|[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]|
Upper Tribunal (Administrative Appeals Chamber)
You are here: BAILII >> Databases >> Upper Tribunal (Administrative Appeals Chamber) >> All Party Group on Extraordinary Rendition v The information Commissioner and Foreign and Commonwealth Office  UKUT 560 (AAC) (11 November 2013)
Cite as:  UKUT 560 (AAC)
[New search] [Context] [View without highlighting] [Printable RTF version] [Help]
IN THE UPPER TRIBUNAL Case No.: GIA/2230/2012
ADMINISTRATIVE APPEALS CHAMBER
Before: Mr Justice Charles CP
Mr Justice Burnett
Upper Tribunal Judge Wikeley
For the Appellant: Mr Timothy Pitt-Payne QC and Miss Joanne Clement, instructed by Hogan Lovells International LLP
For the First Respondent: Mr Robin Hopkins, instructed by the Solicitor to the Information Commissioner
For the Second Respondent: Ms Karen Steyn and Mr Julian Blake, instructed by the Treasury Solicitor
The INTERIM DECISION of the Upper Tribunal is to allow the appeal in part.
The decision of the First-tier Tribunal (General Regulatory Chamber) (Information Rights) dated 03 May 2012, in relation to the Appellant’s appeals against the Information Commissioner’s Decision Notices FS50262409, FS50279042 and FS50296953, involves an error on a point of law (in relation to Ground 4 of the grounds of appeal). The appeal is therefore allowed in part.
This decision is given under section 12(2)(a) and 12(2)(b)(ii) of the Tribunals, Courts and Enforcement Act 2007.
1. This is an appeal by the All Party Parliamentary Group on Extraordinary Rendition (“APPGER”) against the decision of the First-tier Tribunal (“the FTT”), which for the most part dismissed APPGER’s original appeals against three Decision Notices issued by the Information Commissioner (“the IC”). The IC, in turn and again for the most part, had upheld the reliance by the Foreign and Commonwealth Office (“the FCO”) on various exemptions under the Freedom of Information Act 2000 (“FOIA”) in response to three batches of requests for information made by APPGER.
2. This is an interim decision. APPGER has five grounds of appeal against the FTT’s decision in respect of its three appeals. The first two grounds of appeal raise matters which are likely to be affected by the forthcoming decision of the Supreme Court on the appeal from the Court of Appeal’s decision in Kennedy v IC  EWCA Civ 317;  1 WLR 3524. Those two grounds have accordingly been stayed. This interim decision, therefore, deals only with grounds 3, 4 and 5.
The background to the APPGER requests
The role of APPGER
3. APPGER is a cross-party association of MPs, established in 2005 by Mr Andrew Tyrie MP in response to allegations that the UK Government had been involved in the US extraordinary rendition programme. Extraordinary rendition is a euphemism for an extraordinary practice, namely the extra-judicial transfer of detainees, typically individuals “of interest to the security services”, and usually across state boundaries or between different authorities within them, for the purposes of interrogation, often in circumstances where those individuals face a real risk of torture.
4. APPGER’s investigations into this practice have included making several FOIA requests to Government departments and, where disclosure of information has not been forthcoming, pursuing complaints to the IC and appeals to the FTT. An earlier such appeal to the FTT was heard by way of a discretionary transfer in the Upper Tribunal (APPGER v Ministry of Defence  UKUT 153 (AAC)). There is a comprehensive review of the background to the present appeal in the FTT’s decision (EA/2011/0049-0051). Much of the context of this appeal is also in the public domain by way of judgments in the Divisional Court and the Court of Appeal. We do not doubt the importance and gravity of the underlying issues, but this extensive prior judicial treatment means that we can summarise the context of this case rather more shortly than would otherwise be appropriate.
The context to the APPGER requests
5. The APPGER requests in issue in this appeal related to the cases of three individuals, each of whom was subject to extraordinary rendition: Mr Bisher al-Rawi, Mr Jamil el-Banna and Mr Binyam Mohamed.
Mr al-Rawi and Mr el-Banna
6. Mr al-Rawi is an Iraqi citizen but a longstanding British resident. Mr el-Banna is a Jordanian national but with refugee status to remain in Britain. In November 2002 Mr al-Rawi and Mr el-Banna were both detained at Gatwick Airport under the Terrorism Act 2000 before boarding a flight to The Gambia. Shortly afterwards, they were both released and allowed to continue their journey. On arrival in The Gambia, however, both men were arrested on suspicion of having links with al-Qaeda. In December 2002 they were moved to Afghanistan and in March 2003 transferred to Guantanamo Bay. Mr al-Rawi was not released from Guantanamo Bay until March 2007, with Mr el-Banna following in November 2007.
7. These events have led to a total of seven court judgments, culminating in Al Rawi and Others v Security Service and Others  UKSC 34;  1 AC 531. The Supreme Court held unanimously that there is no power at common law to replace public interest immunity (“PII”) – under which a judge decides whether in the public interest certain material should be excluded from a hearing – with a closed material procedure. A majority of the Supreme Court further held that there is no power at common law to introduce a closed material procedure following the conclusion of the normal PII process.
8. Mr Binyam Mohamed is an Ethiopian national who was given leave to remain in the UK (for four years) in 2001. In April 2002 the Pakistani authorities seized him at Karachi airport. Shortly afterwards, the US authorities informed the UK’s security services that they were holding Mr Mohamed and sought information about him. There is now no dispute that (i) Mr Mohamed was subject to serious mistreatment and torture while being “interrogated” in Pakistan; (ii) the UK authorities were sent a report by their US counterparts about Mr Mohamed’s treatment; and (iii) an officer of the UK security services also interviewed Mr Mohamed whilst he was being held in Pakistan. Mr Mohamed was subsequently moved to Morocco in July 2002, to Afghanistan in January 2004, being subject to further mistreatment in both countries, and then to Guantanamo Bay in September 2004. He was detained for a further four years before being charged with terrorism offences under the US Military Commissions Act. However, in November 2009 the District Court for the District of Columbia accepted Mr Mohamed’s account, describing his treatment as torture, and ruled that his confessions could not be used to detain him (Farhi Saeed Bin Mohamed v Obama, Civil Action No. 05.1347).
9. These events have led to an even greater body of litigation in the UK courts (“the BM litigation”), resulting in seven Divisional Court and two Court of Appeal judgments. The first three Divisional Court judgments concerned Mr Mohamed’s application (in proceedings commenced in May 2008) that the UK Government should disclose certain documents on a confidential basis to his US lawyers. In August 2008 the Foreign Secretary provided the High Court with a PII certificate to the effect that it was in the public interest that the documents should not be so provided. However, in October 2008 Mr Mohamed’s US lawyers gained access to the relevant documents through habeas corpus proceedings in that jurisdiction.
10. Thereafter, the only live issue in the BM litigation was whether the Divisional Court should restore seven short paragraphs to its first judgment (at  EWHC 2048 (Admin)). This passage contained the gist of reports from the US authorities to the UK authorities as to Mr Mohamed’s detention and treatment between 2002 and 2004. The Foreign Secretary issued two further PII certificates in August and September 2008, asserting that the position of the US Government was that, if the paragraphs were published, then it would re-evaluate its intelligence sharing relationship with the UK Government, which would itself seriously prejudice UK national security. The Divisional Court in effect accepted those certificates in its fourth judgment, delivered on 4 February 2009 (R (on the application of Mohamed) v Secretary of State for Foreign & Commonwealth Affairs (No 2)  EWHC 152 (Admin);  1 WLR 2653), concluding (at paragraph ) that:
“… In short, whatever views may be held as to the continuing threat made by the US Government to prevent a short summary of the treatment of BM being put into the public domain by this court, it would not, in all the circumstances we have set out and in the light of the action taken, be in the public interest to expose the United Kingdom to what the Foreign Secretary still considers to be the real risk of the loss of intelligence so vital to the safety of our day-to-day life. If the information in the redacted paragraphs which we consider so important to the rule of law, free speech and democratic accountability is to be put into the public domain, it must now be for the US Government to consider changing its position or itself putting that information into the public domain.”
11. Shortly afterwards, on 15 February 2009, The Observer newspaper published a story (under the headline “Foreign Office link to torture cover-up”), citing an unnamed “senior” US source, claiming that “‘Far from being a threat, it was solicited [by the Foreign Office]’”.
12. In October and November 2009 the Divisional Court, in its fifth and sixth judgments ( EWHC 2549 (Admin) and  EWHC 2973 (Admin)), subsequently reversed the effect of its fourth judgment, and ruled that the redacted paragraphs could be included in the open version of its first judgment. In February 2010 the Court of Appeal then dismissed the Foreign Secretary’s appeal against the Divisional Court’s fifth and sixth judgments (R (on the application of Mohamed) v Secretary of State for Foreign & Commonwealth Affairs  EWCA Civ 65;  QB 218). A key factor in the Court of Appeal’s reasoning was that the truth of Mr Mohamed’s allegations had by then both been accepted by, and put in the public domain by, the US District Court (in November 2009).
The APPGER requests, the FCO responses and the IC and FTT decisions
13. APPGER’s requests to the FCO under FOIA fell into three groups. For convenience we call them (1) the al-Rawi requests; (2) the BM letter request; and (3) the BM interrogation requests.
The al-Rawi requests
14. Initially, in May 2008, APPGER made a total of 22 detailed requests to the FCO. The first five requests did not form part of the subsequent appeal to the FTT. Requests 6-13 (R6-R13) related to the UK’s involvement in the extraordinary rendition of Mr al-Rawi and Mr el-Banna from The Gambia to Afghanistan and eventually to Guantanamo Bay. By way of example, R8 sought:
“all information relating to the threat to the security of Britain or any other nation posed by Bisher al-Rawi and Jamil el-Banna [and] the work allegedly carried out for the intelligence services by Bisher al-Rawi...”.
15. The FCO provided some of the information sought but withheld other material, claiming various exemptions under FOIA, including (but not limited to) sections 23(1) (information supplied by, or relating to, bodies dealing with security matters), 27(1)(a) (international relations), 35(1)(a) (formulation or development of government policy) and 42(1) (legal professional privilege).
16. In his first Decision Notice (FS50262409), the IC upheld the FCO’s reliance on the claimed exemptions as regards the majority of the documents, with some limited exceptions. The FTT largely upheld the IC’s decision, save in relation to four specified documents in respect of which the tribunal substituted its own decision, concluding that part or all of the material should be disclosed as no relevant exemption applied.
The BM letter request
17. On 18 February 2009 – three days after The Observer published its ‘scoop’ – APPGER made a further (unnumbered) request to the FCO, referring to the newspaper’s allegations, and asking for “all relevant information on this issue, including correspondence with the US Administration, redacted where necessary”. The FCO’s response was to refer to some relevant material which was already in the public domain, but to refuse to disclose other information on the basis that the four exemptions referred to above in relation to the al-Rawi requests applied.
18. In his second Decision Notice (FS50279042), the IC upheld the FCO’s reliance on the claimed exemptions, except for four documents in respect of which disclosure was ordered, subject to appropriate redactions. The FTT dismissed APPGER’s subsequent appeal and upheld the IC’s Decision Notice, but allowed the FCO’s cross-appeal on a data protection issue which is no longer live.
The BM interrogation requests
19. R14-22 of the original May 2008 requests all concerned the UK’s involvement in the extraordinary rendition of BM and his subsequent treatment, as well as information held on US interrogation practices and on US-sourced information on various alleged terrorist plots. The only request which is relevant for the purposes of this appeal is R14, which was for
“all information relating to any visits by UK intelligence officers to British resident Binyam Mohamed al-Habashi while he was being held in Karachi in 2002”.
20. The FCO’s response to R14 was to refer to the open version of the Parliamentary Intelligence and Security Committee (ISC) report on Rendition (Cm 7171, 28 June 2007), which included relevant, but partly redacted, passages. For example, paragraph  of the open version of the Committee’s report read as follows:
***. In giving evidence to the Committee in 2006, the Director General of the Security Service told us:
(a) ... when we knew he [BM] was in custody, because we had information we believed relevant to the UK from having lived here, ***
(d) ***. That is a case where, with hindsight, we would regret not seeking proper full assurances at the time...”
21. The FCO claimed section 23(1) in relation to all other material held relating to R14.
22. In his third Decision Notice (FS50296953), the IC ruled that the FCO was entitled to refuse to provide the requested information under R14 on the basis of section 23(1) of FOIA. The FTT dismissed APPGER’s subsequent appeal and again upheld the IC’s Decision Notice.
The procedure adopted by the First-tier Tribunal: the hearing
23. On 18 May 2011 Judge Angel issued detailed case management directions for the process leading up to, and the hearing of, APPGER’s appeal before the FTT. These directions included provision for the hearing “to be heard partly in open and partly in closed because of the need to hear evidence and submissions relating to the detail of the disputed information” (paragraph ). The directions also made provision for a closed bundle of documents (i.e. one available to both Respondents and to the tribunal, but not to the Appellant), along with closed witness statements and closed skeleton arguments, as well as open versions of the same documents, available to all. These case management directions were in line with the FTT’s then Practice Note, Protection of Confidential Information in Information Rights Appeals before the First-tier Tribunal in the General Regulatory Tribunal on or after 18 January 2010 (1 February 2010). We note that this document has since been replaced by a new Practice Note, Closed Material in Information Rights Cases (May 2012).
24. The FTT sat to hear APPGER’s appeal on 10-11 and 14-15 November 2011, followed by a telephone case management hearing on 9 December 2011, and then held two further days of hearings on 27 and 28 February 2012. Most of the first two days in November 2011 were spent in open session, followed by two days in closed session when the FTT examined the disputed materials and heard closed evidence and submissions. In the course of the latter two closed days, Mr Hopkins, counsel for the IC, put to the FCO’s witness, Mr Jonathan Sinclair, a range of points raised in advance by Ms Clement, APPGER’s counsel, who was of course excluded from that part of the hearing (Ms Clement’s comprehensive note, setting out the lines of enquiry she would have pursued in cross-examination of Mr Sinclair, had she been able to attend the closed hearing, ran to 14 pages and 69 questions). The FTT heard the parties’ closing submissions at the hearings in February 2012.
The procedure adopted by the First-tier Tribunal: the decision
25. On 14 December 2011, shortly after the telephone case management hearing, Judge Angel issued a ruling for the further conduct of the case, including detailed provision (running to over two sides) for the FTT to issue a “provisional draft open judgment” on a confidential basis, initially to the Respondents to ensure that nothing had been inadvertently included in the open version (but which properly belonged in the closed version), and then generally to all parties by way of a “reconsidered draft open judgment”. All parties were invited “to submit a list of typing corrections and other obvious errors in writing, so that changes can be incorporated, if the Tribunal accepts them, in the handed down open judgment” (paragraph ).
26. The FCO, in response to the “provisional draft open judgment”, submitted a one-page document to the FTT comprising three “points of clarification” and a number of spelling corrections. One of the “points of clarification” related to the FTT’s description of the “control principle” in paragraph  of its reasons. We return to this in more detail below in the context of the second ground of appeal dealt with in this decision. Suffice to say for present purposes that APPGER, in a letter of 4 April 2012, argued that this particular suggestion by the FCO went “well beyond pointing out a clerical mistake or other accidental slip or omission”. Rather, APPGER argued, it sought “to influence and change a fundamental error of fact affecting the foundation of the Tribunal’s reasoning concerning the public interest balancing exercise.”
27. Having made a number of relatively minor amendments to the text of its decision, the FTT issued its final open decision with reasons on 3 May 2012. This was issued alongside a redacted annexe to the open decision, comprising five short paragraphs (three of which were redacted) and a 28-page schedule in the form of a Table, specifying in relation to each document number (a) the particular numbered FOIA request it related to; (b) whether or not there had been partial disclosure; and (c) the relevant exemptions respectively claimed by the FCO, accepted by the IC and found by the FTT; along with (d) a summary of which factors applied in the public interest test (so far as the qualified exemptions were concerned). There was no separate closed reasoned decision. However, the closed annexe included the three paragraphs redacted from the open version, which provided some summary closed reasons. In addition, the closed version of the schedule included the ‘populated’ version of the “description of document” column in the Table, which had been left blank in the open version, as well as restoring several redacted entries in other columns.
APPGER’s grounds of appeal against the First-tier Tribunal’s decision
28. Judge Angel subsequently gave APPGER permission to appeal to the Upper Tribunal. APPGER’s grounds of appeal were five-fold, namely that the FTT had erred:
(1) in its approach to Article 10 of the European Convention on Human Rights;
(2) in its approach to the phrase “relates to” in the exemption under section 23(1) of FOIA (“Information held by a public body is exempt information if it ... relates to any of the [security] bodies specified in subsection (3))”;
(3) in failing to provide adequate reasons for its conclusions on the application of section 23(1);
(4) in its approach to the control principle, and in so doing erred in the balancing exercise to be conducted under section 27(1) of FOIA (international relations);
(5) in its approach to the formulation and development, as opposed to the implementation, of policy under section 35(1)(a) of FOIA.
The proceedings before the Upper Tribunal
29. On 11 February 2013 the Upper Tribunal stayed grounds 1 and 2 pending the outcome of the Supreme Court’s decision in Kennedy v IC. The Upper Tribunal heard oral argument on the remaining three grounds of appeal between 1 and 4 July 2013. The hearing was open except for about half of the first day and most of the last day; during the closed parts of the hearing we heard closed submissions from Mr Hopkins for the IC and Ms Steyn for the FCO, and were taken carefully through the requested information, document by document, by Ms Steyn. All the parties helpfully produced a coloured schedule identifying the documents that they agreed were and were not the subject of the appeal and the documents where there was a dispute as to whether they remain part of the appeal and the nature of the dispute in respect of each document.
Ground 3: the reasons challenge to the section 23(1) findings
30. Section 23(1) of FOIA provides that “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).” The bodies so specified include the Security Service (or SyS/MI5: subs. (3)(a)), the Secret Intelligence Service (SIS/MI6: subs. (3)(b)) and the Government Communications Headquarters (GCHQ: subs. (3)(c)) (for convenience we call these generically “section 23 bodies”). This is an absolute exemption, and so no public interest test applies. This exemption had been claimed by the FCO for at least parts of all of the al-Rawi documents in dispute, except for one, and it was also in issue for a significant number of documents within the scope of the appeal on the BM letter request. Section 23(1) was also claimed for the single document in dispute as regards the BM interrogation requests (namely the closed version of the ISC Rendition report).
The approach of the FTT
31. The FTT dealt with the parties’ submissions on the scope of section 23 at paragraphs - of its Decision. The FTT then concluded as follows:
“70. To sum up we consider that the Tribunal should adopt a broad, although purposive approach to the interpretation of s.23(1). However this should be subject to a remoteness test so that we must ask ourselves whether the disputed information is so remote from the security bodies that s.23(1) does not apply.
71. The Tribunal have considered the disputed information where s.23(1) has been claimed in some detail. We find, following the legal principles set out above, that where the FCO has claimed the s.23(1) exemption that it is engaged. We observe that the FCO could, in our view, have claimed the exemption for even more information in this case.”
The parties’ submissions to the Upper Tribunal
32. Mr Pitt-Payne submitted, in essence, that paragraph 71 was no more than a bare conclusion by way of an assertion, with no explanation as to why the FTT considered that the requested information fell within section 23(1). He acknowledged that it may not be possible in a case such as this for a FTT to provide a fully reasoned open judgment, noting the constraints imposed by rule 14 of the Tribunal Procedure (First-tier Tribunal) (General Regulatory Chamber) Rules 2009 (SI 2009/1976; “the Tribunal Procedure (FTT)(GRC) Rules 2009”). In particular, rule 14(9) provides that in a case “involving matters relating to national security, the Tribunal must ensure that information is not disclosed contrary to the interests of national security”. Furthermore, according to rule 14(10):
“(10) The Tribunal must conduct proceedings and record its decision and reasons appropriately so as not to undermine the effect of an order made under paragraph (1), a direction given under paragraph (2) or (6) or the duty imposed by paragraph (9).”
33. However, Mr Pitt-Payne argued that the rationale for the duty to provide reasons was partly to concentrate the mind of the judge and partly to render practicable the exercise of rights of appeal. This was especially important in a case such as the present, where one of the parties had been excluded from parts of the proceedings. It was therefore incumbent on the FTT, he submitted, to demonstrate that it had applied its own independent judgement to the question of whether or not section 23(1) applied to the disputed documents.
34. Ms Steyn submitted that the adequacy of the FTT’s reasons had to be assessed in the context of the evidence and submissions before the tribunal, and that the degree of particularity required depended on the issues falling for decision. It was plain that the test that the FTT had applied was to ask itself how the FCO had come to hold the disputed information (i.e. whether it had been supplied directly or indirectly by a section 23 body) and/or was the information something to do with such bodies. The answer to that question was a simple binary “yes” or “no”; indeed, in many cases it was, in her terms, “blindingly obvious” (as, for example, in respect of the evidence from the Director-General of the Security Service; see paragraph 105 of the ISC Report, cited at paragraph 20 above). It was wholly unnecessary to impose a burden on the FTT to explain matters that were self-evident.
35. Mr Hopkins essentially adopted Ms Steyn’s submissions, noting also that this was not a case where the public authority had been testing or pushing the boundaries of what can properly be said to “relate to” a security body – indeed, the FTT had expressly stated that the FCO could have claimed the section 23(1) exemption for other information.
The Upper Tribunal’s analysis
36. The process of providing open and closed schedules was agreed or at least not objected to by the parties. As it explained, the FTT spent a day reading the documents before the hearing started and two days in closed session. The format of the schedules and the time taken in reading and in closed session show that, as one would expect, a considerable amount of careful work was done by the FTT on the documents. The FTT considered each disputed document, identified whether it was covered in whole or in part by section 23(1) (and/or indeed other exemptions), and did not simply accept the FCO’s claims.
37. So (should this be needed) there is clear confirmation that the FTT carried out the difficult and laborious task of considering the requested documents with considerable diligence and care. Very properly Mr Pitt-Payne accepted this and confirmed that his criticism was limited to a failure to give adequate reasons following that approach.
38. One of the cardinal principles in assessing the adequacy of any court or tribunal’s reasons is that they should be read as a whole. Mr Pitt-Payne’s analysis focussed almost exclusively on paragraphs 70 and 71, but this passage cannot be read in isolation. Those two paragraphs are preceded by over two pages of discussion and analysis of counsel’s submissions on the construction of section 23(1). We acknowledge that, if it is read in isolation, the summing up in paragraphs 70 and 71 can be said to be one that sets out an approach that uses words of degree and generalities and so does not indicate how the FTT has applied the statutory test set by section 23. But it is expressly a “summing up” and in our view the earlier paragraphs show how the FTT has applied that summary and thus the statutory test as it construed it. The earlier paragraphs show that on the issue of supply, whether direct or indirect, the FTT accepted that the test was essentially “how did the FCO come to have this information?” APPGER’s submissions on this issue were rejected as being too narrow. The FTT also rejected Ms Clement’s arguments on the meaning of “relates to”, in which she had sought to rely, by analogy, on data protection case law. The FTT concluded that the core of the test was whether the information was something to do with a section 23 body, subject to a remoteness test.
39. This is not, by any stretch of the imagination, a case where the FTT’s reasons amount to no more than “a bare traverse” of the relevant legislative provision (see e.g. R v Mental Health Review Tribunal, ex parte Clatworthy  3 All ER 699 at 703), or no more than an assertion that the FTT had read and applied the statutory test, which Mr Pitt-Payne came close to arguing. Rather the FTT has carefully set out the competing submissions, indicated clearly which it has preferred (and why) and then gone on to apply those tests to the disputed information. The methodology adopted was transparent, even if the detailed document-by-document findings in the open schedule could not be fully comprehensive, bearing in mind the constraints imposed by rule 14(9) and (10).
40. In closed session we explored with Ms Steyn whether some form of general description could have been applied to individual documents in the open version (e.g. “letter from minister A to minister B”). However, we were persuaded by her argument that to do so in the context of the absolute exemption provided by section 23 might in itself have the effect of disclosing exempt information by allowing others to “piece the jigsaw together”.
41. Further, and in any event, we have concluded that such a time consuming exercise (which would inevitably involve inviting submissions in closed session on the general description to be used in the open version) would provide little or no added value in the context of the absolute exemption provided by section 23. This is because any general description could do little more than mirror the language of the section. For example, a passage in the open reasons to the effect that “we accordingly found that section 23(1) applied to the information specified in the various individual documents in the attached schedule, which included material such as security assessments by section 23 bodies and counsel’s advice referring in turn to information provided by such agencies” effectively adds nothing to the approach adopted by the FTT of setting out its views on the way in which section 23 is to be applied to the requested documents, reading the documents with care and setting out its conclusion in an open and a closed schedule.
42. More generally, the problem for any decision maker in giving reasons explaining why, for example, a document need not be disclosed because it falls within or outside a statutory test, or is irrelevant to issues in litigation, or is covered by legal professional privilege, is a difficult one because they are inevitably linked to the content of the document that is not to be disclosed to the person to whom (amongst others) the reasons are directed. These problems are exacerbated by the nature of the section 23 exemption and rule 14(10).
43. It seems to us that there is strength in the view that we could have refused to embark on the examination of the documents in closed session that all the parties invited us to carry out, on the basis that the open explanation given by the FTT of its decision is adequate. However, we were persuaded to carry out this closed examination because we agree with Mr Pitt-Payne’s points that the rationale for providing reasons extends beyond the giving of open reasons to a party who is excluded from seeing the relevant documents and, in some cases (we emphasise “some”) there may be a need to provide detailed closed reasons to inform the appeal court or tribunal of the reasoning process by reference to the contents of the documents.
44. However, we pause to add that it seems to us that in many cases permission to appeal on the basis of a “reasons challenge” should not be given simply on the basis that the excluded party has not seen the documents or the closed reasoning (if any) and wants the appeal court or tribunal to check the conclusions reached on the application of an exemption to the requested information. In any event, when dealing with an application for permission the court or tribunal can consider the impact of the closed reasoning.
45. In our examination of the documents we have necessarily applied the test explained and adopted by the FTT in its open Decision and have done so, without prejudice to the parties’ future arguments on the stayed Ground 2. On that approach, and we suspect on any approach, the assessment of whether information “relates to” a section 23 body may have some “fuzzy” boundaries, and/or for other reasons, issues that give rise to a fine balance may arise. In our view, if that is the case it might trigger the need to provide closed reasons as to why the document fell on one side of the boundary rather than the other.
46. Our examination of the documents demonstrated that none of the documents for which the section 23 exemption was claimed came anywhere near such a fuzzy edge or boundary. They were, in Mr Hopkins’s words, “slam dunk” in the middle of section 23. So there is no need to explain to those who know the content of the documents how balanced or finely balanced decisions were reached.
47. Notwithstanding the (for the most part) “blindingly obvious” nature of the documents in question, we repeat that it is clear that the FTT approached its task diligently. Our examination shows that unsurprisingly there were one or two minor typographical glitches such as where an individual entry in a particular column was missed in the Table. For example, the FTT omitted to record any finding in respect of the exemptions applying to Document DN 1/16. However, it was plain on inspecting the document in question that section 27(1)(a) applied to the entire document and section 23 to some parts. This type of glitch did not come remotely near suggesting that the FTT had erred in law in any material way.
48. In carrying out this closed exercise we also considered the documents helpfully identified by the parties as being documents where there was a dispute as to whether the document remains part of the appeal.
49. Naturally we have considered what if any further reasons we should given in an open or closed decision or schedule for our conclusions agreeing with those of the FTT on the application of section 23.
50. We have concluded that there is no need for us to give any further reasons because we have identified the test we have applied in determining the issues and a reading of the document demonstrates in each case that the answer is clear.
Ground 4: the control principle and the public interest test under section 27(1)
51. Section 27(1) and (2) of FOIA provide for a qualified exemption in the following terms:
“27 International relations
(1) Information is exempt information if its disclosure under this Act would, or would be likely to, prejudice—
(a) relations between the United Kingdom and any other State,
(b) relations between the United Kingdom and any international organisation or international court,
(c) the interests of the United Kingdom abroad, or
(d) the promotion or protection by the United Kingdom of its interests abroad.
(2) Information is also exempt information if it is confidential information obtained from a State other than the United Kingdom or from an international organisation or international court.”
53. Although not claimed as widely as section 23(1) for the al-Rawi requests, section 27(1), and in particular usually section 27(1)(a), was claimed for a number of the documents which were subject to the appeal. Unsurprisingly, the FCO had also claimed it for a number of documents relevant to the BM letter request. However, section 27(1) did not arise at all in relation to R14 of the BM interrogation requests.
The approach of the FTT
54. The FTT considered the scope of section 27(1)(a) in the passage commencing at paragraph  of its Decision, and at paragraph  adopted the guidance of the Upper Tribunal (itself derived from case law in the superior courts) to the effect that “appropriate weight needs to be attached to evidence from the executive branch of government about the prejudice likely to be caused to particular relations by disclosure of particular information” in connection with the establishment of the real and significant risk that disclosure would prejudice the relationship between the UK and any other state. Applying that approach it concluded that section 27(1)(a) and (2) were engaged at paragraphs [133 -134].
55. The FTT then considered the public interest test for the qualified exemptions. It first considered and weighed the various public interest factors in favour of disclosure (paragraphs [154-165]), then the factors in favour of maintaining the exemptions (paragraphs -), before turning to their application in the context of each of the claimed qualified exemptions. Paragraphs  and  of the FTT’s provisional draft open judgment read as follows (absent the passages in italics which, as we explain below, were added later)
“166. The public interest factors against disclosure require careful definition. S.27 represents the inherent public interest in the UK having effective and efficient relations with foreign states, particularly the USA. But it goes beyond that in this case. It goes to the willingness of the US to share with the UK all types of material relating to national security. This sharing is subject to what is known as the ‘control principle’ whereby ---- there is an understanding that secret intelligence ---- material provided, on security or diplomatic channels, is not released without the specific consent of the provider. Material may range from warnings of a planned terrorist attack to the routine sharing of small pieces of intelligence. The latter provide a ‘jigsaw’ or ‘mosaic’ enabling a larger and significant picture of a potential threat to be built up from smaller and, by themselves, apparently insignificant pieces of information.
167. We find there is a very strong public interest in the maintenance of the "control principle” governing the use of ---- secret intelligence ---- information supplied to the UK through security and diplomatic channels, so as not to prejudice the supply of intelligence forming part of a "mosaic" enabling a picture of potential terrorist activity, or threats to national security or UK interests abroad to be built up and countered ”
56. When circulated to the parties for comment, this prompted the following “point of clarification” from the FCO as regards paragraph :
“The term ‘control principle’ has only formally been used in connection with sharing of information on intelligence and security liaison channels. We therefore propose the following form of words:
‘This sharing is subject to what is known as the “control principle” whereby material provided, on security or diplomatic channels, is not released without the specific consent of the provider. For convenience, we refer to this understanding as the ‘control principle’ although we recognise that this term has only formally been used in connection with sharing on intelligence and security liaison channels.”
57. APPGER’s response to this suggestion was a 3-page letter of 4 April 2012, which characterised the FCO suggestion as “a form of wording that would equate the understanding that diplomatic materials will not be disclosed without the consent of the provider with the control principle”. This, it was argued, was far more than a “point of clarification”; rather, it infected the FTT’s approach to the application of the public interest test in the context of the section 27 qualified exemption. In short, the FTT had already identified a difference in weighting between the control principle, as conventionally understood, which gave rise to a “very strong public interest” (emphasis added), and the expectation of confidence in diplomatic exchanges, which gave rise only to a “strong public interest” in maintaining the exemption. APPGER further argued that the FTT’s definition of the control principle effectively elided this distinction and undermined its whole approach to the balancing exercise. APPGER then made a request that the FTT hearing be re-opened, while recognising “the extraordinary nature of the request.”
58. The FCO’s response was to reiterate that its suggested amendment was no more than that, and that the matter was for the FTT to decide; “it was put forward as clarification regarding terminology, nothing more”.
59. The FTT did not re-open the hearing. Nor did it accept the FCO’s suggested wording. Instead, the final version of paragraph  was the same as in the draft save for the addition of seven words (shown in italics in the citation at paragraph 55 above and underlined in the following extract): “This sharing is subject to what is known as the “control principle” whereby there is an understanding that secret intelligence material provided, on security or diplomatic channels, is not released without the specific consent of the provider.” Similarly, the “very strong interest” in the maintenance of the control principle was now described as referring to “the use of secret intelligence information supplied to the UK through security and diplomatic channels” (at paragraph ).
60. However, the FTT made no change, and nor did the FCO propose any amendment, to paragraphs , , [180 to 182],  and  of its provisional draft open judgment, and these paragraphs of its Decision read as follows (with our emphasis):
“177 The public interest against disclosure requires careful definition. It goes beyond the inherent public interest in the UK having effective and efficient relations with foreign states, and particularly the US. It goes to the willingness of the US to share with the UK all types of secret intelligence material relating to national security. This sharing is subject to the "control principle" described above, whereby material provided through security or diplomatic channels, is not released without the specific consent of the provider.
178 The absence of even the smallest piece of information could make it harder for the UK secret services to construct, from such small pieces of intelligence, a "jigsaw" or "mosaic" enabling a larger and significant picture of a potential threat to be built up from smaller and, by themselves, apparently insignificant pieces of information.
180 Another aspect is the US view that the release of information, provided through security or diplomatic channels, remains subject to the "control principle", even if it has otherwise been placed in the public domain. In evidence, examples were given of the order of the Court that material should be disclosed and the leaking of diplomatic cables by Wikileaks. In the current case, even if the Tribunal ordered the release of information which appeared to it to be already in the public domain, such release would be likely to be regarded by the US as breaching the control principle.
181 The reason for, or the reasonableness of, the attitude adopted by the US does not form a part of the balancing exercise the Tribunal is required to undertake; it is the fact of the existence of those attitudes which matters. Similarly, it is not the fact that information released might be seen to be innocuous (for example, because it was already in the public domain) that has to be weighed in the balance, but that the release itself will be seen as a further breach of the control principle, and could result in a reduction in access to intelligence material. In striking the balance of the public interest there must be regard to the strong desirability of not damaging the UK's access to intelligence material.
182 We have heard closed evidence and seen the disputed information which further strengthens our view, that like the Court in the BM appeal when faced with PII, that the information should not be disclosed because the public interest balance favours maintaining the exemption despite the very strong public interests expressed by F1 – F3 factors set out above [note: which were in favour of disclosure].
188 In favour of maintaining the exemption is the strong public interest in the UK having access to secret intelligence capable of forming part of a ‘mosaic’ that may be used in identifying and frustrating future terrorist plots. This interest is given added weight by the fact that a further release of material, in breach of the ‘control principle’, could reduce access further. In weighing the public interest, the weight given to maintaining the exemption should be multiplied by the magnitude of the adverse consequences should a terrorist plot aimed at causing loss of life go undetected.
189 All of the material in respect of which the section 27(1)(a) and section 27(2) exemptions are claimed relates to information covered by the ‘control principle’ in that it is either US sourced diplomatic or security material, UK material reporting on US diplomatic or security service views, or UK material responding to US diplomatic or security material. The public interest in maintaining the ‘control principle’ so as not to adversely affect the supply of secret intelligence on national security matters is very high indeed.”
The parties’ submissions to the Upper Tribunal
61. Mr Pitt-Payne’s primary submission was that the FTT had erred in its approach to the control principle and as a result had also erred in conducting the public interest test under section 27. More particularly, he identified three more specific errors of law (or “sub-grounds”) under this head. First, he submitted that the FTT had committed a procedural error by failing to re-open the case for further evidence and argument, following the circulation of the draft judgment. Second, he argued that the FTT had erred in its definition of the control principle. Third, he submitted that the FTT’s findings under section 27 were perverse and unsupported by the evidence.
62. At the heart of the submissions made by Ms Steyn and Mr Hopkins was the point that the use by the FTT of the term the “control principle” was no more than a variation in terminology and so APPGER’s submissions were founded on terminology rather than substance. On that basis the FTT had made clear, or clear enough, particularly in paragraph  (read with paragraphs [180 to 182]) of its Decision how it was using the term the control principle and that it had concluded, as it was entitled to do, on the evidence (open and closed) that disclosure of the material in respect of which the section 27 exemption was claimed could result in a reduction in the access given by the US to the UK of intelligence material.
The Upper Tribunal’s analysis
Preliminary observation on references to the control principle
63. There is ample high authority to the effect that the “control principle” is a convenient shorthand description of a convention, rather than a statement of any legal principle. For example, as Lord Judge CJ explained in R (Mohamed) v Secretary of State for Foreign & Commonwealth Affairs  QB 218 at  (emphasis added):
“44. At the risk of repetition, in general terms it is integral to intelligence sharing arrangements that intelligence material provided by one country to another remains confidential to the country which provides it and that it will never be disclosed, directly or indirectly, by the receiving country, without the permission of the provider of the information. This understanding is rigidly applied to the relationship between the UK and USA. However although confidentiality is essential to the working arrangements between allied intelligence services, the description of it as a "control principle" suggests an element of constitutionality which is lacking. In this jurisdiction the control principle is not a principle of law: it is an apt and no doubt convenient description of the understanding on which intelligence is shared confidentially between the USA services and those in this country, and indeed between both countries and any other allies. If for any reason the court is required to address the question whether the control principle, as understood by the intelligence services, should be disapplied, the decision depends on well understood PII principles...”
64. In a similar vein, Thomas LJ (as he then was) referred to the control principle as a “general principle or convention”, while emphasising that “it is a convention as opposed to a matter of legal obligation” (see R (on the application of Mohamed) v Secretary of State for Foreign & Commonwealth Affairs (No 2)  EWHC 2549 (Admin);  1 WLR 2653 at  and ).
65. Given that the control principle is not a term that carries any specific legal meaning, any use of that label or shorthand in a wider or different way than is usually, or has previously been, the case cannot, in itself, amount to an error of law.
66. However, it is clear from the BM litigation and the FCO’s letter suggesting alterations to paragraph  of the FTT’s provisional draft open judgment, that previously the shorthand or label had only been used in relation to intelligence material (see the citation from the judgment of Lord Judge CJ) and so to the sharing of material on intelligence and security channels (see the letter). It follows that in the BM litigation the risk of harm asserted in the PII certificate had a direct link to the information that was covered by the control principle.
67. The risk asserted and accepted in the BM litigation was that disclosure of intelligence material provided pursuant to the control principle would give rise to a serious and openly expressed risk evidenced by the BM letter that the US would in the future not share so much intelligence material with the UK. This carries the consequential risk that the ability of the UK to build up a mosaic of information which could be used, for example, to identify and frustrate a terrorist plot or other potential harm to national security would be seriously prejudiced. The FTT concluded that disclosure of any of the information in respect of which the section 27(1)(a) exemption was claimed would give rise to such a risk and its view that there was a very strong public interest against such disclosure (as opposed to a strong one) is based on that conclusion. We shall refer to this risk as an Intelligence Information Sharing Risk.
68. A logical reaction by a supplier of category A information to a breach of the basis on which it was supplied is that further supply of category A information will be restricted to avoid the risk of damage that the basis of supply was intended to avoid. So such a common subject matter and direct link will be a relevant factor in determining and explaining the risks that would or would be likely to flow from disclosure of material without the consent of its supplier. In the BM litigation the category A material was intelligence material.
69. It follows that if and when that direct link or common subject matter is absent and so the issue is whether disclosure of category B information (e.g. diplomatic material) supplied on a confidential basis will give rise to a risk that in the future the person or country who supplied it will restrict his or its supply of intelligence material (our category A information), this will not be established by referring only to the risk that would arise, or has arisen, when intelligence material is disclosed in breach of the terms on which it was supplied in confidence (and so without the permission of the supplier). Additional and/or different reasoning will be needed to explain why the asserted Intelligence Information Sharing Risk arises from disclosure of different categories of confidential information provided through different channels and for different purposes. Such reasoning could build on, or be linked to, the position relating to intelligence material and/or the basis on which it and the relevant different types of confidential information are supplied, or it could establish the common result in different ways. But, in our view, it is needed to show why the same, or effectively the same, risk arises from the disclosure of different types of confidential information conveyed through different channels and for different purposes.
The information in respect of which the FCO claimed the exemption under section 27(1)(a) of FOIA (“the Section 27 Information”)
70. The FTT found that section 27(1)(a) (and, where appropriate, section 27(2)) was engaged where it had been claimed by the FCO, accepting Mr Sinclair’s explanation in evidence that the material in question fell into two broad categories (at paragraph ):
“(1) The exemption is claimed over confidential exchanges between US officials and UK officials and certain documents that provide comment on US intentions. The release of these documents would, in his view, prejudice the UK’s relationship with the US. This limb also applies to documents relating to The Gambia.
(2) The exemption is claimed over communications that detail UK views on US policy, or outline steps that the UK has or will take in handling US requests. The release of these documents would be likely, in his view, to have a prejudicial effect.”
71. The FTT went on in paragraph  to state:
“We accept that Mr Sinclair as a member of the Diplomatic Service and a Senior Civil Servant in the FCO has a much better view of the effect of prejudice of disclosure than the Tribunal. We find no evidence in this case to seriously contradict his view notwithstanding the clear and strong public interest in issues around extraordinary rendition. We therefore find that s. 27(1)(a) and s. 27(2) are engaged for the materials where it is claimed. We have applied the appropriate weight as set out in Hogan when applying the public interest test.”
72. In those paragraphs neither the FCO, through Mr Sinclair, nor the FTT identify the prejudice and thus, applying the approach adopted by the FTT by reference to earlier authority, the real and significant risk of harm that is relied on to establish either (a) the prejudice to relations between the UK and any other state, or (b) any more specific, wider or higher risk of harm to the public interest (e.g. the existence of an Intelligence Information Sharing Risk).
73. Further, and importantly, the categories of information within the accepted description of the Section 27 Information do not include or extend beyond intelligence material and material supplied on (or only on) intelligence and security liaison channels. It follows that for the reasons set out under the previous heading the reasoning and evidence to establish the existence of a relevant real and significant Intelligence Information Sharing Risk, if any, if the Section 27 Information is disclosed cannot be confined simply to an assertion of the existence of that risk if and when intelligence material is disclosed in breach of the terms on which it was supplied.
The approach to be taken to the public interest balancing exercise under section 27 of FOIA
74. This assessment gives rise to a process of parallel reasoning that arises in other balancing exercises (e.g. in the BM litigation in respect of PII) and so it involves what is sometimes described as an “apples and pears” comparison.
75. In our view correctly, it was accepted before us by the FCO and the IC that when assessing competing public interests under section 27 of FOIA the correct approach is to identify the actual harm or prejudice that the proposed disclosure would (or would be likely to or may) cause and the actual benefits its disclosure would (or would be likely to or may) confer or promote. This equates to the approach now taken in PII claims. Also, it is in line with the approach that the FTT explained it had adopted in deciding that section 27 was engaged, namely that a real and significant risk of that disclosure would prejudice relations with another State had to be established and the FTT’s acknowledgement in paragraph  that the public interest factors against disclosure require definition.
76. Such an approach requires an appropriately detailed identification, proof, explanation and examination of both (a) the harm or prejudice and (b) benefits that the proposed disclosure of the material in respect of which the section 27 exemption is claimed would (or would be likely to or may) cause or promote. Plainly that includes an identification of the relevant material and the circumstances in which it was provided to or obtained by the body claiming the section 27 exemption.
77. To enable us to understand how this had been done before the FTT, from an early point in the hearing in open and closed session we pressed the FCO for:
(1) clarification of what its case before the FTT was as to the risks of harm that would arise from public disclosure of the two broad categories of material in respect of which it claimed the section 27 exemption, and
(2) whether it was in a position to add anything now.
The answer to (1) by reference to what had been written or said in open and/or closed session before the FTT was not immediately forthcoming. However on 3 July (the third day of the Upper Tribunal hearing) the FCO provided an open statement in the following terms:
“In light of public statements that have been made since the hearing before the First Tier Tribunal, in the context of the passage of the Justice and Security Bill through Parliament, the FCO can now say in OPEN that the disclosures as a result of the Binyam Mohamed litigation caused actual damage. It caused a reduction in the flow of intelligence. The First Tier Tribunal heard evidence to that effect and recorded at §2 of the Closed Annexe that on the evidence before them ‘the damage to the supply of secret intelligence was actual, not merely hypothetical”.
78. On the fourth and last day of the hearing, and so after the main closed session and shortly before Mr Pitt-Payne’s reply, Ms Steyn provided the following open statement:
“The FCO’s principal case before the FTT was that the public disclosure of any of the documents in respect of which section 27 had been claimed would further undermine US confidence in its exchanges with the UK, including in the field of intelligence sharing. The release of such documents would complicate the intelligence-sharing relationship and give rise to a real risk of a further reduction in the flow of intelligence.”
79. This made it clear that, as the FTT found, the FCO was asserting that an Intelligence Information Sharing Risk would arise if any of the Section 27 Information was disclosed.
80. There was no such particularised open statement of the FCO’s case in the material before the FTT.
APPGER’s position in respect of the open statements before us particularising the FCO’s principal case on the risk of harm to the public interest
81. In our view, unsurprisingly Mr Pitt-Payne made clear and we accept that APPGER had not understood that the FCO’s case before the FTT was that disclosure of any of the Section 27 Information would give rise to an Intelligence Information Sharing Risk and thus that any such disclosure would carry a real risk that the sharing or flow of intelligence material between the US and the UK would be reduced to an extent that would be likely to cause real damage to national security or the work and effectiveness of the section 23 bodies. This was not challenged. We say unsurprisingly because:
(1) we agree with APPGER’s submission based on an analysis of the open evidence that, on a fair reading of it and so one that takes a realistic view to de-coding some of the language used and/or to identifying the purpose for which it was being used, there was no open evidence that either disclosure of diplomatic material would (i) breach the control principle, or (ii) lead the US Government to re-evaluate the intelligence sharing relationship with the UK,
(2) the questions that APPGER’s counsel invited the FTT and counsel for the IC to put in closed session do not contain any questions on this topic, and
(3) in our view, it is inconceivable that there would have been no such questions directed to an examination of the reasoning and evidence to support this assertion that there would be a risk of actual harm to intelligence sharing if APPGER had understood that the FCO’s case was as set out in the open statement made to us.
82. As to the point in sub-paragraph (3) of the last paragraph, the risk as particularised to us is expressed in a way that means that to establish the asserted risk of the further reduction in the flow of intelligence the following would have to be demonstrated:
(1) existing problems concerning the flow of intelligence material and their cause, and
(2) why disclosure of all of the types of material falling within the two broad categories describing the Section 27 Information on whatever terms, understandings or channels that information was given, would give rise to that risk.
So, if APPGER had understood that this risk of harm was being asserted they would naturally and in our view inevitably have wanted to challenge or test that two stage process and so why disclosure of any of the Section 27 Information (our category B above) would give rise to a risk that the flow of intelligence material (our category A above) would be affected and reduced.
83. In reaching the conclusions set out in the previous paragraph we have not overlooked the point that the FTT altered paragraphs  and  to limit the material referred to therein to secret intelligence material and so to exclude all other (and so diplomatic) material and arguably to provide a direct link to an Intelligence Information Sharing Risk. This change occurred after the hearing and, in any event, in our view cannot fairly be relied on, and correctly was not relied on, to show that APPGER were informed of, and were so given an opportunity of addressing the FCO’s principal case as formulated to us. Namely, that an Intelligence Information Sharing Risk would arise for the reasons so indicated (or other reasons) if any of the Section 27 Information was disclosed. Indeed, as mentioned later this alteration causes confusion rather than clarity.
84. Also, we have not overlooked the points that the open evidence contained some material that might have been relied on to support the case identified in the open statement made to us in that:
(1) when Mr Sinclair was being questioned on the stances taken by the Bush and Obama administrations respectively he stated as follows (Vol 2 at p. 1289):
“In that while these were different governments in the US, what is very clear from my experience, is that disclosure of either diplomatic or intelligence material is seen as equally bad by any number of administrations.”
(2) in its lengthy (and now partly redacted) letter to the IC the FCO, in setting out its approach to the various requests, expressly made the point in the open version that “the US Government sees little distinction between intelligence material and classified diplomatic correspondence”, and
(3) Mr Sinclair referred to particular sensitivities in connection with The Gambia.
85. But, in our view correctly, it was not asserted that such open evidence, or that any other open evidence or submissions would have fairly informed APPGER that the FCO’s principal case on the section 27 exemption was as set out in the open statement made to us or indeed that the risk being asserted was an Intelligence Information Sharing Risk.
86. Further, if an equivalent statement of this principal case had been made to and been understood by the FTT, it is difficult to see how the FTT could have reasoned its conclusion on this exemption in the terms that it did.
87. The absence of a transcript of the closed sessions means that we are not in a position to follow from it how the FCO’s principal case on the existence of an Intelligence Information Sharing Risk was advanced orally before the FTT and how the FTT and counsel for the IC tested this aspect of the case through questioning and argument. But it has not been asserted that the transcript, or anything else, would show that:
(1) the FCO placed reliance on the “control principle” in the way that the FTT did in its Decision to establish the existence of an Intelligence Information Sharing Risk, and so the very strong public interest against disclosure found by the FTT, or
(2) the FCO advanced reasoning that would have founded or encouraged that reliance or an expansion of the previous use of the shorthand description the “control principle” to the supply of a wider class or classes of information (and so to the Section 27 Information).
88. We were not invited to read the notes of the closed session made by the FTT and have not done so. We have however had the benefit of reading the written closed submissions and of being referred to particular points raised in and by the FTT in closed session. We record that in our view none of them show as clearly as the open statement before us what the FCO’s principal case was on the harm or risk of harm that would be caused by disclosure of any of the Section 27 Information or that such case was subjected to the analysis that we have concluded APPGER would have wanted to pursue in open session and to be pursued by others in closed session.
89. Given the correct common ground set out in paragraphs 75 and 76 above it is an unfortunate result that the party excluded from the closed session was ignorant of the FCO’s primary case on the harm that would be caused if any of the Section 27 Information was disclosed and indeed was unaware that it was asserted that this would give rise to an Intelligence Information Sharing Risk.
90. This result demonstrates that the FCO, the IC and the FTT did not appropriately identify or explain this in open session, for example by providing an open statement along the lines of that given to us, after we had pressed for it, or otherwise.
The lack of an open statement before the FTT particularising the FCO’s principal case
91. The unfairness caused by the result stated above could only be justified if there was a good enough reason why the FCO’s primary case that disclosure of any of the Section 27 Information would give rise to an Intelligence Information Sharing Risk could or should not have been made clear in an open statement as it has been before us.
92. As to that we can appreciate that, for example, the linkage of the material that constituted the background to the Section 27 Information to intelligence material or the intelligence material sought in the BM litigation, and/or the general relationship between the UK and the US at the time, increased the need for some of the FCO’s assertions and evidence on its primary case on the existence of an Intelligence Information Sharing Risk being advanced in closed session.
93. We also appreciate that at that time these factors might have led to an argument that the very disclosure of the existence of such a risk would itself cause actual harm to an aspect of the public interest in promoting international relationships and in particular the relationship between the UK and the US.
94. But, as the FCO raised no objection to paragraphs , [180 to 182],  and  of the FTT’s Decision, and suggested the alteration that it did to paragraph  of the provisional draft open judgment it is apparent that the FCO was not then concerned that the disclosure of an assertion that an Intelligence Information Sharing Risk would exist if any of the Section 27 Information was disclosed would, of itself, cause harm. This is because the FCO submits that by those paragraphs the FTT is saying that the disclosure of any of that information would give rise to that risk.
95. It follows that there was no good enough reason, and indeed none was suggested to us, why the FCO’s case that disclosure of any of the Section 27 Information would give rise to an Intelligence Information Sharing Risk was not made clear to APPGER before or during the open parts of the hearing.
96. The failure of the parties and the FTT to make this clear to APPGER resulted in avoidable substantive and procedural unfairness.
The use by the FTT of the description the “control principle”
97. The terms of the open statement to us of the FCO’s primary case on harm and the FCO’s suggested amendment to paragraph  of the FTT’s provisional draft open judgment indicate (and the FCO confirmed) that the FCO had not argued before the FTT that:
(1) the Section 27 Information had been supplied pursuant to the convention or shorthand known as or previously referred to as the “control principle”, or
(2) this description should be used as a shorthand description of the terms, understandings and channels upon which all of the Section 27 Information was given.
98. The FTT’s introduction of what it says is known as the control principle is found in paragraph  where the FTT expresses the view that the public interest against disclosure went beyond the inherent public interest of the UK having effective and efficient relations with foreign states, because in the view of the FTT it goes to the willingness of the US to share with the UK “all types of material relating to national security”. The FTT explains this by stating that such sharing is subject to “what is known as the control principle”. It repeats this in, for example, paragraph .
99. It is not clear to us whether in paragraphs , , , and  the “sharing” that the FTT is referring to is the sharing of (a) “all types of material relating to national security”, or (b) following the amendment of “secret intelligence material” or (c) a wider class of material including the Section 27 Information.
100. The FTT’s use of the phrase “what is known as” naturally links the FTT’s use of the description or shorthand to one in general use, or at least to one that had been used previously. But, as the FCO points out in its letter suggesting alterations to the provisional draft open judgment, if that was the FTT’s intention it had mis-described what “is known as”, or as the FCO put it “has previously formally been described as” the control principle because the FTT refer to communication on security and diplomatic channels. Also, the FTT’s reference to “material” without any limitation, other than by its reference in the preceding sentence to “all types of material relating to national security” raises doubt as to its use and so its understanding of the description of the “control principle”, particularly when the earlier references to it are read with paragraph .
101. In paragraph  the FTT state that the “control principle” applies to the material in respect of which the section 27 exemptions are claimed. The explanation it gives for this is founded on the description it then gives of the information in respect of which the section 27 exemption is claimed. That description is in different and narrower terms to the description it accepted and set out in paragraph . But both descriptions:
(1) cover (a) material that is not intelligence material or material relating to national security, and (b) communication outside intelligence and security channels, and
(2) do not match the amended (and it seems to us the unamended) references to the material said by the FTT to be subject to the “control principle” in paragraphs ,  and .
102. In our view the FCO’s suggested alteration to paragraph  goes well beyond a variation of terminology as it and the IC have argued because by making it the FCO invited the FTT:
(1) to define how it had used the term the “control principle”,
(2) to acknowledge that it was using it in a new sense to cover wider channels of communication, and possibly a wider range of material, and by so doing
(3) to introduce a line of reasoning that did not accord with the way in which the FCO had put its case on the existence of an Intelligence Information Sharing Risk.
As to point (3), if the FCO’s suggestion had been accepted it would (as no doubt was intended) have arguably provided a link to the application of the “control principle” to the Section 27 Information in paragraph  and so arguably a reason for the existence of an Intelligence Information Sharing Risk if any of the Section 27 Information was disclosed. And, by so doing, it would have provided a better basis for the argument advanced by the FCO that the FTT was using the term in a different and wider sense to the way it had been used in the past and the substantive issue was whether the FTT was entitled to find on the evidence, as it did, that an Intelligence Information Sharing Risk would arise on disclosure of any of the Section 27 Information.
103. However, the FTT did not adopt the FCO’s suggestion. Rather, it altered its description of the “control principle” by adding to paragraphs  and  references to the “control principle” being an understanding governing the use of “secret intelligence material”.
104. This is confusing in at least three ways. First, because the FTT refers to something that it says “is known as” the control principle but describes it differently to its previous formal use by including communication on diplomatic channels. Secondly, the FTT’s amendments to paragraph  and  continue the uncertainty as to how, and in respect of what sharing, it is using the term the “control principle” in the paragraphs leading up to paragraph . And most importantly, the FTT does not address and explain why:
(1) as stated in paragraph , the material in respect of which the section 27 exemption is claimed, as described by it in paragraph  (or differently in paragraph ) is covered by the “control principle” in the sense given to it in paragraphs  and  (as amended), or in any other sense and/or why this, of itself, gives rise to an Intelligence Information Sharing Risk, and further or alternatively and more generally
(2) how the very strong public interest it has taken into account (and thus an Intelligence Information Sharing Risk) arises from disclosure of material that is not within the control principle as that shorthand had previously been used (and thus, for example, as it had been used in respect of the PII claim in the BM litigation), and so
(3) how the very strong public interest it has taken into account (and thus an Intelligence Information Sharing Risk) arises from disclosure of any of the Section 27 Information having regard to its subject matter and the terms, understandings or channels upon which it was given.
Was the FTT’s response to the APPGER letter unfair and an error of law?
105. By its letter dated 4 April 2012 APPGER made an analogy with the position in the Court of Appeal in the BM litigation as to the issue of decisions in draft form, in R (on the application of Mohamed) v Secretary of State for Foreign & Commonwealth Affairs  EWCA Civ 158;  QB 218 Lord Judge CJ observed as follows (at ):
“4. The primary purpose of this practice is to enable any typographical or similar errors in the judgments to be notified to the court. The circulation of the draft judgment in this way is not intended to provide an opportunity to any party (and in particular the unsuccessful party) to reopen or re-argue the case, or to repeat submissions made at the hearing, or to deploy fresh ones. However, on rare occasions, and in exceptional circumstances, the court may properly be invited to reconsider part of the terms of its draft. (see for example Robinson v Fernsby  WTLR 257 and R (Edwards) v The Environment Agency  1WLR 1587). For example, a judgment may contain detrimental observations about an individual or indeed his lawyers, which on the face of it are not necessary to the judgment of the court and appear to be based on a misunderstanding of the evidence, or a concession, or indeed a submission. As we emphasise, an invitation to go beyond the correction of typographical errors and the like is always exceptional, and when such a course is proposed it is a fundamental requirement that the other party or parties should immediately be informed, so as to enable them to make objections to the proposal if there are any.”
106. In our view, this letter makes it clear that in APPGER’s view the FCO’s suggestion was going beyond what was permissible by seeking to change what it regarded as defective reasoning to arrive at a conclusion that an Intelligence Information Sharing Risk applied in this case.
107. The FCO responded to that letter, apparently without being asked. However, the FTT took no further action, beyond making the amendments described above. By implication, APPGER’s request for the hearing to be re-opened or for further submission to be allowed was dismissed. We do not regard this as satisfactory.
108. Procedurally, in substance APPGER’s letter was an application for further directions under rule 6(1) of the Tribunal Procedure (FTT)(GRC) Rules 2009. It included reasons under rule 6(3). The FTT should have dealt with this application, either by expressly rejecting it or by a direction to “hold a hearing to consider any matter” under rule 5(3)(f). Given that the FTT plainly adopted the former course, this refusal could have been done either by way of a separate ruling to that effect or in a passage in the final decision.
109. However the determinative question for us is not whether there was a procedural error but whether the effective refusal of the FTT to hold a further hearing or to allow further submission to be made rendered its process and decision unfair and so was an error of law.
110. In our view, it did because it perpetuated a situation in which APPGER had not been made aware of the FCO’s case on the actual harm or risk of harm to the public interest that would be caused by disclosure of any of the Section 27 Information, and so the substantive and procedural unfairness referred to above (see paragraphs 89, 90 and 96).
111. By the time the FTT was invited to hold a further hearing or allow further submissions to be made, in one sense that unfairness was water under the bridge but the request by APPGER should have alerted the FTT to its existence or the possibility that it existed and, in any event, it flagged issues that of themselves merited a further hearing or further submissions being allowed.
112. If the FTT had held a further hearing or allowed further submissions in our view the prospect that the earlier unfairness would have been identified and rectified would have been high.
113. So, in our view, the failure of the FTT to accede to APPGER’s request that it hold a further hearing or allow further submissions is an error of law that has caused and perpetuated unfairness that warrants the setting aside of its Decision on the application of the section 27 exemption to the Section 27 Information.
Did the FTT err in its approach to the control principle and in so doing to the balancing exercise to be conducted under section 27(1) of FOIA?
114. The conclusion concerning the failure of the FTT to have a further hearing or allow further submissions renders this alternative argument academic. However, we record that this argument also succeeds.
115. As to this we repeat that we acknowledge that :
(1) the FTT could in its reasoning have re-defined and used the shorthand description “the control principle” in a new and wider sense than the way it had previously been used, and that
(2) if it had adopted the alteration suggested by the FCO, it would have provided a better base for the argument advanced to us by the FCO and the IC.
116. As the FTT did not adopt the suggested alteration we have not had to consider this argument. But, we comment that weaknesses in it appear to be:
(1) the absence of any explanation as to (a) why the previous use of the shorthand should be extended to the supply of the wider Section 27 Information and so any consideration of the terms, understandings and channels upon which it was provided and how they link to the provision of intelligence information on intelligence and security liaison channels, and (b) why disclosure of that information (our category B information) would or might give rise to a risk that the supply of information of a different type (i.e. intelligence information - our category A information) would be reduced, and
(2) it relies on a different line of reasoning to that advanced by the FCO, albeit one that might be said to have elided steps in the reasoning reflected by the terms of the open statement to us of the FCO’s primary case.
117. The result of the decision of the FTT not to accept the suggested alteration and to make a different one is that the confusion and deficiencies referred to in paragraph 104 above exist. And, in our view, they found the conclusion that the FTT failed to explain adequately why it reached the conclusion it did that the disclosure of any of the Section 27 Information would give rise an Intelligence Information Sharing Risk and so to the very high public interest against disclosure that it took into account.
118. Alternatively, we have concluded that the decision of the FTT not to accept the alteration suggested by the FCO and to make a different one without carrying it through to its conclusion in paragraph  or matching it to its earlier description of the Section 27 Information in paragraph  means that the FTT did not properly understand and adopt the underlying reasoning advanced by the FCO to establish an Intelligence Information Sharing Risk in respect of the Section 27 Information and so its conclusion on the existence of that risk does not have a proper evidential and reasoned foundation.
119. The rationale for this ground is that there was no evidence on which the FTT could have rationally reached its conclusion. APPGER therefore have the difficulty in advancing it that they are excluded from the closed evidence.
120. As our earlier conclusions render this ground academic and mean that we may have to re-examine the claims for exemption based on section 27 a further consideration of this ground is unnecessary. It would also be inappropriate because, in carrying out that exercise, we will have to consider the existing and any further open and closed evidence advanced by the FCO in the light of the evidence and points advanced by APPGER on the FCO’s primary case on the risk of actual harm. A part of that consideration will be whether any of the evidence or information that the FCO argues should be closed should be provided to APPGER.
Ground 5: the formulation and development test under section 35(1)(a)
121. Section 35(1)(a) of FOIA provides for an exemption with regard to “Information held by a government department ... if it relates to — (a) the formulation or development of government policy”. This qualified exemption potentially applied to only a small number of documents within the scope of this appeal: three documents in relation to the al-Rawi requests (documents DN1/39, DN1/41 and DN1/50) and two documents with regard to the BM letter request (DN2/26 and DN2/35). Section 35(1)(a) was not relevant to the sole document in issue under R14 of the BM interrogation requests.
The approach of the FTT
122. The FTT dealt with section 35(1)(a) at paragraphs - and - of its Decision. In the first passage it summarised Mr Sinclair’s open evidence to the effect that Government policy had evolved from only providing representation for British nationals to also representing British residents. The FTT noted that the issue that arose for decision was whether there was a “further development of the representation policy or formulation of new policies in relation to detainees at the time of the requests which would also be deserving of a safe space” (at paragraph ). The FTT accepted the principle of Ms Clement’s submission that a distinction had to be made between policy formulation and development on the one hand, and implementation on the other, but it did “not consider in this case that such a clear distinction can be made” (at paragraph ).
123. In its conclusions, the FTT acknowledged that there had been a change in policy as regards the groups of persons on behalf of whom the FCO would make representations, but found that at the time of the requests the policy was still being developed, that there remained one detainee in Guantanamo Bay that the Government was trying to get released and that some of the disputed documents were in draft form only (at paragraph ). The FTT found there to be “a very strong public interest in the Government having a safe space to develop its policy” (at paragraph ).
The parties’ submissions to the Upper Tribunal
124. Mr Pitt-Payne had two short points on this ground of appeal. The first was that the FTT had erred in allowing the FCO to rely on section 35(1)(a) in circumstances where, by the time of the requests, the policy had been settled, namely that representations would be extended to British residents as well as nationals. The second was that the FTT had failed to give reasons for its conclusion that the policy was still under development.
125. Ms Steyn referred us to the High Court decisions in Office of Government Commerce v IC (Attorney-General intervening)  EWHC 774 (Admin);  QB 98 and HM Treasury v IC  EWHC 1811 (Admin);  QB 563, in which Stanley Burnton J (as he then was) and Blake J respectively held that section 35(1)(a) should be given a broad construction. She submitted that the evidence before the FTT demonstrated that the detainee policy was not simply concerned with securing the release and return of British nationals and residents. Furthermore, Ms Steyn argued, the FTT had indeed recognised the distinction between the formulation and development of policy as opposed to its implementation.
126. Mr Hopkins supported Ms Steyn’s submissions, arguing that this ground of appeal boiled down to no more than a disagreement by the Appellant with the FTT’s assessment of the evidence before it.
The Upper Tribunal’s analysis
127. There is nothing that we can usefully add to the analyses of the scope of section 35(1)(a) in Office of Government Commerce v IC (Attorney-General intervening) and HM Treasury v IC. In our judgment, as Ms Steyn and Mr Hopkins both submit, this was a relatively straightforward case in which the FTT was amply entitled to find, even on the open evidence before it alone, that the detainee policy remained live at the time of the requests. In the wider context of this appeal, the section 35(1)(a) point was relatively marginal, and certainly did not require more extensive reasoning on the FTT’s part.
128. Our scrutiny of the relevant documents in closed hearing confirms that, as with the section 23 exemption, it is obvious to any reader why the information they contain falls within the scope of section 35(1)(a). Thus by identifying the test it was applying and setting out its conclusions the FTT provided adequate open reasons and any further reasons that did not refer to the actual content of the closed documents would not have added anything useful. Also closed reasons would not give any added value to a decision maker who can read the closed documents.
The issues as to which of documents identified in the schedule produced by the parties remain within the appeal
129. As mentioned earlier the parties provided us with a schedule that identifies these documents and reasons why the disputes existed. The essential theme of these reasons was that the FTT had concluded that a claimed exemption applied to the whole of a document but APPGER was not aware of the bases upon which the FTT had reached these conclusions. However, and in our view having regard to the relevant exemptions (namely sections 42(1) (legal professional privilege) and 35(1)(b) and (c) (ministerial communications and advice from Law Officers) the parties very sensibly agreed that if on our reading of the documents those exemptions applied to the whole document it would fall out of the appeal.
130. We deal with each document in the attached schedule.
131. In the context of sections 42(1) and 35(1)(b) and (c) the problem relating to piecing the jigsaw together from a general description of the document discussed above in respect of the section 23 exemption (see paragraph 40) has no weight and where we have given one it seems to us that this assists in identifying to the excluded party why the exemption applies (e.g. this is Law Officer’s advice or advice from counsel), albeit that it adds little to the definition of the exemption. Also, the general identification in respect of Decision Notice 2, Document 26 shows why the document is not within section 35(1)(b), applying the definition in section 35(5).
132. As is well known and established legal professional privilege is categorised in two classes, namely legal advice privilege and litigation privilege. As to the former it applies whether or not litigation is contemplated or pending but the latter, which extends to communications with non-professional agents and third parties (e.g. witnesses or potential witnesses) only applies when litigation is contemplated or pending.
133. Documents do not become privileged by the mere fact that they have at some time been submitted to a legal adviser for the purpose of obtaining legal advice whether or not litigation is pending.
134. But advice privilege applies to all confidential communications with solicitors and counsel (and some other legal advisers) for the purposes of seeking and giving legal advice. Furthermore, litigation privilege covers communications between a wider class of persons that come into existence for the purposes of obtaining evidence or information in respect of the conduct of pending or anticipated litigation. In both cases, such communications cover documents and information prepared and obtained for those purposes and information contained in or attached to documents that are so prepared and communicated. This is because such documents and information form an integral part of those communications and so engage the underlying reasons for the existence of legal professional privilege.
135. This is not the place to discuss any overlap between legal professional privilege and Law Officer’s advice and requests for it and, for present purposes, it is enough to record that in our view the underlying purposes of the exemptions in sections 35(1)(b) and (c) cover information contained in or attached to such communications and which form an integral part of them.
136. In making and considering claims that more than one exemption applies to a document the FCO, the IC and the FTT have identified passages within documents that are covered by, for example, the section 23, section 27 and section 35(1)(a) exemptions. As indicated above, it is obvious to a reader of the relevant document or passage why the FTT (applying the test it did) concluded that the section 23 exemption applied. There is no challenge to the identification of the information that engages the section 27 exemption. When information has been referred to as an integral part of a communication and thus its purposes, as explained in the last three paragraphs we are of the view that when sections 42(1) or sections 35(1)(b) or (c) apply to it they apply to the whole communication even though other exemptions also apply, or may apply, to parts of it. When this is the case we have stated in the attached schedule that the “overlap point” applies to the relevant document(s).
137. In respect of the documents listed under Decision Notice 1 our conclusions on the application of the exemption in section 42(1) is primarily based on the application of advice privilege although parts relate to the gathering of information for the purposes of the existing litigation. The advice relates to existing litigation. We have summarised the test for the application of these two aspects of the privilege and it is not practicable or necessary to explain the reasons for our conclusions further in an open document and so without referring to the content of the information. In our view, there is no need for closed reasons where we have decided that the exemption applies to the whole document because, as in the case of the section 23 exemption, a reader of the document will see, without further explanation, why the privilege and so the exemption applies. The same applies to the documents we have concluded are covered by the exemptions in section 35(1)(b) and (c). Where we have concluded that the exemption in section 42(1) does not apply to the whole of a document the consequence is that the document remains part of the appeal and, in our view, the time for the giving of any closed reasons for this conclusion should be when the outstanding issues on the appeal are decided in respect of that document, and so if and when orders for disclosure with or without redactions (to include redactions of information that is out of scope) are considered.
138. In respect of the documents listed under Decision Notice 2 our conclusions on the application of the exemption in section 42(1) is primarily based on the application of litigation privilege because it relates to the gathering and use of material for the purpose of existing litigation, although some parts include legal advice relating to the litigation. It seems to us that the application of litigation privilege to this request can come as no surprise to APPGER because it is clear that the BM letter was sought for the purposes of the BM litigation and the claim to PII in it. In our view, a reading of the documents demonstrates clearly that legal professional privilege applies to them and no further open explanation is necessary or practicable and a closed explanation is not needed to inform anyone who can read the documents why litigation privilege applies to them.
139. A further point arises in any consideration of the application of the section 42 exemption and it is whether the document came into existence in furtherance of any iniquity. This is because legal professional privilege does not apply to any such material. We have therefore carefully considered whether the documents indicate that there was any iniquity or impropriety. We agree with the FTT that they do not.
140. Finally, we mention that (a) no issue of dual or dominant purpose arises in respect of the application of the section 42(1) exemption, and (b) we have considered the manuscript annotations on all the documents and have concluded that they do not merit any separate treatment or classification.
The re-consideration of the claims for exemption based on section 27
141. No doubt as indicated at the hearing APPGER will consider whether they still wish to pursue their claims for disclosure of this material now that they know that the FCO asserts that its disclosure will give rise to an Intelligence Information Sharing Risk and the reasoning behind that assertion indicated by the open statement to us. As to that reasoning, we note and agree with the view expressed by the FTT in the first sentence of paragraph  of its Decision to the effect that of central importance is the existence of (rather than an assessment, through other eyes, of the reasonableness of) the attitudes and stance of the US and from that base the existence of the risks they give rise to.
142. We also invite APPGER to take account of the following in respect of the BM letter request:
(1) Our conclusions on the application of the exemption in section 42(1) to the documents in respect of which it was claimed and in respect of which it was found by the FTT although it was not claimed.
(2) APPGER’s sensible approach that if we concluded that all of a document is covered by the section 42(1) exemption it was not part of the appeal.
(3) The impact of those conclusions and that approach on the point that the public interest in the public being able to see or check for itself documents that could support or refute the allegation or inference made in The Observer newspaper on the basis of an unnamed “senior” US source (and without it seems any documentary back-up) that there was some impropriety falls to be assessed against the background that four judges and two members of the FTT, in carrying out their duty under the checks and balances contained in FOIA, have reached the conclusion set out in paragraph 139 above.
143. If APPGER decides to pursue the BM letter request there will be a directions hearing to determine:
(1) whether this should be done before or at the same time as the hearing of the stayed grounds of appeal, and in any event
(2) what further open evidence and argument should be filed, and having regard to the directions made as to this,
(3) whether and when directions should be given concerning (a) whether there should be a closed hearing, (b) what evidence (in addition to the requested documents) and argument should be closed, and (c) what oral evidence should be given.
144. We stress that we accept and acknowledge that FOIA and the public interests consideration under the section 27 exemption and other exemptions cover a wide range of issues and an approach that “one process fits all” is not appropriate to the examination of competing public interests under FOIA. Rather a proportionate approach must be taken and what is or is not fair in a given case will depend on the circumstances of that case. This qualifies the comments we make below.
145. Naturally, we also acknowledge (a) the benefit of hindsight and that the FOIA jurisdiction is relatively new and is developing, (b) that the approach taken by the FTT in this case was in line with the practice guidance then in existence and was one that was often adopted at that time when there was closed material and evidence and (c) the obvious care and diligence that the FTT brought to its task.
146. However, we think that it is appropriate for us to make some general observations because of our conclusions that: (a) the failure by the parties and the FTT to identify with sufficient clarity the case being advanced in open and closed session by the FCO concerning the nature of the actual risks of harm to the public interest it was asserting would flow from disclosure of the material in respect of which it was claiming the section 27 exemption has resulted in unfairness which could have been avoided, and (b) it seems that this was caused, at least in part, by the approach taken to the closed aspects of the proceedings.
147. The FTT (it seems with the consent or acknowledgement of the parties) should not have proceeded with the closed session without it being recorded. We do not understand how this can have been thought to be appropriate in this case. In our view, it reflects an approach that fails to properly recognise the need to exercise appropriate care in determining what should be closed and to ensure that there is a proper record of what happened in closed session to inform the decisions to be made on what the excluded party should be told about this, and to inform any appellate tribunal or court.
148. Also, in cases of this type, and more generally we consider that there is a need to consider whether there should be oral evidence and cross examination in open (and / or closed session if there is one) or whether a combination of the duty of candour, documentary evidence and argument would provide an appropriate and sufficient process to fairly advance and test the risks of harm to the public interest being advanced, as it is in, for example, many PII claims.
149. As mentioned earlier, there was common ground before us that when assessing competing public interests under FOIA the correct approach is to identify the actual harm or prejudice that the proposed disclosure would (or would be likely to or may) cause and the actual benefits its disclosure would (or would be likely to or may) confer or promote. This equates to the approach now taken in PII claims and requires an appropriately detailed identification, proof, explanation and examination of both (a) the harm or prejudice, and (b) benefits that the proposed disclosure of the relevant material in respect of which the exemption is claimed would (or would be likely to or may) cause or promote.
150. How this is to be done fairly depends on the circumstances of the case. In some cases an approach of classifying a public interest as weak, strong or stronger and putting them in competing columns may be appropriate but in others, and particularly when competing public interests of the type and strength involved in this case arise, such an exercise does not go far beyond the starting point and is inevitably generic.
151. In many cases a particularisation of the reasons for those categorisations will be warranted to properly inform a consideration, weighing and balancing of the competing actual harm and benefits.
152. In our view, in all cases in which competing public interests have to be assessed, and in particular in a case when issues such as those that arise here are relevant, care must be taken to ensure that the competing factors are properly and sufficiently identified and so it is important for the parties and the tribunal (or other decision maker) to consider how the actual prejudice or harm on the one side and the actual benefit on the other is to be identified and established.
153. An exchange of witness statements is not best suited to a particularised identification of such issues and competing factors and, when it is needed, such an identification can generally be better achieved by a direction that a document identifying the issues, and so such factors, be produced by the parties. Such documents can take several forms.
154. Problems and fairness issues relating to what should be disclosed before and after a closed session are not new. As pointed out in Browning v IC and Department of Business Innovation and Skills  UKUT 236 (AAC) there is a distinction between the requested and disputed material itself, which cannot be disclosed without undermining the FOIA jurisdiction, and material advanced in respect of a qualified exemption to show that the balance of public interest is against disclosure. As to the latter, and subject to points that arise if such material would also be covered by a FOIA exemption or should not be disclosed for other reasons, fairness dictates that the requester (and so the party excluded from any closed session) is told as much as possible about the argument he has to address and of the evidence and reasoning it is based on.
155. So, when any closed session has examined evidence and/or argument relating to the existence and strength of a public interest against disclosure and so usually, if done properly, evidence and argument concerning the actual risks of harm that it is asserted the disclosure sought will or may bring about, it is highly likely that after the closed session the FTT and the parties should consider:
(1) whether amendments or additions should be made to an open document identifying the actual risks of harm being asserted, and/or
(2) whether such an open document should be prepared, and/or
(3) whether the excluded party should be told in specific or general terms of closed evidence, reasoning or argument.
156. That consideration is directed to ensuring that so far as possible the excluded party is informed of the case he has to meet. Also, it is directed to ensuring that the tribunal and the other parties keep under review the validity of the reasons why evidence and argument and/or the gist of them should be withheld from the excluded party.
Signed on the original on 11 November 2013
Mr Justice Charles CP
Mr Justice Burnett
The number of the document is taken from the second column of the schedule provided by the parties to identify the documents concerning which there was a dispute between the parties as to whether it remains part of the appeal.
This schedule should be read with in particular paragraphs 129 to 140 of this Decision.
Decision Notice 1
27(21) This is advice from a Law Officer and section 35(1)(c) applies to all of it. The overlap point applies to it.
28(22) All of the document is covered by s. 42(1). It is a note of a conference with counsel. The overlap point applies to it.
29(23) All of this document is not covered by section 42(1) because a claim for legal professional privilege could not be maintained in respect of all of it. The overlap point applies to it.
This document remains subject to the appeal.
30(24) All of this document is not covered by section 42(1) because a claim for legal professional privilege could not be maintained in respect of all of it. The overlap point applies to it.
This document remains subject to the appeal.
32(26) This is advice from counsel and section 42(1) applies to all of it. The overlap point applies to it.
34(N/A) These are emails communicating and commenting on advice from counsel and section 42(1) applies to all of them The overlap point applies to them.
37(29) These are emails communicating and commenting on legal advice from counsel and section 42(1) applies to all of them. The overlap point applies to them.
40(31) This is an email communicating advice from counsel and section 42 applies to all of it. The overlap point applies to it.
42(33) This is plainly a communication between ministers and the whole document is covered by section 35(1)(b). The overlap point applies to it.
46(36) This is an extract from a witness statement and so section 32(1) applies to the whole document. The overlap point applies to it.
47(37) This is plainly a communication between ministers and so the whole document is covered by section 35(1)(b). The overlap point applies to it.
49(39) This is plainly a communication between ministers and section 35(1)(b) applies to all of it. The overlap point applies to it.
50(40) All of this document is not covered by section 42(1) because a claim for legal professional privilege could not be maintained in respect of all of it. The overlap point applies to it.
This document remains subject to the appeal.
68(48) This is an advice from counsel and section 42 applies to the whole document. The overlap point applies to it.
77(57) This is plainly a communication between ministers and section 35(1)(b) applies to all of it. The overlap point applies to it.
85(63) All of this document is not covered by section 42(1) because a claim for legal professional privilege could not be maintained in respect of all of it. The overlap point applies to it. The parts that are covered by sections 23 and 42(1) are obviously within those exemptions and the recorded decision of the FTT that the rest is out of scope is not challenged. It is arguable that all or substantial parts of the document are covered by sections 35(1)(a) and/or (b).
This document remains subject to the appeal.
Decision Notice 2
(1) Section 42(1) applies to the whole document. The overlap point applies to it.
(3)(ii) Section 42(1) applies to the whole document. The overlap point applies to it.
(6) Section 42(1) applies to the whole document. The overlap point applies to it.
(7) This document is not in our closed bundle and we have not tracked down a copy to check if it is the same as 3(ii).
(8) Section 42(1) applies to the whole document. The overlap point applies to it.
(9) Section 42(1) applies to the whole document. The overlap point applies to it.
(10) Section 42(1) applies to the whole document. The overlap point applies to it. So we agree with the FTT and it seems that the FCO limited its claim to legal professional privilege to the top part of the document because it claimed exemption to the remainder on a different basis.
(13) Section 42(1) applies to the whole document. The overlap point applies to it.
(14) Section 42(1) applies to the whole document. The overlap point applies to it.
(15) Section 42(1) applies to the whole document. The overlap point applies to it.
(16) Section 42(1) applies to the whole document. The overlap point applies to it.
(17) Section 42(1) applies to the whole document. The overlap point applies to it.
(18) Section 42(1) applies to the whole document. The overlap point applies to it.
(19) Section 42(1) applies to the whole document. The overlap point applies to it.
(20) Section 42(1) applies to the whole document. The overlap point applies to it.
(22) Section 42(1) applies to the whole document. The overlap point applies to it.
(23) Section 42(1) applies to the whole document. The overlap point applies to it.
(26) This is a submission to ministers and so not within the definition of ministerial communications in section 35(5). The overlap point applies to it.
This document remains part of the appeal.
(33) Section 42(1) applies to the whole document. The overlap point applies to it.
(34) This document is not in our closed bundle and we have not tracked down a copy to check if it is the same as (9).
(35) Section 42(1) applies to the whole document. The overlap point applies to it.
(36) Section 42(1) applies to the whole document. The overlap point applies to it.
The FTT found that section 42(1) applied to documents 3(i), (5), (11), (25), (27), (28), (30), (31), and (32) when it had not been claimed by the FCO. We agree that the section 42(1) exemption applies to the whole of each of those documents. The overlap point applies to each of them.