Introduction
In this paper, I propose the concept of a ‘hybrid’ institution, which I define with reference to certain institutions within the UK's constitutional order which provide oversight of national security processes for which the executive has primary responsibility.Footnote 1 I focus in particular on the Commissioners who have overseen and oversee the use of investigatory powers and the work of the intelligence services.Footnote 2 These institutions, as was once said of another hybrid institution – the Independent Reviewer of Terrorism Legislation – operate within situations in which ‘potential conflicts between state power and civil liberties are acute, but information is tightly rationed’.Footnote 3 They are ‘hybrid’ institutions in that they marry certain of the features characteristic of political institutions with others characteristic of legal institutions. I evaluate the relevant institutions and the role they play within the national security constitution, and show that their hybrid status facilitates the performance of a function which neither fully legal nor fully political institutions could fulfil. On the basis of that account, I examine reforms to the status and functions of the Commissioners by the Investigatory Powers Act 2016. These, I suggest, represent a new stage in the history of hybrid institutions, one which risks undermining those of their features which have made them so important to accountability in the domain of national security and which demonstrates the outer limits not only of the Commissioners specifically, but of hybrid institutions in general.
1. Law, politics, secrecy
For a long period now, consideration of the UK's constitutional order has been framed according to the interaction of politics and law.Footnote 4 These spheres of constitutional action are generally distinguished in one of two ways: either according to the differing rationalities each practice employs, or by reference to the distinctive institutions in which they take place. In accounts of the latter type, the archetypal legal institution is a court, in which two parties present to a neutral decision maker (or group thereof) arguments, and a decision is made by him, or her, or them. Courts can determine only those disputes brought before them, which will be limited by rules of standing and jurisdiction, as well as rules as to what arguments have purchase therein; they reflect, that is, a distinction between arguments which are suitably ‘legal’ and those which are not, and so which courts, in the ordinary course of events, can neither hear nor act upon. The remedies which courts might offer are limited, and for the most part can be made use of only in relation to the specific dispute before them, even if the doctrine of precedent allows the effect of a single judgment to ripple through the legal system. On the other hand, the archetype of the political institution is Parliament, in which – at its best, at least – open and wide-ranging debate can take place, without preconditions, and without limit as to either the matters which might be taken into consideration or the ‘remedy’ that might be provided. On paper, at least, the Westminster Parliament can legislate about anything, at any time, with the limitations which exist on it mostly, and more immediately, political rather than legal. The continued presence of even those legal constraints which do exist are ultimately themselves contingent upon the non-existence of the political will to remove them, as the Brexit process has already shown, and the reform of the Human Rights Act may well confirm at some point in the future.
This general distinction between legal and political institutions can be unpacked along three axes, each with an obvious relevance to the specific context of national security. The first is the question of independence: the question of whether, and the extent to which, the institution in question is free from, and insulated against, political influence. In the case of courts in the domestic order, the historic overlap between the judicial and legislative institutions came to an end with the creation of the Supreme Court. New appointees to that court are not made life peers until they leave it,Footnote 5 and so all senior judges are now disqualified from membership of both Houses of Parliament.Footnote 6 But judges are not only separate from other institutions; they are also protected against them. The Constitutional Reform Act 2005 introduced a system of merit-based appointment, heavily limiting executive involvement in appointments.Footnote 7 The security of tenure given to judges for the first time by the Act of Settlement is now found in the Senior Courts Act 1981 and the 2005 Act, both of which provide that judges hold office on good behaviour, and may be removed only on an address presented to both Houses of Parliament.Footnote 8 On the other hand, the defining feature of political institutions is that their members enjoy no such independence – Parliament is a political institution, making political decisions, and the dominant chamber is populated by periodic election and so with people who can be (the point is crucial) removed.Footnote 9 Not all parliamentarians, of course, stand in the same relationship with the executive of the day, and the level of independence which any individual enjoys in respect of the executive varies with, most obviously, party affiliation and the chamber of which the individual in question is a member.Footnote 10 There is, nevertheless, at the aggregate level, a palpable party political bent (both between and within political parties) to much of Parliament's work, and though that fact does not preclude disagreement with the government's position (and may in fact, in the case of the opposition, magnify such disagreement) it is possible that the merits of the issue become (or at least risk becoming) secondary to those political considerations. What this will mean for the treatment of national security issues by Parliament has of course varied over time, and there have been points in history where – for good and sometimes less good reasons – national security has been treated as an essentially non-partisan issue.
The second axis along which institutions can be divided is also in part a function of the degree to which the functioning of each institution is insulated from party politics: what room exists for the presence, and the influence, of appropriate expertise. The most obviously relevant form of expertise is legal, which is a necessary feature of the courts but possessed by politicians only contingently. But expertise means something more: like other areas of policy, national security possesses a technocratic dimension to which the question of general legal expertise is mostly irrelevant.Footnote 11 On this more specific expertise judges rank less highly: though certain judges will develop expertise in national security questions – and some will have experience as counsel in litigation concerning such questions, perhaps even as special advocates (in which case they will have been subject to security vetting at some point in the past)Footnote 12 – their acquisition of relevant experience is likely to take place in an unsystematic fashion. And though such expertise is no doubt taken into account in appointments to the various specialist judicial bodiesFootnote 13 (as it is, it would seem, in the assignment of cases in, at least, the High Court)Footnote 14 there does not seem to be any formal means by which those who lack familiarity with these issues can be educated upon them apart from by hearing and deciding relevant cases. On the other hand, while there is no doubt that there is often considerable legal expertise in Parliament, and in the House of Lords in particular, the nature of political institutions is such as to devalue this expertise; it may inform general policy, and inform too specific decisions, but it will not – unlike in a court – determine them. On the subject-specific expertise, there is no particular reason to expect that any significant number of Members of Parliament outside of the current executive will have experience or knowledge relevant to the question of national security or related matters, though some – particularly those who have previously held Ministerial posts – may well do so.Footnote 15 Given the contemporary trend of those who have left the Great Offices of State leaving the House of Commons soon afterwards, however, there is little basis for confidence that this will be the case. More generally, the continued presence of individuals possessed of relevant experience in the elected chamber is subject also to the vagaries of the electoral process. The House of Lords may fare better in this respect, given the level of executive, military, and judicial experience found there.Footnote 16 This latter, for reasons considered below, often goes hand in hand with experience elsewhere in the national security apparatus of the constitution.
The third axis of differentiation is the jurisdiction of the relevant institutions: where and how they provide accountability in the broadest sense. The jurisdiction point is no different in the context of national security than in other policy areas: courts in an adversarial system resolve the cases brought before them, but have no capacity to seek out (nor jurisdiction to resolve) problems of their own initiative. They are similarly limited in the form of remedy they might provide. At the opposite end of the spectrum, the Parliamentary jurisdiction is in effect limited only by political considerations, including the requirement to prioritise. Parliament can legislate on what it wants and in a manner of its choosing. It can tweak the edges of the legal order, or sweep away hundreds of years of accumulated statutory regulation (and associated common law) with a single enactment. The only limits which exist are the weaker limits found in the Human Rights Act 1998, and the stronger (but soon to be diminished) limits imposed by the European Communities Act 1972's incorporation of EU law into the domestic order.
In the specific context of the national security constitution – the set of institutions and practices by which oversight is carried out in the domain of national securityFootnote 17 – however, there is a fourth facet to our institutional analysis, over and above those elements which essentially distinguish political from legal institutions: that of secrecy. Oversight of national security processes, and of the use of counter-terrorist and investigatory powers, cannot take place in the open, but requires a process which allows for access to sensitive national security material and for, before that, the identification of persons who are trusted to consult it. Secrecy therefore has two interconnected facets: access to and ability to make use of material which is secret and (upon which it is often, but not by definition, contingent) a certainty that the material will not be publicly disclosed. Both legal and political institutions in their archetypal form score badly on both measures. In each case, however, adjustments can be and are made to allow legal and political institutions to make use of secret material as part of their operation. Judges are permitted to view sensitive material, and indeed are often required to do so (even if only to judge that it is sufficiently sensitive that it cannot be disclosed to the parties or the wider public).Footnote 18 Like Ministers of the Crown,Footnote 19 but unlike coroners,Footnote 20 judges of the High Court and above are permitted to see security-cleared material without passing through the relevant vetting procedure.Footnote 21 Admirable though this combination of independence and trustworthiness is, however, the normal judicial process lacks the secrecy that is accepted to be – at least some of the time – a necessary feature of national security accountability. The processes of the courts are in almost all cases open to the press and public and, even where courts sit in camera, to the parties and their representatives.Footnote 22
This problem has been in part addressed by a number of exceptions to the principles which govern ordinary legal proceedings.Footnote 23 Amongst these are the longstanding rules of public interest immunity (PII), whereby a body, usually some element of the executive, may certify that the disclosure of certain specified material would be harmful to the public interest, such that it outweighs the public interest in its disclosure; where the court to which the claim is made accepts that assessment, the material is not disclosed in evidence and cannot be relied upon by any party to proceedings. PII therefore meets our criteria only in part; the material is available to the court, but not relied upon by it, and so of no use in processes of accountability to which secret material is relevant. More recent innovations include, on one hand, the closed material procedure – introduced by the Special Immigration Appeals Commission Act 1997 and now available in a variety of contexts, and (under more restricted circumstances) in civil proceedings generallyFootnote 24 – and, on the other, the creation of an Investigatory Powers Tribunal,Footnote 25 in which the regime of secrecy represents an even greater departure from the common law tradition of open justice.Footnote 26 Here, secret material is not only available to the body in question but can be relied upon by it in determining the distribution of rights and obligations. But while secrecy can be achieved within such extraordinary judicial venues, there is a clear cost in terms of those factors which define the judicial process: though justice might be done, it is not seen to be done,Footnote 27 with the key material dealt with usually in closed judgments, and – in the case of the IPT – operating on a ‘neither confirm nor deny’ basis where assumed, rather than proven, facts are often at issue. And though these innovations address the question of secrecy, they are nevertheless, in the context of the criteria outlined above, resolutely legal institutions, in which independent and legally qualified actors adjudicate upon cases argued before them on adversarial principles, determining them according to a category of considerations which are identifiably legal.Footnote 28 For the most part, the relevant national security tribunals slot seamlessly into the ordinary court structure,Footnote 29 with it being possible to appeal to the Court of Appeal and then, if necessary, to the Supreme Court.Footnote 30
On the political side, it is in the nature of Parliament that it is a site of public contestation: for that reason, to attempt to achieve any sort of secrecy is inimical to its performance of its constitutional role (though very exceptionally, recourse is made to private sessions of the House of Commons).Footnote 31 Nevertheless, details of national security matters will not be given to Parliament. This is not to say that there is no possibility of secrecy whatsoever: judicious distribution of information on ‘privy council terms’ allows, for example, for the leader of the opposition to be kept informed of relevant matters.Footnote 32 Even, however, if Parliament might be capable of performing its immediate constitutional role in conditions of secrecy, it could not hope to perform the broader, legitimating role upon which the continued good health of the constitutional order depends: that role requires that its debate takes place openly and publicly.Footnote 33 Here too, however, there exists a variation on the norm designed to accommodate the needs of national security within Parliament: the Intelligence and Security Committee.Footnote 34 The Committee was established by the Intelligence Services Act 1994. The Justice and Security Act 2013 made it a ‘Committee of Parliament’ and gave the power of appointment to the relevant chamber of Parliament (upon nomination by the Prime Minister and after consultation with the leader of the opposition), with Ministers of the Crown, as before, ineligible.Footnote 35 As a corollary, its members now enjoy a greater security of tenure than before: although the basic position is (as previously) that they hold office only for the lifetime of the Parliament in which appointed, members could previously be removed before the end of their mandate by the Prime Minister.Footnote 36 Now, involuntary removal requires a resolution of the House of Parliament of which the person in question is a member.Footnote 37 Where previously the PM appointed a chairman, the ISC now chooses its own chair.Footnote 38 The aggregate effect of these rules is to constitute it as a body with a considerable (and significantly increased) level of independence from the executive whose activities it works to scrutinise. There are no further statutory requirements as to the qualifications that members of the ISC must possess.Footnote 39
The role of the ISC was also modified by the 2013 Act. While it retains the function of overseeing the ‘expenditure, administration, policy and operations’ of the security and intelligence agencies (a small but significant expansion as compared to the 1994 Act),Footnote 40 its ability to produce ad hoc reports is no longer limited to those functions. Instead it may ‘examine or otherwise oversee such other activities of Her Majesty's Government in relation to intelligence or security matters as are set out in a memorandum of understanding’.Footnote 41 The relevant MoU, published as an annex to the ISC's 2014 report,Footnote 42 identifies a number of activities undertaken by specified departments which the ISC will oversee in addition to the oversight function specified in the 2013 Act, though the relevant activities are described at a high level of generality. These are subject to certain agreed limitations including that ‘[w]here there are legal proceedings (criminal or civil), inquiries or inquest proceedings, the ISC and HMG will consider carefully whether it is appropriate to proceed with an investigation’.Footnote 43 These exist alongside statutory limitations (applying to both statutory and agreed oversight functions).Footnote 44 The MoU justifies the additional oversight functions of the ISC by noting that it is ‘the only committee of Parliament that has regular access to protectively marked information that is sensitive for national security reasons: this means that only the ISC is in a position to scrutinise effectively the work of the Agencies and of those parts of Departments whose work is directly concerned with intelligence and security matters’.Footnote 45 This is perhaps the most important feature of the ISC in terms of distinguishing it from its Parliamentary counterparts: the Committee is entitled by statute to request information from the Directors of the SIAs, who are required to disclose in accordance with arrangements approved by the relevant Secretary of State unless it cannot be disclosed because it is both ‘sensitive’Footnote 46 and ‘information which, in the interests of national security’ should not be disclosed.Footnote 47 A second exception applies to ‘information of such a nature that, if the Secretary of State were requested to produce it before a Departmental Select Committee of the House of Commons, the Secretary of State would consider (on grounds which were not limited to national security) it proper not to do so’.Footnote 48 Though this is – not least because of its breath – a significant power as far as the oversight of national security matters goes (albeit one whose exceptions go a considerable distance towards blunting it), it is weaker than the equivalent powers possessed by courts.
The ISC must make an annual report to Parliament, and may make other reports concerning its functions to Parliament as it considers appropriate: again, it has a roving jurisdiction.Footnote 49 Reports must first be sent to the Prime Minister, and any material which the Prime Minister (after consultation with the ISC) considers ‘would be prejudicial to the continued discharge’ of the functions of the security services or those bodies covered by the MoU must be excluded.Footnote 50 Where material is excised on that basis, the report must state that fact, and in practice ‘the published Report is the same as the classified version sent to the Prime Minister (albeit with redactions)’ so that it is clear not only that redactions have been made, but also whereabouts in the report redaction has taken place.Footnote 51 Reports can also be made to the Prime Minister ‘in relation to matters which would be excluded’ under that same standard if the report was made to Parliament.Footnote 52 Here, not even the fact of the report, never mind its content, will be known to the public. The nature of the exercise is such, however, that whether the provisions in question are being abused is impossible to know: certainly, while the matter would seem to be justiciable in the strict sense, any court called upon to review the decision to exclude sensitive material would apply a standard of review so light in its touch as to be barely discernible.
The overall effect is that institutions such as SIAC and the IPT on one hand and the ISC on the other – institutions which, notwithstanding modifications made in order to reflect the national security context in which they operate, are fundamentally either legal (and so have limited jurisdiction) or political (and so are either insufficiently independent and only contingently possessed of the necessary expertise) – are necessarily limited in their ability to provide oversight in the national security context. For that reason, the national security constitution relies heavily upon institutions which are neither one thing nor the other; which possess some of features characteristic of political institutions alongside others characteristic of legal institutions.Footnote 53 These are ‘hybrid’ institutions.
2. Hybrid institutions
Hybrid institutions possess enjoy some combination of the various features above as distinctive of legal and political bodies; features which permit them to do, in the context of national security law and policy, work for which neither Parliament nor the courts are wholly suited. One such institution is the Independent Reviewer of Terrorism Legislation (IRTL);Footnote 54 others, the focus of the present paper, include the various investigatory powers and intelligence Commissioners (past and present).Footnote 55 Two points must be made immediately about the concept of hybrid institutions (and the hybrid constitutionalism which they might be taken to exemplify) under discussion here. The first is that to identify it as a prominent feature of the national security constitution – that part of the constitutional order which deals with national security as an area of policy – is not to claim the entirety of that space for the concept: as seen above, ‘standard’ institutions, both legal and political, operate in this domain, and no one would wish to exclude them from it entirely. Nor, second, is to say that national security is an area in which hybrid institutions flourish to say that the concept is exclusive to this area. There are many institutions which marry legal and political forms in distinctive and productive ways, some of them relevant to national security only contingently, others not relevant to it at all.
Amongst these latter is the public inquiry, which now usually takes place under the Inquiries Act 2005, often (though not inevitably) chaired by a judge,Footnote 56 and which of course are employed in a vast range of contexts. Though there has been some use of public inquiries in the national security context (broadly understood)Footnote 57 the most important such inquiry in recent history – the Gibson inquiry, established in order to consider ‘whether Britain was implicated in the improper treatment of detainees, held by other countries, that may have occurred in the aftermath of 9/11’Footnote 58 – was not a statutory inquiry,Footnote 59 and was ultimately brought to a premature conclusion as the Crown Prosecution Service considered whether to bring criminal proceedings in relation to the underlying events. By the time, however, that the CPS decided not to bring charges,Footnote 60 the matter had been handed over instead to the ISC,Footnote 61 whose statutory powers to require the provision of material which might otherwise be subject to PII must be balanced against several other of its features. First, its relative lack of independence; secondly, the exceptions to those powers for which the 2013 Act provides; thirdly, its inability to compel witnesses to testify;Footnote 62 and, fourthly, the possibility of withholding its reports in whole or in part according to their type. Public inquiries, however, are ad hoc institutions, whereas those hybrid bodies which are most likely to produce a meaningful and ongoing accountability are standing bodies; relatedly, inquiries lack the roving jurisdiction which is so central to the effectiveness of hybrid institutions as a category. They are therefore less effective in contexts where oversight is contingent, taking place only when the executive decides that it will; where secrecy is the norm, the fact of oversight must not be dependent on public knowledge of the events to which oversight is related. Notwithstanding, therefore, that examples of hybridity can be identified beyond the national security context, it seems undeniable that, if only by reason of the need to maintain secrecy and the tension between secrecy and the archetypal forms of legal and political institutions, the domain of national security is one in which hybrid institutions have a special purchase. They are more numerous and more prominent than is the case elsewhere.
3. The Commissioners
Having introduced the concept of a hybrid institution, it is possible now to turn to a particular group of such institutions as they have existed and evolved over time: the intelligence and investigatory powers Commissioners.Footnote 63
(a) Background
There have existed over the last 40 or so years a number of distinct Commissioner roles in the field of national security, though the work of what were until recently three separate persons is now – following the enactment of the Investigatory Powers Act 2016 – consolidated within the person of a single Investigatory Powers Commissioner, assisted by a number of Judicial Commissioners.Footnote 64 However, even before there was statutory regulation of the interception of communication (within the British Islands, that is),Footnote 65 there was oversight by a senior judge, with Lord Diplock tasked with reporting, in 1981, on the interception of communication as it was carried out by the Post Office on behalf of the police and security services.Footnote 66 The Interception of Communications Act 1985 created the post of Commissioner to review the Secretary of State's functions under that statute (broadly, the issuance of warrants for the interception of communication) and to assist the newly created Interception of Communications Tribunal.Footnote 67 To this landscape was added, by the Security Service Act 1989, the Security Service Commissioner;Footnote 68 he, in turn, was joined by the new Intelligence Services Commissioner introduced by the Intelligence Services Act 1994.Footnote 69 Each was given a variety of oversight duties and the statutory possibility of being provided with ‘such staff as the Secretary of State thinks necessary for the discharge of his functions’.Footnote 70 This original generation of Commissioners were all required to be persons who at the time of appointment held or had held ‘high judicial office’.Footnote 71 Each was to hold office ‘in accordance with the terms of his appointment’,Footnote 72 and so the requirement of judicial experience was not accompanied by any statutory security of tenure, even temporary. In the early years, the reports of the Commissioners are mostly of very little value. The work of the Security Service Commissioner, for example, never rose above the perfunctory:Footnote 73 the Commissioner chose to apply judicial review principles in his work, meaning that as long as he was ‘satisfied that the application for a warrant [fell] within the functions of the Service’ it was not his place ‘to substitute [his] judgement for that of the Secretary of State, but to consider whether the Secretary of State could have properly come to the conclusion which he did’.Footnote 74 When the Security Service Act 1996 expanded MI5's functions to include acting ‘support of the activities of police forces and other law enforcement agencies in the prevention and detection of serious crime’Footnote 75 the Commissioner reconsidered the matter but concluded that his approach should be maintained.Footnote 76 In practice, therefore, the reports are mostly a rehashing of the relevant statutory frameworks, with substantive consideration limited to procedural questions and error-reporting. This approach was carried over by the Commissioner, Lord Justice Stuart-Smith, into his work as Intelligence Services Commissioner when that post was created, and was still evident in the years following the enactment of the Regulation of Investigatory Powers Act 2000. As time has passed (and the issue of investigatory powers become more politically salient) this style has faded, and the reports issued in the final years of this generation of Commissioner were longer and more informative.
RIPA 2000 abolished the previous Commissioners and provided for a renewed Interception of Communications CommissionerFootnote 77 and an Intelligence Services Commissioner (now responsible for all the intelligence services).Footnote 78 It also gave new functions to the Chief Surveillance Commissioner whose role had been created by the Police Act 1997,Footnote 79 and who is responsible for overseeing the use of the RIPA part II powers of surveillance and the use of covert human intelligence sources. The requirement of having held high judicial office applied also to these three roles.Footnote 80 While each of the two RIPA Commissioners held office ‘in accordance with the terms of his appointment’,Footnote 81 the Police Act 1997 treated the Chief Surveillance Commissioner (and those Commissioners appointed to assist him) differently: they were appointed for a (renewable) fixed term of three years, and could be removed from office before the end of the term only if a resolution which approved the removal was passed by both Houses of Parliament.Footnote 82 The RIPA Commissioners were similarly required to produce annual reports, which were to be and were laid before Parliament, with any exclusion of material acknowledged in an accompanying statement. Nevertheless, the most important feature of those of the Commissioners reports which have been made public is that their scope is determined not by the Commissioners themselves but, in the first place, by Parliament and the executive. This has meant that that the powers whose existence would have been most controversial in the period before they were avowed (bulk acquisition of communications data, and the retention and use of bulk personal datasets, both of which – as the ‘bulk’ label suggests – encompassed also persons of no interest to the security services) were either not subject to oversight, or were subject to oversight informally and/or without the fruits of the relevant labour being made available to the public. By far the most important of the reports by the Interception of Communications Commissioner, for instance, is that into the acquisition of communications data under the powers contained in the Telecommunications Act 1984.Footnote 83 The use of the power in question, though long-standing (and known to previous Commissioners) was effectively secret, with the statutory power, as it were, hidden in plain sight. Only after its avowal was the necessary (public) oversight possible, and even then on an ad hoc basis. Though own-initiative reports to the Prime Minister were permitted by RIPA 2000, there was no duty on the Commissioners, the Prime Minister, or anyone else, to make the content (or even the existence) of such a report known. One of the primary advantages of hybrid mechanisms is that – unlike courts – their jurisdiction is not limited by rules of standing or the need to for there to exist some legal dispute. Legislating so as to allow the Commissioners to produce reports on matters relating to their functions and then permitting those reports to end up in an executive black hole undermines this advantage. And, as discussed below, the basic position under RIPA 2000 – as regards both exclusions from the annual reports and the absence of any duties in relation to other reports – carries over into the current law.
Alongside their oversight work, the various Commissioners have always been tasked with working in conjunction with the associated judicial body, being required to assist the Tribunals which have and continue to exist to oversee the systems of warrants etc, receiving from the Tribunals reports of contraventions of the relevant statutory regimes, and being required to report on contraventions which had not been the subject of a report of the Tribunal. Complaints made to the Tribunal relating to interferences with property and acts done by virtue of an authorisation under s 7 of the 1994 Act were in fact to be investigated by the Commissioner – first by the Security Service Commissioner, and later the Intelligence Services Commissioner – with any warrants or authorisations relevant to matter be assessed by the Commissioner applying judicial review principles; in practice, no such complaint seems to have ever resulted in a determination in the complainant's favour. This effectively judicial role was abolished by RIPA 2000, which nevertheless required the various Commissioners to assist the Investigatory Powers Tribunal, including by offering the Tribunal their opinion as to ‘any issue falling to be determined by the Tribunal’.Footnote 84 The Commissioners thus took on a more obviously hybrid role – one unsullied by a very close association with the relevant judicial body.
(b) The Investigatory Powers Act 2016
In his report on the system of investigatory powers, the then Independent Reviewer of Terrorism Legislation, David Anderson QC, recommended the creation of a body – the Independent Surveillance and Intelligence Commission – which would merge the functions of the three then-existing Commissioners and, perhaps, a ‘more general supervisory power’ over the activities of the security and intelligence agencies.Footnote 85 The Commission would be led by a Chief Commissioner. This single body would, by virtue of its size and unified nature, have several advantages over the model it replaced, the inconsistencies of whichFootnote 86 had frequently been remarked upon. In the event, and despite the position taken by the Joint Committee on the Draft Bill,Footnote 87 the Investigatory Powers Act introduced not a Commission but a Commissioner – the Investigatory Powers Commissioner (IPC) – with responsibility for overseeing the whole set of powers and processes previously overseen by the separate Commissioners, but now assisted by a number of other Judicial Commissioners (JCs).Footnote 88 This new body can be mapped against the criteria identified above.
From the point of view of independence, several points can be made about the IPC and JCs. Appointment to all of these roles is carried out by the Prime Minister rather than, say, the House of Commons or an independent body of the sort that is now responsible for senior judicial appointments.Footnote 89 The appointment process provided for by the 2016 Act was criticised in advance of its enactment by the Office of the Interception of Communications Commissioner (IOCCO), on the basis that appointment by the Prime Minister ‘dilutes public confidence and independence’.Footnote 90 Though the Joint Committee on the Bill recommended that the appointments of JCs be carried out by the Lord Chief Justice following consultation with the Prime Minister and the devolved administrations, it did not think that ‘appointment by the Prime Minister would in reality have any impact on the independence of the Investigatory Powers Commissioner and Judicial Commissioners’.Footnote 91 Its reason demonstrates an interesting feature of hybrid institutions, whereby the senior judiciary is drawn upon not only due to its expertise, but because its members are seen as inherently independent, and so capable of balancing out or negating issues of independence which might otherwise arise: ‘In modern times, our senior judges have had an unimpeachable record of independence from the executive and we believe any senior judge appointed to these roles would make his or her decisions unaffected by the manner of appointment’.Footnote 92 The Commissioners operate on the model of the Surveillance Commissioners: they are appointed for a (renewable) term of three years,Footnote 93 and can be removed from office in limited circumstances, either by the Prime Minister under certain specified circumstances – including bankruptcy or imprisonment – or following the passing of a resolution approving the removal by both Houses of Parliament.Footnote 94
The independence of the IPCs and JCs is thereby secured only in some of the ways one might expect where judges are performing a strictly judicial role. The desirability of securing even a (theoretically lesser) degree of independence (and the appropriateness of a statutory security of tenure as the means of doing so) having been recognised, however, the three-year term of appointment is unfortunately short. This point has been made in the context of the IRTL: though appointment to that post is for a renewable term of three years, this is not a statutory requirement, and there is no security of tenure at all. Even there was some such security, the former holder of the post has identified the risk that ‘a reviewer might be perceived to be influenced by the desire to be appointed to a second term – or, indeed, by the desire for additional preferment once the term of office is complete’.Footnote 95 The same point might be made of the IPC and the JCs, and it is not clear why the Commissioners should be appointed for a period so much shorter than the standard length of a Parliament. Extending their term to five or even seven years would seem to make sense, and would create the possibility of moving to non-renewable terms, if it was felt that the prospect of not having one's position renewed risks producing incentives which run counter to the intention of Parliament.Footnote 96 Even as the position stands, however, the Commissioners as an institution bear – from the point of view of independence – far more resemblance to legal than to political institutions.
From the point of view of expertise, a person, again, may be appointed as a Commissioner only if he or she holds or has held a high judicial office;Footnote 97 a person may not be appointed as the IPC unless jointly recommended by the Lord Chancellor, the Lord President of the Court of Session, and the Lord Chief Justices of England and Wales and Northern Ireland,Footnote 98 while appointment as a JC requires the joint recommendation not only of these four, but also of the IPC.Footnote 99 The first IPC appointed subsequent to the enactment of the IPA was Sir Adrian Fulford, a Lord Justice of Appeal. Later in 2017, a total of 13 JCs were appointed, amongst them a former Surveillance Commissioner (Lord Bonomy) and members of the Special Immigration Appeals Commission and the Security Vetting Appeal Tribunal. There is thus both legal expertise of the sort usually associated with courts and – within the limits of that constraint – a degree of expertise in the specific domain of national security. From this point of view, therefore, the new institution again has the aspect of a court rather than a legislature.
(c) Functions and powers of the Judicial Commissioners
As, considered from the point of view of both independence and expertise, the Commissioners possess the characteristics of legal rather than political bodies, the claim that they enjoy a hybrid status by definition represents an assertion about their jurisdiction, which resembles far more closely that of political institutions than that of their legal counterparts. In this regard, the 2016 Act also represents a significant evolution from its predecessors. The primary duty of the IPC is to keep under review (‘including by way of audit, inspection and investigation’)Footnote 100 the exercise by public authorities of specified statutory functions relating to the interception of communications and the acquisition of communications data as well as a number of further matters, including acts – such as ‘the acquisition, retention, use or disclosure of bulk personal datasets by an intelligence service’ – only avowed in the lead up to the enactment of the Investigatory Powers Act.Footnote 101 This, it goes without saying, is the sort of function which legal institutions do not – because they cannot – play in our domestic order.
Within this oversight function – post hoc and general – one difficulty in the Commissioners model has been that it mixes statutory and non-statutory functions in a fashion which threatens to undermine the independence of the model. The prior Commissioners were, at various points, tasked with providing oversight of the use of bulk personal datasets (BPDs) and some elements of the regime for accessing bulk communications data (BCD) acquired in accordance with directions under s 94 of the Telecommunications Act 1984 (but not the giving of those directions in the first place). Because, however, the powers in question had not been openly avowed, the oversight did not have a legal basis and could not be discussed in the versions of their reports laid before Parliament. When the Justice and Security Act 2013 was enacted, it inserted into RIPA 2000 a provision allowing the Prime Minister to give directions to the ISC requiring him to keep under review any specified function of, inter alia, the Intelligence Services.Footnote 102 In accordance with this, there was given a direction formalising the oversight of BPDs,Footnote 103 and another requiring oversight of compliance with the Consolidated Guidance to Intelligence Officers and Service Personnel on the Detention and Interviewing of Detainees Overseas, and on the Passing and Receipt of Intelligence Relating to Detainees. This formalisation represents an improvement on the status quo ante, but remained imperfect: the basic rule was that Directions were to be published, but the Prime Minister was permitted to withhold them if it appeared that publishing them would be contrary to the public interest generally or prejudicial to a number of specified interests, including national security.Footnote 104 For that reason, it is not known how many other Directions might have been issued under the provision. It was, however, reported in late 2016 that there had been disclosed to Privacy International (in the course of its challenge within the IPT to the use of BPDs and BCDs) part of the Confidential Annex to the Intelligence Services Commissioner's 2014 report, and that this indicated the existence of a third Direction, the subject of which remained unknown until 2018, and which is discussed further below.Footnote 105 This possibility of adding to the oversight role remains intact under the 2016 Act, which permits the Prime Minister to direct the IPC to ‘keep under review the carrying out of any aspect of the functions’ of the SIAs, their heads, or ‘any part of Her Majesty's forces, or of the Ministry of Defence, so far as engaging in intelligence activities’.Footnote 106 A direction may, but need not, be given at the request of either the IPC or the ISC.Footnote 107 They are discussed further in the following section, but it is important to note than nothing in the 2016 Act would appear to preclude the Prime Minister from making informal requests to the IPC to carry out additional oversight. The core jurisdiction of the IPC can therefore be augmented and though additions made in exercise of the statutory power are limited to the SIAs and intelligence more broadly, there is no reason that non-statutory directions to the IPC could not relate to other elements of the national security constitution.
A further duty of the IPC is to report ‘relevant errors’ to those to whom those errors relate.Footnote 108 A relevant error is, in the first place, an error ‘by a public authority in complying with any requirements which are imposed on it by virtue of this Act or any other enactment and which are subject to review by a Judicial Commissioner’.Footnote 109 This creates a link between the role of the IPC and the work of the Tribunal, with the IPC obliged to inform the person to whom the error relates both of their right to apply to the Tribunal and such details as he considers necessary to exercise those rights.Footnote 110 The error reporting function is, however, heavily caveated: it bites only where an error is a ‘serious error’ and ‘it is in the public interest for the person to be informed of the error’.Footnote 111 The IPC may not decide that an error falls into the former category unless he ‘considers that the error has caused significant prejudice or harm to the person concerned’Footnote 112 and, in particular, an error is not a serious error simply because there has been a breach of a person's rights under the European Convention on Human Rights as incorporated into domestic law by the HRA 1998.Footnote 113 A number of factors must be considered in deciding whether reporting of a serious error is in the public interest – including ‘the seriousness of the error and its effect on the person concerned’ but also ‘the extent to which disclosing the error would be contrary to the public interest’ or prejudicial to a number of specified interests – and the public authority in question must be asked to make submissions on the point before a decision is made.Footnote 114 This threshold for deeming an error serious is therefore an extremely high one. It will deprive the error-reporting function of much of its impact, meaning that unlawful acts by public authorities will often go unknown to the public at large. The possibility of redress before the IPT may therefore be lost notwithstanding that the Commissioners might know that that the error was of such nature that a claim in respect of it would result in a successful application to the Tribunal. That an error might be serious but nevertheless not reported because it is not in the public interest to do so moves the system of error-reporting from the category of merely inadequate to that of outrageous.Footnote 115 Within our institutional classification, the de facto link between the IPC and the Tribunal created by the former's error-reporting function might be thought to give rise to resemblance between the IPC and a court; permitting considerations of public interest to override questions of legality pulls against that instinct. Though these duties exist alongside one to give assistance to the IPT (including the Commissioners’ opinion on matters falling to be determined by the Tribunal),Footnote 116 that role – which resembles the role of the Commissioners pre-RIPA – does not upset the clear distinction between the judicial institution, which decides specific legal questions on the application of specific claimants, and the hybrid one, whose primary function is to consider the operation of the system at the general level.
The question of jurisdiction arises, finally, in the context of the IPC's duty to make an annual report on the exercise of his statutory functions, as well as such other reports as are required by the Prime Minister.Footnote 117 The 2016 Act prescribes in significant detail the content of such a report – requiring, most notably, that it contain ‘statistics on the use of the investigatory powers which are subject to review by the Investigatory Powers Commissioner’.Footnote 118 More significantly, however, from the institutional point of view, and following the pattern of the earlier statutes, the 2016 Act permits the IPC to make ‘at any time, make any such report to the Prime Minister, on any matter relating to the functions of the Judicial Commissioners, as the Investigatory Powers Commissioner considers appropriate’.Footnote 119 These provisions allow the JCs to exercise a degree of self-direction within the confines of a model which is in the first place oriented towards regular and ongoing oversight. Alongside the post hoc nature of the oversight, their existence again works to perpetuate the jurisdictional distinction between the work of judges qua judges and that qua JC, and so the hybrid nature of the latter.
(d) Secrecy
The final axis along which to consider the IPC and JCs is, as noted above, that of secrecy: access to confidential material, the ability to rely upon it in holding the relevant actors to account, and the mechanisms for ensuring it does not become public where disclosure would be contrary to the public interest. It is secrecy, it was said, which permits hybrid institutions to play, within the national security context, a role for which fully legal and fully political institutions are often inapt. The first thing to note here is that the requirement of high judicial experience which applies to the IPC and JCs means they are deemed to be security-cleared, though it is possible and perhaps likely that additional checks are made on those who are being considered for appointment to one of these roles. Their statutory powers are significant: JCs may ‘carry out such investigations, inspections and audits as the Commissioner considers appropriate for the purposes of the Commissioner's functions’,Footnote 120 while any person working within a public authority (and others) must ‘disclose or provide to a Judicial Commissioner all such documents and information as the Commissioner may require for the purposes of the Commissioner's functions’ and ‘provide a Judicial Commissioner with such assistance as the Commissioner may require in carrying out any investigation, inspection or audit for the purposes of the Commissioner's functions’.Footnote 121 Crucially, this assistance includes ‘such access to apparatus, systems or other facilities or services as the Judicial Commissioner concerned may require in carrying out any investigation, inspection or audit for the purposes of the Commissioner's functions’,Footnote 122 meaning that the JCs can require access to the computer systems on which data is stored and through which it is queried by the intelligence services and others. This should ensure that oversight is based on a direct experience of the systems as they can be and are used in practice, rather than a curated and perhaps unrepresentative sample of the SIAs’ work. Strikingly, these statutory powers are in no way caveated by the needs of secrecy.
The annual reports of the IPC are to be made to the Prime Minister, who must publish and lay them before Parliament. The Prime Minister (after consultation with the IPC) may, though, exclude from publication any part of a report if, in the PM's opinion, its publication would be ‘contrary to the public interest’ or prejudicial to one of a number of specified interests, including national security and ‘the continued discharge of the functions of any public authority whose activities include activities that are subject to review by the Investigatory Powers Commissioner’.Footnote 123 This provision is framed more widely than is justifiable: the fact that the specified interests are additional to, rather than working to delimit, the general reference to the public interest, means that material can be withheld from the public for reasons which are not a compelling justification for secrecy – the harm that might be done to the reputation of the security services, for example, if the public was to become aware of certain of its activities, insofar as such damage would prejudice the continued discharge of its functions. Nor is it made explicit that the reference to the public interest imposes on the Prime Minister a duty to balance the public interest considerations favouring non-disclosure against those favouring disclosure. As it stands, the fact that what is at issue here is a body with a broader jurisdiction than is enjoyed by courts – operating at the macro level and in the absence of a lis inter partes – goes hand in hand with a broader possibility of withholding material than is the case in, say, the use of closed material procedures under the Justice and Security Act 2013. This is difficult to justify given that, in advising the PM, the SIAs are liable to take a very broad view of what would be prejudicial to the discharge of their functions. Wherever material is removed from the annual report of the IPC on this basis, the report must be laid before Parliament together with a statement as to whether any part of it has been excluded from publication in accordance with these rules, but if – as has been the case in the past – all that is said is that further material is found in a confidential annex, then the possibility of political accountability is only minimally heightened by this requirement. The IPC may, the 2016 Act provides, publish any other report (or a part of such a report) ‘if requested to do so by the Prime Minister’.Footnote 124 Not only is there no assumption in favour of such publication, but this provision must mean that the IPC is not otherwise permitted to publish either those reports made at the request of the Prime Minister or, indeed, his own-initiative reports. This also tips the balance too far towards secrecy – an independent oversight mechanism should retain the right, in extreme circumstance, to publish material that it believes the public needs to know. Had previous Commissioners done so, the regime of investigatory powers might have been rationalised long before Edward Snowden precipitated the enactment of the 2016 Act. As it is, the Prime Minister acts as a – literally – inscrutable gatekeeper to the information contained in such a report, and the refusal to request its publication cannot be challenged if the very existence of the report is unknown.
Finally, there is the issue of additional directed oversight functions, the possibility of which was noted above. Such a direction must be published, unless and to the extent that ‘it appears to the Prime Minister that such publication would be contrary to the public interest or prejudicial to’ one of the standard three interests (national security, the prevention or detection of serious crime, and the economic well-being of the UK) or to ‘the continued discharge of the functions of any public authority whose activities include activities that are subject to review by the Investigatory Powers Commissioner’.Footnote 125 At the time of writing, two such Directions have been published. One relates to oversight of the compliance of the SIAs, the armed forces and military intelligence with the Consolidated Guidance.Footnote 126 It is by no means clear why such oversight could not have been required by the 2016 Act itself, given that it predated the 2016 Act, and it appears always to have been the intention that it would carry over in the new regime.Footnote 127 The other was revealed when, in March 2018, the Prime Minister made a written statement to the House of Commons revealing the existence of a direction by which she ‘instructed the Commissioner to keep under review the application of the Security Service guidelines on the use of agents who participate in criminality and the authorisations issued in accordance with them’.Footnote 128 Any further Directions which may exist now or in the future can be kept secret. We see here the outer limit of the utility of the hybrid model: the oversight will be independent and based upon access to secret material, but may be considered so sensitive that not only can the public (and Parliament) not know what are the conclusions of the IPC, but cannot even know that certain oversight is taking place. So significant are these limitations that is justifiable to query whether the oversight in question is better than no oversight at all.
(e) The role of the JCs in warrantry
The Commissioners under the IPA are not, however, merely a continuation of the previous model, providing reactive and general oversight of otherwise secret material and so straddling neatly the boundary between legal and political institutions. The most important innovation of the new IPC and the JCs who work with him is their involvement in the so-called ‘double lock’ process introduced by the 2016 Act in relation to surveillance warrants issued thereunder. The majority of warrants under the IPA are, in keeping with practice going back hundreds of years, issued by a Minister of the Crown, who is to decide whether the application for the warrant fulfils the statutory criteria, which themselves vary depending upon the nature of the warrant in question and in particular whether it is a ‘targeted’ or ‘bulk’ warrant.Footnote 129 For the first time, however, warrants must (except in cases of urgency) be approved also by a JC, who is to review the granting person's conclusions as to ‘whether the warrant is necessary on relevant grounds’ and ‘whether the conduct that would be authorised by the warrant is proportionate to what is sought to be achieved by that conduct’.Footnote 130 In conducting such review, the JC is to, first of all, ‘apply the same principles as would be applied by a court on an application for judicial review’ and, second, consider the relevant matters ‘with a sufficient degree of care as to ensure that the Judicial Commissioner complies with’ the general duties in relation to privacy found elsewhere in the Act.Footnote 131 Refusal by a JC must be in writing and, where it is carried out by a JC other than the IPC himself, can – in effect – be appealed to the IPC.Footnote 132
This new regime reflects – at least in part – indications in the jurisprudence of the Court of Human Rights that secret surveillance measures must be approved by an actor with some degree of independence; a requirement which may in time evolve into a requirement of judicial rather than executive approval.Footnote 133 Several points can be made about it. The first is that it is not, for a number of reasons, wholly novel. As considered above, the Commissioners were in pre-RIPA regimes tasked with reviewing – again, on judicial review principles – warrants and authorisations. But of course that review was not only post hoc, but took place only where a relevant complaint was made to, say, the Security Service Tribunal and then passed by it to the Security Service Commissioner. Here, review by a judicial figure on judicial review principles takes place in all cases and, more importantly, prior to the warrant coming into force and whatever interference with individual rights it authorises taking place. Even these practices, however, have an analogue in prior regimes: under RIPA, for example, certain intrusive surveillance authorisations under Part II of that Act take effect only when approved by one of the Surveillance Commissioners, who are to give approval only where satisfied that ‘there are reasonable grounds for believing’ that the requirements for the making of an authorisation (done in the first place usually by a senior police office) are satisfied.Footnote 134 Nevertheless, the prior, individualised (in that specific warrants are considered) and general (in that all except urgent warrants are so considered) review by the JCs in the context of the IPA marks a considerable evolution of practice in this area.
Two points about the double-lock arrangements stand out. The first is that the JC role is a subordinate one. Its practical significance depends in large part upon the approach taken to review and, in particular, how the 2016 Act's reference to ‘judicial review principles’ is interpreted and, indeed, whether it differs meaningfully from the role of Surveillance Commissioners in assessing – under previous legislation – whether there are ‘reasonable grounds to believe’ that the requirements are met. Much uncertainty on that point was evident during the process leading to the IPA's enactment. It was argued, primarily on the authority of Secretary of State for the Home Department v MB,Footnote 135 that the effect of the statutory reference to judicial review is to require JCs to independently satisfy themselves that the statutory requirements are satisfied. This conclusion, though attractive, is problematic – in MB, the Court of Appeal held that the Convention requirements could be satisfied only if ‘the court made its own independent assessment of whether the requirements for imposing the control order were satisfied’.Footnote 136 But challenges to the imposition of a control order engage Art 6 in a way which the granting of, say, an interception warrant does not, control orders being sufficiently restrictive as to make the procedural requirements of Art 5(4) of the Convention equivalent, in this particular case, to Art 6 in its criminal element. The Court of Human Rights has never determined that Art 6 applies to challenges to interception regimes (see eg Kennedy v United Kingdom, in which the point was explicitly left open)Footnote 137 but it seems very likely that – even if it does – Art 6 does not also apply to the original granting of such a warrant. As such, the correct approach to review would seem – on the face of the 2016 Act – to be the appropriate common law standard. That is, the correct approach would seem to be that of reasonableness, unmodified even by the now-standard references to the sliding scale of intensity or the references to common law proportionality, given that the common law, which until recently did not recognise a right to privacy,Footnote 138 is hardly likely to deem it a fundamental right of the sort which brings the proportionality standard into play.Footnote 139 If that analysis is correct, then the result would have been that the oversight by JCs was weaker than that carried out by Surveillance Commissioners – testing whether a decision was outwith the range of reasonable conclusions open to the decision-maker rather than (as the Surveillance Commissioners do) whether there are reasonable grounds for it. What would have resulted – it would seem – was not independent warranty, but rather a scheme in which the executive's role is given a veneer of independence by a process which in practice changes very little.
It is remarkable, therefore, that the IPCO – in advance of the double-lock provisions coming into force – published an ‘Advisory Notice’ on the ‘Approval of Warrants, Authorisations and Notices by Judicial Commissioners’.Footnote 140 The mere fact of the provision of advice is a step forward for this new hybrid institution as compared to its various predecessors; that the advice was directed not only at public authorities but also ‘to the general public’ marks a particularly important departure, given that the work of those predecessors was addressed very clearly to the former. The key paragraphs of the document are those which relate to the approach to be adopted in the double-lock process:
The purpose of the so-called ‘double lock’ provisions of the Act are to provide an independent, judicial, safeguard as to the legality of warrants, in particular to their necessity and proportionality. In cases engaging fundamental rights, the Judicial Commissioners will not therefore approach their task by asking whether a Secretary of State's decision that a warrant is necessary and proportionate is Wednesbury reasonable, as this would not provide the requisite independent safeguard.
This is the approach that is taken by domestic courts in judicial review cases when reviewing measures and decisions that interfere with fundamental rights under the Human Rights Act 1998 and when applying EU law, and a similar approach is adopted when considering interferences with common law rights. Since the Judicial Commissioners are required to adopt the same approach as would be applied by a court in judicial review proceedings, the Judicial Commissioners will adopt this approach in such cases.Footnote 141
This approach therefore interprets the double-lock requirement in a strong form, though falls short of an approach in which the JCs decide for themselves whether the issue of the warrant is necessary and proportionate, as evidenced by the assertion that ‘in deciding whether to approve a decision to issue a warrant, the Judicial Commissioners will ask themselves whether the Secretary of State's decision to issue a warrant’ is necessary and proportionate.Footnote 142 Even interpreted aggressively, therefore, the subordinate nature of the role – deriving from the clear statutory language – is inescapable. What is striking, however, is the lack of clarity in the Notice as to which cases will, and will not, engage fundamental rights – a question which is difficult enough where the rights in question are found in the Convention or EU law, but which cannot easily be answered in relation to the sorts of common law rights which are also referenced here. Though the answer will surely be that the vast majority of warrants engage one or the other of these sets of rights, the fact that the Notice appears to anticipate that some warrants will not will lead to a situation in which the nature of the oversight by the JCs varies according to subtle assessments of what each does or does not infringe. And so, while the Advisory Notice is welcome – and not only for its content – it leaves unanswered certain crucial questions.Footnote 143
The second point to be made about the role of the Commissioners under the IPA follows from this first one. Notwithstanding its limitations, there has been a shift in the form of oversight here. What is at issue remains a hybrid institution, but now with a dimension that is more akin to the ordinary judicial role, testing the legality of individual warrants in advance of their coming into force rather than merely considering (a sample of) them and related arrangements after the fact. This combination of direct involvement in the warrantry process and responsibility for oversight thereof was the source of considerable comment in the process leading to the Bill's enactment. In his report proposing the creation of the Independent Surveillance and Intelligence Commission, David Anderson QC concluded that it would not be difficult to combine the two functions, noting the precedent of the Surveillance Commissioners discussed above. This conclusion, however, rested on a separation between the JCs (who would issue warrants) and the inspectorate (who would inspect their implementation) which would permit a productive dialogue between the two.Footnote 144 The regime contained in the draft Investigatory Powers Bill was heavily criticised – the Interception of Communication Commissioner's Office said that if the perception was of the JC's ‘marking their own homework’ this would ‘dilute the credibility and independence of the new’ body.Footnote 145 It therefore deemed it necessary that those authorising warrants and those overseeing and auditing the process be operationally distinct.Footnote 146 An intention to that effect was confirmed by the then Home Secretary in evidence to the Joint Committee on the Draft Bill, whose report emphasised the need for there to be ‘a clear delineation of functions within the Judicial Commissioners in order to ensure public confidence in the independence and impartiality in the exercise of the Commissioners’ oversight functions’.Footnote 147
Though such delineation may well exist in practice, it is not required by – nor even apparent from – the 2016 Act, which assigns both sets of functions to the JCs as a group with no formal differentiation between those who carry out the prior authorisation and those responsible for ex post oversight. As with the role of the JCs in the double-lock system, which might easily have been made stronger than it would seem to be, the contingency of this differentiation – the fact that it will be effected by internal arrangements not prescribed by statute – undermines the role of the JCs in promoting public confidence in the regime of investigatory powers instituted by the 2016 Act. It would have been straightforward to task judges with authorising warrants as equal partners in a system with an element of genuine independence and to ensure, separately, a robust system of oversight. As it is, neither of those things has happened: the JCs role in the double-lock does not amount to genuine and independent oversight, but is nevertheless close enough to risk destroying trust in the broader oversight role of the IPC. What might have been a triumph of hybrid institutions within the national security constitution risks instead undermining their work.
Conclusion
The very existence of a national security constitution – the processes and practices by which accountability is secured in the national security context – depends heavily upon hybrid institutions. That is, the fact that accountability for action in this sphere is possible is a function of the existence of institutions which do what neither Parliament nor the courts could do, providing an ongoing oversight of the use of the relevant powers informed by access to the SIAs and the material which directs their work. Without the hybrid institutions, the picture would be far less edifying: despite the value of their work, neither the Intelligence and Security Committee of Parliament nor the various secret Tribunals have the capacity to provide an independent and informed overview of law and practice in sensitive areas. Together, the IRTL on one hand and, on the other, the various Commissioners which have existed or now exist, go a considerable distance to filling the gap, and ensuring that not only is Parliamentary scrutiny (carried out by those who are not members of the Privy Council and so cannot be trusted with the state's secrets) meaningful, but that the wider community has some method by which it might inform itself, should it choose, of the reality of a world which it can never directly experience and about which the law books are liable to mislead.
With the enactment of the Investigatory Powers Act 2016 – and the entry into force, bit by bit, of its provisions – a new phase in the development of the institutional facets of the national security constitution has begun. The IPC and the other JCs carry out a wider range of functions than their predecessors, and do so in the context of much broader public awareness of the national security operations of the state. For both reasons, the minimalist approach of past Commissioners to the information they provided is no longer tenable, and the institution more valuable as a result. Alongside this, the 2016 Act involves the Commissioners for the first time in the prior approval of warrants, though as the subordinate part of a double lock system rather than as the primary (or even sole) decision-maker. The effect is to assign to an institution which already marries legal and political dimensions a function which was traditionally in the constitution reserved to the executive. The ability of the IPC and JCs – now the premier hybrid institution in the national security constitution – to perform successfully the constitutional role assigned to them will be in large part a function of their ability to remain outside of the executive; to not allow their involvement in the process of granting warrants to undermine their work in holding to account, at the macro level, the regime of investigatory powers and those with primary responsibility for its operation. On this point, an incident early in the life of the IPC bodes well for its approach. As part of litigation before the Investigatory Powers Tribunal, there was revealed correspondence between GCHQ and Sir Adrian Fulford – who had earlier in 2017 been appointed as the first IPC – in which the GCHQ suggested to the IPC that the former develop a ‘process or protocol by which we… might better liaise with the [Investigatory Powers Commissioner's Office] to manage any circumstances where a piece of litigation… could raise issues in relation to oversight activity’.Footnote 148 The Commissioner's response to this suggestion was to note its inappropriateness (there being no conceivable situation where the work of the Commissioner ‘could be the subject of any form of prior agreement, however transparent, especially with a party subject to [the IPC's] oversight’), to reassert his independence (which he deemed ‘a crucial aspect of the work of my office’) and to provide a copy of both GCHQ's letter and his response to the IPT, in order to ensure that ‘the present litigation is conducted with appropriate transparency’.Footnote 149 If this response represents the general work of the IPC in years to come, it may be that – notwithstanding the risks inherent in the muddying of the Commissioner's constitutional role – the hybrid institutions of the national security constitution prove to be capable of remedying entirely the weaknesses, in this particular area of policy, of the traditional processes of legal and political accountability.
Author ORCIDs
Paul F Scott, 0000-0003-3933-7826
