JOINT DISSENTING OPINION OF JUDGES RYSSDAL, BERNHARDT, THÓR VILHJÁLMSSON, GÖLCÜKLÜ, PALM, PEKKANEN, SIR JOHN FREELAND, BAKA AND JAMBREK

1. We are unable to subscribe to the opinion of a majority of our colleagues that there has been a violation of Article 2 (art. 2) of the Convention in this case.

2. We will take the main issues in the order in which they are dealt with in the judgment.

3. As to the section which deals with the in terpretation of Article 2 (art. 2), we agree with the conclusion in paragraph 155 that the difference between the Convention standard and the national standard as regards justification for the use of force resulting in deprivation of life is not such that a violation of Article 2 para. 1 (art. 2-1) could be found on that ground alone. We also agree with the conclusion in paragraph 164 that there has been no breach of Article 2 para. 1 (art. 2-1) on the ground of any shortcoming in the examination at national level of the circumstances surrounding the deaths.

4. As to the section dealing with the application of Article 2 (art. 2) to the facts of the case, we fully concur in rejecting as unsubstantiated the applicants ’ allegations that the killing of the three suspects was premeditated or the product of a tacit agreement among those involved in the operation (paragraph 184).

5. We also agree with the conclusion in paragraph 200 that the actions of the four soldiers who carried out the shootings do not, in themselves, give rise to a violation of Article 2 (art. 2). It is rightly accepted that those soldiers honestly believed, in the light of the information which they had been given, that it was necessary to act as they did in order to prevent the suspects from detonating a bomb and causing serious loss of life: the actions which they took were thus perceived by them as absolutely necessary in order to safeguard innocent lives.

6. We disagree, however, with the evaluation made by the majority (paragraphs 202-14) of the way in which the control and organisation of the operation were carried out by the authorities. It is that evaluation which, crucially, leads to the finding of violation.

7. We recall at the outset that the events in this case were examined at the domestic level by an inquest held in Gibraltar over a period of nineteen days between 6 and 30 September 1988. The jury, after hearing the evidence of seventy-nine witnesses (including the soldiers, police officers and surveillance personnel involved in the operation and also pathologists, forensic scientists and experts on the detonation of explosive devices), and after being addressed by the Coroner in respect of the applicable domestic law, reached by a majority of nine to two a verdict of lawful killing. The circumstances were subsequently investigated in depth and evaluated by the Commission, which found in its report, by a majority of eleven to six, that there had been no violation of the Convention.

The finding of the inquest, as a domestic tribunal operating under the relevant domestic law, is not of itself determinative of the Convention issues before the Court. But, having regard to the crucial importance in this case of a proper appreciation of the facts and to the advantage undeniably enjoyed by the jury in having observed the demeanour of the witnesses when giving their evidence under examination and cross-examination, its significance should certainly not be underestimated. Similarly, the Commission ’ s establishment and evaluation of the facts is not conclusive for the Court; but it would be mistaken for the Court, at yet one further remove from the evidence as given by the witnesses, to fail to give due weight to the report of the Commission, the body which is primarily charged under the Convention with the finding of facts and which has, of course, great experience in the discharge of that task.

8. Before turning to the various aspects of the operation which are criticised in the judgment, we would underline three points of a general nature.

First, in undertaking any evaluation of the way in which the operation was organised and controlled, the Court should studiously resist the temptations offered by the benefit of hindsight. The authorities had at the time to plan and make decisions on the basis of incomplete information. Only the suspects knew at all precisely what they intended; and it was part of their purpose, as it had no doubt been part of their training, to ensure that as little as possible of their intentions was revealed. It would be wrong to conclude in retrospect that a particular course would, as things later transpired, have been better than one adopted at the time under the pressures of an ongoing anti-terrorist operation and that the latter course must therefore be regarded as culpably mistaken. It should not be so regarded unless it is established that in the circumstances as they were known at the time another course should have been preferred.

9. Secondly, the need for the authorities to act within the constraints of the law, while the suspects were operating in a state of mind in which members of the security forces were regarded as legitimate targets and incidental death or injury to civilians as of little consequence, would inevitably give the suspects a tactical advantage which should not be allowed to prevail. The consequences of the explosion of a large bomb in the centre of Gibraltar might well be so devastating that the authorities could not responsibly risk giving the suspects the opportunity to set in train the detonation of such a bomb. Of course the obligation of the United Kingdom under Article 2 para. 1 (art. 2-1) of the Convention extended to the lives of the suspects as well as to the lives of all the many others, civilian and military, who were present in Gibraltar at the time. But, quite unlike those others, the purpose of the presence of the suspects in Gibraltar was the furtherance of a criminal enterprise which could be expected to have resulted in the loss of many innocent lives if it had been successful. They had chosen to place themselves in a situation where there was a grave danger that an irreconcilable conflict between the two duties might arise.

10. Thirdly, the Court ’ s evaluation of the conduct of the authorities should throughout take full account of (a) the information which had been received earlier about IRA intentions to mount a major terrorist attack in Gibraltar by an active service unit of three individuals; and (b) the discovery which (according to evidence given to the inquest by Witness O) had been made in Brussels on 21 January 1988 of a car containing a large amount of Semtex explosive and four detonators, with a radio-controlled system - equipment which, taken together, constituted a device familiar in Northern Ireland.

In the light of (a), the decision that members of the SAS should be sent to take part in the operation in response to the request of the Gibraltar Commissioner of Police for military assistance was wholly justifiable. Troops trained in a counter-terrorist role and to operate successfully in small groups would clearly be a suitable choice to meet the threat of an IRA active service unit at large in a densely populated area such as Gibraltar , where there would be an imperative need to limit as far as possible the risk of accidental harm to passers-by.

The detailed operational briefing on 5 March 1988 (paragraphs 22-31) shows the reasonableness, in the circumstances as known at the time, of the assessments then made. The operational order of the Gibraltar Commissioner of Police, which was drawn up on the same day, expressly proscribed the use of more force than necessary and required any recourse to firearms to be had with care for the safety of persons in the vicinity. It described the intention of the operation as being to protect life; to foil the attempt; to arrest the offenders; and the securing and safe custody of the prisoners (paragraphs 17 and 18).

All of this is indicative of appropriate care on the part of the authorities. So, too, is the cautious approach to the eventual passing of control to the military on 6 March 1988 (paragraphs 54-58).

11. As regards the particular criticisms of the conduct of the operation which are made in the judgment, foremost among them is the questioning (in paragraphs 203-05) of the decision not to prevent the three suspects from entering Gibraltar . It is pointed out in paragraph 203 that, with the advance information which the authorities possessed and with the resources of personnel at their disposal, it would have been possible for them "to have mounted an arrest operation" at the border.

The judgment does not, however, go on to say that it would have been practicable for the authorities to have arrested and detained the suspects at that stage. Rightly so, in our view, because at that stage there might not be sufficient evidence to warrant their detention and trial. To release them, after having alerted them to the state of readiness of the authorities, would be to increase the risk that they or other IRA members could successfully mount a renewed terrorist attack on Gibraltar . In the circumstances as then known, it was accordingly not "a serious miscalculation" for the authorities to defer the arrest rather than merely stop the suspects at the border and turn them back into Spain .

12. Paragraph 206 of the judgment then lists certain "key assessments" made by the authorities which, in paragraph 207, are said to have turned out, in the event, to be erroneous, although they are accepted as all being possible hypotheses in a situation where the true facts were unknown and where the authorities were operating on the basis of limited intelligence information. Paragraph 208 goes on to make the criticism that "insufficient allowances appear to have been made for other assumptions".

13. As a first example to substantiate this criticism, the paragraph then states that since the bombing was not expected until 8 March "there was equally the possibility that the ... terrorists were on a reconnaissance mission".

There was, however, nothing unreasonable in the assessment at the operational briefing on 5 March that the car which would be brought into Gibraltar was unlikely, on the grounds then stated, to be a "blocking" car (see paragraph 23, point e). So, when the car had been parked in the assembly area by one of the suspects and all three had been found to be present in Gibraltar, the authorities could quite properly operate on the working assumption that it contained a bomb and that, as the suspects were unlikely to risk two visits, it was not "equally" possible that they were on a reconnaissance mission.

In addition, Soldier F, the senior military adviser to the Gibraltar Commissioner of Police, gave evidence to the inquest that, according to intelligence information, reconnaissance missions had been undertaken many times before: reconnaissance was, he had been told, complete and the operation was ready to be run. In these circumstances, for the authorities to have proceeded otherwise than on the basis of a worst-case scenario that the car contained a bomb which was capable of being detonated by the suspects during their presence in the territory would have been to show a reckless failure of concern for public safety.

14. Secondly, it is suggested in the second sub-paragraph of paragraph 208 that, at the briefings or after the suspects had been spotted, "it might have been thought unlikely that they would have been prepared to explode the bomb, thereby killing many civilians, as Mr McCann and Ms Farrell strolled towards the border area since this would have increased the risk of detection and capture".

Surely, however, the question is rather whether the authorities could safely have operated on the assumption that the suspects would be unlikely to be prepared to explode the bomb when, even if for the time being moving in the direction of the border, they became aware that they had been detected and were faced with the prospect of arrest. In our view, the answer is clear: certainly, previous experience of IRA activities would have afforded no reliable basis for concluding that the killing of many civilians would itself be a sufficient deterrent or that the suspects, when confronted, would have preferred no explosion at all to an explosion causing civilian casualties. It is relevant that, according to Soldier F ’ s evidence at the inquest, part of the intelligence background was that he had been told that the IRA were under pressure to produce a "spectacular". He also gave evidence of his belief that, when cornered, the suspects would have no qualms about pressing the button to achieve some degree of propaganda success: they would try to derive such a success out of having got a bomb into Gibraltar and that would outweigh in their minds the propaganda loss arising from civilian casualties.

15. The second sub-paragraph of paragraph 208 goes on to suggest that it "might also have been thought improbable that at that point" - that is, apparently, as McCann and Farrell "strolled towards the border" - "[the suspects] would have set up the transmitter in anticipation to enable them to detonate the supposed bomb immediately if confronted".

Here, the question ought, we consider, to be whether the authorities could prudently have proceeded otherwise than on the footing that there was at the very least a possibility that, if not before the suspects became aware of detection then immediately afterwards, the transmitter would be in a state of readiness to detonate the bomb.

16. It is next suggested, in the third sub-paragraph of paragraph 208, that "even if allowances are made for the technological skills of the IRA, the description of the detonation device as a `button job ’ without the qualifications subsequently described by the experts at the inquest ..., of which the competent authorities must have been aware, over-simplifies the true nature of these devices". The exact purport of this criticism is perhaps open to some doubt. What is fully clear, however, is that, as the applicants ’ own expert witness accepted at the inquest, a transmitter of the kind which was thought likely to be used in the present case could be set up so as to enable detonation to be caused by pressing a single button; and in the light of past experience it would have been most unwise to discount the possibility of technological advance in this field by the IRA.

17. Paragraph 209 of the judgment expresses disquiet that the assessment made by Soldier G that there was a "suspect car bomb" was conveyed to the soldiers on the ground in such a way as to give them the impression that the presence of a bomb had been definitely identified. But, given the assessments which had been made of the likelihood of a remote control being used, and given the various indicators that the car should indeed be suspected of containing a bomb, the actions which the soldiers must be expected to have taken would be the same whether their understanding of the message was as it apparently was or whether it was in the sense which Soldier G apparently intended. In either case, the existence of the risk to the people of Gibraltar would have been enough, given the nature of that risk, justifiably to prompt the response which followed.

18. Paragraph 209, in referring to the assessment made by Soldier G, also recalls that while he had experience with car bombs, he was not an expert in radio communications or explosives. In considering that assessment, it would, however, be fair to add that, although his inspection of the car was of brief duration, it was enough to enable him to conclude, particularly in view of the unusual appearance of its aerial in relation to the age of the car and the knowledge that the IRA had in the past used cars with aerials specially fitted, that it was to be regarded as a suspect car bomb.

The authorities were, in any event, not acting solely on the basis of Soldier G ’ s assessment. There had also been the earlier assessment, to which we have referred in paragraph 13 above, that a "blocking" car was unlikely to be used. In addition, the car had been seen to be parked by Savage, who was known to be an expert bomb-maker and who had taken some time (two to three minutes, according to one witness) to get out of the car, after fiddling with something between the seats.

19. Paragraph 210 of the judgment asserts, in effect, that the use of lethal force was made "almost unavoidable" by the conveyance to Soldiers A, B, C and D of a series of working hypotheses which were vitiated by the absence of sufficient allowances for alternative possibilities and by "the definite reporting ... of a car bomb which ..., could be detonated at the press of a button".

We have dealt in paragraphs 13-16 with the points advanced in support of the conclusion that insufficient allowance was made for alternative possibilities; and in paragraphs 17 and 18 with the question of reporting as to the presence of a car bomb.

We further question the conclusion that the use of lethal force was made "almost unavoidable" by failings of the authorities in these respects. Quite apart from any other consideration, this conclusion takes insufficient account of the part played by chance in the eventual outcome. Had it not been for the movements which were made by McCann and Farrell as Soldiers A and B closed on them and which may have been prompted by the completely coincidental sounding of a police car siren, there is every possibility that they would have been seized and arrested without a shot being fired; and had it not been for Savage ’ s actions as Soldiers C and D closed on him, which may have been prompted by the sound of gunfire from the McCann and Farrell incident, there is every possibility that he, too, would have been seized and arrested without resort to shooting.

20. The implication at the end of paragraph 211 that the authorities did not exercise sufficient care in evaluating the information at their disposal before transmitting it to soldiers "whose use of firearms automatically involved shooting to kill" appears to be based on no more than "the failure to make provision for a margin of error" to which the beginning of the paragraph refers. We have dealt already with the "insufficient allowances for alternative possibilities" point (see, again, paragraphs 13-16 above), which we take to be the same as the alleged failure to provide for a margin of error which is referred to here. Any assessment of the evaluation by the authorities of the information at their disposal should, in any event, take due account of their need to reckon throughout with the incompleteness of that information (see paragraph 8 above); and there are no cogent grounds for any suggestion that there was information which they ought reasonably to have known but did not.

21. Paragraph 212, after making a glancing reference to the restrictive effect of the public interest certificates and saying that it is not clear "whether the use of firearms to wound their targets may have been warranted by the specific circumstances that confronted them at the moment of arrest", goes on to say that "their reflex action in this vital respect lacks the degree of caution ... to be expected from law-enforcement personnel in a democratic society, even when dealing with dangerous terrorist suspects, and stands in marked contrast to the standard of care reflected in the instructions in the use of firearms by the police". It concludes with the assertion that this "failure by the authorities also suggests a lack of appropriate care in the control and organisation of the arrest operation".

22. As regards any suggestion that, if an assessment on the issue had been required by their training or instruction to be carried out by the soldiers, shooting to wound might have been considered by them to have been warranted by the circumstances at the time, it must be recalled that those circumstances included a genuine belief on their part that the suspects might be about to detonate a bomb by pressing a button. In that situation, to shoot merely to wound would have been a highly dangerous course: wounding alone might well not have immobilised a suspect and might have left him or her capable of pressing a button if determined to do so.

23. More generally as regards the training given, there was in fact ample evidence at the inquest to the effect that soldiers (and not only these soldiers) would be trained to respond to a threat such as that which was thought to be posed by the suspects in this case - all of them dangerous terrorists who were believed to be putting many lives at immediate risk - by opening fire once it was clear that the suspect was not desisting; that the intent of the firing would be to immobilise; and that the way to achieve that was to shoot to kill. There was also evidence at the inquest that soldiers would not be accepted for the SAS unless they displayed discretion and thoughtfulness; that they would not go ahead and shoot without thought, nor did they; but they did have to react very fast. In addition, evidence was given that SAS members had in fact been successful in the past in arresting terrorists in the great majority of cases.

24. We are far from persuaded that the Court has any sufficient basis for concluding, in the face of the evidence at the inquest and the extent of experience in dealing with terrorist activities which the relevant training reflects, that some different and preferable form of training should have been given and that the action of the soldiers in this case "lacks the degree of caution in the use of firearms to be expected of law-enforcement personnel in a democratic society". (We also question, in the light of the evidence, the fairness of the reference to "reflex action in this vital respect" - underlining supplied. To be trained to react rapidly and to do so, when the needs of the situation require, is not to take reflex action.)

Nor do we accept that the differences between the guide to police officers in the use of firearms (paragraph 137 of the judgment) and the "Firearms - rules of engagement" annexed to the Commissioner ’ s operational order (paragraph 136), when the latter are taken together (as they should be) with the Rules of Engagement issued to Soldier F by the Ministry of Defence (paragraph 16), can validly be invoked to support a contention that the standard of care enjoined upon the soldiers was inadequate. Those differences are no doubt attributable to the differences in backgrounds and requirements of the recipients to whom they were addressed, account being taken of relevant training previously given to each group (it is to be noted that, according to the evidence of Soldier F at the inquest, many lectures are given to SAS soldiers on the concepts of the rule of law and the use of minimum force). We fail to see how the instructions for the soldiers could themselves be read as showing a lack of proper caution in the use of firearms.

Accordingly, we consider the concluding stricture, that there was some failure by the authorities in this regard suggesting a lack of appropriate care in the control and organisation of the arrest operation, to be unjustified.

25. The accusation of a breach by a State of its obligation under Article 2 (art. 2) of the Convention to protect the right to life is of the utmost seriousness. For the reasons given above, the evaluation in paragraphs 203 to 213 of the judgment seems to us to fall well short of substantiating the finding that there has been a breach of the Article (art. 2) in this case. We would ourselves follow the reasoning and conclusion of the Commission in its comprehensive, painstaking and notably realistic report. Like the Commission, we are satisfied that no failings have been shown in the organisation and control of the operation by the authorities which could justify a conclusion that force was used against the suspects disproportionately to the purpose of defending innocent persons from unlawful violence. We consider that the use of lethal force in this case, however regrettable the need to resort to such force may be, did not exceed what was, in the circumstances as known at the time, "absolutely necessary" for that purpose and did not amount to a breach by the United Kingdom of its obligations under the Convention.

[1] The case is numbered 17/1994/464/545.  The first number is the case's position on the list of cases referred to the Court in the relevant year (second number).  The last two numbers indicate the case's position on the list of cases referred to the Court since its creation and on the list of the corresponding originating applications to the Commission.

[2] Rules A apply to all cases referred to the Court before the entry into force of Protocol No. 9 (P9) and thereafter only to cases concerning States not bound by that Protocol (P9).  They correspond to the Rules that came into force on 1 January 1983, as amended several times subsequently.

[3] Note by the Registrar: for practical reasons this annex will appear only with the printed version of the judgment (volume 324 of Series A of the Publications of the Court), but a copy of the Commission's report is available from the registry.