PARTLY DISSENTING OPINION OF JUDGE BRATZA

1. While sharing the view of the majority of the Chamber that there was in the present case a violation of Article 2 of the Convention in its procedural aspect, I would have gone further and found that there was additionally a violation of the substantive provisions of that Article.

2. The case is materially different from a number of cases against Turkey which have been examined by both the former Commission and the Court, involving the killing of individuals by unknown perpetrators (see, for example, Yaşa v. Turkey judgment of 2 September 1998, Reports 1998-VI, Tanrıkulu v. Turkey [GC], no.23763/94 judgment of 8 July 1999, Reports 1999-IV; Mahmut Kaya v. Turkey , no.22535/93, judgment of 28 March 2000, Reports 2000-III; Kılıç v. Turkey , no.22492/93, judgment of 28 March 2000, Reports 2000-III; Akkoç v. Turkey , nos. 22947/93 and 22948/93, judgment of 10 October 2000, Reports 2000-X). In each of these cases, it was found that it had not been established to the requisite standard of proof, that is “beyond reasonable doubt”, that police officers, members of the security forces or other agents of the State had been responsible for the deprivation of life of which complaint was made. By contrast, in the present case it is undisputed that police officers shot and killed the applicant’s brother, İrfan Ağdaş. In these circumstances, so far from the burden of proof resting on the applicant, it seems to me that it must in principle be for the respondent State to establish on the evidence before the Court that the deprivation of life resulted from the use of force which was no more than “absolutely necessary” for one or more of the legitimate purposes set out in paragraph 2 of Article 2 – in this case, the “defence of any person from unlawful violence”. This term indicates that a stricter and more compelling test of necessity must be employed than that applicable when determining whether State action is “necessary in a democratic society” under paragraph 2 of Articles 8 to 11 of the Convention: the force used must be shown to be strictly proportionate to the achievement of the permitted aims (see McCann and Others v. the United Kingdom judgment of 27 September 1985, Series A no. 324, p. 46 §§ 148-149; Gül v. Turkey , no.22676/93, judgment of 14 December 2000, § 77).

3. In the present case, the death of İrfan Ağdaş was the subject of criminal proceedings for intentional homicide instituted against the three police officers concerned in November 1996. In April 2001 (some five years after the death and four and a half years after the filing of the indictment) the Eyüp Assize Court acquitted the police officers, finding them to have acted in self-defence when İrfan Ağdaş opened fire on them. The decision of the Assize Court was upheld on appeal by the Court of Cassation in July 2002.

4. I accept that in normal circumstances it would require cogent elements to lead the Court to depart from the reasoned findings of fact reached by national judicial authorities, particularly where, as in the present case, the Court has not itself had the benefit of seeing and examining the relevant witnesses and forming its own assessment of their credibility. However, as is pointed out in the judgment (§ 91) the central importance of the protection afforded under Article 2 is such that the Court is required to subject deprivations of life to the most careful scrutiny even where domestic proceedings and investigations have already taken place. Moreover, in view of the different burden of proof in criminal proceedings as well as the different standards applied in assessing criminal responsibility, the fact that such proceedings resulted in the acquittal of the police officers can in no sense be regarded as decisive of the issue which arises under the Convention, namely whether the use of force has been shown to be absolutely necessary.

5. In paragraph 94 of the judgment, the Court has noted its serious doubts as to how the shooting took place, these doubts stemming largely from the manner in which the investigation at the scene of the incident, the post-mortem examination and the investigation by the criminal court had been conducted. I not only share these doubts but consider that the serious deficiencies in the investigation and in the resulting proceedings were such that it has not been shown that the use of force was no more than absolutely necessary or that it was strictly proportionate to any of the aims permitted under Article 2. I would, in particular, highlight the following troubling features of the case.

a) As noted above, the criminal proceedings against the police officers lasted more than 5 years. The accused police officers were summoned for the first time to give their statements before the Assize Court in May 1997, one year after the killing had taken place. One of the accused appeared before the court in September 1997, four months after the summons had been issued and after five hearings had already taken place; the other two officers appeared before the court to give their statements only in March 1998, nearly a year after the summons had been issued. On the same date, the Assize Court summoned the police officers who had collected the empty cartridges to give their statements before the court; in the result, only one of the police officers from the squad appeared before the court in February 1999, eleven months after the summons had been issued.

b) It appears that the 9 mm Browning gun allegedly used by the applicant’s brother and retrieved by the police officers was never subjected to any fingerprint analysis in order to establish whether it had ever been handled by him. This is the more remarkable having regard to the fact that no nitrate or nitrite ion traces were found in the chemical analysis of the skin samples taken from İrfan’s hands, even though the first ballistic report concluded that seven bullets had been discharged from the gun (see § 25). I note that in its judgment the Eyüp Assize Court does not appear to have remarked on either of these points.

c) Although two ballistic examinations were carried out, there appear to have been significant contradictions between the results of the examinations. In the report of the examination carried out by the Criminal Police Laboratory of Istanbul on 14 May 1996, it is stated that of the seventeen empty 9 mm cartridges submitted for examination from the scene of the incident, seven of the bullets had been fired from the Browning gun and ten from the weapons used by the police officers (judgment, § 25). However, in response to the applicant’s complaint that the empty cartridges kept in the Eyüp Security Directorate were not the same as those mentioned in the ballistics report, the Assize Court in April 1999 ordered that the cartridges be produced. The cartridges were not in fact produced until July 2000, some fifteen months later, the Security Directorate having claimed in September 1999 that the cartridges had been lost during renovations in the police station. On the order of the Assize Court , a further ballistic examination of the three weapons belonging to the accused police officers was conducted on 7 August 2000 by the same Criminal Police Laboratory. In its report of 23 October 2000 the Laboratory concluded that only four of the seventeen empty cartridges matched one of the weapons. Of the remaining cartridges, six came from a different weapon and seven from yet another weapon, neither of which weapons had been submitted for ballistic examination (judgment, § 75). This apparent inconsistency does not appear to have been commented on in the Assize Court ’s judgment.

d) It was the accused officers’ case that they had approached four suspects in order to carry out an identity check and body search, that the suspects had all run away and that during their flight İrfan had opened fire at the police officers. This account would appear difficult to reconcile with the contemporary radio communications in which there is no mention of any person other than İrfan ( “There was a person carrying a plastic bag. We opened fire at him, we captured the wounded person... we shot the man.” ). In the same communication it is also far from clear that the police officers were responding to the opening of fire on them. As is noted in the judgment of the Assize Court , the police headquarters reported that the TEM team had captured a person “in a mixed 502 situation”, that is in a situation of armed conflict with the police. The Assize Court explained this by saying that it was only in the heat of the moment that the police officers had informed the headquarters that such a situation of conflict had occurred. But no explanation is offered for the fact that in the subsequent exchange the headquarters reported that the TEM team had opened fire “without a 502 situation”.

e) It appears that at least two of the three persons alleged to have been with İrfan at the time of the incident were not only known to the authorities but had been taken into police custody at a later date. However, it does not appear that they were ever questioned about the alleged incident. When for the first time in February 2000 the Assize Court requested information about the identity and addresses of the three individuals, the Security Directorate reported that two of the individuals who were “terrorists” had been killed in an operation conducted on 20 August 1996 in Eyüp (judgment, §§ 70-72).

f) Despite B.M.’s evidence that, immediately after the shooting, a crowd of people descended on the officers with the result that they had quickly to put İrfan in the car, the only independent eye-witness called to give evidence was A.B. There is nothing to suggest that any attempt was at any stage made by the authorities to trace and interview other eye-witnesses of the incident. A.B. gave evidence which contradicted the account of the accused police officers that İrfan had opened fire on them. The Assize Court’s conclusion that she had not in fact witnessed the incident and that her evidence was not sincere was based primarily on the fact that in her statement before the court she had placed the incident some two hours before it had in fact occurred and on the fact that she saw only one bullet wound in İrfan’s body and thus appeared to have assumed that the other bullet found to have struck İrfan must have been fired in the car – a fact which was disproved by the autopsy report. While accepting that the Assize Court was better placed to assess AB’s credibility, I do not find either of the grounds relied on by that court as affording convincing reasons for rejecting her evidence in its entirety. In this regard I note that in her statement to the applicant’s representative made a few months after the event, A.B. had given the time of the incident as “around 7 p.m.” when “it was starting to get dark”, an estimate which was closer to the actual time than that made nearly a year later. I also note that it appears that A.B. was not the only person who initially thought that İrfan had been struck by a single bullet: in his evidence to the Assize Court, A.Y., one of the accused police officers, stated that İrfan had told him that he had been shot in the left arm and that “it did not seem to be a serious wound”.

6. Notwithstanding their doubts about the circumstances in which İrfan lost his life, the majority of the Court find that there is an insufficient factual and evidentiary basis on which to conclude that he was deprived of his life as a result of the use of force which was more than absolutely necessary (judgment, § 96). But this to my mind is to apply the wrong test and to reverse the burden of proof. As noted above, the test to be applied is not whether there is a sufficient evidence to satisfy the Court that the use of force was more than absolutely necessary; rather, it is whether the evidence is such as to satisfy the Court that the use of force was no more than absolutely necessary in self-defence. On the basis of the material before the Court, I am not so satisfied and consequently do not find that the killing of İrfan was justified in terms of Article 2 of the Convention.