Chapter 7 - Predetermined and offensive use of lethal force

< Return to Contents | Previous | Next >

7

Predetermined and offensive use of lethal force

Chapter 7

1. As we noted in chapter 3, once Task Force 81 (TF81) had identified those responsible for Lieutenant Tim O’Donnell’s death in the 3 August 2010 attack, successful applications were made to the International Security Assistance Force (ISAF) to have several of the leaders placed on the Joint Prioritised Effects List (JPEL). As we also discussed in chapter 3, Operation Burnham was described as a “deliberate detention” or “kill/capture” operation that was directed at capturing two of those on the JPEL, Abdullah Kalta and Maulawi Neimatullah.
2. In this chapter we address aspects of the Inquiry’s Terms of Reference relating to the predetermined and offensive use of lethal force—often referred to as targeted killings. The Terms of Reference direct us to consider the following questions:1
   1. Did the rules of engagement permit the predetermined and offensive use of lethal force? If so, was this understood by the New Zealand Defence Force (NZDF) and responsible ministers?
   2. Did NZDF’s interpretation and application of the rules of engagement in this respect change over time?
3. As we understand it, these issues arise from the allegations in Hit & Run that in the two years following Operation Burnham, Abdullah Kalta and two other targets who TF81 believed to have been involved in the 3 August 2010 attack were deliberately killed in operations involving TF81.2 As we explained in chapter 1, only the operation in respect of Abdullah Kalta requires consideration. NZDF was not involved in the other two operations identified. In this chapter we consider whether TF81’s rules of engagement permitted targeted killings and whether ministers understood that, and address the death of Abdullah Kalta in an ISAF operation in 2012.

The Joint Prioritised Effects List

4. We begin by explaining the role of the JPEL and its relevance to this aspect of our Terms of Reference, given that they do not mention it. We refer to the JPEL because it was, in effect, the mechanism through which the predetermined and offensive use of lethal force (as well as other forms of targeting, such as detention operations) was coordinated by ISAF. While individual nations’ rules of engagement might permit targeted killings in certain circumstances, in practice a JPEL listing was required to obtain approval for such an operation and the necessary resources from ISAF. This allowed ISAF to ensure its resources were used most effectively, against legitimate targets of high priority.
5. The JPEL was, as the name suggests, “a list by which the joint coalition forces [in Afghanistan] recorded and prioritised various effects that were sought in relation to specific targets, mostly individuals”.3 The effects identified for a particular individual could range from ongoing surveillance through to capture or kill.
6. Brigadier Lisa Ferris, Director of Defence Legal Services, summarised the process by which the JPEL was composed in her presentation to the Inquiry.4 Broadly speaking, the process involved three stages:
   1. Individual ISAF member states, including New Zealand, identified people for potential inclusion on the JPEL. This involved a targeting process through which suspected insurgents, and the desired “effect” to be achieved against them, were identified. As part of that process, targets were vetted from an intelligence perspective and validated to ensure compliance with International Humanitarian Law and rules of engagement. NZDF legal officers were involved in that process. Intelligence reporting relied on had to meet standards of reliability and credibility prescribed by ISAF.
   2. Individual member states then nominated the identified targets for inclusion on the JPEL.5 Nominations were considered within ISAF. Decisions on inclusion and prioritisation were made through a coordination process administered by boards, which included subject matter experts at the operational level of the coalition, such as intelligence and operational staff and legal advisers.
   3. If a nomination was approved, the individual was placed on the JPEL as a target and assigned a priority. The JPEL was regularly reviewed and updated reporting was required to justify the person’s continued inclusion on the list.6
7. The result of an individual being listed on the JPEL as a lethal target was that they were treated as directly participating in hostilities for the duration of the listing and could be attacked by ISAF forces.7 Before such an attack could be launched, however, the target needed to be positively identified as the individual on the JPEL (using visual or certain other specified means) to a level of “reasonable certainty”.8 Positive identification had to be maintained until the time of the engagement. ISAF approval still generally needed to be obtained before such an attack, and any participation by New Zealand forces had to comply with the New Zealand rules of engagement.9 If a lethal JPEL target was captured or indicated a clear wish to surrender, he or she could no longer be killed.10 A lethal JPEL listing also did not permit the killing of individuals who were with the target.11 (Where other individuals were near the target, that would need to be assessed in accordance with the principles of distinction, proportionality and precaution as discussed in chapter 6.)
8. The JPEL was therefore a mechanism through which ISAF task forces could draw on the resources of ISAF as a whole to identify individuals who were members of the insurgency and to carry out operations against them. The Inquiry accepts NZDF’s point that the JPEL was not a “kill list”12 because not all individuals placed on the JPEL were identified as targets to be killed. Capture was generally the intended and desired outcome, even in what were described as “kill/capture” operations. Operations against JPEL targets were conducted in accordance with the restrictions in both the ISAF and the relevant member force’s rules of engagement—including any restrictions as to the use of “minimum force”. As NZDF acknowledges, however, a lethal JPEL designation did authorise the use of ISAF resources to conduct an operation up to and including the use of lethal force against the targeted individual.
9. We acknowledge that the public have understandably shown concern about the use of the JPEL, as is reflected in Hit & Run.13 The reality is, however, that the JPEL itself could not permit the use of predetermined and offensive use of lethal force. The use of such force had to be lawful (in terms of International Humanitarian Law and other relevant law) and consistent with the relevant national and ISAF rules of engagement. The JPEL effectively added a layer of protection or assurance, by helping to ensure that targeted killings were only conducted against high priority targets in respect of whom there was a substantial body of reliable intelligence showing that they were active members of the insurgency.
10. Further, it is important to recall in this context Professor Dapo Akande’s point (highlighted in chapter 6) that unlike conventional warfare where the soldiers on the opposing sides wear uniforms and are readily identifiable, those participating in an insurgency are often indistinguishable from the local civilian population.14 Accordingly, identifying those who are active participants in an insurgency requires the use of a mechanism such as the JPEL. Obviously, such a mechanism must be supported by robust and reliable intelligence-gathering and monitoring processes, as we discuss further below.

The law relating to targeted killings

11. The Inquiry is not required to determine whether targeted killings are morally or legally justified in a general sense. Our Terms of Reference focus on the effect of the rules of engagement and the knowledge of responsible ministers about the possibility of such killings. However, it is useful to briefly summarise the applicable legal framework by way of context.
12. “Targeted killings” are not specifically defined in any treaty or other rule of international law. Professor Akande adopted the following working definition in his advice:15

    … a predetermined and offensive use of lethal force against an individual who was identified in advance of the operation and specifically selected as being liable to have such force used against him or her.

13. That is generally consistent with the definitions adopted in Oxford’s The Handbook of the International Law of Military Operations16 and NZDF’s manual on the Law of Armed Conflict.17
14. The legal, moral and philosophical justification for targeted killings has been the subject of considerable academic and political debate. Much of the debate has centred on targeted killings in the context of the United States’ use of drones as part of the post-9/11 “war on terror”18 and the programme of targeted killings conducted by Israel’s Government.19 The legality of the latter programme was considered by Israel’s High Court of Justice in the 2006 Targeted Killings case,20 which found that such killings were neither “always legal” nor “always illegal” in that context.21
15. As already discussed in chapter 6, International Humanitarian Law permits the use of lethal force against an individual during the conduct of hostilities, subject to the principles of distinction, precaution and proportionality.22 If the use of lethal force against an individual is consistent with those principles, that force is lawful under International Humanitarian Law. And, adopting the approach of the International Court of Justice, where there is no violation of International Humanitarian Law, there will generally be no violation of International Human Rights Law.23
16. Applying those principles, The Handbook of the International Law of Military Operations concludes that the targeted killing of an individual during the conduct of hostilities is lawful only where:24
    1. The targeted individual is a person who is a lawful target of direct attack; and
    2. The operation is:
       1. planned and conducted so as to avoid erroneous targeting, as well as to avoid and minimise incidental civilian harm;
       2. not expected to cause incidental civilian harm that would be excessive in relation to the concrete and direct military advantage anticipated;
       3. suspended when the targeted person surrenders or otherwise falls hors de combat; and
       4. not otherwise conducted by prohibited means or methods of warfare.
17. As explained in chapter 6, insurgent fighters who are members of organised armed groups that are party to the conflict are lawful targets of direct attack under International Humanitarian Law. The accurate identification and verification of individuals to be placed on the JPEL was essential. The Handbook of the International Law of Military Operations states that:25

    It is … of utmost importance that, in practice, no targeted killing be carried out before all feasible precautions have been taken to rule out erroneous or arbitrary targeting.

    That is, all feasible precautions must be taken to ensure that individuals placed on the JPEL are lawful targets of attack and to verify the identity of the proposed target before an operation against them begins.

18. “Feasible precautions” are those precautions that are practicable or practically possible taking into account all relevant circumstances, including both humanitarian and military considerations.26 What is “feasible” to identify and verify a target will depend on the circumstances of the case and is ultimately “a matter of common sense and good faith”.27 The information relied upon must be sufficient to enable the attacking force to satisfy itself that the proposed target does not benefit from civilian protection.28 As Professor Akande notes, this will generally require an assessment of the most recent intelligence regarding the individual in question.29
19. No international case law or other authoritative statement defines the level or quality of intelligence required to verify a target. The UN Special Rapporteur on Extra-Judicial, Summary or Arbitrary Executions recommends that state armed forces should:30

    … use all reasonably available sources (including technological ones such as intelligence and surveillance) to obtain reliable information to verify that the target is lawful …

    And:31

    Targeted killings should never be based solely on “suspicious” conduct or unverified – or unverifiable – information. Intelligence gathering and sharing arrangements must include procedures for reliably vetting targets, and adequately verifying information.

20. Consistent with this, the NZDF’s manual on the Law of Armed Conflict provides that:32

    Commanders have a legal duty to take practicable steps to gather information and intelligence about the targets they are about to attack and the likely incidental consequences of the means and methods of combat they intend to employ. Wilful blindness to facts that argue against an attack does not provide an excuse for the resulting death and destruction.

21. The Inquiry has had access to material about the JPEL process, both in relation to specific targets and more generally. Much of that material is contained in ISAF documents and remains classified. It will be apparent, however, that the JPEL process contained a number of in-built safeguards:
    1. An application was subject to internal checks within TF81 before it could be submitted to ISAF. It was then subject to scrutiny by several distinct processes within ISAF. This was a meaningful process. JPEL nominations were declined relatively often, indicating they were not simply “rubber stamped”.33
    2. The process to have a target placed on the JPEL required significant recent and reliable intelligence about the target’s active involvement in the insurgency. The target pack for Abdullah Kalta, which has been released through the Inquiry’s classification review process, provides an example of the number and frequency of intelligence reports that supported JPEL listings.34
    3. JPEL listings had to be regularly reviewed based on updated intelligence and could be renewed if justified.
22. Once a target was listed on the JPEL, additional requirements had to be met before ISAF would approve an operation involving the predetermined use of lethal force against them. There were strict requirements for positively identifying the target and assessing potential collateral damage. Although the precise nature of these requirements cannot be disclosed, the Inquiry is satisfied they were adequate to ensure the right people were targeted and to minimise the risk of any incidental casualties.

    Obligation to capture rather than kill if possible?

23. In addition to the above requirements, The Handbook of the International Law of Military Operations states that targeted killings may not be carried out where the threat posed by the targeted person could be neutralised through capture or other non-lethal means without additional risk to the operating forces or the civilian population.35
24. As Professor Akande discussed in his opinion,36 this position draws on Recommendation IX of the 2009 International Committee of the Red Cross (ICRC) guidance on Direct Participation in Hostilities, which concluded that:37

    The kind and degree of force which is permissible against persons not entitled to protection against direct attack must not exceed what is actually necessary to accomplish a legitimate military purpose in the prevailing circumstances.

25. As Professor Akande noted, however, this remains one of the most controversial conclusions contained in the ICRC guidance.38 It has been criticised as creating a new “capture rather than kill” standard—which some commentators view as undermining the presumption that lethal force may be used in the first instance against legitimate military targets.39 Professor Akande accordingly expressed considerable caution about any suggestion that targeted killings may only be carried out where the threat posed by the individual could not be disabled by other means.40 The Inquiry considers there is an insufficient basis for applying such a test under the current law.

    Law enforcement operations

26. The Inquiry heard differing views as to whether the same rules apply in situations during an armed conflict that fall outside what could be considered the “conduct of hostilities”. Mr Nicky Hager, in particular, relied on the work of former ICRC Legal Adviser, Nils Melzer, to draw a distinction between the “conduct of hostilities” and “security” or “law enforcement” operations conducted by NZDF.41
27. Even within the context of a situation of armed conflict, armed forces may carry out tasks involving the use of force that do not form part of active hostilities. These are often described as “law enforcement” tasks, and typically include activities such as manning checkpoints or conducting cordon and search operations. Nils Melzer takes the approach that, under the rules of International Humanitarian Law itself, lethal force may only be used against an individual in such circumstances where it would otherwise be permissible under the general law enforcement paradigm recognised by International Human Rights Law.42 That is, where it is both necessary and proportionate to address a threat to human life.43
28. Nils Melzer’s work has been very influential. However, as Professor Akande advised, the factual distinction between what amounts to the “conduct of hostilities” and what amounts to “law enforcement” has not been clearly and objectively articulated.44 In The Handbook of the International Law of Military Operations Nils Melzer offered the following definition of the “conduct of hostilities”:45

    In sum … the conduct of hostilities is best understood as comprising all activities that are specifically designed to support one party to an armed conflict against another, either by directly inflicting death, injury or destruction, or by adversely affecting its military operations.

    However, application of that definition is by no means straightforward in practice.46

29. Nils Melzer has acknowledged that the “conduct of hostilities” and “law enforcement” are not mutually exclusive and may intersect—particularly in a situation of non-international armed conflict.47 As discussed in chapter 6, individuals other than members of a state’s armed forces who are directly participating in hostilities are not entitled to “combatant privilege”. They therefore remain subject to prosecution under domestic criminal law for acts that have been committed during the armed conflict, even if those complied with International Humanitarian Law. Such individuals are therefore at the same time both a lawful military target and a law enforcement suspect. The use of force against them may therefore be considered to fall under both the “conduct of hostilities” and the “law enforcement” paradigms.48
30. The Inquiry further notes in this context that the ICRC convened an expert group in 2013 to discuss the distinction between the “conduct of hostilities” and “law enforcement” in armed conflict.49 The group discussed a number of case studies, including the example of the targeted killing of an isolated “sleeping fighter”. This is a member of an organised armed group that is a party to the armed conflict who is sleeping at home with his family, in an area that is not an active combat zone at the time.
31. Members of the ICRC expert group were divided as to whether a targeted killing operation in those circumstances formed part of the conduct of hostilities. But by a small margin, a majority of experts considered that the situation was sufficiently connected to the armed conflict to form part of the “conduct of hostilities”.50 The “sleeping fighter” could therefore be lawfully attacked provided that the International Humanitarian Law principles of proportionality and precaution were fulfilled.51 Central to that analysis was the fact that the fighter was not a civilian and therefore was a legitimate target of attack under International Humanitarian Law.52 Neither the fact that the fighter was asleep nor that he was outside the active combat zone at the time of the operation changed that view.53
32. For the purposes of the present analysis, the Inquiry is satisfied that targeted killings of individuals on the JPEL did form part of the conduct of hostilities. As we have said, those individuals had been determined to be legitimate targets of attack under International Humanitarian Law. That remained the case even if the same individuals could also have been prosecuted by Afghan authorities under domestic criminal law.

The New Zealand rules of engagement

33. Turning to the New Zealand rules of engagement, rule H is of primary relevance.54 This rule was amended in December 2009 to allow a broader range of individuals to be targeted. Before the amendment, the rules of engagement only permitted attacks against identified members of specified insurgent groups. The amendment to rule H allowed TF81 to attack individuals, forces or groups directly participating in hostilities against the Afghan government.55 The term “directly participating in hostilities” reflected the wording of Additional Protocols I and II of the Geneva Conventions.
34. The reasons for the December 2009 changes to the rules of engagement were set out in a briefing document to the Minister of Defence from the then Chief of Defence Force, Lieutenant General Jerry Mateparae. They were:56
    1. To make the New Zealand rules of engagement consistent with the ISAF rules of engagement, to avoid interoperability difficulties with other coalition forces.
    2. To better enable TF81 to conduct direct action tasks against insurgent networks in support of ISAF and the Afghan government.
    3. To give TF81 the flexibility to plan attacks on individuals and groups which, by use of force, resisted ISAF lawfully extending the authority of the Afghan government. These were not limited to the groups mentioned in the original version of rule H. Some people whom intelligence indicated needed to be attacked to preserve the security situation could not be attacked under the existing rules of engagement.
    4. To allow the same ISAF staffing process and authorisation levels to apply to TF81 operations as applied to other ISAF Special Forces.
35. The briefing also stated that removing the reference to specific insurgent groups would not narrow the rules of engagement, because members of those groups could be placed on the JPEL once they satisfied the criteria for prosecution by deadly force based on available intelligence. The new rules of engagement would allow such targets to be prosecuted once confirmation of direct participation in hostilities was established.
36. In a written submission after the Inquiry’s second public hearing, NZDF further explained the reasons for the December 2009 rules of engagement amendment.57 As noted above, part of the reason lay in the difficulty that the insurgency was not limited to persons who were identifiably members of the groups originally specified in the rules of engagement. If the targets were members of the specified groups, then capture and kill operations would be proper under the rules of engagement. But intelligence used to identify such people was often unable to establish whether they belonged to the specified groups. It was therefore necessary to amend the rules of engagement since the original wording:58

    … affected the ability of NZ forces to plan and conduct direct action missions against insurgent elements in cases where the intended target could not be identified as belonging to the force or forces specified in the text of the rule. It was considered that this impeded the ability of TF 81 to undertake part of its mission, ‘to defeat the insurgency in Afghanistan’. It was also difficult to reconcile with the broad scope of the need to be prepared to conduct direct actions tasks against insurgent networks in support of ISAF and the Government of Afghanistan.

37. The Inquiry also heard from the then Minister of Defence, Hon Dr Wayne Mapp, who had approved the rules of engagement.59 He described the approval process in detail, including his discussions with NZDF. Without setting out the detailed content of Dr Mapp’s evidence, the Inquiry is satisfied that he took an active supervisory interest in the whole process and took his responsibilities seriously.
38. The main points Dr Mapp emphasised to the Inquiry were:
    1. He thought the rules of engagement should make a distinction between al-Qaida and the Taliban. The fundamental reason for the intervention in Afghanistan was to defeat al-Qaida, which had been proscribed by the United Nations as a terrorist organisation. By 2009 al-Qaida was no longer the global threat it had been, and the Taliban posed a more serious threat to reconstruction efforts and the promotion of public order by the Afghan Government. However, ISAF was not at war with the Taliban as a group, which represented a certain segment of the Afghan population and would need to be dealt with constructively.
    2. He was concerned to ensure that the rules of engagement were drafted so that they covered those directly taking hostile action against the Afghan Government, ISAF or NZDF but excluded those who were only members of the Taliban.
    3. In practical terms the New Zealand Special Air Service (NZSAS) would be primarily protecting Kabul from Taliban attacks.
    4. In more than 90 per cent of missions the NZSAS did not fire their weapons. Instead, the Taliban insurgents were arrested by the Afghan Crisis Response Unit and dealt with through the Afghan justice system.
    5. Changes to the rules of engagement needed to be made because the existing rules were too restrictive.
39. In respect of the rules of engagement amendment in December 2009, the Minister said:60

    I ultimately concluded, after the briefings and considering both the legal and operational situation, that I should approve the amendment to the ROE. I set out my reasons on the NZDF Cover Sheet, dated 14 December 2009. I inserted a handwritten comment saying:

    “I have now been fully briefed on the change, including a discussion on the concept of operations. As a result, I am satisfied that the new ROE meets two criteria:

    1. it complies with New Zealand legal requirements
    2. it meets the operational requirements of Op Wātea and NATO/ISAF.”
40. At the public hearing the former Minister was questioned on this point and made it clear he had a detailed understanding of the changes and the reasons for them:61

    I was aware that we could take direct action, that’s why I got the briefing. That’s why I talked to the people. I mean, to some extent that was always permitted even in the first draft … But they have to have an intention to conduct an imminent or immediate attack. That’s the governing test. So, on that basis I thought, yes, that is appropriate … because the alternative would be allowing them to attack you … We knew that there were people moving through that area [Bamyan] that were intending to attack …

    Our assessment

41. We consider the amended version of the rules of engagement, approved by the Minister of Defence and the Prime Minister, did authorise “predetermined and offensive use of lethal force against specified individuals other than in the course of direct battle.” Such operations would have been possible before the change, but in respect of a more limited class of people. We note in particular that the amended rule H:
    1. permitted “attacks” on individuals who were directly participating in hostilities, in contrast to other rules which only permitted the use of force in defence of people or property or to achieve the mission;62
    2. did not contain a “minimum force” requirement, in contrast to other rules;63 and
    3. required a certain authorisation level for “planned attacks”.
42. On its plain wording, then, the amended rule contemplated that planned attacks could be carried out, and that TF81 would not be limited to the use of minimum force in conducting them. Of course, before carrying out a planned attack, TF81 personnel would have to be satisfied that it complied with International Humanitarian Law in light of the circumstances at the time.
43. It is also clear that NZDF and responsible ministers were aware that rule H permitted targeted killings. NZDF requested the change to the rules of engagement. In its briefing to the Minister of Defence, it stated that removing the reference to specific insurgent groups would not narrow the rules of engagement, because members of those groups could be placed on the JPEL once they satisfied the criteria for prosecution by deadly force based on available intelligence.64 The new rules of engagement would allow such targets to be prosecuted once confirmation of direct participation in hostilities was achieved. Thus, the briefing specifically linked rule H to the use of the JPEL and the fact that targets on the JPEL may be subject to deadly force. The briefing also referred to the need for TF81 to be able to conduct planned “direct action” tasks against the insurgent network.65
44. Dr Mapp confirmed in his public evidence that he was aware the amended rules of engagement would permit kill or capture operations.66 While the Inquiry has heard no direct evidence on the point, we consider this must also have been clear to the Prime Minister when he approved the amendment.67 As such, we are satisfied that NZDF’s ability to nominate targets for the JPEL and to participate in targeted killing operations was understood and mandated by the responsible ministers.
45. Our Terms of Reference do not require us to examine whether the predetermined and offensive use of lethal force by NZDF was consistent with International Humanitarian Law. However, it will be apparent from our summary of the legal framework above that the Inquiry considers targeted killings can be consistent with International Humanitarian Law where the target is a member of an organised armed group. For people to be placed on the JPEL, there had to be a significant body of reliable intelligence about them to that effect.
46. As noted in chapter 6, there is some disagreement over the criteria for determining such membership—in particular, whether the individual must have a combat function or whether making some other contribution to the combat effectiveness of the group is sufficient. We do not intend to analyse in detail the legal approach taken by NZDF in that respect, as it is outside our Terms of Reference. In any event, the JPEL targets nominated by TF81 with whom we are concerned did in fact have combat functions.

The death of Abdullah Kalta

47. The case of Abdullah Kalta’s eventual death in an ISAF operation in November 2012 illustrates how the JPEL process worked in practice and how it interacted with the applicable rules of engagement. However, it is important to note that the operation was not ultimately authorised on the basis of Kalta’s JPEL listing. Additionally, New Zealand personnel neither authorised nor conducted the operation. Even so, NZSAS personnel did contribute significantly to it by providing intelligence to the ISAF decision-makers.68
48. As discussed in chapter 4, after Operation Nova in October 2010 Kalta and Maulawi Neimatullah went to Pakistan. In August 2011 NZSAS personnel in Afghanistan met with members of a partner unit that had conducted an air strike targeting another Tala wa Barfak insurgent, Qari Musa, in May 2011.69 The purpose of this meeting was to share information on JPEL targets to “disrupt insurgent groups in Baghlan”.70 It is apparent that the possibility of conducting targeted killings of other Tala wa Barfak-based insurgents who were on the JPEL was being explored at this point.
49. Kalta returned to Tala wa Barfak at some time in early-to-mid 2012. His return coincided with a significant increase in improvised explosive device attacks by the Tala wa Barfak Taliban on Afghan National Security Forces and New Zealand Provincial Reconstruction Team (NZPRT) personnel. NZDF intelligence staff assessed that Kalta was responsible for this increase in activity. Throughout 2012 Kalta remained on the JPEL and was a key figure in the command and control of the Tala wa Barfak insurgent group. NZSAS personnel based in Afghanistan assessed that he was involved in planning and obtaining materiel for two attacks against the NZPRT, which led to the deaths of New Zealand soldiers in August 2012.71
50. Intelligence reporting in November 2012 indicated that Kalta was preparing to conduct an attack against ISAF or Afghan government targets in the near future.72 On 20 November Kalta and other insurgents were reported as being in the village of Karimak in northeast Bamyan, armed with AK-47s and at least one PKM light machine gun.73 This was near the location of the 3 August 2010 attack that killed Lieutenant O’Donnell. The following day, 21 November 2012, intelligence indicated that Kalta and his group were positioned approximately 200 metres from the site of the 3 August attack and were lying in wait to ambush vehicles travelling along the main road below them.74 NZSAS personnel in Afghanistan collected and collated this intelligence and provided it to decision-makers in ISAF Headquarters.75 At the same time, ISAF aircraft identified a group of armed individuals at the location.
51. To authorise a strike on the basis of Kalta’s JPEL listing, the authorising commander76 had to be satisfied to a high degree of certainty that the individual being targeted was the individual on the JPEL. Although there was substantial intelligence that Kalta was in the vicinity and planning to conduct an attack, it could not be confirmed to the required degree of certainty that he was either in the group seen on the surveillance, or that he was an identifiable individual within that group. As a result, a kinetic strike against him could not be authorised based on his JPEL listing.
52. NZSAS personnel were actively trying to gather intelligence that would provide the necessary degree of certainty when the authorising commander concluded that there were sufficient grounds to conduct a strike on the alternative basis that the group under surveillance was directly participating in hostilities. This decision was based on the intelligence received from the NZSAS, the location where the men were, their behaviour and the presence of weapons. The aircraft fired a Hellfire missile and over 200 30mm cannon rounds in two passes over the area, killing four of the men, including Kalta.77 Although there was significant intelligence provided by NZDF, the air strike itself was an ISAF operation.
53. The fact that the strike against Kalta was not able to be authorised on the basis of his JPEL listing demonstrates the strict identification requirements that had to be met before targeted killings could be conducted.
54. Comments to the media from the then Prime Minister John Key following the operation against Kalta indicated that he was aware of the JPEL process and how targeted killings were authorised (even though that particular strike was not ultimately conducted on that basis):78

    You have to make sure you know exactly it’s the right person in the right location, that they’re on the list of people etc etc.

    The main issue here is trying to make the environment safe. So if we have known insurgents who’ve carried out attacks that have killed people and are planning other attacks – and my understanding was this was an example of that where there was planning for further attacks to take place – then, we are in a war zone and ultimately we need to make sure that our men and women are as best protected as we can.

    That means using intelligence, and if required, making sure that those who would undertake those attacks aren’t in a position to do so.

55. In contrast to Operation Burnham, NZDF publicly acknowledged its role in this strike soon after it occurred. The then Chief of Defence Force, Lt Gen Rhys Jones, was interviewed on Radio New Zealand’s Morning Report.79 He spoke about the degree of NZSAS involvement, the reasons for the operation and the decision to conduct a kinetic strike rather than attempting to capture Kalta.

Conclusion and findings

56. In the result, the Inquiry is satisfied that the rules of engagement did authorise the predetermined and offensive use of lethal force, and that this was understood by NZDF and responsible ministers (the Minister of Defence and the Prime Minister). NZDF personnel deployed in Afghanistan were involved in the JPEL process. While we are not required to make findings about the legality of targeted killings, we observe that, on the information available to us, we have seen no cause for concern about the JPEL process itself as it operated within the coalition or how it was applied in relation to the events before the Inquiry.

< Return to Contents | Previous | Next >