Chapter 3 - Operations Burnham and Nova: origin and planning

< Return to Contents | Previous | Next >

3

Operations Burnham and Nova: origin and planning

Chapter 3

1. In this chapter we address clause 7.4 of our terms of reference, which directs us to report on the “planning and justification/basis for the Operations, including the extent to which they were appropriately authorised through the relevant military chains of command”.1 We describe why Operations Burnham and Nova were undertaken and the planning and preparation that went into them.
2. In chapter 4 we set out what occurred on the two operations. This is a necessary precursor to discussing some of the specific issues on which we are required to report, such as whether the New Zealand Defence Force (NZDF) complied with the applicable rules of engagement and with International Humanitarian Law (including NZDF’s assessment that Afghan nationals were taking a direct part in hostilities or were otherwise legitimate targets during Operation Burnham). We will describe the consequences of the operations, in terms of the casualties and property damage that resulted from them, in chapter 5.
3. In this chapter, we will discuss Operations Burnham and Nova under two headings:
   1. Why were the operations carried out?
   2. What planning and preparation was undertaken?
4. Before we embark on this, however, there are four preliminary matters we should mention. First, the names of villages in the Tala wa Barfak District pose a challenge for outsiders as names and spellings differ, often markedly. The area where Operations Burnham and Nova occurred is referred to by NZDF and others in the contemporaneous documents as Tirgiran Valley, Tirgiran Village or sometimes just Tirgiran, and by the authors of Hit & Run as Khak Khuday Dad and Naik villages. Some NZDF documents also refer to Naik as Dahane Nayak.2 In relation to the villages, we will use the names derived from the villagers by the authors—Khak Khuday Dad and Naik—or refer simply to “the villages”. We will refer to the area where the villages are located as Tirgiran Valley.
5. Second, as Thomas Barfield notes in Afghanistan: A Cultural and Political History, Afghanistan’s physical geography has had a profound impact on the country’s history and culture.3 Geography is particularly relevant to the planning and conduct of Operations Burnham and Nova. The Tala wa Barfak District where the operations took place is mountainous and relatively inaccessible. The particular location of the operations was a remote Y-shaped part of a river valley system bounded by steep, rocky mountain-like ridges.4 The valley floor was around 2,600 metres above sea level, and the surrounding ridges were a further 100–500 metres above sea level. The area was not accessible by road. As will become apparent, the area’s remoteness and rugged terrain are important when seeking to understand what happened on the operations.
6. Third, it is important to understand the basic configuration of the New Zealand military force at the time of Operations Burnham and Nova. The force was known within the International Security and Assistance Force (ISAF) as Task Force 81 (TF81). It comprised approximately 70 people, mostly New Zealand Special Air Service (NZSAS) personnel with some attached specialist support personnel.5 Its Commander, also the Senior National Officer for the NZSAS contingent, was a Lieutenant Colonel. We refer to him as the Senior National Officer throughout this report. The nucleus of TF81 was an NZSAS Squadron, commanded by a Major. In Operation Burnham, the Squadron Commander acted as the Ground Force Commander. Under his command were some specialist force elements, or troops, commanded by captains. One of the captains was the Troop Commander, who acted as the Assault Force Commander for Operation Burnham. TF81 was based in Kabul.
7. Finally, we note that ISAF’s command structure was organised around five regional commands, each of which was responsible for a designated geographic area of Afghanistan. The Provincial Reconstruction Teams (PRTs) were assigned under the operational control of a regional command. So, for example, the New Zealand Provincial Reconstruction Team (NZPRT) was located in an area under the operational control of Regional Command East, in respect of which the United States was the lead nation. Alongside the regional commands was a separate ISAF Special Operations Forces command in Kabul. This command grouped similar forces, with similar missions, under a unified command with a whole-of-Afghanistan remit. TF81, being a Special Forces unit with the capacity to carry out complex operations, was assigned under the operational control of ISAF Special Operations Forces.6 The decision-making steps to achieve approval of an operation were intricate and demanding, as we indicate below.

Why were the operations carried out?

8. Hit & Run has been interpreted as alleging that Operations Burnham and Nova were intended to avenge the death of Lieutenant Tim O’Donnell (that is, they were revenge raids aimed at those responsible for the 3 August attack).7 Certainly, many references in the book (some being quotes from sources) support that interpretation of it. For example, Operation Burnham is described as a retaliation or revenge raid;8 particular conduct during the operation is described as being motivated by “retaliation and punishment”;9 the actions of the NZSAS on Operation Nova are attributed to revenge and punishment;10 and the NZSAS is described as being on a “campaign of retaliation and revenge” or similar in relation to the various operations.11
9. Obviously, to accuse a military unit of conducting revenge raids is to strike at the professionalism and integrity of that unit. Such an allegation is particularly grave when made against a unit that sees itself as being a highly trained, elite unit, such as the NZSAS. It suggests that the operations concerned were not conducted for legitimate military objectives and were not subjected to normal oversight controls, but rather were predominantly driven by emotions and vengeful motives.
10. Mr Hager, who was primarily responsible for writing Hit & Run,12 told us that the book did not allege that Operation Burnham was a revenge raid;13 rather, the operation was in “retaliation” (in a military sense) against particular insurgents. However, he said there was “a ‘mood’ to the operation”—some NZDF personnel had “a mood of anger or vengeance over the death of their colleague during the operation”.14 This “mood” affected the way they acted.15 As we have said, this was not how the book was understood when it was released; nor is it how we have understood it, having read it many times.16
11. We address the question whether the conduct of TF81 personnel during Operation Burnham was influenced by a desire for revenge in chapter 5.17 In this chapter, we focus on the reasons for the operation. We begin with the contemporaneous documents: what do they indicate was the purpose of the operations?

    Operation Burnham

12. Bamyan province, where the New Zealand Provincial Reconstruction Team (NZPRT) operated, was relatively peaceful. The bulk of the population were Hazaras, who were Shia Muslims and had been persecuted by the Taliban. Despite its generally peaceful nature however, there was some insurgent activity in Bamyan. When he took over command of the NZPRT in April 2010, the NZPRT Commander was advised by his predecessor of an insurgent threat from the adjoining province, Baghlan. Intelligence from numerous sources18 over at least the preceding year had indicated that some middle-rank insurgent leaders had homes in the Tala wa Barfak District of Baghlan, as did some lower-level insurgents under their command.19 The reporting indicated that “Tirgiran village” was a “significant support area” for the Taliban—Taliban flags were flown and insurgent leaders held meetings there.20
13. It appears the Afghan National Police did not have a significant presence in Tala wa Barfak and local government had limited impact in the area. Further, there was no effective coalition presence. According to intelligence reports, Taliban insurgents used Tala wa Barfak as a “safe haven” and a base for their operations into Bamyan and other areas, and travelled from there to Pakistan from time to time for further training and resources or to keep a low profile for a period.21 In addition, intelligence reporting indicated there was a Taliban training camp in Tirgiran Valley, near where some of the leaders had their family homes. Areas such as this were not unusual and existed throughout Afghanistan, even though coalition forces had been in the country for almost nine years by this time.
14. We point out here that the contemporaneous intelligence reporting of a Taliban or insurgent presence in Baghlan generally, and Tala wa Barfak in particular, in 2008 and later, is confirmed in a recently published book by an expert on the Taliban in Afghanistan, Antonio Giustozzi. In The Taliban at War 2001 – 2018 he writes:22

    In many locations the Taliban had some pockets of core support, which continued to host Taliban even at the most challenging times. For instance, in Baghlan province, which had certainly not been a Taliban stronghold in earlier days, they managed to find roots from 2006 onwards not only in places such as Gadya (Baghlan Jadid), or the Pashtun settlements in Dand-e Ghori, but also among some Uzbek communities in Burka district. These communities even reached out directly to IMU23 leaders in Pakistan for support, allowing the IMU to establish a direct presence in Baghlan by 2008–9. Due to the inactivity of ISAF troops, the Taliban expanded undisturbed in the northern districts of Baghlan-e Markazi, Baghlan-e Jadid and Burqa. Non-Pashtun districts of Baghlan increasingly came under Taliban influence due to their exploitation of disputes over land and pasture rights, which sometimes dated back decades. Increasingly, a quarrelling party would request their involvement to strengthen the disputant’s position. This occurred in the districts of Nahrin, Tala wa Barfak and Burka.

15. On 9 April 2010, soon after the NZPRT Commander had arrived to take command of the NZPRT, an NZPRT patrol was attacked in the Kahmard District of Bamyan province near the border of Baghlan province. There were no casualties, but the attack confirmed the existence of an insurgent threat from across the Baghlan border—a threat which was assessed to be significant. However, Tala wa Barfak was outside the area of operations of NZPRT, which was, in any event, not mandated or organised to undertake operational strikes to deal with insurgent threats. Baghlan fell within the area of operation of ISAF’s Regional Command North, in respect of which Germany was the lead nation. Bamyan, where the NZPRT operated, was within Regional Command East’s area, which was under United States command.
16. The NZPRT Commander monitored the position in Tala wa Barfak closely. He had an initial discussion with the headquarters of Regional Command North about the perceived insurgent build-up in Tala wa Barfak, but they had other priorities. There was a Hungarian PRT based in Baghlan, but it was fully occupied in dealing with significant insurgent activity near its Pol-e Khomri base. Further, like the NZPRT, the Hungarian PRT was not organised for offensive operations. The NZPRT Commander also sought the assistance of Task Force Wolverine. Task Force Wolverine was a United States Army manoeuvre force under Regional Command East, responsible for conducting full spectrum operations and partnering with PRTs and Afghan forces to improve security.24 While those at Task Force Wolverine shared the NZPRT Commander’s concerns, Baghlan was outside their area of operations.
17. In addition, the NZPRT Commander engaged with key leaders in Bamyan province and the Tala wa Barfak District to discuss the insurgency situation and was involved in formal meetings with Afghan national and local government officials. During this period one of the insurgent leaders, Qari Musa, was nominated to the Joint Prioritised Effects List by TF81 on behalf of the NZPRT.
18. In May 2010, following an “in-theatre” visit, the Chief of Defence Force, Lieutenant General Jerry Mateparae, expressed concern to the Director of Special Operations about the developing insurgent situation in Baghlan province. He indicated that if the situation continued to deteriorate and to affect the NZPRT’s area of operations, he might look at tasking TF81 to assist. This was consistent with the Cabinet mandate under which TF81 operated. That allowed national tasks to be given priority, with one such task being to provide support to the NZPRT.25 Around this time, TF81 began providing intelligence and analytical assistance to the NZPRT to assess the insurgent network operating in Bamyan.26
19. The nature and extent of the threat from the insurgents in Baghlan became even more apparent in the attack of 3 August 2010, which resulted in Lieutenant O’Donnell’s death. The NZPRT Commander considered the 3 August attack showed a worrying level of sophistication and expertise on the part of the insurgents. It intensified his concerns about the security of the NZPRT and about its ability to maintain the local population’s trust and confidence that it could provide a secure environment in Bamyan.
20. As we discuss in more detail in the next section, planning for a possible operation began almost immediately, the intended targets being two of the insurgent leaders thought to be responsible for the attack—Abdullah Kalta and Maulawi Neimatullah. For now, we only need note that the concerns which the NZPRT Commander identified throughout the planning process were reflected in the purpose of the operation. So, for example, TF81’s concept of operations27 identified two principal effects for the proposed operation:28
    1. to protect the NZPRT, Afghan security forces and the local population from insurgent activity, both generally and in relation to a possible imminent attack identified by intelligence sources;29 and
    2. to enhance the authority and legitimacy of the Afghan Government.

    The Ground Force Commander’s orders for Operation Burnham identified similar objectives, albeit in different language, namely to disrupt the insurgents’ attack network and to increase security for Bamyan province.30

21. To summarise, then, we consider that the contemporaneous documents and the evidence of the NZPRT Commander show:
    1. From the time the NZPRT Commander took command of the NZPRT in April 2010, he was concerned about insurgents coming into Bamyan province from Tala wa Barfak to attack the NZPRT and Afghan authorities. He was actively trying to address that concern.
    2. The NZPRT Commander’s wish to disrupt insurgent activities in Tala wa Barfak was based on both the force protection of the NZPRT and his desire to build the local population’s confidence in the NZPRT and Afghan authorities by ensuring their security.
    3. These two factors, coupled with the fact that the insurgents had conducted a successful attack on 3 August and that intelligence reporting indicated they planned further attacks in Bamyan in the near future, were the drivers for Operation Burnham.

    The contemporaneous documents do not support the view that Operation Burnham was a revenge raid. Rather, the documents identified objectives that the NZPRT Commander had been pursuing for some time and which were plainly legitimate from a military perspective.

    Operation Nova

22. Within a few weeks of Operation Burnham (which, as we discuss in chapter 4, did not result in the capture of either of the two targeted insurgent leaders), planning began for a return operation, now referred to as Operation Nova. Hit & Run says of the return operation that “it is hard to see their [TF81’s] actions were born of anything but revenge”.31 It goes on to claim that the purpose of the operation was to “wreck the houses again, this time more thoroughly”.32 Hit & Run quotes an anonymous NZSAS source, who stated that the purpose of the operation was to punish the insurgents by destroying the houses, and it was “pure revenge”.33 According to the book another anonymous NZDF source stated that an internal report said the goal of the operation was to scare the insurgents and discourage further attacks.34 Again, the question is what do the documents show was the purpose of Operation Nova?
23. Intelligence indicated that the insurgents had regrouped following a period of hiding after Operation Burnham and were intent on conducting more attacks into Bamyan before winter.35 Operation Nova was another attempt to capture Maulawi Neimatullah, conducted on 3 October 2010, some six weeks after Operation Burnham. There was nothing unusual about conducting a repeat operation in an area such as Tala wa Barfak, where neither the Afghan National Police nor coalition forces had a permanent presence, and where there continued to be intelligence reporting about the presence of insurgents and the threat they posed to the NZPRT and others in Bamyan. The troop orders for Operation Nova describe the purpose of the operation in similar terms to those for Operation Burnham: to disrupt the Baghlan insurgent network and to enhance the security of Bamyan.36
24. The position, then, is that the planning documents do not support the notion that Operations Burnham and Nova were revenge raids. That is not the end of the matter, however. Whatever the view reflected in the documents, the reality on the ground may have been different. Some of those carrying out the operations may have been in an angry or vengeful mood and that may have affected what they did. As a consequence, we explored this topic with NZDF witnesses, particularly when examining certain occurrences during Operation Burnham. We come back to this issue in chapter 5.37

What planning and preparation was undertaken?

25. Given the nature of most modern conflicts, New Zealand forces are likely from time to time to deploy overseas to operate in conjunction with other forces or as a member of a coalition, as they did in Afghanistan. When the NZSAS is deployed overseas as a Special Operations Force within a coalition such as ISAF, it will necessarily be subject to the coalition’s priorities. Further, in some instances New Zealand will not be able to deploy sufficient suitable equipment (assets) to support NZSAS operations—an obvious example being helicopters. In such circumstances, the NZSAS may have to use coalition assets for some operations. When nations deploying to Afghanistan contributed forces and resources, they also determined whether those assets were for national requirements or whether they were to be assigned under ISAF to be used more broadly. ISAF allocated the use of such assets based on its operational priorities.

    Operation Burnham

26. The planning process for an operation such as Operation Burnham followed a well-established pattern.38 While it could be considered complex given factors such as the number of different force elements involved and the range of possible contingencies and risks faced, the planning process followed a standard concept employed generally by Special Operations Forces in Afghanistan. There is no doubt that TF81 was competent and qualified to undertake such an operation.39
27. While TF81 had completed numerous operations since deploying to Afghanistan in October 2009, most were ground-based, vehicle-mounted operations, carried out in and around Kabul and its six surrounding provinces. Operation Burnham was different, in that it involved the use of multiple supporting aircraft and a drone to surveil and access targets in a remote location outside TF81’s usual area of operations. Hon Dr Wayne Mapp, the Minister of Defence at the time, told the Inquiry that he understood that Operation Burnham was the biggest operation that New Zealand forces in Afghanistan had carried out.
28. As well as gathering intelligence and planning the operation, TF81 had to obtain approvals and meet specified requirements. In the case of Operation Burnham, TF81 required approval through both its national (New Zealand) and operational (ISAF) chain of command:
    1. National approval: The Chief of Defence Force had to approve the operation as it was to be conducted outside TF81’s mandated area of operation. When Cabinet agreed to the deployment in mid-2009, it defined TF81’s operating area as Kabul and the six surrounding provinces but allowed operations outside this area if approved by the Chief of Defence Force.40 The Chief of Defence Force approved the conduct of Operation Burnham on 12 August 2010. Apart from authorisation by the Chief of Defence Force for an “out of area” operation, no further national approvals were required.
    2. Operational approval: Gaining approval from ISAF involved a set process. The Afghan Government had to give its approval, which was obtained through the Ministry of Interior.41 Regional Command North had to give its concurrence because it was the “owner” of the area where Operation Burnham was to occur (known as the Battle Space Owner).42 This was to ensure TF81’s planned operation did not conflict with any operations planned by Regional Command North. Having met these requirements, TF81 had to submit a concept of operations (or CONOPS) to ISAF’s Special Operations Forces Headquarters (the approving authority on behalf of the Commander ISAF) to obtain final approval to conduct the operation.43 An ISAF Legal Advisor’s concurrence was also obtained as part of the concept of operations process.44
29. We have had access to TF81’s planning material, which is extensive. By way of illustration, the Ground Force Commander, who is responsible for all that occurs on the ground, gave lengthy written orders that covered all aspects of Operation Burnham.45 The orders were presented in a standard military format, covering situation, mission, execution, administration and logistics, and command and signals. This format enables a concise presentation to troops on key matters, such as the terrain and intelligence; the specified task; how the force will go about achieving that task; the administrative and logistic support available; and how the force will be organised. In addition, the Troop Commander, who was the designated Assault Force Commander for the operation, gave orders covering the roles and responsibilities of the assault force, including how the three buildings identified as associated with the targeted insurgents (referred to as A1, A2 and A3) were to be secured and entered.46 They are quite detailed. The Ground Force Commander referred in his orders to anticipating possible insurgent reinforcements from the surrounding area, although the orders did not identify the direction of any threat.47
30. As we are not experts in military planning, we are not in a position to evaluate the overall competence of the planning process, although it appears to us to have been methodical, detailed and adaptive to changing circumstances. As will be obvious from what we have just described, the requirements and independent scrutiny incorporated into the national and ISAF approval processes indicate that the planning was of an appropriate standard. It aligned to ISAF’s requirements as set out in its Standard Operating Procedures and to NZDF’s doctrine. The Inquiry’s military expert considered that the planning process met the requirements and was of an equivalent standard to comparable forces conducting similar operations. But ultimately a detailed assessment of the planning process is unnecessary for the Inquiry’s purposes. It is sufficient that we focus on those aspects of the planning and preparation that bear directly on the issues that remain in contention.
31. Accordingly, we will give a general outline of the planning process and then address in more detail four aspects that are particularly important, namely:
    1. How did the planning approach and assess risks?
    2. What preparation was there in relation to the protection of civilians?
    3. What were the operational command and control arrangements between the ground force and the air assets?
    4. What information did TF81 have to indicate that the targets would be present when Operation Burnham was carried out?

Overview of the planning process

32. We begin with a general description of how the planning for the operation evolved and the necessary approvals were obtained.
33. As we have said, the NZPRT Commander had been concerned about the insurgent threat to Bamyan emanating from Tala wa Barfak since arriving in Afghanistan in April 2010, and had been in touch with both Regional Command East and Regional Command North about dealing with the situation. It became clear that both regional commands had other more pressing operational priorities. In May 2010, following the unsuccessful insurgent attack of 9 April on the NZPRT, TF81 began to gather intelligence on the insurgent networks operating in Bamyan.48
34. Soon after the 3 August 2010 attack, the NZPRT Commander contacted the TF81 Senior National Officer about a response. Around this time, TF81 began focusing its intelligence collection efforts on determining who specifically had been involved in the attack and would therefore likely be central in organising further attacks.49 Intelligence reporting from a variety of sources soon identified the leaders as including Abdullah Kalta, Maulawi Neimatullah and Qari Miraj.50 The NZPRT and TF81 were already aware of these individuals, as they were part of the Tala wa Barfak insurgent network that the NZPRT had been monitoring for some time.51 Earlier reporting had indicated that Kalta had a home in Naik52 and that Miraj was originally from the same area, although he had since moved to another village called Anadarah, about 16 kilometres north of Naik as the crow flies.53 Additional reporting received in the days following the 3 August attack provided further confirmation that Kalta had a residence in Naik and indicated that Neimatullah lived in the same village.54 The intelligence reporting also said that Miraj and Kalta were planning a further attack on a NZPRT or Afghan security force patrol.55
35. On about 10 August, the TF81 Squadron Commander (who was the Ground Force Commander for Operation Burnham) travelled to Bamyan, accompanied by intelligence officers, to discuss the situation with the NZPRT Commander and his intelligence staff. By 11 August 2010, planning for a possible operation into Tirgiran Valley was gathering pace.
36. From the outset, TF81 personnel in Kabul had kept NZDF Headquarters in Wellington appraised of developments. The Chief of Defence Force was concerned to ensure that TF81 was progressing plans to target the insurgent group responsible for the attack.56 The Senior National Officer sent regular updates to Wellington, mainly through the Director of Special Operations. The Chief of Defence Force granted formal approval for TF81 to conduct an operation outside its mandated area on 12 August.57
37. The operation required approval through the ISAF chain of command as the TF81 troops and necessary supporting air assets were under the operational control of the Commander ISAF and the operation was being conducted within ISAF’s area of operational command. To obtain the necessary air support, TF81 submitted an Air Mission Request to Regional Command North on 11 August 2010. The request:
    1. sought provision of two CH-47 Chinook helicopters to transport personnel and two AH-64 Apache attack helicopters as escorts for the Chinooks;
    2. noted that TF81 would be partnering with the Afghan Crisis Response Unit (CRU);
    3. referred to the operation as a deliberate detention operation;58 and
    4. said the purpose of the operation was to disrupt a threat from Baghlan and demonstrate the security of Bamyan to the local people.

    The focus of the operation was on detention. On operations of this type, no firing was intended or planned, although, of course, it might occur.

38. The Commander Regional Command North needed to give his concurrence for the operation since it would be conducted in his battlespace. Accordingly, the Senior National Officer and Ground Force Commander had at least one, and perhaps more, video conferences with the Commander Regional Command North and one of his deputies, a United States Brigadier General, to discuss the proposed operation. The possibility of Regional Command North conducting the operation itself was raised but, after consideration, the Commander indicated that Regional Command North would not conduct an operation in the area. The Commander ultimately gave consent to the operation on 16 August 2010, attaching three conditions, namely that:59
    1. a liaison officer from TF81 be situated at the Regional Command North’s Tactical Operations Centre;
    2. the ground forces conduct a “soft knock” before entering any “compound of interest” (that is, the buildings associated with the insurgent targets);60 and
    3. there be “key leader engagement”.61

    The first of these conditions ensured that there was a direct line of communication between TF81 and Regional Command North and provided the Commander Regional Command North with an immediate awareness of events should anything go wrong on the operation. The remaining two conditions reflected the Commander ISAF’s guidance on night raids, which is discussed below.

39. On 15 August TF81 submitted applications for Abdullah Kalta (Objective Burnham), Maulawi Neimatullah (Objective Nova) and Qari Miraj (Objective Yamaha) to be placed on the Joint Prioritised Effects List.62 These applications were approved over the following days.63 As the intent of the operation was to arrest and detain Kalta and Neimatullah,64 the Afghan Ministry of the Interior organised the issue of arrest warrants for both men.65 The intention was that the CRU would execute these warrants during the operation if one or both of the insurgents were found at their houses.66
40. On or about 16 August 2010 the concept of operations was submitted for approval to ISAF Special Operations Forces Headquarters, which was responsible for briefing up to ISAF Headquarters as required and providing approval for the operation (under delegated authority from the Commander ISAF).67 The Commander of ISAF Special Forces Operations granted approval around 18/19 August 2010.
41. The concept of operations outlined the reason for and the intent of the operation, provided a scheme of manoeuvre, assessed the risks involved in the operation, explained how and to what extent “Karzai’s 12”68 would be complied with and provided justification for conducting a night raid in accordance with the Commander ISAF’s Night Raids Directive. The TF81 legal officer was involved in preparing the concept of operations to ensure that the planned operation met the necessary legal requirements.
42. As noted, the objective of the operation was to detain the two insurgent leaders who had been identified as living in Naik: Kalta and Neimatullah. To achieve this, TF81 proposed to travel to the area by helicopter during the hours of darkness,69 after receiving intelligence confirming the targets were present. Troops would secure and then enter three buildings that were believed to be associated with the targets, who would be detained and arrested in the first instance if they were present. The buildings would also be searched for explosives and evidence to support a later Afghan prosecution.
43. By chance, the Chief of Defence Force and Minister of Defence were in Afghanistan on a pre-arranged visit in the days leading up to Operation Burnham. The purpose of their visit was to show support for the NZPRT following Lieutenant O’Donnell’s death and to discuss New Zealand’s ongoing commitment to ISAF.70 On 19 August the Senior National Officer and the Ground Force Commander briefed them on the operation.71 The Minister’s approval was not required for the operation—the Chief of Defence Force had authority to approve operations outside Kabul and the surrounding area and had already done so. The briefing occurred because the Minister happened to be present and this was TF81’s current operation. However, given the size and significance of the operation, the Minister made a phone call to the then Prime Minister, Rt Hon John Key, and asked the Chief of Defence Force to inform him of what was planned.72 The Chief of Defence Force did so, and the Prime Minister was apparently comfortable with the proposed course of action.73
44. Also on 19 August, the Ground Force Commander went to Bagram Airfield to brief the commander of the air assets that would be supporting the operation. He was accompanied by the Troop Commander, the Air Liaison Officer (who was responsible for submitting requests for air support), the Joint Tactical Air Controller (JTAC) and helicopter landing zone security personnel. They discussed the nature of the support TF81 would require during the operation. The Ground Force Commander also briefed the General in charge of the CRU, as the operation was to be a partnered one (as most TF81 operations were). CRU personnel would principally form part of the security force for the main helicopter landing zone and the aerial response force.
45. In the days leading up to the operation, TF81 received intelligence to the effect that the two targets were in Naik.74 On 20 August 2010 final approval of air asset support was confirmed, which would allow the operation to proceed on the night of 21/22 August. TF81 made final preparations over the course of 20 August. The Ground Force Commander and the Troop Commander gave the troops their orders, and they conducted rehearsals for the operation. Troops had time to call their families at home, which some did. As we explain below, the operation was considered to be a relatively high-risk one for the TF81 personnel involved. They were going into an area where armed resistance might well be encountered and there was a chance they would not see their families again.

    Risk assessment

46. Operational risk assessment is an important part of the planning process. Commanders need to identify risks and plan for them; those under their command need to be told of likely areas of risk so that they are prepared. Operational risk is part of the concept of operations and is presented as “risk to force” and “risk to mission”. What do the planning documents for Operation Burnham say about risks?
47. In his orders for the operation, the Ground Force Commander referred to intelligence indicating that the insurgents Kalta and Neimatullah were both confident, that Taliban fighters in the area walked about fully armed and that the insurgents had access to rocket-propelled grenades (RPGs) and PK machine guns (general purpose Russian-made weapons which are commonly used by insurgents in Afghanistan).75 He stated that Kalta led a group of 14–20 fighters and Neimatullah was one of his associates. It was assessed that the targets might use RPGs to defend their compounds, in the expectation that reinforcements would come from surrounding areas to inflict casualties on coalition ground forces.
48. The concept of operations for Operation Burnham assessed that the overall risk to the coalition force was “high” due to the likely insurgent presence, the assessed insurgent response and the absence of an ISAF or Afghan security force presence in the area.76 Given the available intelligence and the risk assessment, air support was critical and, as a practical matter, Operation Burnham could not have proceeded without it.
49. As part of the planning for the operation, the Ground Force Commander and the Troop Commander assessed the areas around the valley in which the operation was to take place and identified the most likely areas where a threat could materialise. These were areas where troops would be most vulnerable to incursion by any enemy forces approaching from surrounding areas. Air assets would be tasked to monitor those areas and to protect ground troops, by firing their weapons if necessary. Although, as previously noted, it was anticipated that insurgent reinforcements might appear from the surrounding area, nothing was said about the direction from which they might come, perhaps because the available intelligence was thought to be insufficiently specific in this respect.

    Protection of civilians

50. Turning to the question of how the planning addressed the protection of civilians, we note that the concept of operations indicated that the overall risk of civilian casualties was “low”.77 The basis for that assessment is not stated explicitly. The Inquiry’s military expert suggested this risk assessment would likely have been based on a number of mitigation strategies developed in the course of planning. These include: knowledge obtained from earlier “pattern of life” assessment; conducting the operation at night when the majority of civilians would be inside their houses and coalition forces had a tactical advantage; the fact that pre-emptive or other deliberate air strikes were not planned; the intention to conduct a “soft knock call out”; and the employment of escalation of force measures as prescribed by the Commander ISAF.78 Witness evidence aligned with that assessment and it is further supported by Sir Angus Houston’s response to questions from Counsel for the Villagers.79
51. Reflecting an awareness of the need to protect the civilian population, the orders to ground troops before the operation noted that:
    1. troops were to be prepared to encounter women and children when clearing the compounds of interest;80
    2. if any casualties occurred, the troop medic was to provide medical support while medical evacuation was arranged, if required;81 and
    3. in the event of a death—whether enemy or civilian—troops were directed to collect DNA samples and search and photograph the body (while treating the person with dignity).82
52. A number of directives or requirements relevant to the protection of civilians applied to the operation.83 These included ISAF’s Night Raids Directive, a set of standards or requirements known as “Karzai’s 12” and the Tactical Directive issued by General David Petraeus in August 2010. As noted already, a concept of operations submitted to ISAF needed to address how the Night Raids Directive and Karzai’s 12 would be complied with, as ISAF command took this into account when deciding whether to approve the operation. In addition, “pattern of life” analysis was also relevant to the protection of civilians. We deal with each in turn.

    ISAF’s Night Raids Directive

53. Civilian casualties resulting from night raids appear to have been relatively low84—possibly because civilians were more likely to be inside their homes at night. Despite this, the local Afghan population perceived night raids as more invasive than daytime raids and strongly opposed them. The Night Raids Directive issued by General Stanley McChrystal on 5 March 2010 began by saying that it was preferable for ISAF forces to explore all other feasible options before conducting a night raid targeting compounds and residences.85 The directive required Afghan National Security Forces to be involved in all night raids, including the planning process, and provided that Afghan forces should take the lead in entering compounds and conducting searches. The directive also required troops to record all property seized or damaged, and to leave a receipt with instructions for claiming compensation. Overall, the directive did not forbid the use of night operations; rather, it caused coalition forces to be more thoughtful and discriminating about their use and, where they were used, imposed conditions on how they should be conducted.
54. The reasons given in the Operation Burnham concept of operations for conducting a night raid were to minimise the potential for civilian casualties, to minimise the threat of an insurgent response and to increase the chances of successfully detaining Kalta.86 The document indicated the CRU would be involved in the operation, although it is clear from the orders issued to troops that it was not intended the CRU would take the lead in entering compounds—that would be done by TF81 personnel. Witnesses explained this was because of the security situation and the fact that the CRU personnel were not used to participating in this type of operation (that is, in a remote location where insurgents were expected to be present). This was not strictly in accordance with the Commander ISAF’s intent as expressed in the Night Raids Directive. However, this was not a violation of the directive as there was justifiable reason for not adhering strictly to it—ISAF personnel were always permitted to ensure their own force protection and self-defence.

    Karzai’s 12

55. As noted earlier, Karzai’s 12 was a set of conditions agreed between General Petraeus and Afghan President Hamid Karzai in an attempt to ensure cultural sensitivity during operations and keep civilian casualties to a minimum. The conditions included coordinating with local officials and tribal elders, ensuring troops received training on local culture and customs, attempting a “soft knock” before forcing entry to a compound and using interpreters.87
56. The concept of operations explained that coordination with local officials had not been conducted due to the potential for the mission to be compromised.88 However, troops had received training in local culture and customs and a “soft knock” or call out would be used before entering compounds. Again, while satisfying all 12 conditions was preferable, an operation could still proceed where some conditions were not met, provided there was a sufficient explanation for the non-compliance.

    COMISAF’s Tactical Directive

57. General Petraeus, as the Commander ISAF, issued a Tactical Directive in August 2010 prior to Operation Burnham.89 The aim of the directive was to “reduce the loss of innocent civilian life to an absolute minimum and reinforce the concept of disciplined use of force in the fight against the insurgency”. Among the protections set out in the directive were fire control measures directing when and under what conditions commanders could use fires.90 The directive stated that a commander authorising a strike must determine that no civilians were present. If civilians were present, the directive still permitted the use of fires in certain circumstances, one of which was self-defence. The directive acknowledged that protecting the Afghan people required killing, capturing, or “turning” insurgents. The directive also encouraged partnering with Afghan troops during the planning and execution stages of operations. It said partnering would help to avoid misunderstandings or ignorance of local customs or behaviours (which could result in casualties), ensure greater situational awareness, alleviate the anxiety of the local population and build confidence in Afghan security forces.

    Pattern of life analysis

58. In the published literature, pattern of life analysis is most commonly associated with targeted killings by rockets fired from drones. Before a targeted drone strike, the targeting force carries out a pattern of life analysis of the target’s habits and patterns of daily life. In addition, the targeting force gathers information about the patterns of life of civilians in the particular locality. This enables the force to plan the strike in a way that gives the maximum chance of striking the target while minimising the risk of civilian casualties.
59. In its report Troops in Contact: Airstrikes and Civilian Deaths in Afghanistan, Human Rights Watch made the point that most civilian casualties from air strikes in Afghanistan were caused in what it described as unplanned situations, as when troops under attack from insurgents called in air strikes for tactical support.91 Human Rights Watch said that where air strikes were planned, civilian risk mitigation procedures could be undertaken in advance and these minimised the risk of civilian casualties. One of these procedures was “pattern of life” analysis. The report said:92

    Planned attacks allow the US and NATO to use civilian risk mitigation procedures, including formal risk estimates to model and minimize civilian casualties. This includes a “pattern of life analysis”, which looks for civilians in the area for hours or days before an attack using “eyes on the target” ranging from ground observers to technical reconnaissance.

    The purpose of such analysis is to build a picture of daily life in the relevant area—in relation both to the target and any cohabiting family members, and to civilians more broadly (women and children in particular)—and to identify any facilities in the immediate area that are granted special protections under International Humanitarian Law, such as cultural and religious sites.93

60. While the literature primarily focuses on targeted strikes, pattern of life analysis is also conducted before other types of operations (although the analysis required is particularly detailed in the case of targeted strikes). Such analysis assists in building the intelligence picture of the situation likely to be encountered during the operation and in assessing the risks to troops. It also facilitates planning on how to reduce the likelihood of civilian casualties and other collateral damage. Pattern of life analysis is conducted through a variety of means, including intelligence reporting and aerial surveillance.
61. Pattern of life analysis was used in the planning of Operation Burnham. Before the operation, a drone conducted various reconnaissance trips over the objective area. The purpose of this reconnaissance was to map the layout of Kalta’s and Neimatullah’s compounds so as to gain an understanding of the terrain. This included identifying any areas that may be at risk of collateral damage and observing whether (and where) women and children were present. The drone observed numerous people coming and going from the targets’ compounds.94 Most were men, although children and a probable female were seen at Neimatullah’s residence. Possible mosques were also identified.95 The drone imagery was not detailed enough to identify specific individuals (for example, to confirm whether the targets were at the compounds).
62. In summary, the pattern of life analysis (including intelligence collected) confirmed that the targets’ compounds were located in a village, Naik, and that civilians were likely to be in the compounds, or in their general vicinity. This was reflected in the orders given to the troops, which directed them to be prepared to encounter women and children when clearing compounds.96

    Arrangements between air assets and ground forces

63. Because Operation Burnham was to occur in a remote and inaccessible location, and because of the anticipated risk of an insurgent response, air asset support was essential to allow the safe conduct of the operation and to ensure its timely completion. The air assets required included:97
    1. two Chinook helicopters to transport the main ground force;
    2. two UH-60 Blackhawk utility helicopters to transport the command group and the aerial response force;
    3. two Apache attack helicopters (referred to as the Air Weapons Team) to escort the Chinooks and to provide, in conjunction with an AC-130 Spectre gunship, security for ground troops; and
    4. a remotely piloted aircraft (drone) to monitor developments on the ground before, during and after the operation.
64. As we have said, TF81 submitted a formal request for air support to Regional Command North on 11 August 2010. Ultimately, however, the supporting air assets came primarily from Regional Command East. It appears that the timing of the operation was in part determined by air asset availability: final confirmation of air asset support for the operation was only received on 20 August 2010. As noted above, the Ground Force Commander, the JTAC and others briefed the Air Mission Commander on 19 August, explaining the Ground Force Commander’s intent for the mission and the areas where air support would be required.
65. The air assets were contributed by the United States and operated by United States air crew.98 They would provide support for the TF81 mission but remained under the command and control of their own commanders and subject to their own national rules of engagement.99 As one of the air crew put it: “The Air Mission Commander has the final clearance of fires for air assets conducting the mission.”100
66. In practical terms, this meant the air assets would coordinate with TF81 and seek to act in a manner consistent with the Ground Force Commander’s intent. They would seek clearance before firing at a target, particularly in areas near ground troops. However, they did not require approval from the Ground Force Commander to fire and could choose to engage without clearance if, for example, it was necessary for their own safety or the safety of ground forces. They would act in accordance with the orders of their own Air Mission Commander and could engage a target if they considered it was in accordance with their national rules of engagement, which governed their conduct at all times (whether or not they had clearance to fire).
67. As we describe in more detail in chapter 8, the United States Forces—Afghanistan (USFOR-A) conducted an investigation under Army Regulation 15-6 (the AR 15-6 investigation) into Operation Burnham.101 This was conducted in the last two weeks of September 2010. As part of the AR 15-6 investigation, the aircrews of the Apache helicopters were interviewed. In the course of the interviews, the air crew described the information they were given during their pre-operation briefings.102 Although there are some discrepancies in the descriptions given, it seems from analysing them together that the air crew were told they were the first coalition forces to go to the area in 10 years and that there was likely to be “enemy resistance”.103 The briefings addressed patterns of life in the area. Some air crew said that they knew civilians would be in the locality and that they had to be cautious with fires, while others said they did not remember any discussion about women and children. One of the air crew said: “During the brief it was stated that anyone leaving the objective was declared hostile.”104 It is not clear what the “objective” referred to was in this particular context, although that term was often used to refer to the buildings to be entered and searched during the operation.
68. The complex task of communicating with the air assets and coordinating their movements fell to the JTAC, a TF81 Corporal who was specifically trained for that role. The JTAC would be positioned with the Ground Force Commander during the operation and would relay to the Ground Force Commander any requests that the air assets made for clearance to engage a target. The JTAC would then relay the Ground Force Commander’s decision to the air assets. The Ground Force Commander retained ultimate responsibility for clearing engagements, although he would consider advice from the JTAC and other sources before doing so.
69. While the Ground Force Commander and the JTAC received information from a variety of sources, inevitably, given the location, the air assets would have a better view of what was happening on the ground during the operation than they did. Accordingly, the Ground Force Commander and the JTAC would have to rely in part on information reported by the air assets to assess whether a proposed target was directly participating in hostilities and whether the risk of collateral damage was within acceptable limits, so that clearance to engage could properly be given.105 Even if clearance to engage was granted, the air assets would still need to be satisfied that the engagement complied with their rules of engagement and that friendly forces would not be endangered. In some cases, the Ground Force Commander might explicitly grant clearance on a conditional basis—for example, by saying that air assets were cleared to engage if the target was positively identified and there were no collateral damage issues. As we discuss in chapter 4, this occurred on one occasion during Operation Burnham.

    Information about the whereabouts of the targets

70. As discussed above, Operation Burnham had been planned based on intelligence suggesting that both Kalta and Neimatullah had residences in Naik. However, intelligence also showed that they moved about frequently, whether within Tala wa Barfak, into an adjoining province such as Bamyan or even to Pakistan.106 For the detention operation to be successful, TF81 would need up-to-date intelligence indicating that the targets were likely to be at their compounds (or in the general vicinity) at the time of the operation. The concept of operations indicated there would be a “HUMINT trigger”—that is, TF81 would be relying on human intelligence reporting about the location of the targets.107
71. As we have noted, intelligence reporting in the days leading up to the operation indicated that the targets were in Naik. As far as we have been able to establish, this was the confirmatory intelligence TF81 relied on when beginning the operation. For various reasons evident on the documents we have received, it was not possible in the circumstances to obtain positive identification of the targets at their compounds immediately before the operation. Given that the operation aimed to detain the targets and did not involve any planned strikes, we accept that reliance on human intelligence in this way was justifiable. Further, as we discuss later, the intelligence received was largely accurate.108

    Capture of suspected insurgents

72. The concept of operations indicated that the CRU would be responsible for making any arrests in accordance with Afghan warrants during the course of the operation, and detainees would be handed over to the Afghan Ministry of the Interior for prosecution.109 However, the orders of the Ground Force Commander and the Troop Commander for the operation stated that the assault force (which did not include any CRU) would enter the buildings and detain the targets if they were present.110 Any detainees would be taken first to the NZPRT’s base in Bamyan for identification, then—if they were confirmed as the targets—to TF81’s base in Kabul before being handed over to the Afghan National Directorate of Security in Kabul.111 We return to the issue of detention in chapters 10 and 11.

    Operation Nova

73. Although the coalition forces failed to capture Abdullah Kalta or Maulawi Neimatullah during Operation Burnham, TF81 considered that the Tala wa Barfak insurgent network had been disrupted by the loss of materiel and personnel during the operation. Despite this assessment, by mid-September, intelligence reporting indicated that Neimatullah was living in his house in Tirgiran Valley and was “looking for revenge” for Operation Burnham.112 Accordingly, TF81 senior leadership began planning a return operation to Tirgiran Valley to conduct another deliberate detention operation targeting Neimatullah only, relying on the same warrant that the Afghan Ministry of Interior issued before Operation Burnham.
74. Between at least 23 and 26 September 2010, TF81 leadership engaged in discussions with Regional Command North seeking permission for another operation in Tirgiran Valley to “remove or at least disrupt” Neimatullah.113 Regional Command North appeared concerned that the Taliban would claim civilian casualties following the operation regardless of whether or not they occurred. We do not have a record of the exact date Regional Command North granted permission for the operation, but this seems to have occurred by 29 September 2010, when the Chief of Defence Force approved a deliberate detention operation in Tirgiran Valley targeting Neimatullah on the night of 2–3 October 2010.114 On 1 October 2010, the Chief of Defence Force and Director of Special Operations briefed the acting Prime Minister on the operation.115 The planning process and the risk assessment for Operation Nova were similar to those for Operation Burnham, which we have described above. We will not repeat that detail here. We simply note that we did not see anything in the material available to us to indicate there were inadequacies in the planning undertaken.
75. The operation was to involve a combined TF81 and CRU force landing in the valley to search A3 for Neimatullah.116 The United States again provided transport and supporting air assets. The most significant difference to Operation Burnham was that the helicopter landing zone was to be in the stretch of valley about halfway between A1 and A3, rather than to the north of Khak Khuday Dad as it had been for the previous operation.
76. Overall, the Inquiry is satisfied with the planning process for Operations Burnham and Nova.

Figure 3:
Tirgiran Valley, Baghlan province, Afghanistan

Satellite image dated 25 April 2010, source: Digital Globe