Airwars v Information Commissioner & Anor

1. On 26 March 2018, the United Kingdom conducted a Reaper drone strike in the Syrian Euphrates valley. The strike hit the intended target, an Islamic State tactical unit, but also killed a civilian. The Secretary of State for Defence, Gavin Williamson MP, gave a written statement to Parliament https://questions-statements.parliament.uk/written-statements/detail/2018-05-02/HCWS665 that included the following: As part of our counter-terrorism strategy, the UK is playing a leading role in the Global Coalition to defeat Daesh — a unified body of 75 members. We have committed nearly 1,400 military personnel to the region to provide support to local partners. In the air, the RAF has conducted more than 1,600 air strikes in Iraq and Syria – second only to the US – and provides highly advanced intelligence, surveillance and reconnaissance to Coalition partners. These strikes are undertaken in the collective self-defence of Iraq as part of the global Coalition to defeat Daesh, and at the request of the Government of Iraq. On the ground, British soldiers have trained over 60,000 members of the Iraqi Security Forces in engineering, medical, counter-IED and basic infantry skills. As a result of the Coalition’s action, Daesh has lost more than 98 percent of the territory it once occupied in Iraq and Syria, and 7.7 million people have been liberated from its rule. We do everything we can to minimise the risk to civilian life from UK strikes through our rigorous targeting processes and the professionalism of UK Service personnel. It is therefore deeply regrettable that a UK air strike on 26 March 2018, targeting Daesh fighters in eastern Syria, resulted in an unintentional civilian fatality. During a strike to engage three Daesh fighters, a civilian motorbike crossed into the strike area at the last moment and it is assessed that one civilian was unintentionally killed. We reached this conclusion after undertaking routine and detailed post-strike analysis of all available evidence. There are limits on any further details that can be provided given ongoing operations and consequent national security issues. As with any serious incident the wider Coalition also conducts its own investigation and will report in due course. These events serve to remind us of the consequences of conflict and of the heavy price that the people of Syria have paid. It reminds us that when we undertake military action, we must do so knowing that it can never be completely without risk. Such incidents will not weaken our resolve to defeat Daesh and rid the world of its poisonous ideology of hate and intolerance. The UK’s commitment to the Global Coalition against Daesh and to the people of Iraq and Syria will remain as strong as ever. 2. Airwars was founded in 2014, and describes itself https://airwars.org/about/team as a not-for-profit transparency watchdog that tracks, assesses, archives and investigates civilian harm claims in conflict-affected nations. It claims, we consider with some justification, to be “a leading authority on conflict violence as it affects civilian communities”. 3. Joe Dyke, Airwars’ Head of Investigations, submitted a request to the Ministry of Defence under the Freedom of Information Act 2000 ("FOIA"). The following information was Requested: (i) All details on the March 26, 2018 strike including close geolocation, time of day and other relevant information. (ii) All documents assessing civilian harm in the incident. 4. This appeal proceeds only in relation to part (ii) of the request, and we need say no more about part (i). On 7 April 2021 the MOD responded that the requested information was held, but was exempt from disclosure under FOIA section 24, concerning national security, section 26, concerning defence, and section 27, concerning international relations. These are all qualified exemptions, meaning that they will only entitle the MOD to withhold the information if the public interest in maintaining the exemptions outweighs the public interest in disclosure. The MOD considered that it did. The MOD’s response was maintained following an internal review, in which it additionally relied on the exemption at section 40(2), concerning personal data, to withhold the names of junior civil servants and service personnel. 5. Mr Dyke complained to the Information Commissioner, whose Decision Notice https://ico.org.uk/action-weve-taken/decision-notices/2022/10/ic-114449-b7p7/ of 19 October 2022 agreed with the MOD’s position. Airwars has accordingly exercised its right of appeal to the Tribunal. Since these proceedings began, the MOD has changed its position such that it now also relies upon section 23(1), concerning information supplied by, or that relates to, specified security bodies, and section 24(1) in the alternative. This is in relation to all information in the scope of the request. In relation to some information the MOD now also relies upon section 42(1), concerning legal professional privilege. The appeal 6. We should record that the proper title of the appellant in these proceedings is Airwars. Making the request for information and the complaint to the Commissioner, and lodging this appeal, was all done by Mr Dyke expressly on Airwars’ behalf. The rights contained within FOIA extend to an unincorporated association, which can therefore be the legal party to the appeal notwithstanding its own lack of legal personality. 7. Early on in the proceedings, the MOD made an application to provide closed evidence pursuant to rule 14(6) and for there to be a closed session during the hearing from which everyone but the two respondents would be excluded. In response, Airwars applied for a special advocate to be appointed to represent its interests or, in the alternative, for representation by the Commissioner during the closed session to be required. Judge Neville decided that the latter course was necessary to ensure that proceedings were fair, and the reasons for that decision are at Annex 1 to these reasons. The hearing 8. In the open part of the hearing, we heard evidence from Mr Dyke on behalf of Airwars and, on behalf of the MOD, Mr Alexander Oliver, who is Deputy Director responsible for operational policy in its Security Policy and Operations Directorate (now its Military Strategy Operations and Commitments Directorate). He confirmed the content of a witness statement by his predecessor in that role, Ms Katherine Costelloe. Each witness was cross-examined, and questions asked by the Tribunal in clarification. We shall not set out the witnesses' evidence save where necessary to explain our conclusions. 9. After the parties had concluded their evidence, we convened a closed session from which everyone but the two respondents and their representatives were excluded. Mr Oliver gave further evidence, we received further submissions, and considered additional documentary evidence. Necessarily, we were also provided with the requested information to determine the relevant issues for ourselves. 10. In adopting that closed procedure, we are acutely aware of its derogation from two key principles: first, natural justice, a party having a right to know the full case against it, and to test and challenge that case; and second, open justice, fundamental to the dispensation of justice in a modern, democratic society. As held in Browning v Information Commissioner & DBIS [2014] EWCA Civ 1050, such derogation is nonetheless permissible in a FOIA appeal. Without it, the appeal process would either reveal the requested information, defeating the whole purpose of the appeal, or risk the adverse consequences that the MOD claim would result from its disclosure. In cases where it applies, rule 14(9) also obliges the Tribunal to ensure that information is not disclosed contrary to the interests of national security. 11. We have been careful to minimise any disadvantage to Airwars in accordance with Browning. Where Mr Oliver was unable to answer a question in public he indicated as such, and the topic was explored in closed session together with other matters suggested by Airwars. Once the closed session had ended, a gist containing as much detail as possible of what had taken place was provided so that submissions could be made on Airwars 4 behalf. Keeping the bounds of what evidence and argument must be withheld carefully under review, we are satisfied that the proceedings have been fair and conducted in public so far as could reasonably be accomplished. In doing so, we express our gratitude to Mr Goss for fulfilling the Commissioner 4 s role as "guardian of FOIA44 during the closed session; as was anticipated in Browning, this was vital to ensuring fairness in the absence of a special advocate. 12. The principles above also apply to some of reasons for deciding the appeal which, when taken together, cannot begiven publicly without revealing the requested information or otherwise giving rise to harmful consequences. There is a closed annex giving those additional reasons. 13. In deference to the mutually agreed importance of the issues, we have set out our assessment of the principal points made by the parties. Inevitably we have not been able to address all the evidence and argument we have heard, but have taken it all carefully into account. The Tribunal's approach 14. The Tribunal both decides the relevant facts and decides for itself whether the provisions of FOIA have been correctly applied. But it does not start with a blank sheet: the starting point is the Commissioner's decision., to which the Tribunal should give such weight as it thinks fit in the particular circumstances. The proceedings are inquisitorial, save that the Tribunal is entitled to respect the way in which the issues have been framed by the parties. 15. When assessing the public interest balancing test for a qualified exemption, the Tribunal considers how matters stood at the date of MOD'S response to the request on 7 April 2021. Nonetheless, Airwars seeks to rely on "evidence post-dating the MOD's refusal that further reinforces [Airwars'] contention that transparency and accountability regarding civilian harm policies and practices has consistently resulted in significant reforms and the saving of civilian lives." The basis upon which the Tribunal may do so was put forward by Mr Perry as arising from the judgment of the Supreme Court in R. (Evans) v Attorney General [2015] UKSC 21 at [73], per Lord Neuberger: 73. However, although the question whether to uphold or overturn (under section 50 or sections 57 and 58) a refusal by a public authority must be determined as at the date of the original refusal, facts and matters and even grounds of exemption may, subject to the control of the Commissioner or the tribunal, be admissible even though they were not in the mind of the individual responsible for the refusal or communicated at the time of the refusal to disclose (i) if they existed at the date of the refusal, or (ii) if they did not exist at that date, but only in so far as they throw light on the grounds now given for refusal… 16. We disagree that Lord Neuberger, in the second limb of that sentence, opened the door to entirely new post-response events simply because they are relevant, in a general sense, to the public interest factors in play. We are bound by Montague v IC & DBT [2022] UKUT 104 (AAC) and Smith v IC [2022] UKUT 261 to treat the date of response as the fixed point in time at which the public interest is assessed. 17. Evidence that comes into existence after the date of the response can be considered only insofar as it sheds light on the public interest factors as they existed at that date. An illustrative example can be found in Corderoy v IC & MPS [2025] UKFTT 682 (GRC), where findings and concessions recorded in a post-response decision of the Investigatory Power Tribunal were admissible because they concerned the state of affairs that existed at the date of the response. Navigating the absolute and qualified exemptions 18. The MOD now primarily relies on sections 23(1) and 24(1) in the alternative, without withdrawing its reliance on the other claimed exemptions. As section 24(1) cannot apply to information that falls within 23(1), confirmation by a public authority, the Commissioner or the Tribunal as to which exemption applies risks revealing whether or not the information was supplied by, or relates to, one of the bodies at section 23(3). In FCDO v IC & Williams [2021] UKUT 248, the Upper Tribunal held that the legislative scheme does permit a public authority, in order to protect national security, to ‘‘mask’’ the actual exemption that applies by reliance upon the two sections in the alternative. This practice may well put the requester at a disadvantage, not least in that information may be ruled exempt without the requester or the public ever being told the applicable exemption. 19. Use of such masking must, we consider, be justified in a particular case. Sometimes it may, as a matter of obviousness, serve no legitimate purpose whatsoever. The facts of Williams v IC & MPS [2023] UKFTT 1079 (GRC) may provide an example; it must be doubted whether the public authority in that case could have justified masking the engagement of section 23 where information having been supplied by the Secret Intelligence Service and the Security Service was already a matter of public record. 20. As a matter of necessity, if the Tribunal holds that either section 23(1) or section 24(1) applies to any information then it must still set out its assessment of the PIBT even if -unbeknownst to anyone but the public authority and the Commissioner — that exercise was unnecessary because the absolute exemption at section 23(1) applied. Otherwise, should section 24(1) apply to any information, the Tribunal will have failed to give reasons for its decision. In any event, the Supreme Court in Department of Business & Trade v IC [2025] UKSC 27 confirmed that the public interest in withholding the information raised by each exemption must be aggregated. As only section 24 is incompatible with section 23, we shall set out an omnibus assessment of the public interest arising under all claimed exemptions. Inevitably there is a degree of overlap between them. 21. We first address a dispute over the scope of the request, then assess Airwars’ case on the public interest in disclosure. After that then set out as much detail as we can on the public interest in maintaining each of the exemptions, before setting out our final conclusion. Scope 22. Airwars points to several pieces of evidence as to the nature of investigations into civilian casualties (sometimes termed CIVCAS). This includes an extract from a book, Reaper Force: Inside Britain’s Drone Wars, by Dr Peter Lee: "Max was a key figure on 39 Squadron at that time, with particular responsibility for RAF Reaper tactics, and would go on to be one of the most influential members of the Reaper Force. He recalls: ‘This CIVCAS incident prompted extensive reviews; squadron reviews; RAF review; legal review; coalition process review. The reviews and investigations extended up through the whole RAF command chain to the Permanent Joint Headquarters and to NATO headquarters in Kabul. All of the reports were provided to the UN Special Rapporteur. The investigations concluded that from the point where 39 Squadron became involved in the incident, until after the conclusion of the attack, the crew’s actions were in accordance with the Law of Armed Conflict and their procedures and directives.” 23. Disclosed documents in the USA relating to a USA/Coalition investigation in relation to a 5 October 2015 strike that killed 8 civilians. This displayed four main stages: an Initial Notification, a Civilian Casualty / CIVCAS assessment, a decision to open a formal investigation, and that investigation. It includes completed forms and pro forma memoranda. These give a flavour, argues Airwars, of the type of investigation likely to have been conducted by the UK in relation to the strike referenced in the request. Any documents completed fall in scope of the request and there is public interest in their disclosure, even if all the completed information is exempt. Mr Perry gave the example of the Initial Notification form used in the USA and provided in the bundle, which asks certain questions — what, where, who, etc -against which a person types the answers. These questions would, Mr Perry argued, show how the incident was assessed, even if the answers are withheld. 24. This is a convenient moment to note that we take full account of Airwars' submissions on disaggregation in general. If a document contains a mixture of exempt and non-exempt information we would seek to identify whether it is possible to disclose the non-exempt parts. 25. A final point on scope concerns a “Weapons Release Report”. The MOD confirmed that such a report was completed in relation to the strike, but on considering the report for ourselves we agree with the MOD that it contains only technical data and is not “a document assessing civilian harm”. The public interest in disclosure 26. In his skeleton argument, Mr Perry summarised Airwars’ evidence as disclosing six public interest factors: (1) First, there is simply no transparency when it comes to the processes by which the MoD undertakes assessments of civilian harm, save for general statements of principle and public disclosure of the “Weapon Release Report”: §§11-13 above. (2) Second, the lack of transparency regarding the UK’s civilian harm practices and procedures is out of step with is Coalition allies, most notably the US: §§20-26 above. JD/201-202 explains in summary how “From the United States to the Netherlands to New Zealand and Australia, I have shown how these mechanisms start with clarity on how states process and then report on civilian harm allegations”, whilst “the UK's policies are going the other way — with the amount of information released to the public declining over several years. There is still no clear and open process of civilian harm assessment …”. (3) Third, the UK’s processes and processes for assessing civilian harm are both ineffective and out of kilter with its Coalition allies. This is demonstrated by a consistent relative underreporting of civilian casualties by the UK when compared to its Coalition allies (see §§14-19 above), as well as the “error” made by the UK in records of the March 2018 Strike (see §35 above). (4) Fourth, the March 2018 Strike was the only strike where the UK has admitted to killing a civilian as part of Operation Shader (or, indeed, any other UK military operation in the past decade). It is therefore the only feasible subject of a FOIA request seeking to obtain a practical understanding of how the UK assesses civilian harm following airstrikes. (5) Fifth, the Coalition has clearly explained that it had no involvement in, or record of, the March 2018 Strike: §34 above. If the withheld information reveals that the Strike was conducted unilaterally by the UK then this significantly increases the public interest in disclosure, inter alia, because the UK had publicly adopted a policy of seeking to eliminate the threat from IS in a coordinated, cooperative manner that involves as many stakeholders as possible. If, on the other hand, the March 2018 Strike was conducted as part of the Coalition’s activities, then this casts doubt on (a) the UK’s commitment to ensure all civilian harm incidents are properly recognised by the Coalition (the discrepancy was raised by Airwars more than two years ago), and (b) the robustness of the UK's and Coalition’s procedures for assessing civilian harm. (6) Sixth, because (a) the March 2018 Strike was made against IS, and the families of civilians killed by Coalition strikes in Iraq and Syria have faced stigma and threats due to being wrongly associated with IS, and (b) the bereaved families may be entitled to compensation, transparency would have significant positive impacts for the family members of civilian killed by the Strike. 27. We pay tribute to the care and industry that Airwars has devoted to setting out its case on the public interest, all of which we have carefully considered. Mr Dyke's witness statement covers 56 pages and contains over 20,000 words. Practically, we can only give a relatively brief summary. 28. To do so, we first turn to the following introductory passages in Mr Dyke's witness statement: 9. As I will show in the coming sections, this incident is of vital public interest both individually and for understanding the wider conduct of the UK in warfare. To date, it is the sole civilian the UK has accepted killing in eight years of bombing Syria and Iraq — and more than 4,000 munitions dropped. Dozens of other well-documented cases have been deemed 'non credible' by the Ministry of Defence. This includes three in which the Coalition itself accepted that civilians died and the UK accepted carrying out the strike — but refused to recognise that those who died were civilians. The UK is increasingly an outlier in the Coalition, with nations including the United States, Netherlands and Australia identifying which strikes they conducted and therefore which civilians they killed. 10. I will outline my questions in more detail in the coming sections but believe it is worth summing up some of the fundamental questions this incident poses and which led me to seek answers through this Request. ●What is the process by which the United Kingdom decides whether or not it has killed a civilian in warfare? With whom does the final decision lie? And what, if any, oversight into that process does the UK public have? ●If, as the UK has accepted, it applies different standards of civilian harm to our closest allies, most prominently the United States, how and why do these differ? And why doesn't the public have any understanding of what these standards are? ● If, as I will show later in my submission, the strike was potentially taken outside the Coalition, under what legal basis was it conducted? And has this legal basis ever been articulated to the public? ●What attempts, if any, are made to identify and assist (for example by offering compensation – which the UK regularly paid during its presence in Iraq and Afghanistan) family members of civilians who are killed? 11. The release of the details I have requested would allow for answers to many of these questions. These are, as I will show later, of the utmost public interest and each and every answer is vital for a functioning democracy. It is critical that citizens of the United Kingdom are informed about how their military and their government, acting in their name, decide who is and is not a civilian. This is of particular importance in circumstances where such designation may be a determination of life or death. More fundamentally, the failure to properly account for the number of civilians killed by military actions can, as senior military officials have said multiple times (see paragraphs 104 and 136 below), create the false impression that war can be fought without human consequences. Without an open and transparent system around how civilian harm is monitored and assessed, the UK public isn't able to have a genuine conversation about the costs of war. The devastating impacts of conflict on societies must be front and centre in a society's reasoning when considering the merits of different potential options. 29. To further summarise Mr Dyke's evidence and argument on those points: a. Mr Dyke refers to the 26 March 2018 strike being the only time during Operation Shader where the UK government has acknowledged killing a civilian. This, he argues, makes it a critical lens through which the public can understand how the UK assesses civilian harm. b. Working against that benefit is the continuing opacity of the strike's circumstances. The Coalition previously stated that it had no record of any strike in the area on that date, raising the possibility that the UK acted unilaterally rather than in agreement with its coalition partners. If so, serious questions arise on whether the strike was lawful. If not, then the wider coalition's systems for recording and reporting strikes, and assessing civilian harm, likewise come into question. c. Also relevant is democratic accountability; the British public have a right to know how decisions that can cause civilian fatalities are taken, that such risk is minimised. Failure to be frank about the risk to civilians the growing illusion that the technological focus of modern warfare means it can be conducted without casualties, a myth that damages the public's ability to scrutinise and sanction future operations. A former Deputy Commander Operations at RAF Air Command had told Airwars4 that thinking "war is bloodless is a mistake because we need to be aware that war is nasty and opting for it must be the last resort. Thinking it can be done cleanly etc is a mistake." d. Mr Dyke points to what he would describe as a comparatively liberal approach to disclosure in the United States of America. After the case of Basim Razzo, an Iraqi national falsely thought by his neighbours to be connected with IS after his home was destroyed and his family was killed in a US air strike, was exposed by the New York Times 5, he was exonerated, the strike revealed as a mistake, and the deaths counted in the tally of civilian casualties. It was the release of civilian harm assessment documents under American Freedom of Information legislation that had revealed the error, and this set a precedent which has seen over 1,300 civilian harm assessments have been released — sensitive details being redacted as necessary. This in turn has influenced and improved US policy. As set out in the Grounds of Appeal : 20. The Appellant will set out in evidence in due course how and why an important factor weighing in favour of disclosure of these documents was the safety and wellbeing of family members of civilians killed or injured by the Coalition strikes. Many of these individuals faced discrimination and physical threats following strikes, because it was believed their family members were killed or injured because they were members of, or assisting, IS. Disclosure of details about the strikes was able to help dispel these mistaken beliefs. 21. The New York Times coverage of these disclosures revealed that US strikes had resulted in “hundreds” of civilian deaths. It criticised various systemic failings revealed by the disclosures, including the inability of the US military to identify patterns of failure or lessons learnt, shortcomings in investigations into strikes causing civilian harm, failure to plan strikes in advance, confirmation bias and cultural blind spots resulting in civilian deaths, breakdowns in technology and surveillance (e.g. lack of footage to determine whether the strike had caused civilian harm), and a failure to account for “secondary explosions”. The New York Times reporting on the disclosures concluded that “Taken together, the 5,400 pages of records point to an institutional acceptance of civilian casualties.” 22. The New York Times reporting has precipitated a significant revision of US civilian harm policy. Less than a month after the publication, Secretary of Defense Lloyd Austin directed the Department of Defense (DoD) to develop an action plan for mitigating and reducing civilian harm. In August, the DoD released the Civilian Harm Mitigation and Response Action Plan, a major policy governing all sections of the US military.10 Among the key announcements from the DoD were the need for a specific civilian harm centre of excellence, the creation of more than 100 specific civilian harm focussed roles within the US military, and new rules requiring better investigation of civilian harm in future conflicts. Marc Garlasco, a former Pentagon official who later investigated civilian deaths caused by U.S. military operations for the United Nations, called this a “sea change” that could save thousands of civilians' lives in future conflicts. 23. A Pentagon spokesperson, John Kirby, praised the Times’ reporting and recognised the vital importance of transparency in fostering accountability: "I would also be remiss if I didn’t also congratulate the staff of The New York Times …. That coverage was – and still is – not comfortable, not easy, and not simple to address. We know that we had more work to do to better prevent civilian harm. And we’re doing that work. We knew that we had mistakes, we’re trying to learn from those mistakes. And we knew that we weren’t always as transparent about those mistakes as we should be. But the reporting reinforced those concerns and in some cases gave us cause for additional concerns. And it made us ask ourselves some new and difficult questions or our own, even as it forced us to answer their difficult questions. I cannot say that this process was pleasant. But I guess that’s the whole point." Mr Dyke's evidence also gives a detailed description of the approach taken by the Netherlands, New Zealand and Australia; when it comes to transparency, he argues, the UK is needlessly and counterproductively well behind its allies. e. While the above arguments are put forward on the basis of just one civilian death, Mr Dyke argues that the reliability of that figure must be doubted. In support of the public interest in its figure, he refers to wider debate and challenge: 84. Since 2014, the UK has dropped more than 4,000 munitions in Iraq and Syria — officially declaring that it had killed 4,013 militants while injuring 302 more, as of March 2019. When it comes strikes that killed ISIS militants, the UK has collated and shared granular detail with the public — with the date and near location of each of those 4,013 strikes published. 85. In contrast, the UK has only accepted evidence of killing one single civilian in Iraq and Syria since 2014, the incident which is the subject of my request. When it comes to the harm from this strike, and the harm from many hundreds of strikes which independent monitors contend have caused civilian harm, the UK MoD takes an entirely different approach — with almost no information released. 86. While Airwars estimates the minimum figure of civilians likely killed by Coalition airstrikes at 8,197, the US-led Coalition has accepted responsibility for 1,437 deaths. The UK has accepted responsibility for only one. 87. The stark comparison between these two figures has not been lost. Lydia Wilson, research fellow at Oxford University's Centre for the Resolution of Intractable Conflict, told the BBC29 (emphasis added) it was "absurd" to suggest only one civilian was killed "given the pictures we have from the bombardment of Raqqa, Mosul, and other targets in the Islamic State." She said: "I would ask how the RAF came at their numbers; that is, how they distinguish between fighters and civilians when buildings have collapsed on top of inhabitants, and on-the-ground, independent reporting is impossible." 88. As I will outline in a later section, this position is at best absurd — but at worst, fundamentally undermines public trust in our the Government, the UK armed forces and democratic institutions. 89. The UK itself has acknowledged this challenge. In early 2019 then-Minister for the Armed Forces Mark Lancaster emphasised that it is “not our position that there has only been a single civilian casualty as a result of our military action. What we are saying is that we only have evidence of what we believe to have been a single civilian casualty. That is a rather different position.” 90. Yet it is unclear what the UK Government is doing to improve its evidence gathering tools in recognition of this inability to understand true harm on the ground, and the Government and MoD have both proven unwilling to work with civil society and independent monitors to understand gaps in its monitoring systems and improve its approach. This is particularly noteworthy since many of the UK’s allies, and independent groups such as ourselves, have identified and implemented tools to more accurately track civilian harm. 91. The impossibility of having harmed only one civilian in nine years of war and the MoD’s refusal to show how it is going to improve its evidence gathering mechanism, have not gone unnoticed by British parliamentarians and the media. 92. As early as 2015 The Guardian was asking "Is UK's claim of zero civilian casualties in Iraq airstrikes credible?" quoting Shashank Joshi, a senior research fellow at the RUSI think tank as saying “it would be astonishing if such a long campaign had seen zero civilian casualties." 93. In 2019, Ann Clywd MP called on the UK government to establish a "dedicated civilian casualty mitigation and investigation team with proper resources, to understand the impact of our operations and accurately record civilian harm." Minister Mark Field replied: “the operational end-use monitoring and establishment of a dedicated civilian casualty mitigation and investigation team are an MoD lead”. This suggested that, five years into the conflict, such a unit was still not independently operational. 94. In 2016, Penny Mordaunt publicly stated that allegations and credible allegations of civilian casualties could be sent to the Ministry of Defense’s Ministerial Correspondence unit, via an email address. We at Airwars filed FOIA requests asking for the total number of emails regarding civilian casualties that this email address had received between 2014 and 2016, and separately between 2016 and 2018. We were told no information within the scope of our request was held. 95. In 2020, Johnny Mercer MP told the House of Commons every allegation of a civilian casualty is referred to the Royal Military Police (RMP)34. We filed FOIA requests asking for a breakdown of all offences reported to the RMP between 2016 and 2020. We received a comprehensive breakdown of all cases the RMP had opened on all topics. There were no allegations of civilian casualties included on this lengthy list. 96. Chris Cole, director of the non-governmental organisation Drone Wars UK, was able to get a limited amount of information about some aspects of the weapons release processes through the FOIA system. He was aware that there was a form filled out by individual British pilots or service people in the event of a potential civilian harm request. Having requested one regarding a specific incident that was rejected, he then asked for a blank form (Exhibit JD24). It showed the type of detail that was to be included in a short report submitted by British pilots when an airborne weapon is fired. 97. As well as technical detail — location, time, altitude, direction and angle of attack, weapon fusing and weather conditions — any report should include a narrative description of target; whether target was hit (“Yes/No/Unknown”); number of weapons that hit intended target; miss distance if applicable; CD or CIVCAS (“Yes/No/Unknown”), impact observed (“Yes/No”) and intent met (“Yes/No/Unknown”). The pilot is also encouraged to add a ‘free format narrative of the whole mission.'" 98. In addition, Airwars, along with multiple other civil society organisations, has been in dialogue with the MoD for many years. As recently as 2 February 2022, Airwars’ policy specialist took part in a closed roundtable with a variety of actors including the MoD around the UK’s stated commitment to the EWIPA declaration, mentioned above. 99. However, while general discussions on policy are on-going, MoD officials have been unable to acknowledge or respond both to continued allegations of civilian harm resulting from its actions, or provide any real detail on the civilian harm assessment process. 100. As outlined in our 2016 transparency audit (extracts at JD4), Airwars had previously received acknowledgements from MoD officials at least on whether or not allegations were linked to RAF actions. Unfortunately, even this process appears no longer to be functioning in the MoD. Airwars has tried multiple occasions in recent years to engage with the Ministry of Defence around civilian harm allegations from UK strikes. The Government has consistently refused to engage. 101. As recently as December 2022, the UK conducted a strike that killed an alleged ISIS weapons dealer. Multiple civilians were also reported to have been injured. Defence Secretary Ben Wallace later told MPs the individual was involved in the creation of " chemical and biological weapons." Airwars has tried to discuss the civilian harm allegations multiple times with the MoD without success, and The Guardian reported that the "UK refuses to say if it investigated reports of Syria drone strike casualties" f. The UK likewise classifies civilian deaths differently to its coalition partners, further frustrating effective scrutiny, as set out in an Airwars article https://airwars.org/europes-shame-claims-of-no-civilian-harm-exposed/ . 30. All this, argues Mr Dyke, operates to undermine lessons from previous conflicts: a. Afghanistan — A 2021 report by the Ceasefire Centre for Civilian Rights quotes Lt Col Mark Goodwin-Hudson, a former head of NATO's civilian casualty monitoring cell in Afghanistan, as providing the following evidence to the House of Lords International Relations and Defence Committee: "Proactively mitigating civilian harm and protecting civilians holds the key to achieving military and foreign policy objectives in Afghanistan and other conflict contexts. Until protecting the population becomes, as General Petraeus said in 2007, "the new big idea-”, Western government's attempts in Afghanistan will continue to falter, limiting their capacity to support the Afghan government's efforts to bring forward peach and stability to the country." b. Iraq — The Chilcot report concluded that the UK Government had been reluctant to confront the scale of civilian casualties caused by the conflict, and had recommended as part of its key findings that the UK government: “277. … a Government has a responsibility to make every reasonable effort to identify and understand the likely and actual effects of its military actions on civilians. … 279. It may not be possible, before committing to a course of action, to produce even broad estimates of the number of civilians that would be directly and indirectly affected by it, or to identify all the effects on civilians. 280. The Government should be ready to work with others, in particular NGOs and academic institutions, to develop such assessments and estimates over time. 281. The Government should take account of those assessments and estimates in developing its strategy and plans as well as in its military tactics and use of ordnance, in order to minimise, to the extent possible, the effects on civilians.” 31. What Mr Dyke describes as an exception to UK government opacity previously occurred in relation to a 2011 drone strike in Helmand Province, where an academic Dr Peter Lee was given access to UK armed forces personnel and records. Dr Lee's resulting book, Reaper Force: Inside Britain's Drone Wars, had given what is put forward as significantly more detail than that requested by Mr Dyke, without appearing to give rise to any negative consequences upon which the MOD relies. We were provided with extracts from the book, which we have read. 32. In relation to the 26 March 2018 strike in particular, Mr Dyke summarises his points as follows: 127. To sum up the vital public concern about this case: — The UK's sole confirmed civilian harm incident in eight years of airstrikes in Iraq and Syria was declared on May 2, 2018, six weeks after the strike. — In the UK government's own admission, the assessment into this incident was made outside of the Coalition civilian harm review process. The UK public has no idea which team carried out the assessment, and what information they were using to make it. The public does not even know if the final decision was taken by the military, a civil servant, or a politician. — The UK statement to parliament declared that the incident was part of the wider Coalition against the Islamic State. — Yet that same Coalition reviewed the incident and concluded there was no Coalition strike in the stated location. There is no obvious explanation for why. — The strike was also not included in documents provided by the MoD on all strikes that killed militants, despite the initial statement explicitly saying three militants had been killed. — No Syrian organisation has any record of a civilian killed that day. In fact, beyond the UK's statement, there is no evidence at all of the strike in question. 33. Mr Dyke is then explicit, and we quote: 128. One conclusion is that this incident was conducted outside the Coalition as an off-books strike, or even an extrajudicial killing. If that was the case, under what legal basis was it conducted? The mandate for the bombing campaign in Syria has been the subject of much parliamentary discussion (Exhibit JD32) and any actions occurring outside of the remit would be, at best, legally questionable. These are fundamental questions that are in the vital public interest. 34. Also put forward as a matter of public interest is the UK's adherence to the Law of Armed Combat (LOAC). In the Chilcot Report this was described as follows: (so far as relevant) Battle Damage Assessment Section 6.2 describes the main principles of International Humanitarian Law (IHL), also known as the Law of Armed Conflict (LOAC) or the Law of War, how they were disseminated to those engaged in military action, and how they were reflected in the UK’s Targeting Directive and Rules of Engagement (ROEs). The key elements of IHL which apply to targeting of military objectives during a conflict are set out in the 1977 Protocol Additional to the Geneva Conventions of 1949 (Protocol I). The main principles can be summarised as: • Distinction. The parties to the conflict must at all times distinguish between the civilian population and combatants, and between civilian objects and military objectives, and shall direct their operations only against military objectives (Article 48). •Proportionality. Military objectives must not be attacked if the attack is likely to cause civilian casualties or damage which would be excessive in relation to the concrete and direct military advantage anticipated (Article 57:2:b). • Military Necessity. Offensive operations must be limited to those which are necessary (Article 57:3). • Feasible Precautions. In the conduct of military operations, constant care shall be taken to spare the civilian population, civilians and civilian objects. Those who plan or decide upon an attack must take a number of specified precautions, focusing on the principles outlined above (Article 57). Submissions in response 35. Neither the Commissioner nor the MOD necessarily accepts or denies the underlying accuracy of the various reports and statements upon which Airwars relies, much less the conclusions drawn by Airwars from its evidence. 36. The MOD does, in particular in Ms Costelloe's witness statement, strongly refute that the strike was, as Mr Dyke said might possibly be inferred, “an off-books strike, or even an extra-judicial killing”. None of the evidence we have seen, open or closed, undermines her statement. 37. The MOD further rejects that any breach of LOAC can be inferred from the open evidence, or that the public orcoalition partners were misled. In her witness statement, Ms Costelloe states that the failure to include the strike in the statistics release relied upon by Mr Dyke was an administrative error, and strongly refutes any wrongdoing or misleading conduct by the MOD. This was addressed in Mr Oliver’s evidence, where he confirmed from his own direct knowledge that there was an administrative error and that efforts had since been taken to rectify it. We accept his evidence on that point, and should say that none of the evidence we have seen could justify a conclusion that anyone has been misled. Whether or not any particular action breached LOAC is outside our competence and is unnecessary to determine. 38. With regard to other specific aspects of Airwars’ case: a. In relation to the 2011 strike in Afghanistan, the MOD observes that the present request was made almost a decade afterwards and concerns a different theatre of operation. It cannot be assumed that the prejudice (or lack of prejudice) arising from disclosures in Dr Lee's book would be repeated in relation to the present request. The MOD does not accept that any inference can be drawn that it approved, adopted or admitted the content of Dr Lee's book or elected to place any information in the public domain. b. Inconsistency between states as to what is released can represent different countries’ freedom of information and related legal regimes, as well as different political decisions on the extent of voluntary disclosure. Such inconsistency does not necessarily work against the public interest. 39. The Commissioner also makes the following points: a. The Commissioner rejects that further transparency in relation to the strike would have “significant positive effects for the family members of the civilian killed” as they can already point to the Secretary of State's statement to Parliament. In any event, this is a private rather than public interest, and in relation to a single strike it is difficult to see how disclosure would serve a wider public interest. b. While not accepting that Airwars’ general propositions, such as that the “the UK's processes … for assessing civilian harm are both ineffective and out of kilter with its Coalition allies” and that the strike may have been conducted unilaterally, are established on the open evidence, the Commissioner does accept that there is a strong public interest in proving or disproving such matters. The Commissioner accepts that there is a “considerable” and “significant” public interest in the disclosure of information relating to civilian casualties as a result of UK military action. Consideration 40. It is common ground that there is public interest in disclosure. Like the Commissioner, we consider that proving or disproving Airwars’ more direct allegations or suspicions is a matter of public importance. The importance of avoiding the death of civilians in military operations can hardly be overstated, and assessment of incidents such as that of 26 March 2018 an obvious means to achieve it. We need not repeat all Airwars’ arguments to that effect. We also consider that the public interest in disclosure is increased by the lack of publicly available information on how such incidents are assessed in general. 41. Before turning to the public interest in maintaining the exemptions, we set out our assessment of other points put forward by Airwars: a. The MOD’s evidence does not engage with many aspects of Airwars’ case. For example, Airwars criticises Ms Costelloe's witness statement for containing no response to its scepticism that there has only been one civilian death from UK-conducted strikes. This criticism is misplaced. The only obligation upon the MOD is to provide evidence necessary to support its own case on the issues before the Tribunal. In any event, the lack of a response to any particular point means that (in evidential terms at least) there is nothing to gainsay it, so Airwars is not disadvantaged. b. We should record Mr Oliver's evidence on the procedures used to assess civilian casualties. When questioned as to the forms used, whether there is a standard written procedure where it could be found, and other matters, he had to concede that he had insufficient familiarity with the procedure to answer them. Airwars criticised him for this but the specificity of the questions was somewhat unheralded and his seniority gives him a wider purview than the detail of individual written procedures. There was also a significant amount of information he did provide. For example, in oral evidence in closed session he set out the command structure and operational details of the airstrike in question, including how this affected the misreporting between UK military forces and the Coalition and how it contributed to the late disclosure of additional responsive material. We were satisfied that he was entirely candid and forthcoming about those matters. c. As to the interests of family members of the civilian killed in the strike: i. In principle we agree with Airwars and disagree with the respondents. The ability of bereaved family members to access information about such incidents and how they occurred is a matter of general public interest. The information sought by Airwars, relating as it does to how harm is assessed, supports that public interest. The public can have confidence that victims have a reasonable opportunity to seek answers and accountability for civilian casualties. ii. We further agree with Airwars’ submission that transparency may avoid misidentification of civilian casualties as combatants by third parties, who then might pursue reprisals against their families. Likewise, effective identification and investigation of civilian casualties will enable family support. iii. Insofar as the MOD put forward an argument that the availability of claims for civil damages can meaningfully reduce the weight carried by the above points, we unhesitatingly reject it. d. Considering the evidence overall, we do accept that there is a lack of publicly available information on the procedure usually followed in assessing civilian harm, and that the nature, comprehensiveness and robustness of such a procedure is a matter of public interest. The absence of any published procedure at all has the potential to undermine public confidence as to its integrity and comprehensiveness; while there is no reason to doubt the good faith and competence of everyone involved, high level assurances do not provide the same confidence as a published procedure that can be scrutinised. That lack of open information correspondingly increases the public interest in disclosing information concerning how civilian harm was assessed in this particular strike. We apply that conclusion to our consideration, while recognising that the present request was for the documents assessing harm in this incident rather than for any documents held by the MOD for the purpose of assessing such incidents in general. Assessing the exemptions Sections 23 & 24 42. Section 23 provides that any information held by a public authority is exempt information if it was directly or indirectly supplied to the public authority by, or relates to, any of the bodies specified in subsection (3). 43. In Commissioner of the Police of the Metropolis v Information Commissioner & Rosenbaum (Information rights — Freedom of information — qualified exemptions) [2021] UKUT 5 (AAC), at [35], the Upper Tribunal set out fourteen principles arising from the Act and previous authority. We set them out in full (details of the authorities cited can be found earlier in the Upper Tribunal's decision): 1. Section 23 affords the "widest protection" of any of the exemptions: Cobain at [19(b)] and [29]. 2. The purpose of section 23 is to preserve the operational secrecy necessary for section 23(3) bodies to function: Lownie at [50]. 3. It is "Parliament's clear intention that, because of what they do, there should be no question of using FOIA to obtain information from or about the activities of section 23 bodies at all". The exclusion of the section 23(3) bodies from the scope of FOIA was shutting the front door, and section 23 was "a means of shutting the back door to ensure that this exclusion was not circumvented": APPGER at [16]. 4. The legislative choice of Parliament was that "the exclusionary principle was so fundamental when considering information touching the specified bodies, that even perfectly harmless disclosure would only be made on the initiative or with the consent of the body concerned": Cobain at [28]; Lownie at [53]. 5. Asking whether the information requested is anodyne or revelatory fails to respect the difficulty of identifying what the revelatory nature of the information might be without a detailed understanding of the security context: Lownie at [42]; Corderoy at [59]. 6. When applying the 'relates to' limb of sections 23(1) and (5) , that language is used in "a wide sense": APPGER at [25]; Corderoy at [59] ; Savic at [40]. 7. The first port of call should always be the statutory language without any judicial gloss: APPGER at [23]; Corderoy at [51]; Savic at [40]. 8. With that warning in mind, in the context of 'relates to' in section 23 , it may sometimes be helpful to consider the synonyms of "some connection", or "that it touches or stands in some relation to" (APPGER at [13], [25]) or to consider whether the request is for "information, in a record supplied to one or more of the section 23 bodies, which was for the purpose of the discharge of their statutory functions" (APPGER at [21], [26]; Lownie at [57]). But the 'relates to' limb must not be read as subject to a test of focus (APPGER at [14) or directness (Lownie at [59]- [60]). 9. The scope of the 'relates to' limb is not unlimited and there will come a point when any connection between the information and the section 23(3) body is too remote. Assessing this is a question of judgment on the evidence: Lownie at [62]. 10. The assessment of the degree of relationship may be informed by the context of the information: Lownie at [4] and [67]. 11. The scope of the section 23 exemption is not to be construed or applied by reference to other exemptions, including section 24 : APPGER at [17]; Lownie at [45] and [52]. 12. In a section 23(1) case, regard should be had as to whether or not information can be disaggregated from the exempt information so as to render it non-exempt and still be provided in an intelligible form: Corderoy at [43]. 13. Section 23(5) requires consideration of whether answering 'yes' or 'no' to whether the information requested is held engages any of the limbs of section 23 : Savic at [43], [82] and [92]. 14. The purpose of section 23(5) is a protective concept, to stop inferences being drawn on the existence or types of information and enables an equivalent position to be taken on other occasions: Savic at [60]. 44. It is an absolute exemption. Information to which it applies is exempt from the duty of disclosure whether or not any possible prejudice could arise from its disclosure. In enacting the exemption, Parliament intended to exclude all the listed security bodies and their activities from the duty at section l of the Act. 45. Section 24(1) provides that information which does not fall within s.23(l) is exempt information if exemption is reasonably required for the purpose of safeguarding national security. 46. In approaching the exemption at s.24(1), the Upper Tribunal in Foreign, Commonwealth and Development Office v Information Commissioner, Williams and Others (Sections 23 and 24) (Information rights — Freedom of information — absolute exemptions, Information rights) [2021] UKUT 248 approved six principles. We summarise them as follows: (1) The term national security has been interpreted broadly and encompasses the security of the United Kingdom and its people, the protection of democracy and the legal and constitutional systems of the state. (2) A threat to national security may be direct (the threat of action against the United Kingdom) or indirect. (3) Section 24 is not engaged, unlike the majority of the qualified exemptions, by a consideration of prejudice. Its engagement is deliberately differently worded. (4) The term "required" means "reasonably necessary". (5) National security is a matter of vital national importance in which the Tribunal should pause and reflect very carefully before overriding the sincerely held views of relevant public authorities. (6) Even where the chance of a particular harm occurring is relatively low, the seriousness of the consequences (the nature of the risk) can nonetheless mean that the public interest in avoiding that risk is very strong. The reality is that the public interest in maintaining the qualified national security exemption in section 24(1) is likely to be substantial and to require a compelling competing public interest to equal or outweigh it. That does not mean that the section 24 exemption carries "inherent weight", but is rather a reflection of what is likely to be a fair recognition of the public interests involved in the particular circumstances of a case in which section 24 is properly engaged. 47. As recognised by the final point, the exemption will only apply if the public interest in withholding the information outweighs the public interest in its disclosure. 48. We confirm that, if and where relevant, we have paid heed to Airwars' submissions that the ‘related to’ limb at section 23(1) should not be stretched too far. We also note Mr Perry's reference to APPGER v Information Commissioner and Foreign & Commonwealth Office [2015] UKUT 377 (AAC) at [20], where it is observed that the inclusion of a section 23 body in the wider distribution of a particular document is exempt, but that exemption does not then automatically extend to the content of the document itself. 49. Airwars asks us to draw an inference from the MOD's late reliance on section 23 that it is unlikely to be engaged. The Commissioner is “sympathetic” to this submission having been made, without saying whether or not it is well-founded, and reassures Airwars that reliance on sections 23 and 24 in the alternative will be “thoroughly and robustly tested in closed, including the reasons for late reliance”. We echo these observations. Late reliance on “masking” is apt to reduce confidence in its use by public authorities, and accordingly its utility. Whether or not it does so in this specific appeal, for reasons that can only be given in closed we find that masking is justified. 50. In relation to both exemptions, so far as relevant, we have carefully considered whether any information can properly be disaggregated. 51. We find that either section 23(1) or section 24(1) is engaged in relation to all information save for that which we have found should be disclosed, and that the MOD is entitled to mask which of them does. Nothing in these open reasons should be taken as any indication that either exemption applies to a particular piece of information, or indeed any of the information. To reinforce this point, it may be the case that some information engages section 23 and some information engages section 24, or that either one of them applies to all of it. 52. What can be said openly is that if any or all information does not engage section 23(1), we have conscientiously, independently and critically examined both whether section 24(1) is engaged and the strength of the public interest engaged. 53. The MOD’s open arguments on those matters incorporate their case on section 26. Those are set out below. We are mindful that while many adverse consequences to disclosure could engage both exemptions, some will not. In its internal review of its response to the request, the MOD stated: 9. However, against release, you were advised that release would prejudice the safeguarding of national security if sensitive operational details were revealed, including the tactics, techniques and procedures (TTPs) of relevant organisations contributing to national security. It is likely that release of the information would impact on the UK’s national security objectives. 54. It further stated in its response to the appeal: 46. In the event that s.24(1) FOIA were to apply in respect of the Responsive Information or any of it, it is a qualified exemption. In those circumstances: a. For the reasons advanced at §9 of the internal review dated 9 June 2021, it is likely that disclosure of the Responsive Information would negatively impact on the UK’s national security objectives. b. Further to paragraph 8 above, the Responsive Information reveals specific intelligence, surveillance and reconnaissance techniques used by the UK Government in order to plan and conduct counter-terrorism operations. Should this information become public, it would give known and suspected terrorists damaging insight into how the UK and its allies plan operations to protect national security and could be used by them to negate the tactics and methods used in UK operations. This would likely prejudice the effectiveness of future counter-terrorism operations, reducing the UK Government’s ability to uphold national security. Counter-Daesh operations remain the UK’s top counter-terrorism priority. c. The seriousness of the consequences of adversaries obtaining any or any improved understanding of TTPs means that the public interest in avoiding that risk is very strong. 55. Airwars does not dispute that such consequences would, if established, engage section 24(1), nor does it disagree that the public interest in maintaining the exemption would be very high in those circumstances. It does repeat its submissions on section 26 (considered below). We also take due account of the parties’ submissions on the distinction between the two exemptions . 56. Insofar as section 24(1) may arise, and subject to any possible disaggregation, we are satisfied that disclosure would engage the exemption. Beyond stating our general agreement with the MOD?s submissions on the prejudice disclosure would cause, and referring to the relevant authorities, we can give no further open reasons. Sections 26 & 27 57. These should be taken together, given the degree of interrelationship between the prejudices claimed by the MOD under each; the relevant context in this appeal is that this particular defence function is performed in coalition with allied states. We agree with the MOD that it has consistently claimed that both exemptions arise in relation to all the information in scope of the request. 58. In Cole v ICO and MOD EA/2013/0042, this Tribunal held as follows: 53. The public interest in maintaining the exemption in section 26(1)(b) is exceptionally weighty. There is an exceptionally strong public interest in preventing harm to the UK’s capabilities in an ongoing armed conflict. The security and safety implications carry very strong public interest weight. 54. We agree with the Commissioner that there would need to be very weighty countervailing considerations to outweigh a risk to security and safety of the forces which was of sufficient severity to have engaged section 26(1)(b). While not bound by that authority, we agree with it. 59. In R. (Mohamed) v Secretary of State for Foreign & Commonwealth Affairs [2010] EWCA Civ 65 at [131], the Master of the Rolls held: … In practical terms, the Foreign Secretary has unrestricted access to full and open advice from his experienced advisers, both in the Foreign Office and the intelligence services. He is accordingly far better informed, as well as having more relevant experience, than any judge, for the purpose of assessing the likely attitude and actions of foreign intelligence services as a result of the publication of the redacted paragraphs, and the consequences of any such actions so far as the prevention of terrorism in this country is concerned… 60. With appropriate adaptation, the above principles apply with full force to the Ministry of Defence and its assessment of how disclosure could cause the prejudice described in both section 26 and section 27. 61. We also agree with what was said in Plowden v Information Commissioner & Foreign and Commonwealth Office (EA/2011/0225) at [19], being that one way in which prejudice will arise is if disclosure “makes relations more difficult or calls for a particular diplomatic response to contain or limit damage which would not otherwise have been necessary”. 62. In some circumstances prejudice may arise even where the foreign state’s reaction is uncertain or unpredictable, the risk of harm can itself stand as the prejudice required by the exemptions. 63. We introduce the MOD’s factual case on section 26 by setting out how it was distilled in Mr Kosmin's skeleton argument: a. Operation Shader is the UK operational name for the UK’s contribution to the US-led military campaign against Daesh. The UK have been participating in operations within Iraq and Syria under Operation Shader since 2014. UK Armed Forces and the UK’s coalition allies continue to conduct such operations. b. Reapers are airframes used by the Royal Air Force. The Reaper platform is a Remotely Piloted Air System (“RPAS”). c. The Reaper platform is a strategic asset that fulfils a number of different missions for the UK military. Its use as a surveillance and reconnaissance asset, as well as strike asset, is well documented in the public domain. d. All MOD kinetic engagements, including with Reaper, must be compliant with the Laws of Armed Conflict (“LOAC”). One of the pillars of LOAC is that parties to an armed conflict must at all times distinguish between combatants and non-combatants. All UK Reaper crews receive rigorous training and repeated assessment on their understanding of LOAC. Before conducting an engagement, it is a requirement to establish positive identification of the target as a combatant. Individuals are assumed to be non-combatants unless positively identified to the contrary. The MOD has confirmed the existence of these processes through the publicly available LOAC manual ‘JSP383’. The methodologies employed to positively identify a combatant are necessarily classified to avoid prejudicing the effectiveness of UK operations: OPEN Costelloe W/S §6.c.iii. e. While Operation Shader is a counter-Daesh operation, other and/or future operations may depend on a greater degree of public uncertainty as to the employment of Reaper in order to be successful. It is therefore important to retain a degree of ambiguity regarding the tactics, techniques and procedures deployed in RPAS operations in order to maintain this flexibility in the future. f. By making the Responsive Information public, the effectiveness of the Reaper platform as an important counter-terrorism asset would be prejudiced. g. Small amounts of information, which do not of themselves appear to be particularly damaging (and might even appear anodyne), can be combined by adversaries with other sources and information held by them. The mosaic effect, as a result of the requested information being combined with other information, poses both a short term and long term risk to the protection of UK forces and forces cooperating with UK forces. The MOD regards the overall impact of the mosaic effect to be against the public interest. h. Through both information publicly available about the UK’s input to Operation Shader and other capacities available to adversaries, it is highly likely that some adversaries have a general understanding of Reaper’s operating profile. For those adversaries currently without a complex understanding of Reaper, the additional information requested by way of the Information Request could result in a significant improvement in the accuracy of, and confidence in, their understanding and assessment of UK operational capabilities and the policy restrictions on its use by UK Armed Forces. i. The harm caused by adversaries combining information regarding Reaper falls to be understood in the following context. Future employments of Reaper and similar capabilities will be affected by adversaries’ understanding of Reaper operations. The utility of Reaper and similar capabilities is in part dependent on the adversaries’ uncertainty of both the capabilities and how the UK employs them. The Responsive Information will allow a sophisticated adversary to improve its understanding of the capabilities of the platform and the UK’s Tactics, Techniques and Procedures (“TTPs”) used in the conduct of counter-terrorism operations. The Responsive Information will also increase adversaries’ understanding of the methods employed by UK forces to avoid civilian casualties. Significantly, a sophisticated adversary’s improved understanding might enable it to take steps to develop countermeasures. For less sophisticated adversaries, additional information may provide a step change in their understanding and ability to limit their exposure to the UK’s RPAS operations. This has the potential to limit the operational effectiveness of Reaper and the ability of the UK to use this asset for conducting successful strikes against highly dangerous terrorists posing a threat to UK national security and UK and allies’ security interests overseas. j. The MOD recognises that the Secretary of State for Defence made a statement to Parliament on 2 May 2018 in respect of the Strike (“the Statement”). Inter alia, the Statement provided: “… There are limits on any further details that can be provided given ongoing operations and consequent national security issues. …” That restriction was consistent with the stance the MOD has adopted in response to the Information Request (c.f. ASA §§36-37). See further OPEN Costelloe W/S §9. k. At OPEN Costelloe W/S §10, Ms Costelloe confirmed that: (i) the Strike was conducted within the framework of the Coalition; (ii) the Strike did not feature in the statistics provided by the MOD in response to an information request dated 16 February 2022 due to an administrative error; and (iii) “The Strike was not “an off-books strike, or even an extra-judicial killing”.” [OB/E910] l. The emphasis placed on the administrative error at ASA §35 and 85(3) is unwarranted: (i) the suggestion that a wider public interest in disclosure arises from the identification and later correction of an administrative error by the MOD is significantly overstated; (ii) the Appellant’s assertion that any systemic “carelessness on the MoD’s part in relation to records of civilian harm” is demonstrated by the identification and correction of this single error is without foundation; and (iii) that the administrative error was candidly accepted exemplifies the MOD acting transparently when exemptions under FOIA do not apply (c.f. ASA §64, final sentence). However, on the facts of this appeal, exemptions apply to all the Responsive Information. 64. This is supported by Ms Costelloe's witness statement: Section 26(1)(b): Defence 21. The information requested by Mr Dyke relates to an air strike conducted by UK forces in Syria on 26 March 2018. The strike was conducted using a remotely piloted air system known as Reaper. Reaper is a sensitive military capability used for intelligence, surveillance and reconnaissance, and kinetic strikes. Crucial to these activities is the ability of the MOD to limit the amount of public knowledge about how UK Armed Forces personnel and others operate Reaper, to avoid prejudicing its effectiveness. 22. Given the nuances in the way Reaper is operated, the details and signature of its operations are of significant interest for hostile actors, including terrorists. This is because the release of any information that confirms previously released information, or provides a new insight, about how the UK uses Reaper would assist hostile actors in developing counter-measures or changing their behaviour to undermine its use or to increase the risk to UK and allied service personnel. 23. The requested information would reveal sensitive operational details, including tactics, techniques and procedures about how UK forces identify, track, and target known terrorists. Of specific relevance to this request is how UK personnel employ tactics to try to avoid civilian casualties. If these were made public, suspected terrorists could adjust how they operate to better protect themselves and reduce the effectiveness of Reaper as a counter-terrorism asset. More detail will be provided in my CLOSED witness statement. 24. Through both information publicly available about the UK’s input to Operation Shader and other capacities available to adversaries, it is highly likely that some adversaries, including Daesh fighters, have a general understanding of Reaper’s operational profile. For those adversaries without a complex understanding of Reaper, the additional information requested could result in a significant improvement in the accuracy of, and confidence in, their understanding and assessment of UK operational capabilities and the policy restrictions on its use by UK Armed Forces. Section 27(1)(a): International relations 25. In my knowledge and experience of working in an international diplomacy and policy environment, I can confirm that the relationships the UK has with partners, including members of the Global Coalition against Daesh, are built on trust. Our allies and partners have a general understanding that information relating to military or security activities are handled in a confidential and a secure manner. Any loss of trust between the UK Government and international partners brought about by publicly releasing classified information shared in confidence would likely negatively impact the UK’s ability to work together closely on current and future shared defence and security objectives. 26. Operation Shader is (and was on 7 April 2021) an ongoing operation and the risk of damaging the partnerships needed to support UK operations is a real one, not hypothetical. The lack of public messaging is essential in maintaining our close collaboration with the other Coalition partners. That close collaboration, including on the command and control architecture used for Reaper, enables the RAF to use its mission-qualified Reaper crews to maximum operational advantage and efficiency. Armed Reapers are also used by the United States’ military and disclosing sensitive details about how they operate may also impact their forces’ ability to use Reaper against terrorists. Doing so could therefore impact the US-UK relationship, which underpins a large amount of UK national security and defence cooperation. I will discuss in more detail the specific relationships that could be impacted by releasing the requested information in my CLOSED statement. 27. Even outside of counter-terrorism operations in Syria and Iraq, other allies may be less likely to work alongside and support the UK in the future if they believed that classified information shared in confidence may later be released to the public. In my time overseeing operational policy in the MOD, there have been other real examples, the details of which cannot be discussed in OPEN, where unauthorised disclosure of sensitive information has led to some negative impacts for the UK ‘s relationships with other states. 28. In order to maintain the positive relations with our international allies and partners on which our Reaper operations frequently depend, we must safeguard our reputation as a trustworthy partner. It is also important to maintain a reputation for trustworthiness to enable us to operate with other allies in the future. The MOD has therefore always been careful to minimise public statements on Reaper operations in order to protect those relationships. 65. In Mr Oliver's closed evidence, he set out specific features of Reaper operations, both in general and in this particular incident, that would diminish their utility and capability if disclosed. This included specific examples of likely adaptations of enemy tactics in the event of disclosure. 66. We accept this evidence and argument as establishing the exemption in principle, and (subject to it being relevant in the case of particular information, as discussed below) find that disclosure would prejudice defence and the capability, effectiveness and security of UK armed forces and cooperating allied forces. 67. As to section 27, in its internal review the MOD stated: 16. It is recognised that the release of the requested would help inform public debate about the way that the UK conducts operations in conjunction with allied forces. However, you were advised that the relationship that UK Armed Forces have with its partners in the global coalition against Daesh are built on trust; and a general understanding that information relating to military or security activities conducted under that partnership are handled in a confidential and secure manner. Any loss of trust between the UK Government and international partners, or any other allied or partner nation, would negatively impact upon the UK’s ability to work together closely on current and future shared defence and security objectives, such as Counter Terrorism and regional stability. 17. It has been assessed that the release of the information in scope of your request would harm relations between the United Kingdom and any other state. This could inhibit the willingness of the other nations to participate or share information about the deployment of these assets in the future. Unwillingness to share intelligence or data may also impact on current and future operations, Therefore, I find that the balance of public interest lies in favour of withholding the information under section 27(1)(a) of the Act. 68. In response to a number of points made in Airwars’ Grounds of Appeal, its Response stated: b. A specific UK Government relationship that could be negatively affected by releasing the Responsive Information is that with the Government of Iraq. The legal basis for the Coalition’s operations in Syria and Iraq is the collective defence of Iraq, at the request of the Iraqi Government. Releasing the Responsive Information would place in the public domain greater detail concerning the nature of the UK’s military support to Iraq, which could risk damaging the relationship of the UK Government and the Iraqi Government and in turn the UK’s ability to conduct counter-terrorism operations in the region in the future. Further details of the specific relationships that could be negatively impacted by releasing the Responsive Information will be addressed in the MOD’s CLOSED evidence. c. Further, UK counter-terrorism operations in Syria and Iraq are conducted as part of the global coalition against Daesh. Contrary to the Appellant’s submissions, the support the UK receives on these operations from coalition partners is based on trust and an understanding that confidential and classified information will remain as such. Crucial enablers to UK counter-terrorism operations such as intelligence sharing, partnered operations, and use of coalition assets are all based on trust, and an understanding that specific information relating to national security techniques and capabilities will not be released in public. Therefore releasing the Responsive Information into the public domain could prejudice this trust and the relationships that are built upon it. Compromise of these relationships would also significantly diminish the UK Armed Forces’ ability to successfully conduct counter-terrorism operations in the future, which also goes to the application of s.26(1)(b) FOIA. d. Even outside the context of counter-terrorism operations in Syria and Iraq conducted as part of the global coalition against Daesh, other allies may be less likely to work alongside and support UK Armed Forces if they believed that classified information shared in confidence may later be released to the public. 69. In relation to both countries, Mr Oliver gave further evidence in closed. He accepted that harm to international relations was likely to be reduced in the case of an airstrike that had already been declared, but argued that in his experience it was unlikely to be mitigated by countries’ own practices as regards disclosure of information. 70. On the evidence, we find sections 26 and 27 to be engaged. 71. On the extent to which disclosure would prejudice the interests described in sections 26 and 27, and accordingly the balance of the public interest, our conclusions on two other points made by the parties are as follows: a. Parallel disclosure i. Airwars argues that prejudice to defence can hardly arise from disclosure of information (or types of information) that has already been released in far greater volumes by other states such as the USA. While acknowledging the MOD’s point on differing FOIA regimes, we agree with Airwars that the revelatory consequences of disclosure will often be lesser if identical or similar disclosed material has already been disclosed elsewhere. ii. In particular, if (for example in the Khan / New York Times disclosures) material originating from the UK government has been released by the USA, in some cases release of that same information by the UK will have little or no revelatory consequence. iii. The MOD is nonetheless right to argue that the opposite may also sometimes be the case: information revealed twice may strengthen its provenance, and additional nuances in the second disclosure, whether in formatting, timing, context or the like, may tell an adversary something of use. This must be assessed on a case-by-case basis, and the Tribunal expects some sort of rational possibility of that occurring to be demonstrated. There is also the ‘jigsaw’ effect, but again this must not be too casually invoked; it will only weigh in favour of withholding requested information where there is a reason to rationally suppose that a risk may exist. iv. We have approached this issue in light of what is said above, and our resulting consideration of the withheld information cannot be set out openly. It is nonetheless proper to record, as a very broad characterisation, that while the MOD’s concerns over revelatory consequences do have some justification, in some cases they do arise from a blanket assumption that serious consequences might flow from parallel disclosure. b. Relevance to the exemptions i. Airwars argues that some documents likely to fall in scope of the request cannot rationally engage the exemptions: (1) The simple fact of whether the strike caused civilian injuries (as well as the single recorded fatality); (2) Most importantly, the terms of reference and procedure followed for any assessment of civilian harm caused by the March 2018 Strike, including the standard of proof applied,14 the government/military bodies involved (e.g. whether there is any ministerial involvement), the involvement of Coalition partners in any investigation, the existence of independent oversight or other safeguards, and the position(s) held by the final decision maker(s); (3) Certain conclusions reached in the course of any investigation, such as whether internal operational failures (such as failure to follow a given procedure or cultural blind spots) contributed to the civilian death; and (4) Whether the MoD considered the safety and wellbeing of the family members of the civilian killed following the strike (in particular on risks that they would incorrectly be associated with IS). ii. The MOD highlights the speculative nature of that argument in what we consider to be rather uncharitable terms. It is entirely proper (and indeed often necessary) for a requester to speculate what information may be in scope, and to make hypothetical submissions in case it is right. We also reject the MOD s assertion that matters such as those described above should have been particularised in the request if Airwars expected them to be disaggregated. It is for the public authority to identify where partial disclosure can be made in response to a request. We have fully undertaken that task in our own consideration. c. Confidentiality i. Mr Perry rightly observed that the MOD has never sought to rely on section 27(2), which exempts information “if it is confidential information obtained from a State”. We agree that section 27(2) does not fall to be considered, and the mere fact that information is confidential does not provide an absolute answer to section 27(1). ii. We nonetheless reject that the existence of section 27(2) informs our approach to section 27(1). The purpose of section 27(2) is to confer exemption on information received in confidence from another state without any need to show that harm to international relations would result from its disclosure (of course, the absence of such harm might still be relevant to the public interest balancing test). Under section 27(1), confidentiality is simply a potential aspect of the factual matrix considered when assessing prejudice to international relations. 72. We address the submissions made in relation to particular countries as follows: a. The United States of America and the coalition i. Airwars argues the coalition partners are mature democracies with respect for the rule of law and with their own Freedom of Information regimes. They will expect that from time to time a body independent of the government may order disclosure of information, so it is less likely to harm diplomatic relations. This is starkly illustrated by the wide-ranging Khan / NYT disclosures, ordered under Freedom of Information legislation. The USA is unlikely, argues Airwars, to take offence at the same happening here but on a more limited basis. ii. We agree that, in general and with all else being equal, a country with its own tradition of government transparency is less likely to be concerned at its operation here. b. Iraq Iraq does not have a tradition of government transparency, so accordingly may have less sympathy with disclosure. While no one sought to disagree with that characterisation, in fairness it should be qualified as an assumption rather than based on any evidence provided to us. We accept Airwars’ argument that Iraq would still understand that it was dealing with a country that does have a Freedom of Information regime, and would (or should) have appreciated that there is always a risk that information will be disclosed. Moreover, the public interest in maintaining the exemption may be reduced if the effect is to impose the other state’s political norms on the UK. Section 42 73. Section 42(1) provides that information in respect of which a claim to legal professional privilege or, in Scotland, to confidentiality of communications could be maintained in legal proceedings is exempt information. This includes legal advice privilege, which attaches to all communications passing between a client and its lawyers, acting in their professional capacity, in connection with the provision of advice which "relates to the rights, liabilities, obligations or remedies of the client either under private law or under public law": Three Rivers (No 6) [2005] 1 AC 610. Such advice must nonetheless be the "dominant" purpose of the document or relevant information. As explained in Jet2 v Civil Aviation Authority [2020] EWCA Civ 35 at [93], privilege cannot prevent disclosure: …simply because the relevant material has, as simply one, minor purpose, the obtaining of legal advice, without consideration of the respective weight of purpose … otherwise … swathes of internal and external material could be excluded from disclosure simply because a lawyer had been copied in and asked for his legal advice as and when he considered it appropriate to give it. 74. Mr Oliver confirmed in his open evidence that MOD legal advisers are involved at all levels, down to the actual conduct of operations. Mr Kosmin argued that “it is plain on the face of [the information] that it attracts legal advice privilege”; Airwars questions whether it can really be so plain, given that engagement of section 42 was not identified by the MOD in its initial response, internal review or during the Commissioner’s investigation. 75. Two passages within the requested information were claimed by the MOD to attract legal professional privilege. Having considered the relevant material for ourselves in closed session, we agree that one plainly contains legal advice and meets the criteria above. The other simply refers to legal advice having been taken and the exemption is not engaged. 76. As with sections 24, 26 and 27, the operation of the exemption depends on the outcome of a public interest balancing test. In both DBERR v OJ Brien [2009] EWHC 164 and Callender Smith v IC and CPS [2022] UKUT 60 (AAC), the fundamental importance of legal professional privilege to the administration of justice was held to carry significant weight in favour of maintaining the exemption. 77. The MOD argues that the usual weight in maintaining this exemption described above is further increased where, as put by Mr Kosmin, “the advice went to the conduct of the strike and the personnel who undertook the strike”. We agree. The weight carried in the public interest balancing test is otherwise informed by the matters already rehearsed. Conclusion on the public interest balancing test 78. We carry forward all the above evidence and argument into the balancing exercise, save where we have given reasons for rejecting it. While in formal terms we have resolved the issue of timing against Airwars’ reliance on post-response events (see [15] -[17] above), on careful consideration we conclude that we would reach the same overall conclusion whichever party is right. Likewise, due to the degree of overlap between the exemption at section 24 and those at sections 26 and 27, whether or not the former is actually engaged is immaterial to the final balance. 79. One matter upon which we can be more specific is legal professional privilege. The ability to obtain frank and fully informed legal advice is vital for the UK government and its armed forces’ compliance with LOAC and their other public and private legal obligations, and assessment of civilian harm in a military operation is a context in which it carries great importance. 80. In conclusion, we reiterate that we are not concerned with determining the rights and wrongs of UK military action abroad, participation in the Coalition, the use of drones as a combat technique, or whether the UK has complied with LOAC. Nor should we be taken as making findings of fact in relation to any of the matters discussed; this is not our function, nor is it necessary to decide the issues before us. The import of the evidence is in its illustration of the issues and debate surrounding the assessment of civilian harm arising from UK drone strikes. 81. Our overall conclusion is that the entirety of the requested information is exempt. In relation to the qualified exemptions engaged, the aggregated public interest in maintaining those exemptions outweighs the public interest in disclosing the information. We recognise that the reader of this decision, having persevered through over 10,000 words describing the evidence and arguments, may still feel rather in the dark as to what we made of them. This is an unfortunate but necessary consequence of the issues engaged, and all we can do is repeat our assurances that we have given the appeal our most anxious consideration. SignedDate: UTJ Neville8 December 2025 (sitting as Judge of the First-tier Tribunal) Appendix 1 — Reasons given by Judge Neville for the case management decision refusing to appoint a special advocate 1. The appellant made an application on 10 May 2023 that the Tribunal appoint a special advocate to represent the appellant’s interests closed evidence and submissions is considered. I need not set out the background or substantive issues in this case, which are already described in the documents filed by the parties. It is sufficient to observe that the MOD resists disclosure of the requested information on the basis of the exemptions at s.23, s.24(1), s.26(1)(b), 27(1)(a), s.40(2) and s.42(1) of FOIA. The first of those two exemptions are relied upon as applying in the alternative without it being disclosed which of them actually does, as discussed in Foreign, Commonwealth and Development Office v Information Commissioner, Williams and others [2021] UKUT 248 (AAC). Insofar as a balancing exercise must be conducted in relation to any of those exemptions, the appellant’s case on the public interest in disclose of the requested information is summarised at paragraph 9 of the application. It is common ground that there is significant public interest in disclosure. 2. The MOD will rely on closed evidence and submissions, and ask the Tribunal to sit in closed session in order to do so. Everyone apart from the two respondents and their representatives will be excluded from the hearing, including the appellant and his representatives. 3. The covering email to the application states that it is made with “the aim of ensuring the appeal can proceed fairly given the absence of the Information Commissioner”. This is because the Commissioner had proposed that he not be represented at the hearing. A lack of such representation, argued the appellant, would remove the basis upon which the Court of Appeal, in Browning v The Information Commissioner & Anor [2014] EWCA Civ 1050, had held that consideration of closed material is fair. 4. In response to the application I made directions asking, in brief summary, for submissions on whether any procedural unfairness would be remedied by ordering the Commissioner to be represented, whether the First-tier Tribunal has power to appoint a special advocate (or request that a Law Officer do so), and on the practicalities of the appointment. I address those issues as follows. Should the Tribunal direct the Commissioner to attend, and would this remedy any procedural unfairness arising from a closed material procedure? 5. The Commissioner accepted that the Tribunal has the power to order him to be represented at the hearing. He nonetheless argued that this was unnecessary for a fair hearing, as the Tribunal could exercise its own independent scrutiny. Support for this could be found in the FtT case of British Union for the Abolition of Vivisection v ICO and Newcastle University EA 2010/0064 (BUAV), (and affirmed in Browning): However, irrespective of the assistance of the Commissioner, the Tribunal, as a specialist tribunal, can be expected to be able, at least in some cases, to assess for itself the application of the provisions of FOIA to the closed material…the extent to which the tribunal will be in a position to do this will depend upon the particular circumstances. 6. The Commissioner reminded the Tribunal of the current pressure upon his resources, asking that it be balanced against the benefits of his attendance. 7. Both the appellant and the MOD argued that a direction should be made requiring the Information Commission to be represented, and that this would assure a fair hearing of the appeal in which the evidence and submissions in closed session can be tested appropriately. The MOD’s written submissions made 5 numbered arguments on why such a direction should be made in this particular case. Subject to one qualification, I agree with each of them. The presence of a representative for the Commissioner was central to the reasoning in Browning concerning fairness. It is particularly worth emphasising the role of the Commissioner in FOIA appeals, both as a matter of principle (Lubicz v Information Commissioner & King's College London [2015] UKUT 555 at [51]; Browning at [33]) and practical experience (Corderoy v Information Commissioner [2017] UKUT 495 (AAC) at [85]; Lownie v Information Commissioner & FCDO [2023] UKFTT 397 (GRC) at [56]). 8. The qualification I mention above is that Browning cannot be read as requiring the Commissioner’s presence as a fundamental pre-requisite for fairness in every case where there is closed material. For example, it is difficult to see how fair consideration of the closed material in Stephen Campbell v Information Commissioner & HM Treasury [2023] UKFTT 885 (GRC), being the exact figures put forward by the public authority as being over the s.12 cost limit, could require the Commissioner to be represented. In the present appeal however, a considerable amount of closed evidence will be considered, the public interest in disclosure is acknowledged to be significant, and the respondents argue for the s.23/s.24 ‘masking’ principle to be applied. Such issues also arose in the case of Lownie, from which the following illustrative extracts can be provided: 6. The effect, and indeed the acknowledged purpose, of reliance upon two alternative exemptions is to mask which one of them actually applies. So, Dr Lownie was not only denied the information to which he claims the law entitles him but was kept in the dark as to why. When lodging this appeal, he argued that this approach was unfair and unlawful. That argument was considered as a preliminary issue in this appeal and two other appeals together, and ultimately by the Upper Tribunal in Foreign, Commonwealth and Development Office v Information Commissioner, Williams and others [2021] UKUT 248 (AAC) (we shall refer to that authority as Williams to avoid confusion with another Upper Tribunal case concerning Dr Lownie). The Upper Tribunal held that the Act does permit a public authority, in order to protect national security, to ‘mask’ the actual exemption that applies by reliance upon ss.23 and 24 in the alternative. While this practice may well put the requester at a disadvantage, the remedy lies with the Commissioner (and, on appeal, the Tribunal) taking steps to be satisfied that one of the exemptions has been properly claimed. As well as stating those matters of principle, at [58] the Upper Tribunal held that in this specific appeal the FCDO is entitled to argue both exemptions in the alternative. That preliminary issue having been decided, the appeal has returned to this Tribunal. […] 54. We have been ably assisted in relation to the applicable legal principles by all three parties. 55. As to their application to the requested information, in his complaint to the Commissioner Dr Lownie said: I cannot believe, having released so much on the Burgess & Maclean case, which is of historical importance and public interest, and almost seventy years after they fled to Russian [and with them all] dead that s.24 applies. 56. There is little else he can say, of course, because neither he nor Mr Callus have been permitted to see the files nor any of the respondents’ submissions on their contents. The input of the Commissioner in the CLOSED hearing, by way of Mr Knight’s involvement and as predicted in Browning at [33], was of assistance in ensuring that we were best able to fulfil our own independent investigatory role and take into account such points as Dr Lownie might have made if he were there. We anticipate that attendance of a representative for the Commissioner will be essential in all cases where s.23 and s.24 are argued in the alternative. 57. The remainder of the arguments made cannot be disclosed in these OPEN reasons. … 58. We have decided that all the contents of both files are exempt from disclosure on the basis of section 23(1) or section 24(1) of FOIA. None of our reasoning can be openly disclosed without compromising the ability of the FCDO to rely on the two exemptions in the alternative, so it is set out in our CLOSED reasons. We can nonetheless assure Dr Lownie that we have taken the utmost pains to ensure that this outcome is correct, and that FOIA has been correctly applied. 9. The Tribunal’s experience of the contribution that representation of the Commissioner makes towards proper scrutiny and fairness is re-emphasised, and is not limited to that appeal. 10. Whether to order representation of the Commissioner in a particular appeal will be case-sensitive, and a matter to which the parties should direct early attention. I am entirely satisfied that it is necessary in this appeal. If Campbell is at one end of the spectrum, then this appeal is at the other. I therefore direct that the Commissioner is represented at the hearing. Does the Tribunal have the power to appoint (or request a Law Officer to appoint) a special advocate? 11. The appellant accepts that directing the Commissioner to be represented is sufficient to address his concerns so, that direction having been made, the appointment of a special advocate can be taken as no longer pursued. Applying the overriding objective, representation by the Commissioner is clearly preferable to that latter option in these circumstances. Appointment of a special advocate would have required numerous practical issues to be addressed, likely with significant cost and delay. 12. For the above reasons a decision on whether there is a power to appoint a special advocate does not strictly arise. In deference to the parties’ full and helpful submissions, I can indicate that my decision would be that the Tribunal can made such an appointment. 13. The Court of Appeal in Browning left the matter unresolved, no doubt due to its central conclusion that the closed material procedure now ordered in this appeal was sufficient, but directed itself at [1] by the absence of an “express” power of appointment in the relevant procedure rules. I drew the parties’ attention to AM (Afghanistan) v SSHD [2017] EWCA Civ 1123, where the Court of Appeal held where a Tribunal’s procedure rules lacked an express power to appoint a litigation friend, this was no barrier where such an appointment was required for a fair hearing. Contrary to the MOD’s submissions, there is no reason why the Tribunal’s power to regulate its own procedure would not likewise empower it to appoint a special advocate. If, as the MOD argues, the Court of Appeal did not properly identify the source of the power to appoint a litigation friend, then that is hardly a matter for this Tribunal; I am bound by the outcome that it does, and by its existence depending on the requirement for a fair hearing. 14. While the MOD is correct that appointment of a special advocate has been put on a formal statutory footing in many cases, it also acknowledges at para 23 of its submissions that other courts have identified an implied power of appointment where necessary; that they have also set a high threshold for appointment does not answer the question of jurisdiction. This Tribunal’s predecessor, the Information Tribunal, appointed a special advocate in Campaign Against the Arms Trade v IC and MOD (EA/2006/0040). There is nothing in the two Tribunals’ respective procedure rules to justify a conclusion that an express power existed in the former but not the latter, notwithstanding the MOD’s dissection of the various differences in wording between them. The 2014 procedure rules, in particular at r.5(1), are wider in scope. Finally, it is of minor persuasive relevance that the Attorney General’s guidance “A Guide to the Role of Special Advocates & the Special Advocates’ Support Office” cites Freedom of Information as an area of law where special advocates can be appointed. This reflects both the guidance’s overall treatment of the issue and that of the authorities, being that the power is implicit and arises in any adjudicative process where such a step becomes necessary for a fair hearing. 15. I need not address practicalities and funding, which would need careful attention in the context of a specific case.