1 Monday, 3 September 2007

2 [Open session]

3 [The accused entered court]

4 --- Upon commencing at 2.20 p.m.

5 JUDGE ANTONETTI: [Interpretation] Registrar, could you please call

6 the case.

7 THE REGISTRAR: Good afternoon, Your Honours. This is case number

8 IT-04-74-T, the Prosecutor versus Prlic et al.

9 JUDGE ANTONETTI: [Interpretation] Thank you. This is Monday,

10 September 3rd, and I welcome Mr. Stringer for Prosecution, counsels for

11 Defence, and the accused in this room. We're now to continue the hearing

12 on the cross-examination of this witness, but I think Mr. Kovacic has

13 something to say first.

14 MR. KOVACIC: [Interpretation] Thank you, Your Honour. Yes. Good

15 afternoon to the Trial Chamber and everybody in the courtroom. I

16 apologise for taking two minutes of the Court's time but I just have to do

17 so. On Thursday the Defence counsels discussed the situation. We

18 analysed it and we came to the conclusion that, with the best will in the

19 world, although that had been our intention that we simply don't have is

20 enough time to prepare a response to the Prosecution's submission of the

21 27th of August requesting the introduction of 380 documents with respect

22 to the military structure of the HVO. As the Trial Chamber already knows,

23 we are working on seven similar requests by the Prosecution. There is

24 1.600 documents in all, and 45 statements, witness statements. The Trial

25 Chamber also knows that we have limited resources, and so we cannot cater

1 to such large amounts of material.

2 We were given until the 8th of September to respond to the

3 Prosecution motion, but since we have to respond to other motions as well

4 between the 10th and 24th of September, with this additional amount on the

5 table we came to the conclusion, Your Honour, quite simply that we could

6 not do that. So we would like to ask you additional time, ideally until

7 the 8th of October, but the normal deadline according to the rules would

8 be the 10th of September.

9 Now, in view of the other six motions we're working on, we fear

10 that we simply don't have enough time. Most of the Defence teams worked

11 during the weekend, and we just can't keep abreast with all this

12 administrative material, so we ask for an extension in time.

13 Perhaps I wasn't clear enough have, Your Honours. It is the

14 request tabled by all the Defence teams, and the Trial Chamber has given

15 us the instructions that we should all pool our resources when that is

16 necessary and possible.

17 JUDGE ANTONETTI: [Interpretation] I will give the floor to the

18 Prosecution for its observations, but if I understood you well, this

19 request deals with the Prosecution's submission on the submission of 380

20 documents dealing with the military structure of HVO. The deadline was

21 September 10th. So you said there's a lot of work ahead of you. There's

22 a lot of submissions. Many motions that you have to respond to and it

23 seems you will not be able to make it by September 10th and you're asking

24 for an extension until October 8. And this is a joint request for all

25 Defence teams.

1 So, Mr. Stringer, what do you have to say to this?

2 MR. STRINGER: We have no objection, Mr. President.

3 [Trial Chamber confers]

4 JUDGE ANTONETTI: [Interpretation] The Trial Chamber has

5 deliberated immediately and allows the motion presented by Mr. Kovacic

6 given that Prosecution made no objection. Defence will have until October

7 8 to respond to the -- to the Prosecution's motion on the 398 documents

8 dealing with the military structure of the HVO. So you have until October

9 8.

10 Now, regarding the next witness, let's take stock of the time

11 left. Mr. Praljak used up 78 minutes, and you have 42 minutes left.

12 After this Mr. Prlic will have 30 minutes and Mr. Coric's Defence will

13 have 30 minutes. According to the time.

14 MS. TOMASEGOVIC TOMIC: [Interpretation] Your Honour, we have ceded

15 our time to the Prlic Defence. We're not going to need to cross-examine

16 this witness. Thank you.

17 JUDGE ANTONETTI: [Interpretation] Therefore, Mr. Coric's Defence

18 time has been allotted to Mr. Prlic's Defence.

19 Let's bring the witness into the courtroom, please.

20 [The witness entered court]

21 WITNESS: EWA TABEAU [Resumed]

22 JUDGE ANTONETTI: [Interpretation] Good afternoon, Mrs. Tabeau.

23 Your hearing will continue today and we will have the cross -- the rest of

24 the cross-examinations and the redirect, and I will thus give the floor to

25 General Praljak.

1 THE ACCUSED PRALJAK: [Interpretation] Good afternoon,

2 Your Honours.

3 Cross-examination by the Accused Praljak: [Continued]

4 Q. [Interpretation] Good afternoon, Doctor.

5 A. Good afternoon.

6 Q. Because of the shortness of time once again could you please give

7 me brief answers wherever possible. Of course, we don't have enough time

8 to go through your very lengthy report, but anyway, would you turn to the

9 portion where you speak about deaths or violent deaths in East Mostar.

10 Page 10 of 32 pages. That is in the Croatian, of course, the Croatian

11 translation. It's table 3. We've already mentioned that. It's page 7 of

12 the English where you state that 373 cases fall within the relevancy test.

13 And then I'd like to refer to the following Croatian page and table 4.

14 Table 4 in the Croatian text. And in table 4 you give the month of death,

15 the number of deaths in East Mostar, unknown causes in view of date and

16 place. And April 1993, let's take a look at that and look at the 7th --

17 or, rather, July, August, and September where the number of both known and

18 unknown deaths increases.

19 Now, this disruption, disturbance, in the number of deaths

20 according to the months and the observations you make here, do you know

21 why under a siege, as you put it, there would be such a disruption in the

22 distribution of deaths if we look at the months of the siege? Did you

23 ever wonder why, why there was a disturbance or why there were

24 differences, whereas the siege was an ongoing state?

25 A. Well, I didn't study the causes of the killing, so I can't say

1 that I know why. Well, cause is general, I mean external factors, the

2 siege activities. I'm not speaking of the causes of death as such.

3 Q. Thank you. So you've never heard of something called Neretva 93

4 offensive?

5 A. Well, I don't know what -- what is this.

6 Q. Thank you. Now, please turn to page 19 of the 32 pages. It's

7 table number 10 and 11 -- or, rather, table 11. Let's look at table 11.

8 Table 11. You say status of siege-related victims reported in the Mostar

9 war hospital books. It's page 13 of the English.

10 A. Yes, I see the table.

11 Q. And you state here that the victims, according to the Mostar war

12 books as follows: There were 160 civilians, 12 detainees, and military

13 158, and unknown 142. Is that correct?

14 A. Yes, it is.

15 Q. Right. Now, am I right in noticing once again in this instance

16 that you didn't take the unknown into account when you calculated the

17 percentages because the civilians, detainees and military you took as a

18 sum, whereas you have left out the unknown. So it's 47. -- 47.9 per cent

19 the military and 48.5 per cent civilian because you've left out the

20 unknown; is that correct?

21 A. Yes, the unknowns are left out.

22 Q. Thank you. Now I'd like to look at table 15.

23 JUDGE ANTONETTI: [Interpretation] Before table 15 let's move to --

24 let's stay on table 11. Here you have 12 detainees that are deceased, in

25 table 11. So as far as the research that you've done, could you tell us

1 whether you studied the situation of these 12 people? Were these people

2 who were wounded when they were admitted into the hospital and then died

3 because of their wounds, or are there other causes to their death? This

4 figure of 12 appears here statistically so could you please answer my

5 question.

6 THE WITNESS: I didn't study these particular 12 cases. This can

7 be done. We can check the record of these 12 persons, but I can't say

8 what kind of cases are these 12 cases right now.

9 JUDGE ANTONETTI: [Interpretation] Thank you very much.

10 Mr. Praljak.

11 THE ACCUSED PRALJAK: [Interpretation]

12 Q. Page 26 of pages 32 -- or rather, table 15 is what I'd like to

13 look at. It's page 17 of the English. The source. You have a sentence

14 here that reads as follows: "Records of deaths which coincide with the

15 records of fallen fighters means that they can be taken into account as

16 soldiers, where as the rest can be taken to be civilians."

17 So these records from the war hospital, did you compare that data

18 from the data that you received from the Ministry of Defence about the

19 killed in the BH army and then change the number of soldiers killed, the

20 number of deaths? Is that correct?

21 A. Yes, I did compare these records with the matched records, and

22 there is a table in a -- in the report on the wounded persons which

23 presents the results of this comparison. The table is unavailable from

24 the report on the killed persons. There is a table in the report on the

25 wounded persons to which we can refer.

1 Q. We'll deal with that in due course, but look at the remaining part

2 of that sentence. And then you said the rest -- "And all remaining

3 records as civilians."

4 Is it true that you did not compare any other list of those -- of

5 the deaths from the list of the armed forces, which means the civilian

6 police as well, all the armed forces and civilian protection and everybody

7 else helping the army? Quite simply, those who did not die as soldiers or

8 militaries you termed civilians. So you say all the remaining are

9 civilians without going in to see whether according to some other sources

10 some more soldiers or militaries would be found; is that correct?

11 A. Well, first of all, I kept the original reporting of civilian

12 military status available from the war hospital records. This is the most

13 important thing.

14 In the second step we matched both, war hospital records and also

15 death registries with the military lists. From these lists we could see

16 that a number of additional cases could be marked as militaries. We also

17 saw that for a number of civilians reported in war hospital, the -- they

18 were matched with the list of fallen soldiers. So we took the most

19 conservative approach, and we remarked the civilians that were found in

20 the military lists and took them as soldiers in the analysis. But -- this

21 is what we have done, but your question is about whether we used any

22 additional sources next to the sources I mentioned in my report to compare

23 the -- the marking of the military/civilian status, and my answer is no

24 other additional sources were used except for the military lists.

25 Q. Very well. Thank you. That seems satisfactory, because there are

1 other sources of -- of course, too, such as profits for people who took

2 part in the army remuneration, whatever. Certain advantages, at any rate.

3 Now let's move on to your second topic, the number of wounded.

4 JUDGE ANTONETTI: [Interpretation] Before moving on to another

5 topic, following your question, madam witness, this is absolutely

6 essential for the Bench, this distinction between, on the one hand, the

7 civilians, and the military on the other hand. If civilians were killed

8 this will have consequences.

9 We looked at the hospital record, the admission record, and you

10 said that you worked with this record, and we have it as an annex. It's

11 2007 and 86, and obviously when somebody is wounded or killed there is

12 something written in a column if this person has military status.

13 However, it may occur that a civilian would be admitted into the hospital

14 in civilian clothes even though it's a military person, because this

15 person was on leave, for example, and changed clothes. But maybe this

16 person was -- eventually got killed during combat because he's a military,

17 but in the record here he will be listed as a civilian.

18 Furthermore, when hospital notes that this person is a military

19 status, it may be somebody who just as military clothes, in a military

20 jacket, and that is why he is noted as being a military. He might just be

21 a civilian who just wore a military jacket. We've seen evidence that has

22 shown that sometimes civilians wore military clothes, and the opposite was

23 also true.

24 So when I was listening to the question and to your answer, a

25 question came to my mind. I wondered whether you looked into the dates of

1 birth of these people and whether you looked into the status of the

2 able-bodied men, adult men, who according to law were considered as

3 integrated into the armed forces when they were between 18 and 60.

4 So if we take an assumption of a 45-year-old man wounded, who is

5 suddenly wounded and is admitted into the war hospital in civilian

6 clothes, what do you make of this man? Is he a military for you or is he

7 a civilian?

8 This is a very lengthy question. I apologise for this, but when

9 you matched the data, did you look at the age of these people to be almost

10 sure that in the tables when you note either civilian or military you also

11 took that parameter into account, age, knowing that you haven't matched

12 this with the listings of the ABiH, which according to the evidence

13 presented by General Praljak seemed to prove that there were a great

14 number of military in the 4th Corps who were killed. So did you first,

15 did you -- did you take into account the age of the victims to make your

16 distribution between civilians on the one hand and military on the other?

17 THE WITNESS: Well, the short answer is no, we didn't look at the

18 age as the criterion for the decision whether or not a person could be

19 considered a military or not. Of course I have seen approaches based on

20 age only. Basically speaking in this age-based approaches, men at

21 military ages are all excluded from the list of civilians. There are,

22 simply speaking, no men, disregarding the actual situation in which the

23 person would die. Only women, all ages, and children, and the elderly or

24 taken and placed on a list of civilians.

25 We didn't follow this approach. Our approach was data driven. We

1 had in one of the sources, war hospital, the military/civilian status

2 available for a number of patients. This was the basis, the starting

3 point for us in our analysis. We considered these cases for whom we had

4 the status available as a subsample coming from a bigger sample of all war

5 hospital records. And based on this subsample, we estimated the

6 frequencies of civilians and militaries, reported civilians, and reported

7 militaries.

8 In the second step, we applied the subsample-based proportions of

9 civilians and militaries to a larger list, the so-called master list of

10 all victims and ended with an estimated number or numbers of civilians and

11 militaries. That was method one. Method number two was also data driven

12 but was based in addition to what was available in the war hospital, the

13 reported status available from the war hospital books. We also included

14 the results of matching of other sources with the military lists. From

15 the matching we could have more information about the militaries, the

16 persons reported as being killed as members of the army or certain other

17 groups associated with the army.

18 Well, this of course doesn't answer the question in what

19 circumstances these persons were killed, whether a soldier in civilian

20 clothes killed while being on a leave or perhaps a civilian having gone

21 and fighting. This is not possible. This cannot be seen in the sources.

22 The individual circumstances of every death cannot be seen in the sources.

23 That is exactly the same with the age-based approach. Whatever we do with

24 age, it is always some related resulting in some bias because not all men

25 at military ages are combatants killed in combat and not all civilians --

1 not all civilians can be seen as civilians killed in non-combat

2 situations.

3 So all these approaches are, in some sense, deficient.

4 JUDGE ANTONETTI: [Interpretation] Well, thank you for your answer.

5 THE ACCUSED PRALJAK: [Interpretation]

6 Q. Now, the second part of your analysis or your expert report,

7 that's what I'd like to look at. Page 6 of the 60 pages of the entire

8 report. That's the B/C/S. In English, it's probably a few pages before

9 that. And it says here that some records were introduced or entered

10 perhaps after -- after the person arrived, because the medical staff were

11 too busy.

12 Now, this term "too busy" and similar qualifications, I'll deal

13 with that for a few moments. Let's see what was recorded at the war

14 hospital. The identification number; the date the person was admitted;

15 the surname; the year of birth; the status, civilian or military; the

16 military unit at the time of the incident; the place of the incident; and

17 how it occurred. So all these things should have been recorded.

18 Was it -- were they indeed recorded in the war hospital records?

19 Is that right? Were those the basic facts that had to be entered and

20 recorded?

21 A. When I was looking for this table. You are saying it's page 6 in

22 the B/C/S report.

23 Q. Yes. Four and 5 of the English. It's not a table. It just lists

24 what was in the records, what should have been recorded. Is that right?

25 A. We are talking about the death registries; right?

1 Q. ID given, first name, father's name, et cetera.

2 A. Yes, yes.

3 Q. Of the wounded. The wounded. We've moved on to discuss the

4 wounded.

5 Can you tell me whether it's correct that these are the records

6 entered into the hospital books of the war hospital?

7 A. Well, indeed. On that page 4 and 5, English version of the

8 report, there is a list of items that were registered for the wounded

9 persons in the war hospital books.

10 Q. Very well. And in footnote 5 you then say that someone, native

11 speakers of B/C/S, entered the sex, age, and nationality. You improved

12 the list by getting some people to enter sex, age, and nationality. As

13 forthright to change one's nationality, I won't go into that any further.

14 We've already mentioned the case of the Yugoslavs and so on who had the

15 right to declare themselves differently, but how do -- how did these

16 people discover the age of the wounded? They had a book with 5 or 6.000

17 entries, and then they determined the age based on what? You can tell the

18 gender by the name, but how do they know the age? If the doctors had on

19 the average 14 patients and were so busy that they were unable to enter

20 that information, they were unable to enter the details such as age, how

21 did that be done subsequently? I'm really interested to know that.

22 A. Well, the year of birth was reported for the patients in the

23 books, and the age is calculated as a difference between the year of

24 wounding and the year of birth. We do this in demography very often. We

25 do it all the time. So not only the data entry clerks, the B/C/S native

1 speakers did their own estimate of age. We did the same later by

2 calculating the difference.

3 And regarding ethnicity, I want to stress this is a study of

4 sources reporting on -- reporting on Bosniak victims. So it wasn't a

5 purpose of this study to include all victims of the siege. On the

6 contrary. The purpose was -- I was tasked to prepare a report that would

7 be summarising casualties among the Bosniak victims. This is why the war

8 hospital books were used as a source representing these victims.

9 JUDGE ANTONETTI: [Interpretation] Madam -- I'm terribly sorry,

10 Mr. Praljak. But Mr. Praljak raised an issue that triggered my curiosity

11 on a particular point. This register, when we look at it very closely we

12 can notice that it was a register that existed for the JNA, that was in

13 the JNA army, because on column 6 you can see that we can see military

14 rank, and there must be as well a reference to column number 5, which is

15 military district of the person who was admitted. In fact, this register

16 was enforced in the JNA. So one may think that this military hospital

17 used this register in order to evaluate all this, but they did not take

18 into account these various columns.

19 Now, Mr. Praljak posed a very interesting question since he was

20 asking you how were they able to know what the date of birth of the person

21 was. If they put down the date of birth, that is either because the

22 injured or the wounded person had his military papers on him or maybe

23 because that person was identified subsequently.

24 So my question is as follows: Did you and your colleagues look

25 into the classification method that was used to admit people by using the

1 criteria in the former JNA, or did you use -- do you think that they used

2 it as a work tool without being particularly precise, without a great deal

3 of precision, let's say.

4 THE WITNESS: Well, regarding the JNA form which was used to

5 register the patients, it is just what they had in the hospital. There is

6 no link between this form and how and what was registered. In the siege

7 of Sarajevo, for instance, one of the important sources was printed on the

8 sandwich paper, because that was the only paper available, and we see the

9 same -- exactly same situation here.

10 This JNA, former registration of JNA members, these forms were

11 just used because they were available. There was no other paper

12 available. And the list of items I reported are the actual items reported

13 in the books as such. I don't think that the method of registration was a

14 very complicated one. These are all self-reported items as mentioned by

15 the patients at the admission to those who were taking these records.

16 So regarding the date of birth, it is just the reported date of

17 birth or year of birth in most cases that is available. There is no

18 documents underlying these answers. There is no military document that

19 would tell us that the person was a military or a soldier.

20 I hope I answered the question.

21 JUDGE ANTONETTI: [Interpretation] Yes, Mr. Praljak, please

22 proceed.

23 THE ACCUSED PRALJAK: [Interpretation]

24 Q. Madam, please turn to page 15, table number 5. Table number 5.

25 My question is very simple. Or, rather, page 10 in the English text,

1 table 5. And you say: "Cases treated in the war hospital according to

2 indication of wounding." You say 2.549 are marked as wounded, and 2.844

3 are not marked as wounded. And two or three lines below you say the

4 remaining 2.844 cases or 52.7 per cent of the cases were not marked as

5 wounded, which means that in the books of the war hospital the diagnosis

6 was not mentioned.

7 And then a line below that you say that strictly speaking,

8 2.500 -- 2.549 persons could be counted in a strict approach as wounded,

9 and you say this number is a minimum that has little to do with the

10 reality of the siege.

11 You said, just a little while ago, when I asked you why the

12 distribution of deaths in those three months was different, you say that

13 you did not deal with the war situation, and here, without any scholarly

14 analysis, you say quite calmly that this number is a minimum that has

15 little to do with the reality of the siege. This would imply that you

16 were there during the siege and that you knew what it was like. So what

17 was your scientific and scholarly basis for writing this sentence?

18 A. Well, the basis for this is the source from which these records

19 come. It is the war hospital books for approximately 2.500 of records, I

20 have the diagnosis available and I could study this diagnosis. Among the

21 cases available as you see, as you know, because I mentioned this, there

22 is mention of shelling victims, gunshot victims, other violent victims.

23 So that is my basis.

24 Q. Madam, please. Do forgive me, but please try to be brief, because

25 unless Their Honours grant me a little more time I won't be able to get

1 through my questions.

2 You say cases not marked as wounded, which means that in the books

3 of the war hospital the diagnosis was not mentioned. You're accusing the

4 doctors of not doing their job properly and then you take on their job and

5 mark the diagnosis. I'm simply asking you on the basis of what scientific

6 or scholarly foundation.

7 The doctors did not enter the diagnosis and you say, I don't care

8 because the siege and the war situation, which I'm not familiar with. I

9 will say that these two groups are similar, the ones marked as wounded and

10 those not marked as wounded.

11 Just in a nutshell, please tell me do you have any grounds for

12 this and, if so, what grounds, or do you not have any basis for this?

13 A. Well, I have very good grounds. I'm extrapolating the results

14 obtained for the patients with diagnosis available or with the patients

15 with diagnosis unavailable, and my ground is an extensive study we

16 completed of differences between these two groups: The group of patients

17 with diagnosis and the group of patients without the diagnosis.

18 As you see in the report, there is quite some room devoted to this

19 issue. We test compared these two groups, separately for civilians and

20 for soldiers using their age and sex distributions. We tested the

21 differences statistically and the differences were not significant

22 significantly, so there is no reason to believe that these two groups are

23 different. Well, we tested the age and sex distribution because these are

24 the basic demographic characteristics which, in this particular case, are

25 responsible for human behaviour.

1 If you have a woman and if you are age whatever but you are a

2 woman, it is unlikely, highly unlikely that you can be seen as a military

3 victim, as a combatant. If you are a man at military ages, then, well, it

4 is much more likely that this death or wounding can be considered of a

5 military person.

6 So the sex and age differences between the patients with diagnosis

7 and without the diagnosis are very important. And we studied them and

8 tested them and the differences are, statistically speaking,

9 insignificant. So there is no reason to believe that these two groups

10 are -- ended in the hospital for totally different reasons, especially the

11 timing of the arrival as patients was largely similar.

12 So these are my reasons for my extrapolation.

13 Q. So you say there is a set where the doctors entered a diagnosis,

14 another set where the doctors did not enter any diagnosis, and you say

15 that these two can be compared purely because there are some similarities

16 in sex and so on, age and so on, as if there is no difference between the

17 total number of the population and those who served in the army in this

18 respect, but I won't go any further into this. Somebody else will deal

19 with the clarity of this comparison of two totally different groups.

20 I would now like to move on and go to page 17, table 8 or, rather,

21 page 11 in your English original. That's table 8. And again we have

22 internal definition of status and we have marked as wounded and not marked

23 as wounded in the war hospital books. And again we have marked as

24 wounded, civilians, 831. And not marked as wounded, 384 civilians. We

25 have 29 detainees, 952 militaries, and 2.541 unknown. And then there is

1 something that is unknown. Neither civilian nor detainee nor military but

2 they are marked as wounded. That's a special subset. Some are marked as

3 wounded, and they are neither civilians nor militaries nor detainees.

4 What else is there? What other group is there, madam, in a town

5 if you are dividing them into civilians and militaries and possibly

6 detainees? Who are these unknown, and how did you arrive at this number

7 of unknown? It is quite unclear to me how you came up with this number.

8 A. Well, sir, you should be reading this table as a report on

9 reporting problems in the source. It is -- unknowns are just cases for

10 whom the status was unavailable; right? On one hand. But there was also

11 another group for whom there is now two distributions, this two

12 dimensional table, sir. You should see that in the last row, the unknown

13 is relating to the status. So these are cases for whom the status was

14 unknown. But for the cases for whom the status was unknown, there is a

15 group of a number of them with diagnosis available, and this is 737. And

16 the other group, 2.051 cases are the cases of unknown status for whom also

17 the diagnosis was unavailable.

18 This is a reporting issue. Obviously in this source there are

19 many reporting issues and there are many unknowns or, say, unspecified

20 categories. This is how you should be reading this table. And this

21 section is summarising the problems in the source. This is not a section

22 in which I discuss final results.

23 Q. But would it not then be logical to have knowns and unknowns and

24 then military, civilians, and so on, but this is mixing apples and oranges

25 or do you know how many civilians wore military uniforms in Mostar simply

1 because it was available and it was good quality and they had nothing else

2 to wear? We have had witnesses here testifying to the fact that many

3 civilians wore military clothes, and they would be then be the unknown who

4 were wounded. Did you investigate that?

5 A. Well, this is an analysis of -- of the sources that were available

6 to me which I am presenting here in this court. So I couldn't be making

7 statements about things that I didn't study.

8 Q. Thank you. Let's move on to page 19, table 10. In English it's

9 page 13, table 10. And again you have external definition of status,

10 marked as wounded. The number of civilians marked as wounded is now

11 defined as 819. Civilians marked as wounded account for 819. Detainees,

12 23, and militaries, 1.006. And you say in this table very clearly that

13 the number of civilians marked as wounded is 819. Is that what it says

14 here, madam? You say quite clearly marked as wounded civilians, according

15 to the war hospital records, 819. This is a clear, firm, precise number;

16 is that correct?

17 A. This table summarises the results of -- of using the military

18 lists in order to improve the reporting of the military status. So the

19 number 819 can be compared with the number of -- from table, say, 10, can

20 be compared with the number 831 from table 8.

21 Q. Correct.

22 A. So it is lower number because certain cases were reassigned the

23 status. And there was a small number of civilians that were also matched,

24 as I said, with the military lists, and to be conservative we changed the

25 status of these reported civilians into militaries because they were

1 matched with the military lists.

2 So this is how these two tables should be compared and should be

3 read.

4 Q. Yes, madam, I've read that, and we have 92 as the number of

5 soldiers. Is this table correct? Does it correctly show the real numbers

6 you have available, improved according to the records of the war hospital?

7 Can we consider this table to be true, correct, precise according to the

8 war hospital books, improved according to your later checking of the lists

9 from the Ministry of Defence which you received? Is that correct?

10 A. Well, of course it is correct. As a result of the analysis that

11 we completed for this particular part of the report, it's a correct table.

12 Q. Thank you. Thank you very much, madam.

13 Now, please open page 22 of 60. It's probably some 2 pages

14 further on in English. It says here in chapter 5: "I deal with the lack

15 of availability of diagnosis and military status." Have you found this?

16 A. Not yet.

17 Q. You say the following: "The missing values are most probably the

18 result of a failure to report woundings." So again you are dealing with

19 somebody that is probably in your science probabilities marked by a

20 number. So would you agree with me that probability theory with respect

21 to a certain set or event is 5 per cent, 7 per cent, and that words such

22 as "possibly," "very likely," and so on are avoided? So am I right in

23 saying we don't know what "most probably" means here and that we can say

24 here, for example, we can assume it could be assumed in the condition of?

25 A. Well, if you're asking whether it can be measured, no. It cannot

1 be measured in terms of a concrete number.

2 Q. Thank you. Table 13, please. It's page 16 in English. That's

3 another table that's important here.

4 The sex patterns among marked and not marked as wounded are

5 reported here, and was this table created by you or is it a correct table?

6 Is it a created table?

7 A. If you mean whether I made this table, then the answer is yes, I

8 made this table.

9 Q. But was it made according to the original data, or is it created

10 according to the system of possibilities and so on? You can't answer

11 that. It doesn't matter. But what I really want to know is: Is it

12 correct --

13 MR. STRINGER: Excuse me, Mr. President. He did put a question to

14 the witness and I think that she should be allowed to answer it. She was

15 not given an opportunity. She was still waiting for the translation.

16 THE ACCUSED PRALJAK:

17 Q. Okay.

18 A. Well, if your question is related to the sex that was originally

19 not reported for many patients but was created on the basis of first

20 names, so in this sense I can say the distribution is created in the sense

21 that sex is not an item that is available for every single patient in the

22 books.

23 Q. [Interpretation] Thank you very much. Do you agree that mark of

24 wounding, no, in that column the relation of men and women is 3.3:1, 630?

25 That means 3.3 less women. And 3.8:1 is the ratio in favour of men with

1 the mark of wounding. You can see that from your calculations of

2 percentages. So do you agree with my calculation that there are

3 approximately four times more men than women who were wounded?

4 A. Well, you can see from the total in the first part of the table

5 that there are 4.038 men versus 1.134 women.

6 Q. Please, madam.

7 A. It is 4 to 1 approximately. This is what I was trying to say.

8 Q. Thank you. Now -- now we're going to go on to page 37 of the 60

9 in the report and look at what is written underneath table 16. I'm going

10 to read one of your sentences again. I'm not going to look at the table.

11 We don't have time. You say, third line -- it's page 25 of the English.

12 Third line from the bottom. "The percentages were obtained by including

13 all the cases dealt with in the war hospital, adapted with statistical

14 data regardless of the existence of diagnosis. Or, rather, disregarding

15 the availability of their diagnosis. The estimates were obtained through

16 including, in the adjusted statistics, all cases of treatments in or

17 hospital, disregarding the availability of their diagnosis."

18 Now, do you stand by that and that you included the rest in the

19 system of further investigation regardless of the fact that no diagnosis

20 existed?

21 A. Well, I was trying to find this sentence you were reading, but

22 generally I stand behind the fact that I made an estimate in relation to

23 the absolute minimum number, and the estimate, the more complete minimum

24 number included the cases without the diagnosis.

25 Q. Now table 18b on the B/C/S page 39. It is page 27 of the English.

1 Table 18b.

2 Look at the distribution there. Minimum observed. That's what we

3 know. You have men and women, and the total number of wounded militaries.

4 Here we're dealing with militaries alone, soldiers. Judge Antonetti asked

5 you what that distribution was. Now, does this show that from the ages of

6 20 to 44 you have double-digit numbers for the wounded or, rather,

7 three-digit numbers for the wounded? And then the curve moves toward --

8 moves precipitously downwards when the older age groups appear, and

9 younger ones, and that derivation curve is great. So is that right, what

10 it says there?

11 A. Well, it is. There is figure 8, next page. So this is exactly

12 what it shows.

13 Q. Thank you. And can we also observe that both women in that total

14 number of 34 were wounded as members of the BH army or some other unit; is

15 that correct? Would that be correct?

16 A. Yes, there are 34 women reported as militaries.

17 Q. Now let's look at the last table.

18 THE ACCUSED PRALJAK: [Interpretation] Your Honours, can you tell

19 me how much more time I have? I would need another 10 minutes to get

20 through all this.

21 THE INTERPRETER: Microphone, Your Honour, please.

22 JUDGE ANTONETTI: [Interpretation] Yes, I think you have about

23 eight or nine minutes left, please.

24 THE ACCUSED PRALJAK: [Interpretation] Maybe until the break

25 perhaps so that I can complete my examination?

1 [Trial Chamber confers]

2 JUDGE ANTONETTI: [Interpretation] The Trial Chamber is a bit

3 skeptical. We'll take a break at half past. So you have 10 minutes to

4 finish.

5 THE ACCUSED PRALJAK: [Interpretation]

6 Q. Madam, take a look at table 22 now, please. It is page 46 in the

7 B/C/S and 33 in the English.

8 You have the wounding distribution according to place, locality.

9 Can you in any way interpret this table? Can you interpret it, and do you

10 know the military situation in Mostar? You already said you do not know

11 or did not know. But what I want to ask you is the following: Those

12 wounded in South Camp, in North Camp, you've divided them into three

13 places. North Camp appears a number of times. So we have North Camp

14 mentioned three times, three North Camps, one with 20 wounded, then

15 North Camp again with 11 wounded, and then North Camp with 10 wounded. So

16 North Camp is mentioned three times. Santiceva is mentioned twice. Why

17 didn't you group all this together so that we have the right figure?

18 A. Well, these are the places as reported in the source. This is why

19 we doesn't group them. We could have taken certain categories together.

20 It was not a table that was meant to be interpreted and compared with any

21 military actions in the area, but because these are the places with the

22 largest numbers of wounding you can use this table in any other

23 interpretation. So this report is not only meant to provide certain

24 statistics for our --

25 Q. Very well, fine.

1 A. -- purpose but also can serve broader purposes.

2 Q. Yes. You think that in analysing these data one should know

3 whether in Santiceva Street, for example, there were any civilians at all

4 for us to know what this is all about. For our interpretation we would

5 have to know about the distribution of the units, whether there were

6 civilians in Santiceva, in South Camp, North Camp, or whether they were

7 purely military facilities; would that be correct?

8 A. Sir, I didn't analyse this table from any other perspective, just

9 by -- I did it to show the locations. So this is the actual contents of

10 the source, no more purposes.

11 Q. Yes, I know that, madam. I'm just asking you whether one would

12 have to know that for further analysis.

13 And my next question is as follows: Is it true and correct that

14 in the 1992 population census this was done by 17.000 recorded census

15 takers, 17.000 census takers?

16 A. We are not talking about the census right now. We are talking

17 about war hospital records, and we are talking about --

18 Q. Doctor, we're not discussing this, talking about anything. I'm

19 not your student. I am asking you the questions, and what I'm asking you

20 is this: Do you know that more than 17.000 different handwritings for the

21 record-takers are found working on the 1991 population census in

22 Bosnia-Herzegovina?

23 A. Well, I think it comes probably from my report on the IDPs and

24 refugees. There was approximately 17.000 interviewers who were taking the

25 census records. That's correct.

1 Q. You could have said yes, that's right, without the explanations,

2 because I have very little time, just five minutes.

3 Is it true that the original population census records don't exist

4 because they were burnt in a fire?

5 A. Well, this is a new theory to me. I am unaware of any fire that

6 would damage the census records.

7 Q. Well, you can ask about that, receive information.

8 Now, the second question. These records written down in different

9 handwritings, and there's no subsequent checking of the material, do you

10 know how many mistakes can occur in the optical record, alpha recorder,

11 reader?

12 A. The census questionnaire was coded. It is not that every question

13 was scanned as originally reported. Answers, responses were assigned

14 numerical codes and those were scanned. And of course, the optical

15 character scanning is not a perfect method and there are mistakes related

16 to this. But how many? If you want a number from me, I can't tell.

17 Q. That's what I'm asking you. It wasn't scanned. This was

18 introduced afterwards, and it was an optical scanner. So I'm asking you

19 whether you know what the error rate can be and when you go on to correct

20 the names and introduce them. So you put right the mistakes and you don't

21 know how many of the records and facts were burnt. This is important data

22 for data comparison. So do you know what is the error margin in

23 percentages? Yes or no?

24 A. I said I can't express the error margin related to optical

25 scanning as a number. I can't tell you how large is the error. I can

1 only say errors are there and in every survey. It is in the census.

2 Every other survey we have to deal with errors due to computerisation of

3 the material. Whether it is optical scanning or it is data entry, there

4 will be errors in any material, survey material that is computerised.

5 Q. Doctor, madam, my question is a very simple one. You don't say

6 there can be errors, will be errors, and so on. Do you know what the

7 margin of error is? What is the possible error that one can make? Do you

8 know that or not?

9 MR. STRINGER: Excuse me, Mr. President, the witness has already

10 answered the question. This is just an argument now.

11 JUDGE ANTONETTI: [Interpretation] Yes. Maybe there's a

12 misunderstanding. Mr. Praljak wants to know what is the error percentage.

13 Can you answer what is the percentage of error possible?

14 THE WITNESS: Well, 100 per cent is the maximum error always.

15 There can't be more than 100 per cent of errors. But I can't assess the

16 percentage of error related to optical scanning. I can't assess this. I

17 can't say.

18 THE ACCUSED PRALJAK: [Interpretation]

19 Q. Now, when entries are subsequently made, that people entered some

20 facts on the basis of names, let's take a group of Serb names and then a

21 Muslim group name and a Croatian name group.

22 Now, can you give us the percentage where there is overlapping

23 between Croats and Serbs and so on? How many names in common? How many

24 names do you find in both groups? For example, in the Serbo-Croatian

25 group. Take the Serbs and the Croats. How many names in common do they

1 have or what is the cross-section of this group of names?

2 MR. STRINGER: Excuse me, Mr. President. I'm going to object to

3 the question. I believe it goes beyond the competence of this witness.

4 She's testified that native speakers of the language are those who did the

5 assessment of names and what ethnicities would be assigned names and that

6 she didn't do any independent assessment of that on her own, so I believe

7 the question is improper.

8 JUDGE ANTONETTI: [Interpretation] Mr. Praljak, Mr. Stringer has

9 just raised an objection. This witness told us that it is the native

10 B/C/S speakers who allotted the ethnicity to the names, according to the

11 names whether they were Serbs, Croats, or Muslims. It seems she has no

12 knowledge of this. So she only took into account what others told her.

13 And I think that's what you wanted her to say, whether there's a margin of

14 error there.

15 THE ACCUSED PRALJAK: [Interpretation] Your Honours, that is my

16 basic question. The lady has signed other people's research, and on the

17 basis of other people's research that she does not know about or cannot

18 testify to their relevance, scholarly relevance, she goes on to compile a

19 series of tables which I want to show are based on assumptions. Which,

20 quite simply, don't stand scientifically. For example, Serbs and Croats,

21 they have 50 per cent names in common. Slobodan is more a Serb name than

22 it is a Croatian, so I would probably be ranked as a Serb voter or Serb

23 refugee. I'm talking about the relevance of the expert team study.

24 Because if you take over somebody else's data and information, you must

25 stand by it if you haven't rejected it. And as another important group,

1 although I don't have now more time, and that's the two groups in fact.

2 One more minute may I have? It is a change data on the '91 census and the

3 group of interviewees, voters, voluntary voters, that is to say, people

4 who want to go to the elections. But I think Mr. Karnavas will address

5 that issue further and I have completed my cross-examination. Thank you,

6 madam. Through Doctor.

7 JUDGE ANTONETTI: [Interpretation] A follow-up question after what

8 Mr. Praljak said, please. Let's imagine that Mr. Praljak was admitted to

9 the hospital because he was wounded. So he's wounded. He's admitted into

10 the war hospital. His first name is Slobodan. If someone who has no idea

11 who Mr. Slobodan Praljak might be, seeing this first name, which seems to

12 be more common among the Serbs, could the conclusion drawn be that he was

13 a Serb even though he's not a Serb? And out of this type of statement

14 could you extrapolate the margin of error?

15 THE WITNESS: Well, I didn't do the assessment of ethnicity based

16 on names, so I don't know what the others did, but if I was doing this --

17 JUDGE ANTONETTI: [Interpretation] Yes, we've understood that you

18 haven't done it. But you are responsible for this report.

19 THE WITNESS: Yes.

20 JUDGE ANTONETTI: [Interpretation] You are testifying, not the

21 B/C/S speakers. And the figures in your charts are based on these figures

22 given by these B/C/S speakers. So any curious mind with a few

23 mathematical or statistical knowledge might wonder and ask these

24 questions, you know, whether out of a sample where consequences were drawn

25 from the first name or the name in order to make the distribution, it

1 would be interesting to know what the margin of error might be.

2 A. Well there is quite a number of unknown cases, unknown on

3 ethnicity, and I believe the -- this is my observation from the data that

4 cases that were relatively straightforward for the B/C/S speakers doing

5 the job were recorded as the Bosniaks. When it came to the cases that

6 were obviously mixed cases like Slobodan Praljak, then these cases, most

7 of them, were left like as unknowns. So this is not the case here that we

8 have for every single record in the war hospital books ethnicity

9 determined. We are not speaking of this situation. And I myself would

10 behave exactly the same. So I would leave out a group of record of which

11 I wouldn't be certain of ethnicity, just out. I would leave it out and

12 not use it in -- in the analysis as determined cases.

13 JUDGE ANTONETTI: [Interpretation] Mr. Praljak, we must now come to

14 a close. One last question.

15 THE ACCUSED PRALJAK: [Interpretation]

16 Q. But, madam, you don't know about this, but anyway there are six of

17 us here, and Serbs have the same first and last names. So with the

18 Bosniaks the number is little less, but if you combine these names then

19 you get 70 per cent. So that's what I'm asking you. Do you know this

20 overlapping of names and what errors -- and this leads to errors and how

21 much that error is when you have the same name in both groups, and then

22 those errors are multiplied. That's what I'm asking you about, this

23 margin of error and the number of errors in the census on the basis of

24 these names.

25 A. Well, I am, of course, aware of overlap of names and that

1 different ethnic groups may have the same names. And as I said, this can

2 be studied in the census, and there is a whole distribution of ethnicities

3 with the same names. But in most cases - this is my observation - there

4 is a dominant group, the biggest group that is -- for which the -- a given

5 name is typical. So in my view I can't say generally what is the margin

6 error. If you are interested in a name, give me the name, and I will

7 bring you the distribution tomorrow from my query in the census. That is

8 so simple. It is a different margin for a different name. There is

9 nothing like a common margin error for all names. There are names that

10 are more common and more people are using these names and there are other

11 names. We also know names, first, surname, that there is just one person

12 having this name in the census. So it is very different depending on the

13 name. This is why I cannot give an assessment for all of them.

14 Q. Thank you.

15 THE ACCUSED PRALJAK: [Interpretation] And thank you, Your Honours.

16 I regret that I cannot continue at least for another 15 minutes.

17 JUDGE ANTONETTI: [Interpretation] Fine. We will now take a

18 20-minute break, and after the break Mr. Karnavas will take the floor on

19 behalf of Mr. Prlic.

20 --- Recess taken at 3.37 p.m.

21 --- On resuming at 3.59 p.m.

22 JUDGE ANTONETTI: [Interpretation] Mr. Prlic is represented by

23 Mr. Karnavas, and therefore Mr. Karnavas has the floor for the

24 cross-examination. You may proceed.

25 MR. KARNAVAS: Thank you, Mr. President, Mr. President. Thank

1 you, Your Honours. Good afternoon, everyone. I would like to thank the

2 Coric team for giving us the time.

3 Cross-examination by Mr. Karnavas:

4 Q. Let me just pick up where we left off with respect to margin of

5 errors. There are all sorts of issues with respect to the margin of error

6 that we can't identify in this particular case. For instance, what was

7 the margin of error for the 1991 census. We don't know that, do we?

8 A. What are you talking about?

9 Q. You know what the margin of error was for the results of the 1991

10 census? Say yes or no. What is your margin of error?

11 A. This is not a yes and no question. Margin of error, if you think

12 of coverage, well, census, was meant as a complete survey of the

13 population, and I think this purpose has been largely achieved.

14 Q. What was the margin of error of all the names that you changed,

15 you or your folks, because you're a member of the OTP, so we can't just

16 pass the buck on to the others, but OTP members changed names. What was

17 the margin of error of all the names that were changed to fit the formula

18 of whether they fit into a Muslim, Serb or Croat? What is the margin of

19 error?

20 A. We did not change to names to put into a Muslim, Croat, or Serb

21 formula. We modified the names in order to eliminate spelling mistakes.

22 Q. How many names were modified; give me a figure?

23 A. I can't say. I can give you if you give me time.

24 Q. Excuse me. You had all the time to prepare. You testified on

25 direct. You came here for cross-examination. Do you have a concrete

1 figure that you can look in your report and say, that out of all of these

2 numbers these are the number of names that we changed. Do you have that

3 figure? Yes, no, I don't know.

4 A. Not at hand. Can be produced.

5 Q. Okay. Is there a reason why it's not in your report?

6 A. Well, we --

7 Q. Is there a reason it's not in your report?

8 A. No, because we don't have it. We don't use it in any report. If

9 you look through all my reports, you won't find this figure in any of

10 these reports.

11 Q. Now, what was the margin of error of all the names that were

12 changed; do you know? Or do you wish this Trial Chamber to assume and

13 presume that every one of the changes was absolutely correct?

14 A. Here we speak of a misunderstanding. Changing the name, modifying

15 a given name, has no impact on the reported ethnicity associated with this

16 name. So there is no margin of error as all. Ethnicity remains the same

17 for every name modified or not.

18 Q. There was some -- there was some discussion earlier on about

19 certain names being similar, whether they're Serb or Croat or Muslim, and

20 you indicated that your staff or the investigators, the B/C/S speakers,

21 changed certain names because there were some problems with them.

22 A. This is wrong. We haven't changed any names because there were

23 some problems with them. I don't know what you are talking about.

24 Q. What did you do, ma'am?

25 A. We assigned ethnicity to the names reported in a source.

1 Q. Now, what is the margin of error in the assignment --

2 THE INTERPRETER: Would the speakers kindly not overlap for the

3 sake of the interpreters. Thank you very much.

4 MR. KARNAVAS:

5 Q. What is the margin of error, ma'am?

6 THE INTERPRETER: Please slow down for the record.

7 THE WITNESS: Yes, thank you. Margin of error for which ethnicity

8 was assigned.

9 Q. Yeah, what --

10 A. I can give you --

11 Q. Ma'am?

12 A. -- distribution.

13 Q. Excuse me. My question is very precise.

14 A. I can't give you a number. A single number.

15 Q. Okay.

16 JUDGE TRECHSEL: You are overlapping again, please.

17 MR. KARNAVAS: Your Honour, I would be more than happy to spend

18 two or three days with this witness. That's part of the problem.

19 JUDGE TRECHSEL: But it's certainly not helpful to make it all

20 disappear because then you have no record. Better to have a record than

21 none, Mr. Karnavas.

22 MR. KARNAVAS: I understand and I appreciate the cautionary

23 remarks from the Bench and from the translators. I apologise.

24 Q. Now, what is your salary, ma'am?

25 A. I'm sorry, but this is not a question related to my reports.

1 Q. Well, it's a question that goes to your bias.

2 A. I don't think so.

3 Q. What is your --

4 MR. STRINGER: Excuse me, Mr. President. Counsel wants to ask

5 what her level is, we could go into private session.

6 MR. KARNAVAS: Why?

7 MR. STRINGER: But I think any greater specificity or disclosure

8 on her part is not warranted.

9 MR. KARNAVAS: Mr. President, I don't see why this should be a

10 secret. I'm a UN taxpayer. I mean, my -- my US taxes go to fund this

11 place. I'm entitled to know, the public is entitled to know and it goes

12 to the bias. If this woman was an employee of the OTP, has a high salary

13 with benefits, obviously it goes to bias.

14 MR. STRINGER: Mr. President, it is unfair to ask any witness,

15 while this is being broadcast over the internet, to tell the world what

16 their level of salary is. Now, if we want to go into private session,

17 counsel can ask what level she's employed at here within the Office of the

18 Prosecutor. It's no secret and there's no dispute; she's an employee of

19 the Office of the Prosecutor. And I think that really establishes the

20 point right there. No greater detail is really required.

21 JUDGE ANTONETTI: [Interpretation] Very well. Without asking any

22 particular questions about the salary but, madam, all the employees of

23 this Tribunal have either short-term contracts or long-term contracts, and

24 they belong to a certain status, the P-1, P-2, P-3, P-4. Where do you

25 belong exactly with regard to the P levels and the question posed by the

1 Defence is not so relevant because the hierarchy shows the theoretical

2 competencies of people, of employees. So somebody who is at a level P-5,

3 for instance, has more, perhaps, theoretical knowledge than a person who

4 is a P-2. So you tell us if you are a P-2, P-3, P-4, P-5.

5 THE WITNESS: Well, I still believe this is not an issue that

6 should be discussed in the context of my report. This is not a secret but

7 I would prefer not to speak about it but because I have been asked by you,

8 I'm a P-3 and I have a fixed-term contract.

9 MR. KARNAVAS:

10 Q. Now, does that also --

11 JUDGE ANTONETTI: [Interpretation] Very well, you are a P-3. Thank

12 you very much.

13 MR. KARNAVAS:

14 Q. Does that mean you enjoy other benefit such as health care,

15 pension? Are you into the pension plan?

16 JUDGE ANTONETTI: [Interpretation] Mr. Karnavas, you know very well

17 that UN staff members have a certain status. There are certain rules. So

18 it's a waste of time to ask this witness if she has pension benefits or

19 other.

20 MR. KARNAVAS: Very well.

21 JUDGE ANTONETTI: [Interpretation] All we need to do is just read

22 the documents that are available to all of us.

23 MR. KARNAVAS: [Previous translation continues] ... Very well.

24 Well, Your Honour, with all due respect, there is a difference if you are

25 on a six-month contract or a one-year contract and you are on a contract

1 that allows you to get rather generous benefits, pension benefits out of

2 the UN, because, again, it goes to bias. This particular witness has

3 worked for the Office of the Prosecution for several years, and I think

4 that's one of reasons why I'm asking. So I -- it's a simple question. Is

5 she, you know, vesting into the pension system. It calls for a yes or no.

6 MR. STRINGER: Mr. President, again I object to the question. The

7 witness has testified that she's been an OTP employee, she's done other

8 reports, she's testified in other cases. We will stipulate, we will agree

9 that she's been an OTP employee, that she's currently at the level of P-3,

10 and that she has a contract that would give her all the benefits that a

11 P-3 would be entitled to for a person having a contract greater than six

12 months, and anything really further is --

13 MR. KARNAVAS: That's fine.

14 THE WITNESS: If I may say something. I am not entitled to all

15 benefits because I'm employed here as a local person. I have right to

16 have very good Dutch insurance for much less than the UN Van Breda

17 insurance costs. And I have no benefits like tax-free purchasing and

18 stuff like that.

19 MR. KARNAVAS:

20 Q. Okay. Thank you. Now, as I understand it, prior to -- this is

21 the first time that demographers are being used in this context, and I'm

22 speaking because of Mr. Brunborg, and during the Blagojevic case, he

23 indicated that use of demographers by the OTP was an experiment. Is that

24 correct?

25 A. Well, I don't know that, but what do you mean an experiment?

1 There were posts created, two posts for professional demographers and

2 still two demographers are employed by the OTP. I don't see it as an

3 experiment any more.

4 Q. Very well. Were the methods that you used and Brunborg used, were

5 those methods used by other demographers prior to your employment with the

6 OTP and prior to his employment with the OTP?

7 A. Well, the methods we use are not new and non-standard. There is

8 one statistic and there are statistical methods. The context the method

9 are used is new. I think it is for the first time that these kind of

10 studies were made as the ones by Brunborg and myself, but also

11 Patrick Ball made many reports of this kind in the context of other

12 conflicts. It is not only the ICTY where these kind of studies that are

13 made.

14 Q. But the method you used, the methodology, the way that you took

15 certain sources, the way that you cleaned up the sources, if I can use

16 that term, the way you applied the sources thereafter, the matching or

17 grafting, whatever you want to call it, that methodology, was that used

18 with other conflicts, in other courtrooms in order to come up with some

19 figure the way you folks have?

20 A. Well, I wouldn't think of other courtrooms but there were

21 certainly reports made for truth commissions. Patrick Ball has made

22 several these kind of records using the same methodology.

23 Q. The same methodology that you used?

24 A. Methodology you were referring to the matching methodology in the

25 context of conflict studies. So my answer is yes. Outside this courtroom

1 and this Tribunal, studies were made, but generally the methodology is not

2 new. It's not non-standard. It's been years of experience in matching

3 that we benefit from.

4 Q. All right.

5 A. It's a method that is used throughout many countries of the world.

6 Q. All right. And the manner in which you did the matching, that is

7 a time-tested procedure?

8 A. Well.

9 Q. It's a yes or no, ma'am?

10 A. No, it is not a yes or no question.

11 Q. Okay, I'll withdraw the question?

12 A. I will be brief.

13 Q. I'll withdraw the question because obviously you don't wish to

14 answer my questions.

15 A. I would disagree, sir. I really want to answer your questions.

16 Q. Very well. Then tell me: In the field of demography, what is the

17 definition of siege?

18 A. There is no definition of siege within the field of demography

19 and --

20 Q. What about ethnic cleansing? What is the definition of ethnic

21 cleansing within the field of demography as applied by your colleagues?

22 A. Well, I don't think there is a definition of ethnic cleansing in

23 demography.

24 Q. Whose idea was it, ma'am, to use the term "internally displaced

25 persons" or "refugees," as you have done in your report, to identify

1 people who moved within the country or moved outside the country,

2 irrespective, irrespective of the reasons? Whose idea was it?

3 A. Whose idea was it?

4 Q. Yes.

5 A. We just used the term in order to make it clear that we are trying

6 to statistically measure the displacement and we believe that is the best

7 term to resemble this idea. Well, I wouldn't be able to tell you who

8 figured out that this term should be used. We have been using this term

9 since the very beginning of this project.

10 Q. Now, when you say "we," who is "we"?

11 A. "We" is the demographic unit, and people who cooperate with this

12 unit.

13 Q. Okay. So that would be Brunborg who preceded you?

14 A. Certainly, yes.

15 Q. You?

16 A. Certainly, I am part of the unit. And many people who worked in

17 the unit throughout the years.

18 Q. I read some of your articles, and in one article at least you

19 talked about internal and external migration which is a rather neutral way

20 of putting it. Why was it necessary to put in, to use the term,

21 "internally displaced persons" or "refugees" especially when you do not

22 give us a definition of how you're using those terms right up front?

23 A. Well, the reason was as I explained. We tried to measure

24 something that was a consequence of a conflict, and of course terms

25 normally used in demography like "internal migration", "external

1 migration" do not express the fact that certain consequences were related

2 to the conflict. So the reason was we wanted to express the demographic

3 consequences of a conflict the best we could and this is why we used the

4 term.

5 And secondly, I disagree there is no definition in my report, what

6 I mean by internal displacement or migration -- well, refugees. There is

7 a definition and we discuss this definition here.

8 Q. Which page is it on, ma'am, because I notice -- before you answer

9 that question, I notice that you have the presence of mind in your

10 topic -- we're speaking of the second report, P 09836 titled "Ethnic

11 Composition, Internally Displaced Persons and Refugees". There is a

12 footnote. So you had the presence of mind to put a footnote there but

13 it's not until, I believe, it's page 10 or 11 that you finally get around

14 to telling us how you interpret or how you're going to use these terms.

15 Was there a reason why you couldn't put your definition right up front and

16 say that these are not normative terms, these are not terms as they're

17 normally used in the courtrooms, but these are terms that we decided to

18 use for convenient purposes?

19 A. Well, it is page 11 indeed where we explain what's our definition

20 of internally displaced persons and refugees, and we explicitly say these

21 are statistical definitions and no legal definitions.

22 Q. Thank you. Thank you, ma'am. Now with respect to siege, what is

23 your definition of siege, number one. I'm asking a compound question but

24 number two, how is it that you determined, you the demographer, that

25 indeed what was happening in Mostar at that conflict, it was a siege or

1 was this something that you decided to use, sort of, for convenient

2 purposes?

3 A. Well, I didn't use it for convenient purposes. I rather follow

4 the common definition, definition, or common use, say better. Well, in

5 many press reports, videos, from this period the term "siege" is used and

6 there is no other connotations behind my use this term. I just took it

7 from the common use in the media in this period.

8 Q. Okay, I take it you believe what you read in the media?

9 A. Well, I don't.

10 Q. Yes or no?

11 A. This is not a yes and no question.

12 Q. Okay, very well.

13 A. The quality of information in the media is very different, very

14 very different, and there is no reason to trust everything you hear in the

15 media.

16 Q. Is there anything in your report where you qualify the term

17 "siege" or "ethnic cleansing" to suggest, at least to the Court, trying

18 to be an objective and impartial and an independent expert, that you are

19 using these terms merely because this is something that you read in the

20 press, versus, say, the impression that I got from reading your report

21 that you believed that there was a siege and that there was ethnic

22 cleansing? Is there anywhere in your report where you -- where you alert

23 the reader as to your intentions behind using these particular terms?

24 A. Well, there is no definitions of the terms "siege" and "ethnic

25 cleansing," you are right. There is no definitions as such. But -- well,

1 I am surprised you are talking about it as the ethnic cleansing is a term

2 that has been used also in many legal documents in the Statute of this

3 Tribunal.

4 Q. Really?

5 A. Yes, really it is.

6 Q. There is a crime called ethnic cleansing, as far as you know?

7 A. Okay, there is no definition in my report. Are you happy?

8 Q. But you would agree with me, ma'am, that it's not your job to

9 determine whether ethnic cleansing occurred, assuming that it did happen

10 and we know what it is, or that a siege was taking place. That is not

11 your job. Your job is to deal with figures and statistics; correct?

12 A. Yes, it is, but as far as I remember I didn't make any - in my

13 final conclusions - any statements related to the fact that there has been

14 ethnic cleansing, there has been siege and what is the meaning of this.

15 Q. Okay.

16 A. In my final conclusions, I speak of the numbers and statistics.

17 Q. Okay very well. Now, getting back to your --

18 THE INTERPRETER: Please slow down for the transcript.

19 MR. KARNAVAS: I apologise.

20 Q. Getting to your -- to your work the way I see it, and help me out

21 here if I'm wrong, you know, we need to look at three basic things or the

22 Trial Chamber needs to look at three basic things, one are the sources

23 that you used; two, the methodology; and then three, the analysis or the

24 interpretation of the results of that. Is that -- am I correct?

25 A. Well, generally, yes.

1 Q. Okay. Now, speaking of the sources, we know that basically you're

2 using, with this particular report, I'm sticking with this one since

3 General Praljak covered the other two, you -- you rely primarily on two

4 sources. One was the 1991 census and the other one was the voter

5 registration in 1997, 1998; correct?

6 A. Well, primarily, yes.

7 Q. Okay. And now let's speak a little bit about the sources. Both

8 sources have their limitations; correct?

9 A. Yes.

10 Q. Okay. And that's why I asked initially if you could tell us the

11 margin of error with the 1991 census, but let's talk a little bit about

12 that. The data that you got from the 1991 census somebody else had

13 prepared?

14 A. Yes, that's right.

15 Q. Okay. And prior to the 1991 census there was an effort in the

16 former Yugoslavia to figure out how many households there were and how

17 many members within these households; correct?

18 A. Well, probably several attempts. It is the task of every

19 statistical office in every country to have this kind of --

20 Q. [Overlapping speakers] 1991, just stick with me, ma'am, work with

21 me here, will you? 1991, that was the purpose of the 1991 census; right?

22 A. Yes. Part of it was related to households.

23 Q. All right. The purpose of it was not to measure how many Muslims

24 or how many Yugoslavs or how many Serbs, Croats, or Montenegrins. That

25 wasn't the real purpose behind the census, was it?

1 A. I think the main purpose of every census in every country is to

2 provide the count of the population. That is the major purpose but of

3 course much more information is collected than just the number of

4 individuals.

5 Q. I didn't ask what information was collected, ma'am. The

6 purpose -- I'll move on.

7 Let's stick a little bit with the issue of nationality. That's a

8 subjective criteria, isn't it?

9 A. It is a self-perceived item in the census questionnaire. It is

10 self-perception.

11 Q. And anybody could answer it however they wished?

12 A. Yes, that's right.

13 Q. And there was no way of checking that?

14 A. What do you mean by checking? Self-perception and responses are

15 just the way things are.

16 Q. Exactly. So somebody could be, say, for instance, Montenegrin and

17 claim to be a Serb. And there's no way to verify whether that's

18 accurate.

19 A. If the person sees himself or herself as a Serb then it is a Serb.

20 Q. Okay.

21 THE INTERPRETER: Please do not overlap. Thank you.

22 MR. KARNAVAS:

23 Q. Now, as I understand it, there was no -- prior to the census,

24 there was no testing done in advance; that is correct?

25 A. Oh, no, you are wrong. There was a pilot census in 1988 [Realtime

1 transcript read in error "1998"].

2 Q. All right. Because you were asked about that, I believe in the

3 Stakic case and at that point you did not know that. So I take it --

4 A. I might have not. I can't exclude that.

5 Q. Okay. So as of 2001 you weren't aware that there was a pilot

6 project?

7 A. Well, I would be surprised at that time if there was no pilot

8 census, because it's normally done. But I might at that time not know

9 exactly about the pilot.

10 Q. [Overlapping speakers] Just as you did not know when you testified

11 in Simic that there were international standards. That had to be pointed

12 out to you in Simic, but then by Stakic you learned that there were indeed

13 international standards?

14 A. Sir, this is not correct. I am aware of the existence of

15 international standards for large surveys like the population census.

16 Q. That wasn't your answer in Simic though; was it?

17 A. Well.

18 Q. Okay?

19 A. You know better than me. I can't confirm. I testified many

20 times. I can't remember every testimony and every transcript.

21 Q. Now, did you check -- take a look and check the results of the

22 pilot project?

23 A. Myself, no, I didn't. This was the role that -- of the

24 authorities preparing the census.

25 Q. Okay.

1 MR. STRINGER: Excuse me, counsel, just a clarification. The

2 record indicates, just on line 24, moving up that the pilot census was in

3 1998.

4 A. 88, 88 of course.

5 MR. STRINGER: I wasn't sure if that was correct or not.

6 THE INTERPRETER: Please do not overlap for the record. Thank

7 you. And speak one at a time. Thank you.

8 MR. KARNAVAS:

9 Q. Now, when did you learn, when did you learn that there was this

10 pilot census?

11 A. Well, I requested at some point specific reports about the

12 preparations of the 1991 census from statistical authorities in Sarajevo.

13 When exactly was it? It was probably around 2002, I believe. I am not

14 sure that.

15 Q. Okay. By 2002 how many reports had you -- had you used where you

16 relied on the 1991 census?

17 A. Well, sir, I have been working with the population census since

18 the time I arrived here. That is September 2000.

19 Q. Ma'am, how many reports did you write? I mean, it's a pretty

20 simple question. How many reports?

21 A. Well, probably just one but we can check this in my --

22 Q. Just one?

23 A. I -- I believe, sir. I don't remember exactly the dates of all my

24 reports.

25 Q. Okay?

1 A. There were that many that I can't.

2 Q. Very well, I don't want to waste my time on that. After the

3 census was conducted --

4 A. I see just one - sorry to disrupt you - in 2001.

5 Q. All right, and that was which one?

6 A. Oh, sorry, two. That would be two.

7 Q. Because you testified in Stakic and it was pointed out to you and

8 learned after Stakic. Simic comes before Stakic, so of course, there's

9 two.

10 A. But do you find it wrong that I learned?

11 Q. No.

12 A. I think it is better that I learned --

13 Q. Yeah.

14 A. -- than I didn't; right?

15 Q. Ma'am, I hope that this is not a trial-and-error process. You

16 come into court, you learn in court and then afterwards, after you've done

17 the damage, then you go back and try to correct it. I hope that's not the

18 process.

19 A. I wish it is not as well.

20 Q. Okay. Now, after the -- after the census was conducted, normally

21 there's also some sort of a post-data-gathering process that's normally

22 done with the census to make sure that it's -- it's correct. Is that

23 right?

24 A. Yes, that's right.

25 Q. And in this case, the 1991 census, that wasn't done?

1 A. It was done. There was a control census after the census.

2 Actually, directly, even, after the census.

3 Q. Okay.

4 A. And it is discussed in my report.

5 Q. Okay, so as far as you know, it was done?

6 A. It was done.

7 Q. And is that why you put in one of your articles that it was done

8 at the municipal level, they -- I believe you stated something to the

9 effect that the municipal authorities cleaned up -- up. I have it right

10 here. You indicate in one of your articles that the municipal census

11 commission searched for and eliminated the duplicates within

12 municipalities. That's on 200 and 201 of your article, one of your

13 articles that you wrote and I'll get the title of it right now. "War

14 related deaths in 1992 - 1995". That was published in the European

15 Journal of Population, 2005.

16 A. I think it is --

17 Q. Now, is it your understanding, really -- it's 1D 01658. It must

18 be in your packet. That's what I'm referring to. I just want to pin you

19 down here, ma'am. Is it your testimony under oath -- could you please

20 help.

21 Is it your testimony under oath, ma'am, that the municipal

22 commissions searched for and eliminated the duplicates within the

23 municipalities?

24 A. Could you please refer to the article where I'm discussing this.

25 Q. Yes it's on page 200, 201. Let's me just ask you the question

1 though. Is it your understanding that the municipal census commissions

2 eliminated duplicates within the municipalities?

3 A. Yes, it is, but this is different than the control census.

4 Q. Okay. I'm just going step-by-step. Okay? So where there were

5 duplicates before -- and then you say in your article in the next step the

6 republican statistical office was supposed to clean up the duplicates

7 observed between the municipalities.

8 A. Yes, as well.

9 Q. Okay. May I ask where did you learn this information?

10 A. Well, if you would go through the report of the --

11 Q. Ma'am, just kindly answer my question. If you would like for us

12 to get a translator to speak in your native tongue, I can make that

13 request. Otherwise, kindly answer my question. Where did you learn this

14 information?

15 A. From the report by Nora Selimovic, included as part of my report

16 on the IDPs and refugees.

17 Q. Were you aware there was an organisational manual concerning the

18 census; yes or no?

19 A. Do you mean like guidelines for interviewers, for municipal

20 commissions --

21 Q. No, no, no?

22 A. -- For instructors?

23 Q. No, an organisational manual at the federal level, at the

24 republican level, as far as what to do with the data once it was gathered?

25 Were you aware that there was such a manual? I'm asking the question.

1 Yes or no. Or in the alternative, were you merely relying on this person

2 who you claim was an expert on the census, the B/C/S person?

3 A. Well, with regard to the organisation of the census and conduct of

4 census, there was census law. With regard to data processing and

5 preparation of the publications based on census data, there, of course,

6 this is part of regular work of every statistical office.

7 Q. Okay, hold it.

8 A. So for this they don't know -- they don't need manuals.

9 Q. Okay, I just want to make sure I understand you correctly. What

10 you're stating is: Either a manual does not exist, or you do not know

11 whether one exists, a manual, that is, with respect to how to process and

12 analyse the data after the census. And let me make sure that we're clear.

13 I am not speaking about publishing the raw data as it was done sometime

14 thereafter in an official publication. Those are two different issues.

15 Are you aware whether there was an organisational manual? Yes,

16 no, I don't know, I can't remember?

17 A. None of these three answers that fits here because I am aware of

18 several documents with several types of guidelines.

19 Q. Okay.

20 A. Also for post-census data processing.

21 Q. Right.

22 A. And you can find all this in my report and in Nora Selimovic's

23 report. I'm not exactly sure what you are talking about, organisational

24 manual. I wouldn't really know what you mean.

25 Q. Well, I explained it.

1 A. About what? What should be regulated about this manual?

2 Q. Very well. I don't want to overlap. I am told that, at least at

3 the municipal level, the municipal census commissions as you stated could

4 absolutely do nothing with respect to eliminate duplicates. All they

5 could do is send the raw data and it was at the republican level that that

6 would be cleaned up. Because after all how can one, as you put it, how

7 can one municipal census commission determine the duplicates within

8 another municipal census commission or municipal census data, if they

9 don't have all of that? So according to you, the municipalities are

10 trying to search for and eliminate duplicates from the municipality, and

11 after that they send all this material to the republican level?

12 A. So the --

13 Q. Is that what you're saying?

14 A. You didn't get it right.

15 Q. Okay.

16 A. In the municipalities, there were municipal census commissions.

17 In every municipality there was a municipal census commission, and there

18 was a group, interviewers and instructors, census instructors, who were

19 jointly responsible for providing a complete material. "Complete" means

20 that the coverage of a given municipality would be full, that there

21 wouldn't be any households left out, not covered by the census, and in

22 order to ensure that the completeness of the census, interview census

23 commissions, municipal census commissions, and the interviewers received

24 specific maps of the enumeration areas. Based on these maps, they were

25 first of all supposed to check every household that was located in their

1 areas and at the end of the census when they were providing their

2 interviewers their materials to the municipal instructors, the instructors

3 were obliged to collect the completeness of the questionnaires collected

4 by the interviewers.

5 Q. Very well.

6 A. And of course, the municipal census commissions had to approve

7 this information and had to make sure that the completeness is assured,

8 and at the same time, it was their role to assure that none of the

9 households is counted twice.

10 Q. Okay. This is what --

11 A. Within -- within the given municipality, of course. This is the

12 municipal checks of the completeness and of duplicates --

13 Q. All right.

14 A. -- in the census response.

15 Q. Well, let me just read here what you say on page 200 of your

16 article which you authored. At the bottom of 200, going to 201, you

17 say, "Duplicates are not unusual in statistical surveys and therefore,

18 during the 1991 census in Bosnia, this problem received the necessary

19 attention too, and thus the municipal census commissions searched for and

20 eliminated the duplicates within municipalities."

21 This is what you state. Now, you stand behind that statement.

22 A. Yes, I do. I just explained.

23 Q. All right. "In the next step the republican statistical office

24 was supposed to clean the duplicates observed between municipalities."

25 So in other words, before the municipalities send the raw data to

1 the republican statistical office, it is your belief that at the municipal

2 level, they were eliminating the duplicates.

3 A. Within the municipalities.

4 Q. Within the municipalities?

5 A. Within the municipalities by comparing the enumeration areas and

6 the collected questionnaires and checking whether the households reported

7 within the municipalities were not reported twice.

8 Q. Okay, very well. And the 1991 census was never officially

9 certified because of the break-up of the war; correct?

10 A. There is no need to certify a census, as such. How do you think

11 it is done?

12 Q. So it's your understanding that under the law that they had passed

13 prior to the census on how the census should be conducted, the criteria,

14 what -- how -- how to analyse the data. It is your understanding that it

15 did not provide for some sort of final certification process?

16 A. Well.

17 Q. Just answer my question, ma'am.

18 A. I am trying.

19 Q. [Overlapping speakers] Say yes or no?

20 A. No, this is not a yes or no question, sir.

21 MR. KARNAVAS: I'll move on, Your Honours.

22 THE WITNESS: No, no, no. There was a census law.

23 MR. KARNAVAS: Your Honours, the question is very clear and very

24 precise.

25 THE WITNESS: And it is not possible.

1 MR. KARNAVAS:

2 Q. Let me restate it.

3 A. Yeah.

4 Q. Let me restate it for you. Kindly listen to my question. Is it

5 your testimony, under oath, that the law and the procedures set forth for

6 the 1991 census did not call for some sort of end -- end-of-the-process

7 certification by the official authorities? It's a yes or no.

8 A. The law, census law, requested the authorities, statistical

9 authority, for providing reports on the conduct of the census, including

10 the financial aspects of the conduct, and such reports were provided to

11 the Assembly of Bosnia and Herzegovina as requested and were never

12 disapproved. So do you think that there is a need of a special law which

13 would say that the reports were -- have been approved? They were never

14 rejected. They were never disapproved. It is not enough as a

15 certification?

16 Q. Ma'am, could you give me the actual number of all the duplicates

17 that were ultimately found by the local authorities?

18 A. Within the municipalities, you mean? Oh, I can't give you this

19 number.

20 Q. [Overlapping speakers] I'm saying for the entire census. Could

21 you give me that figure?

22 A. Of course not. I can't give you this figure.

23 Q. [Overlapping speakers]

24 A. I have that census data in my office; right? And not the study

25 they made, the analysis, the cleaning they made.

1 Q. So you're telling us, if I understand you correctly, that in

2 Bosnia-Herzegovina, they did a final analysis and they came up with a

3 concrete figure of these are the duplicates. This is the number of

4 duplicates. Can we see that? Yes or no?

5 A. In every municipality they probably have this figure.

6 Q. You say probably, ma'am, I don't mean to overlap.

7 A. How --

8 Q. You see -- excuse me. You know, Mark Twain popularised a saying

9 that was coined by Benjamin Disraeli, who as we all know was one of the

10 greatest prime ministers of the UK, and Benjamin Disraeli said, you know,

11 "There are three kinds of lies. Lies, damn lies, and statistics." Now

12 we're dealing with an area called statistics because demographics is based

13 on statistics. And that's why I'm asking you these very precise

14 questions. So kindly answer my questions.

15 Are you aware whether Bosnia-Herzegovina had a figure that said

16 this is the number of duplicates? Yes or no?

17 A. I am unaware of this figure.

18 Q. Okay. Now, did you come up with a particular number as to how

19 many duplicates were within Bosnia-Herzegovina in the census? Did you,

20 your office, your staff, OTP?

21 A. Well, yes we did some checks for duplicates and came up with a

22 number. I don't exactly remember, but it is in my article, I believe.

23 Q. Okay. Now when you say "some," does that mean you did a sample or

24 you looked at all of them?

25 A. We look throughout all of the census records.

1 Q. Okay. And so you have a concrete number.

2 A. Yes. It's several thousand duplicated records.

3 Q. Well, several thousands well, several thousands could be 3, 6, 5,

4 10, could be 20. Do you have an exact number?

5 A. Well, I will try to find this number for you right now.

6 Q. And tell me also while you're looking at it, does that number --

7 is that number the raw number, or is it the number that you and your staff

8 cleaned up as a result of trying to figure out the names and the

9 nationalities?

10 A. Well, I can't find it right now but I'm pretty sure it is here

11 somewhere.

12 Q. If you could answer my last question?

13 A. It is in the range of, say, 8.000 duplicated records.

14 Q. Eight thousand?

15 A. Yes.

16 JUDGE TRECHSEL: If I can be of assistance, the figure of 8.500

17 can be found at the end of the first paragraph on page 201.

18 MR. KARNAVAS:

19 Q. Now, does that figure include -- that's the one that -- does this

20 number include duplicates that were -- let me rephrase it.

21 In arriving at this number, is this after the raw data was checked

22 by the B/C/S folks working with the investigative unit and you and were

23 able to eliminate certain names?

24 A. Yes, well. I want to clarify. We used three different methods to

25 correct the spelling mistakes in the names.

1 Q. Okay.

2 A. Not only B/C/S native speakers who were studying the names, but

3 also, for instance, household method that was a search within household

4 for one correct name and if such a name was found such a name was assigned

5 to all other members of the household.

6 Q. Okay.

7 A. And this method was very efficient in improving the spelling

8 mistakes and --

9 Q. Let me ask you --

10 A. -- the duplicates were checked after the names had been cleaned

11 from the spelling mistakes.

12 Q. Okay. Now you say the spelling mistakes were -- were corrected.

13 As I understand it, there were folks that went household to household, sat

14 there and filled out this data. Correct?

15 A. Yes.

16 Q. So presumably whoever was talking to whomever was spelling their

17 names; right?

18 A. Well, they were just giving the names and the interviewers were

19 writing them down.

20 Q. And then -- and then somehow your office decided that there were

21 misspellings and therefore someone employed by the Office of the

22 Prosecution decided to come up with correct spelling of these particular

23 names?

24 A. Yes, that's right.

25 Q. Now, can you tell me what the margin of error in that process is?

1 A. I wouldn't understand what is the error here.

2 Q. Well, are you saying that every single one that was corrected is

3 absolutely accurate?

4 A. I can't say --

5 Q. Do you know whether for each one that was corrected, assuming that

6 it was an address, that it was checked to make sure that that person was

7 the name reflected now that it was corrected with the spelling? Was any

8 checking done?

9 A. I can't give you a number.

10 Q. Was the checking done, ma'am? Did anybody go back into the field?

11 A. Of course not. Not after our corrections.

12 Q. Okay, now, when you corrected those names, did you send those --

13 that data to the official authorities in Bosnia-Herzegovina, so at least

14 officially they could look at it to see whether indeed what you had done,

15 you and the others, on behalf of the Office of the Prosecution, could be

16 checked and verified and maybe given -- maybe even given some sort of

17 imprimatur by the government of Bosnia-Herzegovina?

18 A. Well, not as such, not for the purpose of comparing and telling us

19 what is the error made, but on several occasions samples of the data were

20 returned to authorities in Bosnia, and we requested that these names would

21 be checked in context of other sources reporting on the same persons, and

22 surprisingly perhaps for you, not for me, but we received quite many

23 confirmations related to the corrected names.

24 Q. All right. And I take it all of that information is available so

25 if my expert wants to check all of that to see how many actually you sent

1 to Bosnia-Herzegovina and what their responses were, I would be able to --

2 my expert would be able to verify that information?

3 A. Well, this was done on several occasions --

4 Q. No, no, ma'am?

5 A. -- throughout all these years.

6 Q. Ma'am, ma'am, please, I kindly beg you to answer my question only.

7 A. But you didn't listen to my answer.

8 Q. Well, it says -- you could say yes. She's welcome to come, she

9 could look at the data. You could say that.

10 A. Your expert is always welcome to come. I will make anything

11 accessible to your expert. I have no secrets but I told you that I didn't

12 send samples with corrected names to statistical authorities to compare

13 with the actual census questionnaires. I said it --

14 Q. Okay I --

15 A. -- A few minutes ago.

16 Q. I understand and I listen very carefully and I -- obviously you

17 did not send --

18 A. No.

19 Q. -- all of the names you corrected. Okay that's number one.

20 Number two, the ones that you did send, do you have that data available?

21 In other words, I would know that this list of names were sent to Bosnia

22 and Herzegovina, to the authorities, and then they came back with some

23 sort of affirmation or something verifying your work product. Is that

24 information available?

25 A. Of course. It would be available if your expert would like to

1 study --

2 Q. It's available?

3 A. Yes, but do you understand what you are talking about?

4 Q. I understand. I'm just asking the question, ma'am, that's all.

5 A. Okay, okay.

6 Q. There's no need to read into my questions; they're pretty simple.

7 All right. Now with respect to -- I don't have too much time to go into

8 all the problems with the sources, but I want to talk a little bit about

9 your choice between 1991, and 1997/1998, and I take it and this is no

10 disrespect to you or criticism, but basically you -- you -- you dealt with

11 the sort of sources that were available; right?

12 A. Well, this is -- if you are talking about 1997/8, I believe that

13 the voters registers is the best source that can be used, and available.

14 And it is available, of course, yes.

15 Q. Maybe I am just tired and I'm not phrasing my questions correctly.

16 You have no data for 1992?

17 A. No, I don't.

18 Q. No data for 1993?

19 A. Yes, I can --

20 Q. Just answer my question, ma'am?

21 A. Yes.

22 Q. You don't have data for 1993; right?

23 A. No, no.

24 Q. 1994?

25 A. No.

1 Q. 1995?

2 A. Yes, sir.

3 Q. Okay. Now, you do have this 1997/1998, and that source also has

4 its own limitations; correct?

5 A. Of course.

6 Q. All right. And again, we don't have that much time, but now, let

7 me break it down even more. For 1991 prior to the outbreak of the

8 conflict you don't know -- you don't have any statistics as to how many

9 folks, for instance, might have participated in the census but left

10 Bosnia-Herzegovina for whatever reason, employment, education, you know,

11 whatever the reason might be. You don't have any data for that?

12 A. Well, prior, yes, we do. This is part of the census

13 questionnaire, the presence or absence --

14 Q. Ma'am?

15 A. -- at the moment of the census.

16 Q. Okay.

17 A. And this can -- this is reported.

18 Q. Okay, please, listen to my question. They're not trick questions.

19 After the census, okay, somebody -- and incidentally, speaking of the

20 census there were folks that filled out those forms that were living

21 abroad; correct? You're shaking --

22 A. Yes.

23 Q. Okay. Now, you don't know the actual reasons why those folks were

24 living abroad, do you?

25 A. I think there is a reason reported in the census questionnaire but

1 not after the census.

2 Q. Not after the census?

3 A. Yes, yes, yes.

4 Q. Now, do you have a concrete figure of how many people participated

5 in the census but then left, say, in 1991, 1992, 1993, 1994 that had

6 nothing to do with the conflict? Do you have any figures?

7 A. Well, I have no figures at all throughout all these years, so --

8 Q. Okay. So as you sit here - and I'm not being critical, I'm just

9 merely trying to establish certain points - you don't know how many people

10 during that period -- and the reason I picked 1991 to 1994 is because

11 that's the period of the indictment. Okay?

12 A. Okay.

13 Q. So you don't know how many people left voluntarily, do you?

14 A. Well, I don't.

15 Q. Okay. You don't know how many of those people left because, for

16 instance, they had employment opportunities?

17 A. I don't.

18 Q. Or for instance, perhaps some could see that perhaps the situation

19 was unraveling and there might be a conflict and so they left in advance

20 and it had nothing to do with being ethnically cleansed but merely they

21 had options and they chose to exercise those options; correct?

22 A. Yes, I don't.

23 Q. And you could say the same for 1992, 1993, 1994,; correct?

24 A. Yes.

25 Q. Now speak of 1995/1996 -- I have to slow down a little bit. We'll

1 give the -- I picked 1995 because of the Dayton Peace Accords although we

2 also have 1994, which is the -- the Washington Agreement, but for 1995 and

3 thereafter, you have no statistics as far as how many folks decided to

4 stay where they were because simply they wanted to be where they were,

5 because the quality of life was much better than going back to Bosnia?

6 A. Well, I have some statistics by UNHCR on refugees, of course.

7 Q. Okay. Now, when we talk about refugees, again you just told me

8 that for 1991, 1992, 1993, 1994, you can't tell me concretely how many

9 people left or even moved within Bosnia-Herzegovina because they were

10 exercising their own free will versus those who actually were chased out.

11 Now, my question is: 1995 and thereafter, you do not have any concrete

12 information, not just you, or anyone else for that matter, as to how many

13 folks found themselves in the United States, Australia, Canada,

14 Switzerland, Norway, Sweden, whatever, who had left for instance because

15 of educational or employment opportunities, or because they left because

16 of the war, because they had to, and simply did not want to return because

17 now they had established themselves someplace else. You don't have a

18 figure for that, do you?

19 A. No, I don't.

20 Q. Okay. And again, I'm not being critical of you. It's just those

21 are the facts of life; okay?

22 Now, with respect to the movements there was a question by

23 Judge Trechsel with respect to whether, you know, how many people were

24 moving in and about Yugoslavia and maybe outside of Yugoslavia prior to

25 the war and whether we could make some comparison. Do you know, you know,

1 it calls for a yes or no answer, do you know whether there's any

2 statistical data that would demonstrate 1991 or, say, 1981 and up to 1991

3 what was the movement within Yugoslavia, internal migration? Are there

4 figures?

5 A. I actually brought a publication for Bosnia, and I have prepared

6 some statistics, so any time we can discuss this as well.

7 Q. Ma'am, you can help me out here. Does that mean yes?

8 A. Yes, I do.

9 Q. Thank you. Thank you very much. And I take it in those -- they

10 would reflect the reasons why somebody's moving, say, for instance, Serbia

11 and now, they're going to, say, Croatia?

12 A. Well, I brought figures for Bosnia, as I said, and you will see

13 internal migration there, even within municipalities, and you will see

14 external migration as well. That means into other countries. I believe

15 other regions in republics, former republics --

16 Q. Okay.

17 A. -- would be considered to be at that time still one the same

18 country.

19 Q. Right. And at that time there was no war?

20 A. It was 1991 statistics, so at the time no war.

21 Q. Okay it would be fair to say, and maybe this is outside the field

22 of demographics, but it would be fair to say that when there is a

23 conflict, at least some people who have options decide to get out of the

24 way of harm, get out of harm's way; correct?

25 A. Of course.

1 Q. All right. Has nothing to do with ethnic cleansing. It simply

2 has to do with wanting to go someplace where the family, the children can

3 be safe; correct?

4 A. Well, not really because as you know, ethnic cleansing --

5 Q. Hold on. Just listen to my question because --

6 A. It's a very strong push factor. Let me finish; right?

7 Q. Okay.

8 A. And we know it from Bosnia and we know it from Rwanda as well, so

9 it is for whatever reason, conflict as such, conflict as such is a push

10 factor for people to move out.

11 Q. I agree.

12 A. So don't say --

13 Q. Some are moving out either before the conflict comes, when they

14 know it's coming. Some may be leaving during the conflict, even though

15 the conflict hasn't arrived at their particular municipality; correct?

16 A. Yes, certainly.

17 Q. Okay. All right. Let me just, because my time is running out a

18 little bit, were you aware of the law that existed in Bosnia, it's called,

19 I believe, the law on return of abandoned property. Do you know anything

20 about that?

21 A. Not really.

22 Q. Were you aware, ma'am, that part of the 1995 Dayton Accords call

23 for anybody within Bosnia-Herzegovina, they could remain wherever they

24 wished. They could remain where they found themselves in 1995 or go back

25 to their place of birth or where they were prior to the conflict or move

1 someplace else. Were you aware of that?

2 A. Not particularly. I didn't study this, but I am aware of the

3 existence of the International Commission for Real Property in Bosnia.

4 Q. Okay. But let me -- help me out here?

5 A. So they probably know everything about it.

6 Q. Okay. What if I were to tell you that, and this can be checked

7 out, the Office of the High Representative was overlooking and still

8 overlooks Bosnia and Herzegovina, but part of the Dayton Accords was that

9 people could settle where they were or move about. In other words, there

10 was freedom of movement.

11 A. Well, I -- I don't know simply --

12 Q. Okay.

13 A. -- if there is such a law you are saying so.

14 Q. Okay. Now what if I were to also tell you -- or let me ask you

15 this: Was abandoned property -- as folks were abandoning their property

16 and moving elsewhere, were you aware whether that abandoned property was

17 used by other, say, refugees coming in to those areas?

18 A. I can imagine this property could be used by --

19 Q. Okay.

20 A. -- others that lost their houses.

21 Q. Okay. Were you aware that the Office of the High Representative

22 that was overseeing this, and there was an Ombudsman's office as well

23 dealing with property issues, but were you aware that the policy then,

24 post-Dayton, that folks that were occupying somebody else's residence

25 would not simply be evicted but there was a need to find alternative

1 housing if those individuals chose to remain in that particular location;

2 were you aware of that?

3 A. I'm not aware of that. I can imagine this could be.

4 Q. And I take it you have no statistics.

5 A. No, I don't have the statistics but there is the CRPC, and I

6 believe they have the statistics.

7 Q. And what I'm driving at is: Let's say folks want to go back into

8 a particular area but now their abandoned property is being occupied by

9 other refugees. Do you know whether - we don't have a figure - how many

10 people chose not to go back to their abandoned properties simply because

11 there was no available housing, say, as late as 1997, 1998?

12 A. I have -- I have no statistics.

13 Q. Okay?

14 A. But if you look at the UNHCR statistics on internal displacement

15 there was a high increase in the years 1997-8 because there were returns

16 and people were unable to return to their houses.

17 Q. Because others were occupying them.

18 A. Others were occupying or houses were damaged. They were -- they

19 had no place to go to.

20 Q. And so we don't have those figures of how many people -- for

21 instance, we can't say that a certain amount of folks chose not to go back

22 to their place of residence because there was no housing available, not

23 because they were being kept out, but simply there was no housing

24 available and under the OHR policy and under the Dayton Accords folks

25 could remain wherever they found themselves or wherever they wished.

1 A. I don't have the figures but I'm sure you can obtain these figures

2 from authorities in Bosnia. These figures must be available, I'm sure of

3 that.

4 Q. I'm just because -- I'm asking the question because you as a

5 demographer I would suspect to have known that, that perhaps when you look

6 at the raw data, if you're going to go to that step of making a sort of

7 legal or political statement that there is ethnic cleansing here, one

8 would at least have some data to show whether indeed the low return was as

9 a result of people actually not being able to come back because there's no

10 housing available, or they don't want to go back because now the

11 demographic has changed to the point that they don't feel safer any more.

12 And we don't have that sort of data, do we?

13 A. No, but what's your question actually?

14 Q. Well?

15 A. I said several times I don't have the statistics on that.

16 Q. That's all that I want. I just want an admission on your part.

17 MR. KARNAVAS: Your Honours, how much time do I have left? I'm

18 moving around. I understand that it's not fluid, but ...

19 JUDGE ANTONETTI: [Interpretation] Mr. Registrar, how much time do

20 we have left?

21 You have nine or 10 minutes more.

22 MR. KARNAVAS: Thank you, Mr. President. With your indulgence if

23 I could take 30 seconds to look at my notes to see what I could possibly

24 do in that time.

25 Q. With the voters registration list, going back to that, that was

1 the only available source post 1991; correct?

2 A. Of this scale, yes, it was.

3 Q. Now, just to be sure, help me out here, that was -- people

4 volunteered actually to registered, to register to vote; right?

5 A. Right.

6 Q. It wasn't mandatory.

7 A. No.

8 Q. And it's not the sort of source that one could look at a

9 demographer would normally look at; correct?

10 A. No.

11 Q. And in this particular -- this particular source also had some

12 limitations; right?

13 A. Right.

14 Q. One of the limitations was that, for instance, if we're trying to

15 determine ethnicity, for instance, the voters were not required and indeed

16 did not indicate their ethnicity; correct?

17 A. The ethnicity wasn't available.

18 Q. And we know that prior to -- prior to the conflict, at least when

19 the 1991 census was done, I believe there were 29 different ways that one

20 could describe themselves?

21 A. There were 91 ways.

22 Q. 91?

23 A. Yes, this is what we see in the census results.

24 Q. So I was told differently but 91. And just out of curiosity,

25 there were ciphers or I think that's how it's called like a number. So a

1 Croat would be, say, a 1, a Serb a 2; is that correct?

2 A. Yes, name, categories, but then combination of categories,

3 regional groups, other nations, et cetera. Yugoslavs, of course, and some

4 unknown.

5 Q. Okay. And some unknown. All right. And it's your understanding

6 that in those combinations that were provided in the census I just want

7 to -- I just want you to -- a clarification, they even had combinations

8 Serbo-Croat, Croat Serb, a Croat Muslim. They had those combinations and

9 they could sort of write in a number that would denote that particular

10 combination. That's your understanding?

11 A. Well, that is now it was reported, yes, combinations are examples

12 of those you mentioned.

13 Q. Okay, when you say this is how it was reported, I'm asking whether

14 the forms provided that?

15 A. The actual -- yes, yes.

16 Q. Okay.

17 A. Yes, yes.

18 Q. Now the voter registration didn't provide any of that; right? One

19 did --

20 A. No, no, no, no.

21 Q. And by this point of course we no longer have Muslims in

22 Bosnia-Herzegovina. Now they're Bosniaks. Right?

23 A. Well, Bosniaks were also reported in the census so it is not that

24 it was invented after the conflict.

25 Q. I didn't say that it was invented, ma'am, and I'm not trying to go

1 there. What I'm saying is that after this -- that we don't have that any

2 more now in 1997/1998?

3 A. Well, in the voters, generally there is no ethnicity reported as

4 such.

5 Q. All right, the reason I'm asking however is at some point you're

6 trying to go backwards and trying match names because you don't have all

7 this other data, and that's the best that you can do?

8 A. Right.

9 Q. Now, one of the things that we -- that you did have was a

10 particular identification number that everybody was supposed to have;

11 right?

12 A. Right.

13 Q. And that was the -- it's got this -- I think it's maticni broj or

14 something like that?

15 A. That's right.

16 Q. Okay. All right. Now, help me out here. As I understand it,

17 that had 13 numbers; right?

18 A. That's right.

19 Q. The first seven numbers were the date of birth?

20 A. Right.

21 Q. Okay. And the other six were sort of personal numbers.

22 A. Well, there were some codes. Like sex was coded, region was

23 coded, and within each sex there was a subsequent number issued to the

24 individuals.

25 Q. But those were -- those were personal numbers basically?

1 A. Yes, of course.

2 Q. All right. And not everybody had a maticni broj. Not everyone

3 had that, right?

4 A. Well, there was a percentage of individuals who didn't report the

5 number in the census.

6 Q. Okay. Hold on. There's a difference between reporting it and not

7 having it. When the 1991 census was done not everyone had one; correct?

8 A. You're right. Not everybody had one.

9 Q. Okay. All right.

10 A. Yes. A small fraction didn't.

11 Q. When you say a small fraction --

12 A. Well, 20 per cent say approximately.

13 Q. Well --

14 A. Approximately.

15 Q. Okay?

16 A. Or I'm wrong. 13 per cent. Now I remember better, sorry.

17 Q. 13.

18 A. So say between 13 and 20 and we can check it later and correct.

19 Q. Okay. And the same thing when -- when people were registering to

20 vote, one of the things that they were supposed to do was also provide

21 this identification number; correct?

22 A. Well, the thing is that for the registration of voters the actual

23 census data were used including the maticni broj and the names of the

24 individuals. So it was, strictly speaking, the same electronic material

25 that we have in our office that was also used for the registration of

1 voters and is available in the electronic lists of voters.

2 Q. All right. Well, in the -- in the previous case that I -- I

3 employed an expert to look into this and had in fact had access to your

4 data, apparently in BiH, in the BiH census, and this was as of 2004 when

5 those figures were checked, 1.402.099 inhabitants, which would have been

6 about 32 per cent of the total population of BiH which was 4.4, had an

7 incomplete -- I guess they call it -- the short term is JMBG number.

8 Would that being correct?

9 A. That might be correct. I don't remember the numbers exactly but

10 some people didn't, some had incorrect, but the point is that exactly the

11 same numbers were listed on one hand in the census material, and on the

12 other hand in the voters registers, so it really doesn't matter if a

13 number is correct or not because if you do the matching through the JMBG

14 it is the same.

15 Q. Exactly. But what I'm saying is there was at least 32 per cent

16 that had incomplete JMBGs. So at least 32 per cent of the 4.4 million you

17 couldn't match.

18 A. Not true. Not true.

19 Q. At least through not through this identification?

20 A. No, I said --

21 Q. Okay.

22 A. The point is that if you have two numbers in two sources and both

23 are incorrect but they are identical in the two sources it doesn't matter.

24 If you have them you can match because they are all the same in the two

25 sources.

1 Q. Okay. Very well.

2 A. This is the important thing about the marching of the census and

3 the voters.

4 Q. Okay.

5 A. The census material was used for the register of voters exactly as

6 available from the census.

7 Q. All right.

8 A. Which is not the case with many other sources that we used but in

9 this particular case it was the big advantage we had.

10 Q. Do you know how many folks that were in Bosnia-Herzegovina chose

11 not to register to vote?

12 A. I wouldn't know exactly, but it is estimated that 82 per cent

13 registered to vote so 18 wouldn't register.

14 Q. Okay, and how many people?

15 A. This is the OSCE estimate, sorry, not mine. Yes.

16 Q. Okay, and how many people did not register to vote that were

17 living outside of BiH?

18 A. I believe this is a difficult question, and I don't know that.

19 Q. Okay.

20 A. And I don't think anybody knows that.

21 Q. All right. Very well.

22 MR. KARNAVAS: One last couple of questions, just another three,

23 Your Honours.

24 Q. And these are technical questions and I've just -- maybe you can

25 help me out here. Dealing with Mostar, we know that after Dayton, Mostar

1 became what is it six municipalities -- or eight municipalities; right?

2 A. Eight municipalities.

3 Q. And prior -- prior to Dayton had been one municipality; right?

4 A. Right.

5 Q. Now when you -- and I believe you said something about bridging

6 over during -- you tried to bridge over in order to figure out who was

7 where and how many and from that determine how many failed to return or

8 how many would have been "ethnically cleansed" or refugees or internally

9 displaced persons; am I correct on that?

10 A. On the bridging and I mentioned the term "bridging," "bridging

11 system" in the context of the new municipalities. In order to be able to

12 produce statistics for the new municipalities, a bridging system is needed

13 that would link the small statistical areas within the Mostar, pre-war

14 Mostar big municipalities with the eight new post-war Mostar

15 municipalities.

16 Q. Okay. And now, did you start your calculations from the census

17 circles?

18 A. Well, we -- I don't understand exactly what you mean here. Do

19 you -- the question is whether we used the census data to --

20 Q. No, no, no, no. Did you start your calculations from the census

21 circles? And I'm told that statistics have them and have sent you or

22 should have sent you this information. What's a census circle? The

23 lowest territorial unit by which population is registered.

24 A. Yes, statistical area, or in the census it is enumeration area.

25 This is how it called.

1 Q. Okay and how many are there for Mostar?

2 A. Oh, for Mostar, very many. I wouldn't know exactly how many but

3 we can check this.

4 Q. Okay, well.

5 A. These are very small areas, really, yes. The bridging system, I

6 want to stress, I didn't make it myself. I received it from statistical

7 authorities.

8 Q. Statistical authorities in BiH?

9 A. In BiH.

10 Q. So, at least, there's no sense in me questioning you on that

11 because that's not your work product. That's somebody else --

12 A. That is not my work product, it is the product of the statistical

13 authority of Bosnia and Herzegovina.

14 Q. All right. Okay. Very well, and did you double-check it to see,

15 to make sure that it's accurate?

16 A. Well, I -- what can I do is making maps and seeing how it works in

17 practice. But I can't check in terms that I can go to Mostar --

18 Q. Why can't you?

19 A. -- Area by area and check every statistical area and check.

20 Q. Well, you can do a sample, couldn't you?

21 A. Sir.

22 Q. You could do a sample, couldn't you?

23 A. The bridging system is an official table that comes from the

24 statistical authority. This is not my task to challenge the bridging

25 system somebody, a statistical authority, developed in Bosnia for their

1 use in the country. It is something that comes to me and I use it.

2 That's it.

3 Q. Okay.

4 MR. KARNAVAS: Your Honours, I believe that's all I have. I

5 appreciate the extra time. Unfortunately, it would have been better to

6 have two or three days, I could have gone at a slower pace. I want to

7 thank you for coming here and testifying. And I want to again thank the

8 Coric team for giving me the 30 minutes. I believe that Ms. Alaburic did

9 have a question that she wished to put, and it was so -- rather technical

10 and I was unable to articulate it, so perhaps if the Court could allow

11 Ms. Alaburic two or three minutes. Otherwise I don't know what -- if she

12 still thinks she needs the time but I'll leave it up to her.

13 MS. ALABURIC: [Interpretation] Today -- they say there is no

14 interpretation so I'll start from the beginning. Today's question of

15 His Honour Judge Antonetti recorded on pages 9 and 10 and the reply was as

16 to the age of the victims as a possible criterion for establishing

17 military or civilian status. The witness's reply was that that was one of

18 the possible methods but that she did not use that method. I believe that

19 based on table 18 and figure 6 from the report on those killed in Mostar

20 we could establish numbers which might assist us in approximating

21 ourselves to a reply to Judge Antonetti's question, and if I am allowed a

22 few minutes I would like to explore this issue.

23 JUDGE ANTONETTI: [Interpretation] Yes. Please go ahead.

24 Cross-examination by Ms. Alaburic:

25 Q. [Interpretation] Before I put my specific question, madam, can we

1 agree that the population in Mostar in 1992 and 1993, and this includes

2 East Mostar, was increased because of a large number of refugees? Can we

3 agree on that?

4 A. Well, I agree that it was increased but refugees I wouldn't think

5 of refugees. These were people from other areas in Bosnia, so I wouldn't

6 call them refugees.

7 Q. Yes, I agree with you. We can speak of displaced persons. I'm

8 not going into the legal definition of the people who arrived in

9 East Mostar at the time, but tell me, among the people who arrived,

10 whether they were internally displaced or refugees, can we agree that the

11 vast majority of them were women, children, and elderly?

12 A. I wouldn't know that really because I -- I wouldn't know, no.

13 Q. Have you any knowledge as to whether the arrival of displaced

14 persons and/or refugees in a certain area, the number of military-able

15 persons increases or does it, on the contrary, decrease as a proportion of

16 the population?

17 A. Well, if -- if it is just one population and one group decreases,

18 the other one must increase, of course.

19 Q. Very well. If we now look at your table 18 and figure 6, and you

20 also mentioned this in the text, however, it's clear from figure 6 that

21 there is a steep increase in persons over 65 years of age, both in the

22 male and in the female populations. Although you did not analyse the

23 causes of death for these people, would it be logical to conclude that

24 regardless of sex, this information means that persons aged over 65 died

25 of natural causes?

1 A. I don't think it would be a correct conclusion because this age

2 group is larger than any other age group shown in this chart. It is just

3 a very broad age group. All persons, 65 plus, are included in this group.

4 So this is a category that is uncomparable with other categories. I would

5 have to redo the chart and include all the ages starting 65 by five-year

6 age groups.

7 Q. Madam, all your groups more or less cover five years, whereas the

8 65 plus age group can also include persons with a 30-year age span. The

9 difference can be up to 30 years. Can we agree they can be from 65 to 95?

10 A. This is what I said. This is what I said, that the length of this

11 interval is much larger than all other five-year age groups. Therefore,

12 it is uncomparable with other age groups. Therefore, there is this jump

13 you see in the chart because it is the aggregate age group. Large age

14 interval.

15 Q. Very well. In my opinion this is a group for which it might be

16 logical to expect that a larger proportion of that group would die of

17 natural causes, but I won't go into this any further. I wanted to explore

18 this category of over 65 year olds. If we analyse the figures, however,

19 we will see that the have a majority of those killed were in fact

20 military-age men, and that is also your conclusion in the text that

21 accompanies the table. Is that correct?

22 A. Yes, it is.

23 Q. If we were to exclude people aged over 65 for which it would be

24 logical to expect that they died of natural causes to a greater extent

25 than others, then the proportion of military-able men in the total

1 population of those killed would be even greater. Would we agree on that?

2 A. Yes.

3 Q. And if we analyse according to various criteria and calculate

4 percentages, adding or leaving out certain categories of the population,

5 we can conclude that there is a significantly greater proportion of

6 military-age men who were killed in this population as compared women or

7 men who were below or above military age. Can we agree on that?

8 A. Yes.

9 Q. If the number of military-age men who were killed is linked to the

10 percentage of military-age men in the total population, which decreased

11 because of the arrival of displaced persons or refugees, women, children,

12 and elderly, then we can conclude that the proportion of military-age men

13 who were killed in this same population is considerably greater than the

14 proportion of women, children, and elderly, and men who were not of

15 military age in that category of the population. Can we agree on that?

16 A. Well, I -- I need to read your statement because you -- you made

17 some statements, but I -- I think I am -- I made it clear in the report

18 that majority of these victims are men at military age. So there is no

19 doubt about it, of course. But I think that age-based approach is just a

20 different approach. We didn't use this approach. We used data-driven

21 approach. So ...

22 Q. Very well.

23 MS. ALABURIC: [Interpretation] Your Honours, I just wanted to

24 point to this information.

25 JUDGE ANTONETTI: [Interpretation] Mr. Stringer, do you have any

1 additional questions in redirect?

2 MR. STRINGER: Yes, Mr. President.

3 JUDGE ANTONETTI: [Interpretation] Do you have many questions?

4 MR. STRINGER: I have perhaps 15 minutes. I'm in your hands,

5 Mr. President, if you prefer to take a break.

6 JUDGE ANTONETTI: [Interpretation] And I also have questions for

7 the witness. So theoretically we should take a break at 5.30 because we

8 started at 4.00 at any rate. So let's take a break now and we will start

9 with my questions and we will have 15 minutes for you. A 20-minute break.

10 --- Recess taken at 5.24 p.m.

11 --- On resuming at 5.46 p.m.

12 JUDGE ANTONETTI: [Interpretation] I have a few questions, but

13 first I would like to be absolutely sure, and I'm asking a question to

14 Mr. Karnavas. 1D 01658 which is demographics, is that what you talked to

15 with -- talked about with the witness, this --

16 MR. KARNAVAS: Yes, Your Honour. There are two articles that the

17 witness wrote, but the one that I referred to was 1D 01658, as I recall.

18 The other article you see in our bundle is 1657. That's just sort of an

19 overview of -- of this conference. The title of that one is "Demography

20 of Conflicts and Violence: An emerging field".

21 JUDGE ANTONETTI: [Interpretation] Fine. Thank you.

22 Questioned by the Court:

23 JUDGE ANTONETTI: [Interpretation] Defence mentioned two articles

24 that you wrote, which I am finding out about during this hearing. One of

25 this article was written with Mr. Jakub Bijak.

1 A. Yes.

2 JUDGE ANTONETTI: [Interpretation] When I read this article, I

3 noted that some of your colleagues, demographer colleagues, made studies

4 on missing persons or persons killed, and you actually write this in your

5 article, especially in tables. For example, on page 197 there's a general

6 chart giving the number of missing or killed persons between 1991 and

7 1995. As sources we have Mr. Bassiouni, an institution in Stockholm

8 called SIPRI, a Mr. Boyle, Mr. Thomas, and Mr. Kenney [phoen]. So we find

9 out that a great number of people have studied this topic of missing or

10 killed people.

11 Out of curiosity, I looked at your three reports because I was

12 wondering whether you referred to other demographers in your report, and

13 then I discover that you actually never made any comparisons with work

14 that could have been made by other people like Mr. Bassiouni.

15 So when studying your three reports it looks like you're the only

16 expert in this field, but when reading this article we find out that there

17 are other people who studied this problem, and on page 199, discover that

18 these people, yourself and these people supported your work on different

19 sources, notably sources from military reports of the ABiH. This was a

20 question put to you by General Praljak as to whether you used the number

21 of killed produced by the ABiH, and I note that in this article you

22 referred to it, whereas you said that you did not have any access to these

23 documents. So I'd like to understand why it is that in an article you

24 quote this as one of your sources and why earlier you told us that you had

25 no knowledge of ABiH archives regarding the number of people killed within

1 the army of ABiH. Could you answer my question, please?

2 A. Well, question number one related to the overview of estimates of

3 war-related deaths in Bosnia by others. I want to make clear that those

4 other names in the table 2 on page 197 are not demographers. This is

5 one -- just one of them was produced by researchers. The other ones come

6 from totally different persons.

7 I, actually, in the article criticise these estimates and some

8 other locally made estimates for the lack of transparency and for unclear

9 sources and some of them even for being politically motivated.

10 I think when it comes to Bosnia then, indeed, it is hard to find

11 transparent scientific estimates of the -- of the victim -- of the number

12 of victims. Actually, the number we produced and published in this

13 article with Jakub Bijak is one, is the probably first one of this sort.

14 Well, why didn't I make references to the work of others in my

15 three reports on Herceg-Bosna? Well, that is serious difference between

16 the Herceg-Bosna studies and the study of Bosnia-Herzegovina. There has

17 been a lot of interest in the number of victims in the Bosnian wars, that

18 means the victims from Bosnia and Herzegovina as such, and many people

19 produced these kind of numbers, both in the former Yugoslavia and outside

20 it, including the commission by Cherif Bassiouni in the first place.

21 When it comes to a small area like Herceg-Bosna it is hard to find

22 comparable figures. Of course, I could have referred to certain figures I

23 know but none of these figures could be of the same sort as our estimates.

24 There is a figure by Edward Vulliamy, for instance, who speaks of 1.100

25 to 1.200 victims of the siege of East Mostar. Well, this number is in

1 line with our number of killed persons. The more complete number of the

2 persons killed.

3 There are also some other estimates or, say, better -- well, I

4 would call them estimates but non-scientific estimates by people from

5 Mostar itself. I know that a person from the company that was responsible

6 for burials and cleaning up the streets from the parties who was in charge

7 of these kind of activities estimated the number of victims at 1.700

8 persons, but -- and of course there are a few others. There is a military

9 estimate that is approximately 1.000, a little bit and -- 1.000 of

10 victims. The military estimate, I mean coming from a military who served

11 in the Bosnian army during the conflict.

12 Well, I didn't refer to them because to me it's hard to refer to

13 figures that are not entirely transparent in terms of methodology and in

14 terms of sources. So this is the reason why. But there are such figures,

15 definitely. But the figures I produced are -- can be justified in terms

16 of sources, in terms of method, and in terms of deficiencies and in terms

17 of coverage. The others would be much harder from this point of view.

18 This is why.

19 Regarding the military lists that I used for the estimate for

20 Bosnia, exactly the same lists were used in this project in the reports I

21 produced in this report -- in this case -- for this case. These are the

22 same military lists of fallen soldiers and other military personnel, the

23 same version, the same source, the ministries of the defence. Everything

24 exactly the same. So I used -- I did use this source in my Mostar

25 studies.

1 JUDGE ANTONETTI: [Interpretation] Thank you very much. I have

2 another question. Last week when Defence asked you on what basis you made

3 your research, at one point in time it seemed that some figures had been

4 given to you by Sarajevo, by an intelligence department which had provided

5 figures on the number of killed and missing persons. And when I look at

6 the article you wrote, which I mentioned earlier and which was presented

7 by Mr. Karnavas, I note that you also made reference to a CIA report

8 called "Military History of Yugoslavia."

9 So this report from the CIA, which must probably be based on

10 elements, I would like to know whether you read this article or whether

11 the references you make in this article are purely formal. Did you read

12 all the references included in the footnotes of the CIA report?

13 A. Well, I certainly didn't read the entire report. It is a very

14 considerable report. I did read parts of the report related to general

15 history of the conflict but certainly not all of it. But at the same time

16 because I read these parts I can't say this is a formal reference that I

17 just gave for the sake of having a reference.

18 Well, I was using it for historical -- for the introduction, for

19 the history of the conflict.

20 JUDGE ANTONETTI: [Interpretation] My last question now. Could the

21 registrar please call up document 9837, figure 6 on page 20 of the English

22 version. And on page 30 of the B/C/S version.

23 Figure 6 gives us the distribution of victims according to age and

24 sex. So we have two curves on the screen.

25 A. May I ask, it is the report on the killed?

1 JUDGE ANTONETTI: [Interpretation] Absolutely, the killed. It's

2 gone. It's not on the screen any more. Figure 6 on page 20 of the

3 English version, please. No, this is not it. It was the other one.

4 MR. STRINGER: Mr. President, I believe for the record the report

5 on the killed is P 9835, not 9837. She may be looking at the wrong

6 exhibit.

7 JUDGE ANTONETTI: [Interpretation] Fine. Well, in the document you

8 gave us, Mr. Stringer, it's 9837.

9 MR. STRINGER: Then I apologise, Your Honour. It's -- I could be

10 wrong, but I have --

11 JUDGE ANTONETTI: [Interpretation] Well, it's on the screen now.

12 It's on the screen. It's on the screen on the B/C/S version, so we'll

13 work with the B/C/S version.

14 Witness, please, this is a chart giving the distribution of

15 victims according to age. From 0 to 4, that's the first age, and then it

16 ends up with 65 plus. In red we have the women, females, and in black the

17 males, or the men.

18 When looking at this -- these curves, we see peaks. We see that

19 there's obviously a peak in the men killed in the 20 to 24, 25 to 29, 30

20 to 34, and 35 to 39 age ranges, and then it decreases.

21 If there were only civilian victims, given that statistically

22 normally there's as many men as women, the curve for men should be similar

23 to the curve for women, but here we have a big difference between the men

24 and the women. And one could draw from this that there's a peak because

25 of the military victims. If there were only civilians who were killed, as

1 women, then the two curves should be identical or very close to each

2 other, or almost -- they should almost overlap, whereas here we have this

3 peak.

4 So how do you explain this peak in the number of killed men?

5 Isn't because in the 20- to 40-year range there are a lot of men

6 who are in uniform, who are military, and that's why we have so many men?

7 And these men could very well be military. If these men were civilians,

8 there the curve of -- there shouldn't be any discrepancies between the two

9 curves between men and women, between the two genders. Could you please

10 answer my question?

11 A. Right. Well, first of all we don't know exactly how many were

12 there, how many men, how many women in the population in East Mostar, but

13 generally what Your Honour said about -- if the numbers would be the same,

14 then the -- and -- then the age distribution of the victims should be

15 largely similar. That is a correct observation.

16 Well, it's important to remember that this particular chart is

17 showing both civilians and militaries jointly. There is no distinction

18 here. It is a combination of these two categories. These are all

19 victims, the minimum number, 539. Well, it is obvious that a majority of

20 victims are men at military age. Likely that many of them were military,

21 simply speaking. How many? Well, we only have the estimate. We cannot

22 say specifically by age how many because that is not possible in this

23 particular case.

24 Well, next thing I want to stress is that that in conflict

25 situation is a result of as well not only military status, civilian

1 military status, but also risk-taking behaviour. This is a very important

2 factor.

3 In populations in conflict, the weakest are usually saved from

4 being exposed to risk. Women and children are saved by others, by men,

5 who tend to take the risk, the higher risk behaviour on themselves. It

6 is, of course, combat on the first place, but on the second place there

7 are other behaviours, simple things; collecting food or water or wood for

8 fire, these kind of things.

9 So I believe that generally men are more exposed to risk in

10 conflict situation. This is another factor. Not all of them are

11 militaries. There will be always more men seen among the victims.

12 JUDGE ANTONETTI: [Interpretation] Yes, but this chart, you know,

13 on the 65 plus age range, this was a question that was asked by Defence,

14 but here what is really strange is the mortality rate is higher for the 65

15 plus age range, and a number of women killed, there were 23 or 24, 25

16 women over 65 killed, and in that age range also there are about 40 men

17 who were killed, and there both curves are almost identical except there's

18 more men dying than women. But in the 65 plus age range the curve picks

19 up. But you know women tend to live longer than men.

20 So we have this age range 65 plus. Isn't it -- does it mean that

21 when we see the age range 20 -- 20 to 50 age range. Couldn't that prove

22 that there were a lot of military in the 20 to 50 age range?

23 A. I wouldn't relate the peak for age group 65 plus with other peaks

24 in these curves. Well, first of all these are not rates, not relative

25 measures. These are just absolute numbers shown in this chart. As I

1 noted earlier today, the 65 plus is a larger age group, larger then any

2 other age group. That means it is to be expected that the number of

3 victims in this age group is higher than in the, for instance, preceding

4 age group, 60-64. I wouldn't pay too much attention to this because that

5 is -- in order to have a better understanding of how many died at age 65

6 plus, it's really necessary to redo this chart and show the data by

7 five-year age groups also after age 65.

8 So my first explanation is this is a larger age group than all

9 other age groups. So the biggest -- most likely not there, I'm pretty

10 sure it's not there. If we would have redo -- redone this chart and also

11 for this age group show five-year age intervals, there wouldn't be peak as

12 such. Although, generally speaking, of course, mortality increases with

13 age and is much higher at higher ages than at lower ages.

14 JUDGE ANTONETTI: [Interpretation] And one last question now.

15 Mr. Karnavas asked you some questions about this, but I'm not quite

16 satisfied with your answers.

17 On the title of your report, document 9835, people wounded in

18 relation to the siege of Mostar. The word "siege" that you use, is it the

19 senior officials at the Prosecution that asked you to do that or did you

20 come up with the title yourself? How did you come up with this title?

21 Was it the OTP who told you to study the victims of the "siege," spelling

22 this word out for you, or is it a demographer who said, "I will come up

23 with this title and call it a siege"?

24 So Mr. Karnavas asked you a few questions about that, but I would

25 like more information on this.

1 A. Right. The Prosecution didn't tell me what title to give to -- to

2 my reports. This is a title I gave to this reports.

3 Why the siege? I have done a lot of work on the siege of

4 Sarajevo, and I believe the idea of using the word "siege" comes from my

5 work on Sarajevo, on the siege of Sarajevo.

6 Well -- and in the case of Sarajevo work, I didn't use this term

7 in any other way than in the context of the common use of this term.

8 There is no special connotations that are behind using this term at all.

9 This is how this episode of the war was often called in the media and in

10 the reports of international observers, and this is how I ended using it,

11 and this is the same reason that I used that term for my work on Mostar.

12 I saw certain similarities between the two situations.

13 JUDGE ANTONETTI: [Interpretation] Fine. I saw Mr. Praljak getting

14 up.

15 Mr. Praljak, you have the floor.

16 THE ACCUSED PRALJAK: [Interpretation] Your Honour, I don't want to

17 take up too much time, but I have noted 18 concepts which have nothing to

18 do with demography: Water, electricity, Ustasha, Orlovac, siege, narrow

19 belt on the west side, cellar, old, generator.

20 JUDGE ANTONETTI: [Interpretation] Fine. Another follow-up

21 question. In the reports that you made on Sarajevo were they also called

22 siege of Sarajevo? I haven't read those reports on Sarajevo, but did you

23 use the same title, Siege of Sarajevo.

24 A. Yes. The word "siege" was used in the title.

25 JUDGE ANTONETTI: [Interpretation] Very well.

1 JUDGE TRECHSEL: I have also a few questions. The first relates

2 to the table, to this drawing we have seen a minute ago where we also have

3 a peak with the age group around 30 of men. Do you have inkling, any idea

4 of whether that could also at least partly, partly, be explained by a

5 higher proportion of men in that age group in the whole population at the

6 time?

7 A. I -- I don't think I have any data that could be used to answer

8 this question. Not in relation to the population in East Mostar.

9 JUDGE TRECHSEL: Thank you. Now, in your article you speak about

10 samples and about the double mentions with -- Mr. Karnavas, I think it

11 was, has interrogated you about that, and at one point the figure of 8.500

12 duplicated records is mentioned. Eight thousand five hundred related to

13 which total?

14 A. The entire population of Bosnia and Herzegovina, that is, 4.4

15 million individuals.

16 JUDGE TRECHSEL: Is that a figure which statistically has

17 significance?

18 A. I believe certainly not.

19 JUDGE TRECHSEL: Thank you.

20 JUDGE ANTONETTI: [Interpretation] Mr. Stringer.

21 MR. STRINGER: Thank you, Mr. President.

22 Re-examination by Mr. Stringer:

23 Q. Witness, just a few follow-up questions. You were asked a series

24 of questions by counsel for Dr. Prlic about census data, guidelines made

25 for how to incorporate data, procedures, deficiencies with the census

1 data, and just is -- is all of this dealt with in Annex B3 of your report

2 on the refugees and the internally displaced persons?

3 A. Yes, it is discussed at length in this annex.

4 Q. Okay. And in particular, I just want to direct your attention to

5 page 66 of that report. Sorry, 66 being the first page of Annex B3. And

6 here you actually refer in the footnote to Nora Selimovic, who is the

7 person I believe you talked about as having been the source of information

8 you received.

9 A. Yes, she is.

10 Q. Okay.

11 A. I want to note that I actually asked for two reports, one from

12 Nora Selimovic. In parallel, I also independently asked for another

13 report from another person who was also involved in the preparation and

14 conduct of the population census. I compared these two reports. I

15 received both, compared, and included in this report just one because the

16 consistency of these two reports was perfect.

17 Q. Okay. Now, just a question or two about the article that

18 Mr. Karnavas asked you about in the cross-examination. And again, just so

19 we're clear, if I could direct you -- in fact, it was the President who

20 directed you to page 199, the reference to military records of fallen

21 soldiers. Do you see that?

22 A. Yes.

23 Q. Okay. The references there military records BH government army,

24 the ABiH; and then VRS; and then thirdly HVO, are those in fact the

25 records that you did rely on in part in making this report?

1 A. Yes, I did.

2 Q. Okay. You were asked a few questions on cross-examination about

3 whether the census had been officially completed or -- or approved; do you

4 remember that?

5 A. Yes.

6 Q. And the census, just to sort of take one step back from it, we've

7 been talking about population data from the census; is that right?

8 A. Yes, that's right.

9 Q. Now, do you know whether this census or censuses in general,

10 whether this census was intended to assemble information or data about

11 other things apart strictly from census data?

12 A. Well, population was just one part of it. It was also a census on

13 households, on dwellings and agricultural farms.

14 Q. Now, as between those different groupings what level of completion

15 did the population data attain that you've based your report on?

16 A. Well, in terms of data checks, logical checks, the population

17 was -- had achieved, had achieved the highest level of attention and

18 success. That means this particular part of the census data has been

19 practically finished and publications were made in relation to

20 publications, several publications by authorities in Bosnia-Herzegovina

21 and also in Croatia.

22 Q. You were asked about whatever data or information may have been

23 available to you regarding pre-conflict patterns of voluntary migration

24 within Bosnia-Herzegovina. Do you recall that?

25 A. Yes, yes.

1 Q. And just referring to page 10 of your report. The very last

2 paragraph at the bottom of that page, page 10. Moving down to the fourth

3 line you say: "Pre-conflict internal migration in Bosnia and Herzegovina

4 was negligible," and then you talk about usual causes of internal

5 migration.

6 Could you just briefly comment on what one would expect in terms

7 of normal patterns and numbers as compared to the numbers that you arrive

8 at in your report?

9 A. Well, I mentioned earlier today that I tried to find data on

10 internal migration, in particular for Bosnia and Herzegovina, and I

11 contacted the statistical authority in Sarajevo and requested what kind of

12 data they had. They sent me a bulletin, statistical bulletin number 271

13 that I also had in my office, and I studied data from this bulletin. This

14 data comes from the 1991 population census and shows the internal

15 migration and external migration at the outbreak of the conflict. That

16 means in 1991. I have extracted the relevant data and made a table with

17 this data. I have the table with me. I also placed the data on internal

18 migration in the context of the census population in Herceg-Bosna

19 municipalities and calculated some rates. These rates I obtained confirm

20 what is in the report and confirm that the internal migration was

21 negligible, and at that time also the external migration from Herceg-Bosna

22 to other countries was also negligible practically.

23 Q. Okay.

24 A. I have this publication registered with the evidence unit, and

25 also the table I made and I can submit this material if --

1 Q. It's available if the Trial Chamber --

2 A. It's available, yeah.

3 Q. -- wants it. Witness, just one last question on that point then.

4 Can you comment briefly on your assessment or your observation in terms of

5 the scale, the level of migration pre-conflict as compared to the data

6 then that you've come up with based on the post-conflict migration?

7 MR. KARNAVAS: If I may just very briefly. I don't object to

8 hearing the answer. However, clearly, and I assume Mr. Stringer did not

9 have any contact with the witness and this is the witness's own doing on

10 her own, but certainly this information was not provided to us. And if

11 you're going to go -- if the Prosecution wishes to go into that area, one,

12 it is outside the scope in the sense that now we're going to specific data

13 that was looked at by the demographer on her own over the weekend and has

14 not been made available to the Defence.

15 While I do not object to the Prosecutor opening up his direct

16 examination in this area, it may require for us at some other point to

17 request for the witness to come back to -- to provide additional

18 questions -- provide additional answers in relation to this particular

19 topic because of the new information that's being -- that we are not --

20 was not made available to us by the -- by the expert. So, again, I don't

21 want to appear that they were trying to keep it out, but I think it's only

22 fair that if it becomes necessary for us to go to question further, I

23 certainly am not prepared today. I would like some time and I would give

24 the Court notice and the witness is in The Hague so it's not going to be a

25 problem.

1 JUDGE ANTONETTI: [Interpretation] If I understood you correctly,

2 Mr. Stringer, you want the witness to talk to you about migration after

3 the conflict, that is to say, after the Dayton agreements and after the

4 war. And Mr. Karnavas, if I'm not mistaken, deems that this is not

5 information that they had and therefore he would like to object to your

6 question.

7 Am I mistaken? I don't know if I'm mistaken, but, Mr. Stringer,

8 you wanted to put this question to the witness as to how did people move

9 after the conflict.

10 Q. [Previous translation continues]... No, not just that,

11 Mr. President. I --

12 JUDGE ANTONETTI: [Interpretation] Yes, not only that but that as

13 well, the migration patterns; right.

14 MR. STRINGER: Let me perhaps re-ask the question because I'm

15 trying to tie this into something that is found in the witness's report

16 and simply ask her to elaborate on that. And let me say that certainly

17 I've not had any contact whatsoever with the witness.

18 JUDGE ANTONETTI: [Interpretation] You may then rephrase your

19 question.

20 MR. STRINGER:

21 Q. [Previous translation continues] ... Again, just drawing your

22 attention, Witness, to this last paragraph on page 10. You refer to

23 minimal internal migration. Could you elaborate on that? What was the

24 basis for that statement? And then how -- what does that tell you about

25 the level of migration that you determined post-conflict?

1 MR. KARNAVAS: Mr. President, again I don't mean to be

2 obstreperous but now we have the witness looking at a piece -- at a chart

3 that was generated that is not being that is not part of the question.

4 And I know Mr. Stringer is an honourable man and didn't contact, but --

5 and Mr. Stringer is doing correct redirect by going back to the -- the

6 report which we all have, but if the witness is going to be looking at

7 charts she prepared that is not available to the Defence, I would object

8 to that. And Mr. Stringer has not asked her to look at that but she has

9 been looking at that. So I would ask the witness to kindly put aside

10 anything that she has that is not part of her report and just focus on

11 Mr. Stringer's questions.

12 JUDGE ANTONETTI: [Interpretation] Ms. Tabeau, if I'm not mistaken,

13 and usually I have a good memory, at some point, and I don't have the

14 transcript before me, but didn't you say that you and your staff studied

15 documents stemming from the relocation commission of the people who were

16 coming back and that you examined the charts of that commission? Is that

17 what you said earlier?

18 THE WITNESS: I wouldn't know exactly what the relocation

19 commission is. I did study some figures by UNHCR, but the whole

20 discussion now is about the issue of pre-war internal migration in Bosnia,

21 pre-war internal migration in Bosnia, in order to have some context for

22 the figures on internal migration, displaced persons, and refugees

23 produced in my report. And I have collected these figures, and I looked

24 at the figures, and I can only say that I fully stand behind my statement

25 from the report that before the conflict, before the conflict, as seen

1 from the data available in the 1991 census, the internal migration in this

2 particular region of Bosnia is negligible. And not only internal but also

3 external. And I can confirm this if Your Honours and everybody else

4 wishes by providing additional analysis, but I -- for now I can only say

5 this. It has all been confirmed.

6 JUDGE ANTONETTI: [Interpretation] Yes, I understand, but

7 Mr. Stringer wanted to talk about the migration patterns post-conflict.

8 Did you study that as well? In other words, how many people came back to

9 live in their former houses, domiciles? How many people came back to

10 their dwellings from a foreign country? Did you study that? It does

11 not -- it's not included in your report, but just as a curiosity point.

12 Did you study that particular post-migration pattern, the return of these

13 people back to their dwellings.

14 THE WITNESS: Well, on several occasions I looked at UNHCR

15 statistics on returns and on displaced persons and refugees as well.

16 Well, I don't think it is possible to obtain statistics, UNHCR statistics,

17 for such a small region as Herceg-Bosna that is available. We can speak

18 of these figures for the entire country, Bosnia and Herzegovina, but I

19 don't think we can have these statistics for comparative purposes for the

20 eight municipalities studied in this report.

21 JUDGE ANTONETTI: [Interpretation] Yes. Mr. Prlic wanted to say

22 something.

23 THE ACCUSED PRLIC: [Interpretation] I'd like to say something to

24 redirect, a question put by Mr. Stringer and also linked to

25 Ms. Nora Selimovic, and I found a piece of information there that I

1 consider could be of interest.

2 On page 76 of this report the last two lines read as follows, and

3 what is referred to is her data and her curriculum vitae. [In English] "In

4 October 1998 after the creation of the state agency for statistics in

5 Bosnia and Herzegovina, she commenced working in this institution. She is

6 still employed there at present."

7 [Interpretation] This sentence reminded me of the fact that I was

8 the person who adopted the law on the statistical office at the level of

9 Bosnia-Herzegovina, because at that time in the government of

10 Bosnia-Herzegovina there were just three ministries. So in the space of

11 time between 1992 to 1998, I don't know which institution existed, who

12 checked the data, and who published the data.

13 I also know that in the budget for Bosnia-Herzegovina, not for

14 1996 or 1997 did there exist an area of the existence of the statistical

15 institute at the level of Bosnia-Herzegovina, reference to that.

16 So I don't know who published that data and who verified that

17 data.

18 THE WITNESS: Can I answer this? Well, the republican statistical

19 office has always been in Sarajevo at exactly the same as before April

20 1992. This is the statistical authority who processed the census data and

21 who submitted to the Assembly of Bosnia and Herzegovina reports on the

22 conduct of the census. The reports were required by the census law to be

23 provided by the end of March for the year, for preceding year, and at

24 least three such reports were submitted. I have copies of this report.

25 If required, I can submit copies.

1 THE ACCUSED PRLIC: [Interpretation] May I just say that that does

2 not answer my question.

3 JUDGE ANTONETTI: [Interpretation] Madam, Mr. Prlic asked you a

4 question. Would you be able to answer that question?

5 THE WITNESS: Yes. This is the statistical office of the

6 Federation of Bosnia And Herzegovina that processed the census data. The

7 state agency is an umbrella that covers two statistical authorities; one

8 in the federation and one in Republika Srpska. They themselves didn't

9 have anything to do with the data from the census as such. So the federal

10 authority, the federal statistical authority was the one that processed

11 the census data and made publications.

12 THE ACCUSED PRLIC: [Interpretation] Of course I'm not satisfied

13 with the answer because part of the state which is half a state cannot

14 deal with information that relates to the state as whole and I think the

15 question steps outside the limits of this examination, but let's stay with

16 what has been recorded in the transcript.

17 MR. KARNAVAS: Well, and just for the sake of clarity,

18 Your Honour, the witness indicated federation and then she said federal.

19 Now, because there is the Federation of Bosnia-Herzegovina, I mean

20 sometimes lawyers and Judges have confused it when they hear federal, at

21 least in my context it's state. So if the witness could elaborate. Is

22 there -- how is she using this term?

23 THE WITNESS: I'm using this as the Federation of Bosnia and

24 Herzegovina.

25 JUDGE ANTONETTI: [Interpretation] Very well. Mr. Stringer.

1 MR. STRINGER: Thank you, Mr. President. Just for the record,

2 Ms. Winner has called up a part of the transcript from last week, and a

3 question from Judge Trechsel, which I think Mr. Karnavas referred to

4 earlier.

5 Judge Trechsel asked, said: "I wanted to ask you whether you

6 have, whether you have science -- whether your science has any figures on

7 what a normal migration is in given area and over a period of six, seven

8 years."

9 And then the witness said, "Not that I have these kind of figures

10 at hand now, but I believe the figures on normal migration in the region

11 of the former Yugoslavia do exist, and for reference purposes, I certainly

12 can look for those figures and I can provide this Chamber with those

13 figures for comparison."

14 And so it's my impression that the witness then went and found

15 those figures and has brought them back should the Trial Chamber wish to

16 see those today.

17 That's just for the record, and I have only one or two questions

18 left, Mr. President.

19 Q. The first is, Witness, whether you know if in fact the census

20 data, the census figures from Bosnia and Herzegovina, 1991, whether they

21 were published by any authorities outside of Bosnia and Herzegovina as

22 well.

23 A. Well, yes, certainly. The -- Croatia published the census data,

24 the ethnic composition by settlements. And Mr. Praljak was using this

25 publication during his cross-examination. This includes beautiful maps

1 for every municipality showing the composition in settlements.

2 Q. Okay. Now, the last issue, and I'm going to ask if the technical

3 people could put up the PowerPoint on the screens. I've just got one

4 slide that I'd like to ask the witness to comment on. It's one that we

5 skipped over in her direct in order to move more quickly.

6 But, Witness, today and I believe last week, during the course of

7 your testimony the word "extrapolation" was used, and I know that this

8 applies not just to this specific example but to others as well. This is

9 table 16 from your report on the wounded people in East Mostar.

10 And if you could with this table perhaps explain to us the manner

11 of extrapolation that you've referred to, specifically how the percentages

12 then were applied from smaller numbers to larger numbers.

13 A. Yes, of course. This is a table that contains information about

14 causes, distribution of victims by cause of wounding. In the first column

15 the overall total is 5.867 of which 2.000, approximately 2.500 are cases

16 with diagnosis available and explicitly reported in the hospital books,

17 and there is a number of unknown cases. This is the 2.844, 2.844.

18 Q. Okay, now let me just jump in there. When it says unknown, what

19 exactly are you talking about here when you say unknown?

20 A. The diagnosis is not reported.

21 Q. So in that book we looked at nothing's written. It doesn't say

22 vulnus explosiva or anything like that?

23 A. That's right. This is unavailable. This is unknown.

24 So --

25 Q. I'm sorry. Just to continue with that because some questions came

1 out suggesting that the doctor didn't know what was the nature of the

2 wound and I wanted to clarify between whether the doctor -- whether it was

3 unknown because the doctor didn't know or whether this was unknown to you

4 because it wasn't in the record?

5 MR. KARNAVAS: Your Honour, Just for the sake of the record, I

6 object. I think the Prosecutor misunderstood, but I don't think anyone

7 made that suggestion, and I think that in fairness to the Defence the

8 question should be withdrawn.

9 MR. STRINGER: Well, I'm just wanting to make a hundred per cent

10 sure. If I misunderstood, then I misunderstood.

11 MR. KARNAVAS: Ask a non-leading question and you'll get the

12 answer. We can move on, thank you.

13 MR. STRINGER: Thank you, Your Honour, for that ruling.

14 THE WITNESS: Okay. So there is a number of cases with unknown

15 diagnosis. Unknown means not reported in the books. And in this table we

16 present minimum numbers in the first -- in the first column in the table.

17 And there is a percentage distribution in the next column. In the

18 calculation of the percentages two categories have been skipped. This is

19 the 2.844 unknowns and 474 cases from missing pages or too incomplete to

20 include in the analysis.

21 Well, in the next column there is an estimated value of wounded

22 persons by the same categories of wounding. Below the table there is a

23 formula that explains how the calculation was made. The calculation

24 takes, as a beginning, the minimum numbers of reported cases. For

25 instance if we take shelling as an example, there were 2.088 cases of

1 victims -- wounded persons of whom we know these were victims of shelling.

2 So we take this as the first component of the estimate and as the second

3 component we take a percentage of the unknowns. The per cent that is

4 given in the second column of the table. 81.9 per cent are the cases of

5 shelling. So we simply take the 81.9 per cent out of 2.844 and added to

6 the minimum numbers observed in the source. And as a result of this

7 procedure we have the estimated value of victims or number of victims of

8 shelling which is 4.418. The percentage distribution remains the same.

9 In the last part of the table we give confidence intervals of this

10 procedure.

11 Q. So that the 81.9 per cent, which is known from the observed --

12 from those records that you could see, you simply applied that same

13 percentage then to the number of unknowns, which were 2.844?

14 A. Yes, this is what was done consequently in all these estimations.

15 Q. Thank you.

16 MR. STRINGER: Mr. President, I have no further questions.

17 JUDGE ANTONETTI: [Interpretation] Very well. Witness, we just

18 ended your testimony. Thank you very much for coming to testify at the

19 request of the OTP. Thank you for testifying, therefore, and thank you

20 for bringing your comments to the reports that you've drafted.

21 No, no.

22 So I would like to ask the usher to escort you out of this

23 courtroom.

24 THE WITNESS: Thank you very much.

25 [The witness withdrew]

1 JUDGE ANTONETTI: [Interpretation] Mr. Stringer, for the rest of

2 the week we don't have any other witnesses because it was not possible to

3 have a witness, and therefore there won't be a hearing tomorrow, no

4 hearings on Wednesday, nor on Thursday.

5 Now, next week we have a witness. I'm not going to give the

6 witness's name because I don't know if it is a protected witness or not,

7 but it's a witness who is scheduled for four days, Monday through

8 Thursday. The legal officer will send you the information with regard to

9 the time, the division of time, because I can't tell you that now since

10 the witness is here for the first -- for four days, the first day will be

11 dedicated to the Prosecution, and the day after as well, and the Defence

12 will probably have two or two days and a half to cross-examine this

13 witness.

14 This is what I wanted to tell you. And if there are no further

15 questions, I will therefore adjourn this hearing and we will see each

16 other next week.

17 --- Whereupon the hearing adjourned at 6.41 p.m.,

18 to be reconvened on Monday, the 10th day

19 of September, 2007, at 2.15 p.m.

20

21

22

23

24

25