uncertain responses with all sorts of qualifications make it extremely

6 difficult to interpret what they mean. What does a person mean who says

7 I -- this person looks familiar but I don't know where I've seen him.

8 Well, the question is to identify a person he has seen at the scene of

9 the crime and he described this person. Now he seems to -- someone looks

10 familiar, but he doesn't look, apparently, similar to the one person he

11 described. So why would that helpful, such points? It just raises

12 questions and instructions should be proper. You should tell him, do not

13 point to anyone who looks familiar; do point to someone who according to

14 you, is the same person you described as so-and-so.

15 Q. Now, are you aware that Gary Wells, we've spoken about Gary Wells

16 before, that he recommends that the instructions include an instruction

17 that the witness state in his or her own words how certain he or she is

18 of any identification?

19 A. Yes.

20 Q. Would you agree with that?

21 A. No, no. I think that's a confusing procedure. I've never

22 recommended that. And in fact, I've never seen cases where a much more

23 -- a stronger instruction has led a problems.

24 Q. Can you --

25 A. No, I don't agree with that procedure. I think it's -- in

1 principle it's inviting people to guess because you're telling them, if

2 you're not certain, if you're only 70 per cent someone looks familiar but

3 you don't know why, you may still respond. That means you're lowering

4 the threshold of a positive response and that lowering of a threshold

5 will also increase the number of false positives, and that's exactly what

6 we don't want.

7 Q. Well, Professor, aren't those matters that are properly up to the

8 court to evaluate, that is uncertain versus certain responses, to take

9 that all into consideration when weighing the evidence?

10 A. Absolutely and they will have to, because I read in the

11 transcript that some of those responses are of that kind.

12 Q. But your rule would foreclose all but the certain responses.

13 A. My rule if applied -- not faced the Court with such problems.

14 Q. Now, you're aware also that the PACE rules include a provision

15 that if a person selects a photograph but is unable to confirm the

16 identification, the person showing the photographs shall ask that witness

17 how sure they are that the photograph they have indicated is the person

18 saw on the specified earlier occasion.

19 A. Yes, I know that. Now, what if the witness says I'm 70 per cent

20 sure? What do we do with them or what do you do? I don't have to do

21 anything. What does the Court do?

22 Q. Isn't that a matter for the Court?

23 A. I think my task is to design a procedure, to advise about a

24 procedure, that gives the Court as much certainty as is possible. I

25 mean, I think therefore it's much better to tell witnesses, Only point as

1 someone if you're certain. It's not only certain, it's also immediate.

2 Q. Let's move to Rule 6, paragraph 16. You say that: "It is

3 essential that the test be administered by an investigator who is not

4 involved in the case."

5 Now, I take it you are aware that this is a rule that is followed

6 in some countries, for example, in the United Kingdom, is not followed in

7 other countries; for example, is not followed in the United States as a

8 matter of routine. Are you aware of that?

9 A. Yes, I am. I saw this was -- this was missing in the rules of

10 the Department of Justice of the United States, and I was surprised.

11 Q. Now, you would agree that if this rule is not followed, then it

12 doesn't mean that there was suggestiveness, it just means that there was

13 a possibility of suggestiveness. Correct?

14 A. Yes.

15 Q. Okay.

16 A. It also means that we cannot decide whether it has happened or

17 not, whereas if the person who administers the test is simply not aware

18 of the identity, that problem does not arise.

19 Q. Would you also agree that for suggestiveness to be a problem you

20 need two components: You need suggestiveness on the part of the

21 investigator and you need a witness who is susceptible to suggestiveness.

22 Correct?

23 A. Yes.

24 Q. So I want to read to you from page 153 of your book "Identifying

25 Ivan" where you say -- and tell me if you still agree with this: "But in

1

2

3

4

5

6

7

8

9

10

11

12 Blank page inserted to ensure pagination corresponds between the French and

13 English transcripts.

14

15

16

17

18

19

20

21

22

23

24

25

1 no way can the expert determine whether the witnesses were in fact

2 influenced by the suggestive properties of the line-up. It is still

3 possible that the witnesses recognise the suspect straightaway, that they

4 would also have recognised the suspect if there had been no suggestion.

5 The court can be told about the reliability of identifications in

6 general, but it must decide itself whether it will accept the specific

7 identification as reliable or not."

8 Do you still agree with that?

9 A. Yes.

10 Q. And that's a case where suggestiveness is actually present in the

11 line-up?

12 A. Yeah, but I'm not advocating that the person who conducts the

13 test should be aware of the identity of the suspect.

14 Q. I think that's clear.

15 A. I'm only saying if he is aware, then something is happening that

16 can only be evaluated by the court and we don't know how they should do

17 that but then they must do that. They have an extra problem. Whereas if

18 you follow Rule 6, then the court doesn't have that problem because that

19 sort of suggestion simply cannot have happened.

20 Q. Let's move to Rule 7, paragraph 17, where you say that "all

21 attempts at identification, including negative ones, must be reported."

22 Now, you would agree, I take it that while all identification attempts,

23 whether successful or not or positive or negative, must be reported and

24 disclosed to the defence, that it does not follow that you can just add

25 up the score and come out with a result. Right? You cannot just say,

1 for example, Well, here we had seven out of ten identifications;

2 therefore that's reliable, whereas seven out of 15 is not an acceptable

3 number. Would you agree with that?

4 A. There's no rule for determining the number of positive

5 identifications that you need. There is another rule saying if you have

6 a negative identification, at least there should be some explanation why

7 that has happened. You cannot simply ignore the negative one and say,

8 Well, we have enough positive ones.

9 Q. The court has to look at all the circumstances of the positive

10 identifications and the negative identifications --

11 A. And the negative, yeah.

12 Q. -- and consider all of the circumstances.

13 A. Yeah, it's quite possible that you have a negative identification

14 and that you understand why that has happened and that it doesn't -- it

15 doesn't harm, it doesn't mean anything for the positive ones. There are

16 also conditions where there is a negative identification by someone who

17 was in an eminent position to view the culprit and if he just says

18 positively, Well, the man I saw is not in this line-up, well that's a

19 very significant statement.

20 Q. And you would have to take into account not just the person's

21 ability to view or their stated ability to view, but all other factors

22 with relation to that.

23 A. Everything that happened in the meantime --

24 Q. And --

25 A. Sure.

1 Q. And that witness's attitude and that --

2 A. Absolutely. But negative -- negative results of a line-up test

3 are in principle as meaningful as positive ones. And if you want to

4 accept the positive ones as evidence, at least you should be concerned

5 about understanding where the negative ones have come from.

6 Q. And that's the court's job?

7 A. Absolutely.

8 Q. Now -- but again, the numbers themselves don't tell you very much

9 without knowing all of the circumstances. Correct?

10 A. The numbers -- it's not -- it's not a voting game. It's not 5

11 against 4, so the 5 win or something.

12 Q. Okay. Let's move to paragraph 18, Rule number 8 where you say:

13 "Witnesses are not allowed to transfer any information about line-ups to

14 other potential witnesses."

15 Now, this is an example of one of those rules, is it not, where

16 if you have a fair line-up procedure and fair line-up, that this rule --

17 a violation of this rule or a potential violation of this rule may not be

18 so significant?

19 A. Yeah. And I expressly said "if that is not practically

20 feasible," because it cannot always be done. At least I think you can

21 request that the best possible measures are taken. So if it's not

22 possible to separate witnesses for a long period, at least you should

23 tell them that they should not discuss the line-up; that's one thing.

24 Another thing is, very simple, that if you have -- if you present

25 several witnesses with this -- in principle, the same selection of

1 pictures, you may rotate the positions of where the persons are located

2 on that line-up. So that communication, Well, I saw him, he was in the

3 top row, might not be helpful to the next witness. You may even envisage

4 that if you have a new witness you find new foils. So there are lots of

5 precautions you can take if it can be understood that the witnesses are

6 in contact with one another.

7 Q. But just to get back to my question: If you are faced with that

8 situation and you take precautions and you have otherwise a fair

9 procedure and a fair test, then a violation of this rule may not be so

10 significant. Correct?

11 A. It is thinkable that this rule is violated and it does not lead

12 to the conclusion that the identification test is therefore not valid and

13 should be nullified, yeah. It's possible that this cannot be prevented.

14 The least a court should do is get -- be informed about such things and

15 then evaluate itself what it means to them; that's all you can do about

16 it. But very explicitly I added here, which I did not with many of the

17 other rules, I added the sentence "if it is not practically feasible"

18 because, I agree with you, it is not always possible.

19 Q. And where it's not possible, then if things otherwise --

20 A. Then at least you can take some precautions and for the rest it's

21 up to the court to decide how important that is.

22 Q. And your own view is that a violation of that rule may not be so

23 significant in isolation? This is one of those rules in isolation.

24 A. Yeah, if you took all the precautions, then the effect of it is

25 also is diminished because then the information -- although the

1 information -- if one witness says, I took part in an identification. He

2 was there, I saw him. That's of course information that may convince the

3 next witness that in fact the person who looks most like the person he

4 remembers, that's him. And the warning in this instruction may be the

5 person is not in the line-up at all, may not be effective anymore. So if

6 it's possible, he should prevent those contacts.

7 Q. Of course. And of course in that scenario that you just

8 described, if everything else was fair, a random guess by the witness

9 would show up on the test. That would be revealed by a correctly done

10 test.

11 A. If the test is run properly and only the threshold to give a

12 positive response is lowered but nothing else, then you would still know

13 it; that's the beauty of the test. But combined with other factors, this

14 again -- this is especially harmful in combination with other factors.

15 In itself, it might not be so harmful.

16 Q. Now, let's move to paragraph 19, Rule number 9, and you set out

17 the different options. And here today you described this as a kind of

18 scale of options with respect to a line-up, an identity parade, or a

19 photo-spread procedure. Now, isn't it true that recognition of an

20 individual is almost always entirely done by recognising the face?

21 A. Well, you are of course aware of the fact that there is some

22 research on, say, the diagnostic value of these various forms of tests,

23 where a test in which you see as much of the person as you experienced,

24 say, during the encounter at the scene of the crime, and that might

25 involve seeing the face from different angles, seeing how his face moves,

1 hearing him speak, seeing how he moves, how he limps, or whatever. All

2 those things may add -- and I only said it's a kind of graded scale. I

3 did not say that at the end of the scale the photographs of faces only is

4 inferior, is not to be accepted. I only said this sort of a graded

5 scale, but I've also said that although the message with the best

6 diagnostic value also have some disadvantage, like it's very difficult to

7 organise, it's much easier and you can in fact do many more of those

8 procedures if you can use photographs. So in fact I have never

9 recommended and I don't think you can find that in my publications that

10 you should not use photographs.

11 Q. Well, in fact, let's look -- if we could look at page 43 of

12 "Identifying Ivan" and you say -- you're talking about different options

13 and you say: "Another factor was presentation by stills, video, or live

14 line-ups," and then you go on to say: "The difference between the

15 recognition rates in these conditions was marginal or non-existent."

16 Do you still hold that view today?

17 A. Yeah, marginal in the sense that the diagnostic value of -- and

18 all of my own work, all my studies, are done with photographs of faces.

19 I have obtained very high diagnostic values. So there's simply no way I

20 could say that photographs are not good enough, if used properly. I'm

21 not saying that.

22 Q. Okay.

23 A. So if that's the point of this discussion, yes, photographs are,

24 if used in the proper manner, perfectly acceptable.

25 Q. Okay. Now, let's move to paragraph 21, the -- what if the rules

1 are violated. And if we could look at page 83 of the -- no, I'm sorry,

2 it's not 83. I'll have to -- if you could look at the Tadic transcript.

3 Do you still have it in front of you?

4 A. Yes.

5 Q. And if I could draw your attention to page 5779. And the

6 question that's put to you, and I just want to see if you still agree

7 with this position: "If I may proceed to Rule 49, should a breach of

8 rules as we have been discussing so far," and just to pause for a moment,

9 in Tadic you had 50 -- you talked about 50 rules in Tadic. Correct,

10 Professor? In the Tadic --

11 A. Wait a minute. You are in line 7 of this page?

12 Q. Yes.

13 A. Okay.

14 Q. I just want to pause for a minute because there's a reference to

15 Rule 49 and I just want to clarify for the benefit of the Court that in

16 Tadic --

17 A. I'm not sure what this refers to.

18 Q. We'll get to this in a moment, in Tadic and in your book

19 "Identifying Ivan" you have 50 rules; correct? And for the purposes of

20 this case you've reduced them to ten rules. Is that correct?

21 A. Yes, that's correct. But on this page you're referring to me

22 there's even mention of a Rule 52. So that cannot be the listing from

23 the Ivan book.

24 Q. Well --

25 A. So that's why I'm saying I'm not exactly sure what this numbering

1 refers to.

2 Q. Well, it's possible that that's a typographical error. Don't

3 worry about rule 52. Let's look at line 7, Rule 49 and I'll just go back

4 to this question. "If I may proceed to Rule 49, should a breach of rules

5 as we have been discussing so far were not justified in a written report

6 lead to a rejection of identification in terms of quality of your field

7 of expertise?"

8 And your answer is: "There I might be invading, say, what is

9 called the domain of the Court, I think, because it is not for me to

10 reject any tests. So I know that rule is maybe not appropriate for an

11 expert to express. But of course it contains a clear warning that there

12 are cases in which rules were being breached to such an extent that

13 really the logical impact of the recognition is totally absent. In that

14 case, at least, I think it should be made known to the Court that there

15 is a real problem. I do not want to express here that I would prescribe

16 -- want to prescribe to the Court what to decide in such a case."

17 I take it that you still agree with what you have expressed there

18 in that case.

19 A. Of course I'm not prescribing the Court anything at all.

20 Q. Now, you were asked on direct examination about what happens in

21 England when -- if the Court finds a violation of the rules. Are you

22 aware that -- just to pursue that point a little bit, are you aware that

23 in England if there is a violation of the PACE rules, it is not -- the

24 identification is not automatically thrown out, the Court evaluates all

25 of the circumstances to decide whether the identification can still be

1 received, whether it goes to weight, and so forth.

2 A. Yes. Exactly the same things happens in the Netherlands where

3 the Minister of Justice prescribes some rules. It's still for the Court

4 whether to accept identifications.

5 Q. The same is true in the United States, is it not? And this is

6 expressed in a case which you discuss in your book, the Ivan book, Neils

7 versus Biggers and it's later amplified in a decision called Manson. And

8 in the United States if there's a breach of the rules, if there's a

9 finding of suggestiveness in the procedure or in the photo spread, then

10 what the Court does in deciding what to do with the identification is it

11 essentially looks at the estimator variables. Is that correct? So it

12 looks at things like the outward --

13 A. I'm not an expert on American law.

14 Q. Well, this is a decision that you cited in your book, so I

15 thought you were familiar with it.

16 A. But you describe -- you seem to describe a present practice and

17 I'm not an expert on that.

18 Q. I understand.

19 A. But I -- I think if you describe it that way, I can easily accept

20 that interpretation of American law from you, but I'm not an expert.

21 Q. I understand --

22 MR. TOPOLSKI: I'm sorry to interrupt. If Mr. Whiting is, and I

23 think he is, is going to pursue this a little further, I say nothing

24 about the American circumstance. I'm slightly concerned about the way

25 he's left the English position with the witness at page 102, line 10/11.

1 He's talking about the court's approach, looking at the circumstances to

2 decide whether the identification can still be received, whether it goes

3 to weight and so forth.

4 That, in my understanding, with respect is not the position in

5 England. The judge deciding on admissibility will decide on

6 admissibility, taking into account any breaches he or she finds have

7 taken place with regard to the procedures. If the judge finds the

8 evidence admissibility, notwithstanding breaches, it is admissible. It

9 would be a matter for the jury as to what weight it attached. So the

10 judicial exercise that a judge is engaged upon in England has nothing

11 whatever to do with weight. For once it is admissible, it is a matter

12 for the fact-finder to decide what weight to attach to it. I say, I say

13 nothing about the American experience. It's not within my sphere.

14 MR. GUY-SMITH: But I am sure, as Mr. Whiting knows, this is an

15 issue in American law of admissibility, not an issue of weight.

16 MR. WHITING: Well, with respect to the English law, I was

17 relying on a case Regina versus Forbes and it was my understanding,

18 although I this is an area I am certainly not an expert, my understanding

19 is if there was a violation of the rules and the court admitted the

20 evidence, that the jury in some circumstances could be informed of the

21 violation of the rules. That was what -- it is that I meant could go to

22 weight.

23 MR. TOPOLSKI: Oh yes.

24 MR. WHITING: It was that that I meant could go to weight.

25 MR. TOPOLSKI: Absolutely. In assessing -- yes, it's -- it is

1 Queen against Forbes in the House of Lords. In assessing what weight the

2 fact-finder should give to the evidence, then the judge should direct the

3 jury that he or she has found that there may have been breaches in the

4 procedures and the jury must take those breaches into account when

5 assessing weight. I was merely relating to the judge's exercise, not the

6 once it gets beyond the judge and before the jury. Mr. Whiting is

7 absolutely right, if that's what he meant.

8 MR. WHITING: That's what I meant when I put the question.

9 Q. In any event, with respect to the Neils test, the court -- some

10 of the things the court will look at, if there is a finding that

11 suggestiveness has in fact occurred, not just the possibility of

12 suggestiveness, but suggestiveness or bias has in fact occurred, the

13 court will look at the opportunity of the witness to view the offender,

14 the witness's degree of attention, the accuracy of the prior description,

15 the degree of certainty, and the length of time between the crime and the

16 identification procedure. Those are the factors that are set out. Are

17 you aware of that?

18 A. Well, I must say, this discussion is a little bit beyond me. But

19 the point I tried to make in section 21 is in fact two points. One is

20 that a violation of the rules usually makes a test weaker, not stronger

21 at all. In the end, you can have weakened the test so much that it

22 doesn't prove anything anymore; that's all.

23 Q. Well, I didn't want to get caught up in the niceties of the Neils

24 decision. The point I was trying to put to you and see if you would

25 agree with me is that if there is a problem -- put it broadly like

1 that -- a problem with the system variables, it would be reasonable,

2 would it not, for a court to consider the estimator variables in

3 determining the weight and significance of that problem. Would you agree

4 with that?

5 A. No, no. Not generally. That might be informative also, but

6 there are situations in which there is nothing wrong with the estimator

7 variables. The witness had a perfect view long enough, there's no reason

8 to doubt whether he remembers the culprit, but still with the system

9 variables so much went wrong that he was led to the wrong response.

10 That's still possible.

11 Q. That's possible. That's possible. What I'm asking is: If a

12 court generally could get some confidence in evaluating problems with the

13 system variables by taking into consideration, at least, the estimator

14 variables.

15 A. Yeah --

16 Q. Would you agree with that?

17 A. No. Maybe this is what you want to hear, but my opinion is that

18 a court should always look at the estimator variables, and some estimator

19 variables are necessary conditions for a good recognition but never

20 sufficient conditions to explain the results of a test. Because even if

21 there is no doubt in the mind of the witness what the culprit looked

22 like, then still it's possible to direct this witness to a picture of a

23 person -- of an innocent person. So estimator variables are necessary

24 conditions; they're not sufficient conditions. You can never say: Since

25 the estimator variables were good, we don't need the system variables

1 anymore.

2 Q. No, I'm not suggesting that.

3 A. Okay.

4 Q. What I'm suggesting, Professor, is that it's proper to consider

5 in evaluating problems with the system variables the estimator variables.

6 That could make a big deference in how you might look at a particular

7 violation in a -- in system variables. You don't agree with that?

8 A. No, I think it's the wrong way of putting it. It's always

9 important for a court to look at the estimator variables, but it cannot

10 -- in my mind, it cannot logically - I'm not talking about legally - but

11 you cannot logically accept the results of a line-up test if the

12 estimator variables are okay but the system variables are massively

13 flawed. So yeah, my way of expressing it is: Of course a Court should

14 always look at the system variables; that's extremely important. But you

15 cannot let the one be a compensation for the other.

16 Q. Going back to Gary Wells, Gary Wells has written about -- and I

17 provided you with the article. It's -- the article is: What's wrong

18 with the Manson test.

19 A. Yeah.

20 Q. And he's written about some problems, and you don't need to look

21 through it --

22 A. That's a web site sort of article, isn't it?

23 Q. That's an article that's on his web site. It's the same article

24 that's cited in your report.

25 A. I'm just trying to help me remember.

1 Q. It is the web site article and it's a draft of an article that

2 he's writing.

3 A. Yes.

4 Q. And if Mr. Younis could provide that. It's this article here.

5 A. He again provides some rules, right?

6 Q. Well, he is talking about this test that we have just been

7 talking about in the United States about if there's -- that's in the

8 Neils test and it's also expressed in the Manson test that if there is a

9 finding of suggestiveness then you look at the conditions of the

10 identification to evaluate whether the identification is admissible. And

11 -- he has some criticisms of the test and then he considers an

12 alternative rule where if there is a finding of suggestiveness in the

13 procedure, that the identification should, per se, be excluded. I just

14 want to read what he says about that.

15 A. It's about the referral.

16 Q. I see this has no page numbers. It is -- it's towards the end.

17 It's on the page that has the words "final notes" at the very bottom of

18 the page.

19 A. Yes, that's about the referral of the proof. Right?

20 Q. Well, it's -- yes, right. And on that page where it says "final

21 notes" I'm just going to read the paragraph, the first full paragraph

22 that starts there. He says: "However" -- and I just want to get your

23 view on it after I've read it.

24 "However appealing the per se exclusions might be at some levels

25 it neglects the fact that there was a logic behind the court's holding in

1 Manson that should be reserved. Consider, for instance, a case where we

2 can be very sure that the eyewitness has a reliable memory. Perhaps it

3 is an abduction case in which the victim eyewitness was held for days by

4 the culprit. The culprit never wore a mask. There was good lighting.

5 The culprit's face was in full view for hours at a time. And the witness

6 was able to describe in detail the size, shape, and location of unique

7 scars on the culprit's face.

8 "Suppose now that a line-up was held only hours after the victim

9 was rescued, but the line-up used unnecessarily suggestive procedures

10 because the fillers did not fit the description of the culprit. Given

11 the profoundly favourable circumstances of witnessing, would it be

12 reasonable to suppress the identification merely because there were not

13 good fillers for the line-up? There is an extreme example, but it

14 illustrates a basic point. The court was correct in Manson to argue the

15 issue of whether identification is nevertheless reliable in spite of the

16 suggestiveness, per se exclusion based only on the finding of

17 suggestiveness would not be in the interests of justice."

18 Do you agree or disagree with that view expressed by Gary Wells?

19 A. I disagree. I disagree with the initial stage where we say:

20 "Let us assume," consider a hypothetical case, where he seems to argue

21 that a case under such condition - saw the culprit for a long time,

22 didn't wear a mask, and so on, saw the scars and everything - the witness

23 simply cannot be wrong. That's the sort of assumption we are asked to

24 make. The witness cannot be wrong. Now does it matter how we test such

25 a witness? It seems to suggest that in cases of abduction, in longer

1 abduction and in close proximity to the perpetrators, there never have

2 been mistaken identifications.

3 I can tell you, that is not so. I gave you the example of Valuse

4 [phoen] from my book where the witnesses saw Mr. Valuse for more than a

5 year and still they were demonstrably wrong. So the assumption here is

6 not reasonable. So the assumption here that in such a condition the

7 witness cannot be wrong is misleading; a witness can be wrong. And if

8 there is no evidence, if you really needed identification evidence right?

9 Because there is no evidence to link the witness to this abduction and

10 everything really depends on the identification test, there's every

11 reason to conform to the rules because you simply cannot argue that a

12 mistake under such conditions is not humanly possible; it is. We know.

13 And if you look at the list of -- say, the list of the 40 cases described

14 by Gary Wells and others, some of them are about abductions and lots of

15 things, as well it looks not unlike this and still there was a mistake.

16 So --

17 Q. Well --

18 A. If the point of the test is that this evidence is essential for

19 linking the suspects to a crime, then there is every reason to follow the

20 rules.

21 Q. Just on that last point, you're not suggesting that any of the

22 examples in those 40 cases, that you have any information that they

23 were anything like this example.

24 A. No, they might well be of that type. I have met myself many of

25 those cases and I see no particular reason why none of them should be

1 there. And in fact, in fact the acceptance of unsafe identification

2 procedures is enhanced by some sort of belief that it doesn't matter

3 because the suspect [sic] can't be wrong anyway. Suspects [sic] can be

4 wrong in an incredible manner when it comes to identifying people.

5 Q. With respect to those 40 cases, since you brought them up, you

6 believe that in 30 cases there was just a single witness. A single

7 witness in 30 of those 40 cases.

8 A. Well, the victims. They're called victims in that paper.

9 Q. Right.

10 A. This is exactly the position he describes. It is the victim that

11 is so certain that she can't be wrong.

12 Q. Well --

13 A. Well, for that we have a test. Why shouldn't we run the test

14 properly?

15 Q. I don't have any further questions. Thank you.

16 JUDGE PARKER: Thank you, Mr. Whiting.

17 MR. TOPOLSKI: I have no re-examination.

18 JUDGE PARKER: Thank you.

19 MR. TOPOLSKI: As I understand it, no one else does either.

20 JUDGE PARKER: They don't have the chance, perhaps.

21 Professor, you will perhaps have discerned from that that counsel

22 has lived up to their promise of trying to conclude your evidence today.

23 We have reached the end of your questioning. The Chamber is grateful for

24 your assistance and thanks you for being here today. You may leave and

25 return to your other interests. Thank you.

1 [The witness withdrew]

2 JUDGE PARKER: That brings us to the questions whether there are

3 any other matters that --

4 MR. GUY-SMITH: Just ever so briefly, I was reviewing the

5 transcript and I believe it's page 109, line 25, I believe the Professor

6 said "suspects" -- he said "suspects" can be wrong in an incredible

7 manner.

8 JUDGE PARKER: Witnesses was my understanding.

9 MR. GUY-SMITH: Mine, too. Mine, too.

10 JUDGE PARKER: That probably means we're both wrong.

11 MR. GUY-SMITH: Probably so, but I wouldn't want to identify

12 that.

13 JUDGE PARKER: The question is whether the Chamber will need to

14 sit later this week. That, I take it, closes the Defence case then?

15 MR. TOPOLSKI: Your Honour, it does.

16 JUDGE PARKER: In all three facets.

17 MR. TOPOLSKI: Well, apart from --

18 JUDGE PARKER: Some 92 bis potentially.

19 MR. GUY-SMITH: Yes.

20 JUDGE PARKER: Mr. Whiting.

21 MR. WHITING: Your Honour, we have three statements to put in

22 rebuttal evidence. Two of the statements -- and they will be all 92 bis

23 statements and they will be ready in 92 bis form tomorrow. The first two

24 statements are agreed with the Defence; there's no dispute. The third,

25 there is a dispute and the dispute is really with counsel for Mr. Limaj,

1 and we've agreed, subject to the agreement of the Court, that we'll do

2 that in writing. I can file a very brief motion explaining why we submit

3 that it's proper rebuttal and Defence counsel can respond and then that

4 can be considered. But that can all be filed this week.

5 I would, finally, suggest that perhaps we set some kind of a

6 deadline for all the 92 bis statements to come into evidence just so that

7 it doesn't drift. I don't know what would be appropriate. There's no

8 surprises, I think, but just so that it gets done.

9 JUDGE PARKER: Does that mean that you agree, Mr. Whiting, that

10 there is nothing more for the Chamber to need to hear in the form of oral

11 evidence, rebuttal, or matters of that nature, and that whatever else

12 needs to be done can be done in written form?

13 MR. WHITING: Yes.

14 JUDGE PARKER: Very well.

15 And you would anticipate your written material being before the

16 Chamber perhaps even tomorrow. Is that right?

17 MR. WHITING: Yes, tomorrow. The -- I can submit the two agreed

18 rebuttal statements tomorrow and the -- and then I -- I'm quite confident

19 I can submit the motion with the third contested rebuttal statement

20 tomorrow.

21 JUDGE PARKER: Thank you.

22 Mr. Khan, if that is done, will Friday be time enough?

23 MR. KHAN: It will be. It is a short matter and focussing on

24 what is rebuttal evidence and whether or not it ought to be heard at the

25 moment.

1 JUDGE PARKER: The Friday following, would that be time enough

2 for your 92 bis formalities to be concluded?

3 MR. KHAN: Well, Your Honour, all the formalities, as I

4 understand it, and perhaps the registrar can help, that need to be

5 conducted by the Defence have been. You've been seized of some time, of

6 course, in relation to the written motion. A formal request has gone to

7 the registry. It's a matter for the registrar, as far as I understand

8 it, to complete the administrative requirements.

9 JUDGE PARKER: Very well.

10 MR. KHAN: As soon as it can. And, Your Honour, whenever the

11 registrar is able to organise the logistics, the procedure can be carried

12 out. Your Honour, of course, it's a matter for you. At first blush it

13 doesn't seem necessary to impose time limits. I'm sure one can trust the

14 registrar to discharge his responsibilities promptly in this matter.

15 JUDGE PARKER: Yes.

16 Mr. Guy-Smith.

17 MR. GUY-SMITH: Yes, I believe that Mr. Whiting may actually have

18 a fourth rebuttal statement that he is not thinking through. It's been a

19 matter of some argument back and forth, but in the case that does become

20 an issue, we will have the matter taken care of within the week. He for

21 sure has three; he may actually have a fourth which is in some

22 discussion.

23 JUDGE PARKER: You're --

24 MR. GUY-SMITH: I'm trying to give him some hope here.

25 JUDGE PARKER: The question is: You have completed your

1 procedures?

2 MR. GUY-SMITH: My procedures are complete. We don't have a

3 rejoinder case -- I notice you mentioned rebuttal but I wanted to make

4 sure you were not contemplating a rejoinder.

5 JUDGE PARKER: Mr. Topolski, there any matter?

6 MR. TOPOLSKI: Yes. I'm conscious of the time. I don't want to

7 make life difficult for the defendants back at the UNDU. I'll be very

8 brief.

9 We have a number of 92 bis statements that are, as it were, down

10 in Kosovo being approved by the witnesses. I had hoped to make an

11 application on that well known piece of legislation, the Ways and Means

12 Act to see if I could persuade this Chamber to be not so allied to 92

13 bis, but in light of what you said to Mr. Khan earlier on, I dare not go

14 near that water because I felt its temperature already.

15 Your Honour, in that sense we are in the hands of others. Your

16 Honour appreciates it's now the registry and either somebody going down

17 to Kosovo from here and complying with and completing the procedures. I

18 would be very grateful if you please didn't tie me to a date that I would

19 be honour bound to give you now and have this dealt with by then for fear

20 that my position to comply is not entirely within my control.

21 JUDGE PARKER: Does this mean that the procedures on your part

22 have been completed?

23 MR. TOPOLSKI: We are awaiting confirmation of the contents of

24 some of the statements in Kosovo. So the straight answer to your

25 question is as I speak now no. But I will ensure before I leave The

1 Hague tomorrow, Wednesday, that it is, because I can direct those down

2 there to do it --

3 JUDGE PARKER: Thereafter your position will be the same as Mr.

4 Khan's?

5 MR. TOPOLSKI: It will, except I'm slight reluctant to, as it

6 were, step aside and let a date be given where I'm not in control of the

7 process. I hope Your Honour follows what I mean --

8 JUDGE PARKER: We were fixing a date for Mr. Khan so don't worry.

9 MR. TOPOLSKI: It would be on a best-endeavours basis.

10 JUDGE PARKER: What I want to be sure is that you reach a

11 position in the next day or two in which it's merely for the registrar

12 after that.

13 MR. TOPOLSKI: Your Honour, that assurance I can give, that I

14 will do all I can to ensure that is the case.

15 MR. KHAN: I do apologise. Your Honour, just one very brief

16 matter before you and Your Honours rise. Your Honours may recollect that

17 a while back Mr. Mansfield alluded to the fact that Mr. Churcher would be

18 preparing a short addendum report in response to Mr. Clark's 92 bis

19 witness statement. Your Honour, that addendum was filed this morning

20 with the registry. I wonder for just the administrative purposes if that

21 can be appended to DL13, which is the exhibit of Mr. Churcher's report.

22 JUDGE PARKER: Yes, that will be done. So it will become a

23 further part of DL13.

24 We should now break for the concerns that have been mentioned.

25 The timetable ahead has been set and I believe is clear so we have no

1 need to repeat that. Should there be any unforeseen difficulty with

2 statements or submissions, an approach to the Chamber staff can

3 facilitate a hearing at short notice.

4 We will adjourn now until we complete the processes of written

5 submissions and then the hearing of the final oral submissions.

6 MR. TOPOLSKI: Your Honour, I think it follows, therefore, that

7 we do not gather together again until the 20 --

8 JUDGE PARKER: End of August. Yes.

9 MR. TOPOLSKI: We wish the Tribunal a pleasant summer break if

10 it's to happen.

11 JUDGE PARKER: We are way off yet, but we are grateful. Yes. We

12 hope you recover from your recent splendid news.

13 --- Whereupon the hearing adjourned at 7.08 p.m.

14

15

16

17

18

19

20

21

22

23

24

25