(OMR0005)

Written evidence submitted by Max Brookman-Byrne

Lecturer in Law at the University of Lincoln and steering committee member of Drone Wars UK.

The reporting of air strikes during Operation Shader

1. This evidence is produced as a result of on-going research into air operations against non-state actors and international law. This evidence has an international law orientation, but its recommendations go beyond that and emphasise issues of transparency generally. The basis of the evidence is a soon-to-be-published report funded by the NGO Drone Wars UK which analysed the narrative reporting of air strikes carried out by the UK in Iraq and Syria.

EXECUTIVE SUMMARY

2. This evidence focuses on the publication by the Ministry of Defence (MOD) of narrative accounts of air strikes carried out by the Royal Air Force (RAF) as part of Operation Shader. These accounts are brief summaries of air strikes based on edited flight reports compiled by pilots after missions have been flown.

3. While the narrative accounts are a step towards transparency around Operation Shader, they fall short in a number of key areas.

4. There is a disconnect between the statistical data surrounding these air strikes and the narrative accounts, such that the impression given is of information being withheld.

5. The factual information contained within the narrative accounts varies in level of detail between individual reports. While many demonstrate the basis upon which a person or object was targeted in an air strike, 39% of reports do not, missing out important information. Failing to provide sufficient information means that it is not clear from the reports whether or not an air strike adhered to international law.

6. Terms such as ‘terrorist’ and ‘extremist’ are prevalent within the reports. These are inappropriate in this context, bearing no legal meaning and obscuring the actual facts that made an individual a lawful target.

7. Details of how risks to civilians have been mitigated during air strikes are absent from all but 3% of the reports, providing an incomplete picture of operations. The absence of this information undermines the MOD’s claim that civilians have (in all but one instance) not been harmed by air strikes under Operation Shader.

8. On the basis of these findings, this evidence makes the following recommendations:

1. There must be a commitment to the provision of a narrative account of every air strike. The apparent exclusion of some air strikes from the reports is extremely problematic in terms of transparency and must not be allowed to continue.

2. Sufficient facts should be given in each report to provide a picture of why a given individual is targeted, detailing, albeit briefly, what it was about their conduct that demonstrated direct participation in hostilities, as required under international humanitarian law (IHL) and the British understanding of relevant law, detailed in the Joint Services Manual.

3. Terms such as ‘terrorist’ and ‘extremist’ are unhelpful. They have no legal function and actively serve to obscure the basis upon which an air strike was carried out. They should not be used as a routine part of the narrative reporting of air strikes.

4. Going forward, the presence of civilians in relation to air strikes must be acknowledged and depicted in every narrative account. This should go beyond simply stating that no civilians were harmed, and actively include a statement as to the proximity of civilians to the strike.

INTRODUCTION

9. This evidence focuses on the way that the Ministry of Defence (MOD) has published a narrative account of air strikes carried out by the Royal Air Force (RAF) as part of Operation Shader.

10. These reports are short accounts of the air strikes carried out on a given day. Beginning with the start of Operation Shader in August 2014, the MOD produces updates periodically (between twice a week and once a month depending on the level of activity) which are then edited and compiled by the MOD into monthly summaries and are available on the MOD website.[1] The current evidence is primarily based on an analysis of reports between August 2014 and December 2017.

11. The narrative reports available on the MOD website vary in length from 8 to 138 words. Generally the reports detail the broad location of a strike, the person or object targeted, the aircraft involved and the munition used, although, as will be shown, the level of detail can vary considerably.

12. The process of how these updates are written has been gleaned through a Freedom of Information (FOI) request made to the MOD. After each mission, the pilots write a flight report, with additional detailed reports where weapons have been fired. These are classified for reasons of operational security as well as containing a large amount of military jargon, so they are reviewed and a new narrative version of each attack is written separately.[2]

13. That the reports represent an attempt at transparency around air operations is positive, but they fall short in many respects, with some strikes apparently being left out, and the reports themselves often lacking clarity. At times the reader is left to assume that a strike was lawful, though with insufficient data to confirm this assumption.

14. The evidence will proceed as follows: first, problems in the reconciliation of statistical and narrative accounts of the Operation Shader air strikes will be discussed. Second, a brief overview will be given of relevant international law applicable to Operation Shader air strikes. Third, it will be demonstrated that nearly half of the reports available online fail to provide sufficient information to demonstrate, on the face of it, that the strike adhered to relevant international law. Fourth, the presence of civilians within the report will be discussed, with emphasis on their almost total absence. The report will conclude with a summary, including recommendations.

INCONSISTENCIES BETWEEN STATISTICAL AND NARRATIVE ACCOUNTS

15. While it is assumed that the reports present the full range of operations carried out as part of Operation Shader, there are inconsistencies between the narrative reports and statistical data.

16. For instance, MOD updates in March 2018 reported three separate UK Reaper attacks, targeting an enemy drone, two people on a motorbike and a ‘group of terrorists’.[3] However, data released under FOI rules show that UK Reapers in fact launched 19 Hellfire missiles in March 2018,[4] which appears to be more than are accounted for in the narrative reports.

17. Additionally, the highly publicised strike that killed British citizens Reyaad Khan and Ruhul Amin in Raqqa in August 2015 do not appear within the MOD’s narrative account of Operation Shader.

18. As a result of this, the impression is given that information in the narrative reports has been left out, which severely undermines the extent to which the narrative reports promote transparency.

19. The government should commit to providing a narrative account of every air strike it carries out in order to promote and maintain transparency. Publicising some while apparently suppressing others has a seriously damaging effect on the overall picture produced.

INTERNATIONAL LAW AND THE NARRATIVE REPORTS

Relevant international law

20. There are many areas of international law implicated by a state’s use of force extraterritorially. However, as the MOD’s narrative reports detail the use of force during an armed conflict,[5] the most immediately relevant is ‘international humanitarian law’ (IHL) which governs how force is used in military engagements. Before embarking on a discussion of the reports and IHL, a quick outline of relevant IHL will be given.

21. A ‘non-international armed conflict’ (NIAC) is a conflict between a state and a non-state actor (NSA), such as ISIS. This is, therefore, the nature of the conflict of which Operation Shader forms a part. During a NIAC, individuals who are not members of a state’s armed forces continue to be recognised as civilian by IHL, meaning that they are generally protected[6] and may not be made the object of attack.[7] However, civilians will lose this protection if, and for as long as, they take a direct part in hostilities.[8] This has the result that members of non-state armed groups remain legally classified as civilians due to the fact that they are not members of state armed forces, but they will lose their protection if they act in a manner that counts as direct participation in hostilities.

22. The type of conduct that will constitute direct participation in hostilities and thereby allow a person to be targeted is controversial and is not universally agreed upon. This idea of direct participation has been interpreted by the International Committee of the Red Cross (ICRC) such that members of non-state armed groups may be targeted at any time due to having lost their civilian protection, but ‘membership’ is specifically defined as the carrying out of a ‘continuous combat function’.[9] Therefore, individuals that a lay person might call a ‘member’ of such a group may not be a member in the eyes of international law under this interpretation, due to the function they carry out for the group, e.g. if they deal solely with propaganda or financial transactions.[10] Conversely, for instance, the United States has interpreted ‘membership’ on the basis of an individual’s formal tie to a group, regardless of whether they have carried out any hostile act.[11]

23. The UK appears to adopt a more narrow interpretation of direct participation. The Joint Service Manual on Armed Conflict (JSM) does not deal with individuals who are members of non-state armed groups, instead asserting that under IHL civilians may not be targeted unless they are taking a direct part in hostilities.[12] The question of whether an individual is directly participating is viewed as ‘a question of fact’, illustrating that the UK’s approach only contemplates attacks against those who are actually participating,[13] thereby excluding the possibility of strikes against individuals purely due to their formal membership of an armed group. Importantly, the JSM states that ‘use of the words “are taking” emphasizes that a potential or future fighter may not be attacked as such’.[14] This presents a picture of the UK as maintaining a robust interpretation of this aspect of IHL which serves to limit the incidence of civilians being mistakenly targeted.

24. Therefore, under the UK’s interpretation of IHL, it is necessary for a person to be carrying out acts that can, in and of themselves, be described as direct participation in hostilities. Under this interpretation of IHL it is not permissible to target a person solely due to their membership of a group. In providing an overview of air strikes conducted and detailing each of their facts and context, the reports have the capacity to demonstrate why (at least in a broad sense) each air strike was lawful, by depicting the conduct that demonstrates direct participation in hostilities.

The extent to which lawfulness can be determined from the reports

25. Unfortunately, while many of the reports of air strikes during Operation Shader contain sufficient factual information to make an apparent case that a strike was carried in accordance with the relevant law on targeting, a large portion do not provide sufficient information to make such a determination. Of 677 reports analysed[15] 263 (39%) did not contain sufficient information to enable a basic, broad determination of lawfulness under IHL. When these figures are broken down by aircraft type it is interesting to note that 45% of reports of drone strikes did not contain sufficient information (162 out of 361 reports), while 32% of conventional air strikes (e.g. using Typhoon or Tornado jets) also failed to provide adequate detail. This suggests a tendency for lower transparency in the reporting of drone strikes as opposed to those of conventional aircraft.

26. To put the failure to provide sufficient information into context, two example reports are given. On 28 September 2017, a Reaper drone is described as having attacked a sniper team, which is clearly a lawful target, but also ‘another group of terrorists, [and] a Daesh held building’.[16] Similarly, on 15 September 2017 a Reaper is described as having ‘conducted two attacks … with Hellfire missiles against terrorists moving through the area.’[17] There is no indication given in these reports on what basis it was decided that the people targeted could be lawfully killed. There is no reference to surrounding facts that demonstrate their participation in hostilities. It is not even stated that the individuals targeted were armed (which, regardless, may well not be sufficient to render someone a lawful target) much less that they were engaged in hostile acts.

27. A related problem is the use of terms like ‘terrorist’ and ‘extremist’ within the reports. These terms are very commonly used to describe targets, comprising 23% of targets depicted in reports of drone strikes and 8% of targets depicted in reports of conventional air strikes.

28. These terms suggest that a person is a lawful target while actually giving no information whatsoever that suggests why that person was targetable under IHL. There is no category of ‘terrorist’ or ‘extremist’ under IHL and they are entirely inappropriate for use within narrative accounts of air strikes. By standing in for the facts of an individual’s participation in hostilities these terms actively stymie transparency.

29. It is not suggested that the MOD should provide a watertight case detailing the lawfulness under IHL of each strike, but they should present sufficient relevant facts to at least support the presumption that the strike was lawful.

THE PRESENCE OF CIVILIANS IN THE REPORTS

30. Related to the issue of international law but going beyond that is the issue of the presence of civilians within the reports.

31. Civilians should be referenced in the reports, as doing so demonstrates a number of things. First, where the facts allow, it shows that civilians have not been killed or hurt in a strike. Second, it demonstrates that the impact of a strike upon civilians is a key consideration when deciding to carry out a particular strike—it shows that the civilian population is taken seriously. Third, it creates the presumption that obligations under IHL in relation to civilians have been taken seriously.

32. Put simply, the inclusion of civilians within the reports—in terms the impact of the strikes upon them, the steps taken to avoid harming them, or even the fact that there were no civilians nearby—demonstrates that they are taken seriously.

33. Within the narrative reports of Operation Shader, reference to civilians in this manner occurs in less than 3% of accounts (19 out 677 analysed). Generally, these instances refer to a drone crew checking an area for civilians before firing, though others are less explicit, instead referring to, for instance, the ‘crew wait[ing] patiently until the targets were in open countryside’.[18] There are reports that refer to civilians more generally, i.e. in relation to them being impeded by a checkpoint which is then targeted,[19] but this is not the same as a clear statement of how risks to civilians were minimised.

34. It is highly concerning that 97% of reports make no mention of possible risk of harm to civilians. Not only does this provide no indication as to whether or not civilians have been negatively affected by air strikes, it also fails to demonstrate that these air strikes have been carried out with requisite precaution, which is an additional requirement of IHL.

35. This is particularly problematic in view of the blanket assertion by the British Government that there is ‘no credible evidence’ of civilian casualties,[20] held until the recent acknowledgement of the accidental killing of a civilian.[21] It is also incompatible with the volume of civilian casualty reports from those present in regions affected by the air strikes, compiled by groups such as Airwars.[22] As a recently retired senior RAF officer stated earlier this year, it is not ‘credible to the average listener that we have not caused any civilian casualties’.[23]

36. The removal of civilians from the narrative portrayal of air strikes under Operation Shader is highly problematic. It provides an incomplete picture of the strikes, both from a legal and transparency perspective. Transparency about the presence or not of civilians would increase the credibility of UK reporting.  It cannot be presumed that civilians are simply not harmed by air strikes.

37. Going forward, the presence of civilians in relation to air strikes should form a part of every narrative account. This should go beyond simply stating that no civilians were harmed, and actively include a statement as to the proximity of civilians to the strike.

CONCLUSIONS AND RECOMMENDATIONS

38. That the MOD produces narrative accounts of its air strikes under Operation Shader is very positive, but the overwhelming picture that emerges when these reports are analysed in detail is one of an attempt at transparency that often falls short.

39. The reports are a step towards ensuring that UK military operations are capable of proper scrutiny and it is one that few, if any, other states also carry out. However, scrutiny is only possible where sufficient information is given, and in a significant number of cases this does not happen.

40. It is recommended that the MOD changes its approach to the reports going forward in the following key ways:

1. There must be a commitment to the provision of a narrative account of every air strike. The apparent exclusion of some air strikes from the reports is extremely problematic in terms of transparency and must not be allowed to continue.

2. Sufficient facts should be given in each report to provide a picture of why a given individual is targeted, detailing, albeit briefly, what it was about their conduct that demonstrated direct participation in hostilities, as required under IHL and the British understanding of relevant law, detailed in the Joint Services Manual.

3. Terms such as ‘terrorist’ and ‘extremist’ are unhelpful. They have no legal function and actively serve to obscure the basis upon which an air strike was carried out. They should not be used as a routine part of the narrative reporting of air strikes.

4. Going forward, the presence of civilians in relation to air strikes must be acknowledged and depicted in every narrative account. This should go beyond simply stating that no civilians were harmed, and actively include a statement as to the proximity of civilians to the strike.

28 June 2018