Privacy International & Ors v Secretary of State for Foreign And Commonwealth Affairs & Ors

This is the judgment of the court.

The principal question – although not the only question – raised on this appeal is both important and difficult. Its importance is shown by stating it: Is a policy which the appellants say “authorises” officers of the Security Service (MI5) to run undercover agents who participate in the commission of criminal offences unlawful? Its difficulty is shown by the very fact that, on that question, two of the five members of the Investigatory Powers Tribunal (“the Tribunal”), whose decision is under appeal to this court, dissented from the majority view. That question ultimately has to be decided by reference to the provisions of the Security Service Act 1989 (“the 1989 Act”), on its correct interpretation. We note at this stage that the word “authorise” requires considerable qualification: as we will come on to explain. And we also emphasise at this stage that the Security Service is not, and has never been, above the law, and it has not sought to suggest otherwise.

Before us, the appellants (which are all non-governmental organisations) were represented, as below, by Mr Ben Jaffey QC and Ms Celia Rooney. The respondents were represented, as below, by Sir James Eadie QC, Mr David Perry QC, Ms Victoria Wakefield QC, Ms Natasha Barnes and Mr Will Hays. Also present, by video link, at the appeal hearing were Mr Angus McCullough QC and Mr Tim Buley QC as Special Advocates. The written and oral arguments presented to us were most helpful.

The Tribunal had been invited to consider, and had considered, a quantity of closed materials and had received submissions at a closed hearing. In due course it delivered a closed judgment as well as an open judgment. This court, in advance of the appeal hearing, had itself been invited to consider and had considered closed materials and the closed judgment of the Tribunal. Having done that, we agreed with the view of the parties, and as proposed by the Special Advocates, that the appeal itself could, at least at this stage, be dealt with on an open basis.

These proceedings were commenced in the Tribunal in June 2017. They have been substantially amended from time to time. In point of form, they are primarily directed at a policy document issued by the Security Service in March 2011 and reviewed (but not changed) in January 2014. The document is entitled “Guidelines on the Use of Agents who participate in Criminality – Official Guidance”. We will in this judgment call that policy “the Guidance”. The appellants only became aware of the Guidance and its terms (in redacted form) during the course of the proceedings. The relief currently sought is a Declaration that the respondents’ conduct is unlawful; an order quashing the Guidance; and an injunction restraining further unlawful conduct. Although the Guidance is a very important feature of the proceedings, as we will come on to say it is in fact to be considered as subordinate to the even more central point as to whether there is any power on the part of the Security Service to “authorise” agents’ participation in criminality.

In its judgment dated 17 December 2019, the Tribunal undertook an extensive review of the background to the Guidance. For present purposes, we think that we can take those aspects relatively shortly.

There has been a Security Service for very many years. Its creation and, until the 1989 Act, its functions and operations were, as is common ground, governed by the Royal Prerogative. From 1952 until 1989 the Security Service (and related Agencies) were required to conduct their operations in accordance with the Directive issued by Sir David Maxwell Fyfe, the then Home Secretary. This had been issued (but not made public at the time) on 24 September 1952. Amongst other things the Directive stated as follows:

“The Security Service is part of the Defence Forces of the country. Its task is the Defence of the Realm as a whole, from external and internal dangers arising from attempts at espionage and sabotage, or from actions of persons and organisations whether directed from within or without the country, which may be judged to be subversive to the State.

You will take special care to see that the work of the Security Service is strictly limited to what is necessary for the purposes of this task.”

In order to fulfil its task, the Security Service necessarily has had to run undercover agents. Such agents, who will usually not themselves be members of the Security Service, may need to participate in conduct which may or would be criminal or tortious in order to maintain their cover and in order better to ascertain the activities and intentions of the organisations in which they operate. Failure to do so might render their intended role ineffective and indeed might expose them to suspicion and personal danger. In modern jargon, such agents are styled “Covert Human Intelligence Sources” (CHIS); we will continue to style them “agents” in this judgment. There can be no dispute (and the Tribunal found as a fact) that such agents are indispensable to the work of the Security Service. They play a vital role in gathering intelligence with a view to protecting the State and the public from serious harm.

A limited degree of oversight of certain of the Security Service’s functions was undertaken by the Intelligence Services Commissioner under the provisions of the 1989 Act and the Regulation of Investigatory Powers Act 2000 (“RIPA”). By amendment to RIPA made in 2013, there was conferred power on the Prime Minister to direct the Intelligence Services Commissioner to keep under review the carrying out of any aspect of the functions of the Security Service (and also other Agencies). Such a direction was given to the Intelligence Services Commissioner (Sir Mark Waller) by the then Prime Minister, David Cameron, on 27 November 2014. It was not made public at the time. It was only revealed in the Tribunal proceedings in 2017.

From 1 September 2017, the Intelligence Services Commissioner was replaced by the Investigatory Powers Commissioner: s. 227 of the Investigatory Powers Act 2016. Section 230 of that Act contained a similar power to give directions: and such directions were given to the Investigatory Powers Commissioner by the then Prime Minister, Theresa May, on 22 August 2017. In the relevant respects that direction provided as follows:

“3.

The Investigatory Powers Commissioner shall keep under review the application of Security Service Guidelines on the use of agents who participate in criminality and the authorisations issued in accordance with them.”

Mrs May made a written statement on 1 March 2018 with regard to this direction (and also with regard to an entirely different direction).

What, then, were the “Security Service Guidelines” there referred to?

It appears that a guidance policy of this sort had first appeared in 1995 (or perhaps earlier). Thereafter, there were various “iterations”, as they have been styled. The current iteration of the Guidance was issued in March 2011. It was not then made public. The Director of Public Prosecutions was notified of it on 3 September 2012. Sir Mark Waller had himself been invited by Mr Cameron on 27 November 2012 to keep the application of the Guidance under review; and he agreed to do so. The terms of such invitation had, however, expressly excluded any suggestion that such oversight would provide endorsement of the legality of the Guidance or the provision of any view as to whether any particular case should be referred to the prosecuting authorities. This direction was then superseded by the direction to Sir Mark Waller of 27 November 2014, referred to above.

The terms of the Guidance have only been made public and revealed to the appellants (in redacted form) during the course of these proceedings. At the outset, the Guidance stated its Policy Aim in these terms:

“The aim of this policy is to provide guidance to agent-running sections on the use of agents who participate in criminality.” The (redacted) principles were to this effect:

“The guidance explains the circumstances in which agentrunning sections may use agents who participate in criminality and sets out relevant procedures.”

There are then 13 paragraphs to the Guidance. These (as redacted) provide as follows:

“1.

These Guidelines are intended to provide guidance to agentrunning sections on the use of agents who participate in criminality.

2.

Part II of the Regulation of Investigatory Powers Act 2000 ("RIPA") creates a regime for authorising the conduct and use of Covert Human Intelligence Sources ("CHISs"). This regime applies to the Service's use of agents, and the Service conducts its agent operations in accordance with RIPA, its subordinate legislation and the CHIS Code of Practice issued under it.

3.

RIPA does not provide any immunity from prosecution for agents or others who participate in crime. Section 27 of RIPA provides that conduct specifically authorised under a CHIS authorisation is “lawful for all purposes”, [redacted]

4.

Subject to this, neither RIPA nor the Code of Practice provides for CHISs to be authorised to participate in criminality. However, the Service has established its own procedure for authorising the use of agents participating in crime, which it operates in parallel with the RIPA authorisation [redacted]. [which governs the use and conduct of CHIS]

5.

[Redacted] the nature of the work of the Service is such that its agents are frequently tasked to report on sophisticated terrorist and other individuals and organisations whose activities may pose a threat to national security and/or involve the commission of serious offences. In those circumstances it may sometimes be necessary and proportionate for agents to participate in criminality in order to secure or maintain access to intelligence that can be used to save life or disrupt more serious criminality, or to ensure the agent's continued safety, security and ability to pass such intelligence.

Authorisation of use of participating agent

6.

An officer empowered to issue a CHIS authorisation under RIPA (an "authorising officer'') may in appropriate cases authorise the use of an agent participating in crime [redacted].

7.

[Redacted] the authorising officer may authorise the use of the agent [redacted] if [redacted].

a.

there is a real prospect that the agent will be able to provide information concerning serious crime [redacted].

b.

the required information cannot readily be obtained by any other means; and

c.

the need for the information that may be obtained by the use of the agent justifies his use notwithstanding the criminal activity in which the agent is or will be participating.

8.

The criterion at paragraph 7(c) is not satisfied unless the authorising officer is satisfied that the potential harm to the public interest from the criminal activity of the agent is outweighed by the benefit to the public interest from the information it is anticipated that the agent may provide and that the benefit is proportionate to the criminal activity in question.

Effect of an authorisation

9.

An authorisation of the use of a participating agent has no legal effect and does not confer on either the agent or those involved in the authorisation process any immunity from prosecution. Rather, the authorisation will be the Service's explanation and justification of its decisions should the criminal activity of the agent come under scrutiny by an external body, e.g. the police or prosecuting authorities. In particular, the authorisation process and associated records may form the basis of representations by the Service to the prosecuting authorities that prosecution is not in the public interest. Accordingly, any such authorisation should, on its face, clearly establish that the criteria for authorisation are met, in terms which will be readily understood by a prosecutor.

Procedure

[An authorisation form is completed which records the

Authorising Officer’s decision]

[Redacted]

11.

It is vital that full and accurate records are made of everything said to an agent on the subject of participation and of his response. [Redacted] it should be clearly explained to the agent that the authorisation process does not bestow on them any immunity from prosecution [redacted].

[MI5 officers involved in the authorisation process should understand that they may be called to account for their decisions and actions about the lawfulness of the agent’s conduct].

Commission of criminal offences not covered by an authorisation

[This paragraph explains what steps MI5’s officers should take in these circumstances]

12.

[Redacted].

Agent Handlers

13.

No member of the Service shall encourage, counsel or procure the commission by an agent of a criminal offence, save and to the extent that the offence is covered by an authorisation issued under these Guidelines.”

It is this Guidance which is the focus of challenge in the proceedings as pleaded. But the fundamental challenge has, during the course of the proceedings, shifted somewhat to examining the meaning and effect of the 1989 Act in terms of the powers (or lack of powers) which it confers.

The appellants, for understandable reasons, have also focused on certain commentaries on the participation of undercover agents in criminality.

Particular emphasis thus was placed before us, as before the Tribunal, on the report of Sir Desmond de Silva QC with regard to the infamous murder of Pat Finucane. Pat Finucane was a leading Northern Irish solicitor, much involved in the defence of those accused of terrorist activities. In February 1989, gunmen broke into his home and shot him dead. This was in the presence of his family; indeed his wife was herself injured. There were strong indications, referred to by David Cameron as Prime Minister in a statement made in 2012, of what was described as “state collusion” in the murder.

Sir Desmond de Silva had been asked to conduct a review of the case. His report was published in December 2012. Definitive conclusions were by no means always possible. But among other things he expressed “significant doubt” if the murder would have occurred without the involvement of “elements of the State”; and suggested that there were “positive actions by employees of the State” to further and facilitate the murder.

At paragraphs 112 and 113 of the Executive Summary of the report, moreover, this was said:

“112.

There are … some broad themes that may still have relevance to the world of intelligence-gathering. I have not concluded that the running of agents within terrorist groups is an illegitimate or unnecessary activity. On the contrary, it is clear that the proper use of such agents goes to the very heart of tackling terrorism. The principal lesson to be learned from my report, however, is that agent-running must be carried out within a rigorous framework. The system itself must be so structured as to ensure adequate oversight and accountability. Structures to ensure accountability are essential in cases where one organisation passes its intelligence to another organisation which then becomes responsible for its exploitation.

113.

It is essential that the involvement of agents in serious criminal offences can always be reviewed and investigated and that allegations of collusion with terrorist groups are rigorously pursued. Perhaps the most obvious and significant lesson of all, however, is that it should not take another 23 years to properly examine, unravel and publish a full account of collusion in the murder of a solicitor that took place in the United Kingdom.”

The appellants placed reliance on the Finucane case and its description in the de Silva report as a worked (albeit extreme) example, even if not entirely proven, of how their objections to the Guidance cannot be confined to the realm of theory and speculation. They also relied on the various other commentaries and debates (post-dating the 1989 Act) expressing concern about the position. Thus Sir John Chilcot, then Permanent Secretary in the Northern Ireland Office, was recorded in the de Silva report as having advised that the non-statutory position (with regard to the authorisation of agents’ criminality) was “unsatisfactory in practice and arguably unacceptable in principle” and that a “satisfactory way forward … could only … be achieved by new legislation”. That view seems to have had the support at the time of several (although by no means all) senior Ministers. For example, the then Attorney General apparently had refused to endorse the legality of any guideline “which appeared to condone in advance the commission of serious criminal acts”. But matters were not followed through by legislation.

Another example on which the appellants seek to rely is the fall-out from the activities of an agent known as “Stakeknife”, said to have been an agent within the IRA. Very serious criminality involving that agent and State personnel has been alleged. It has been the subject of police investigation under the name of “Operation Kenova”, as has been publicised. Nothing has yet been proved, we gather; but the appellants here too seek to rely on the current investigation as illustrative of how serious the potential situation could be with regard to alleged state-authorised participation in serious criminality.

These are, as we see it, cogent points but essentially forensic. The perceived need in some quarters for legislation cannot determine – or, indeed, ultimately be relevant to – the issue of what the 1989 Act, on its true interpretation, actually has (or has not) brought about. The same can be said about the current Covert Human Intelligence Sources (Criminal Conduct) Bill 2020, to which we were referred. That Bill was introduced on 24 September 2020. Its general format is that of substantial amendment of RIPA, by inserting various new sections. It is currently making its way through Parliament. We were told at the hearing that it is thought likely that an Act will be passed into law relatively soon. If so, that may, for the future, to a very considerable extent resolve on a statutory basis some of the issues and uncertainties thrown up by these proceedings (and by the previous debates). But very interesting and important though it is, the Bill cannot of itself, as we see it, legitimately be used to cast light on the meaning and effect of the 1989 Act.

We turn, then, to the provisions of the 1989 Act. They are central to this appeal. It can be taken that a significant motivation for the 1989 Act was the, at the time well known, “Spycatcher” litigation.

The long title to the 1989 Act provides as follows:

“An Act to place the Security Service on a statutory basis; to enable certain actions to be taken on the authority of warrants issued by the Secretary of State, with provision for the issue of such warrants to be kept under review by a Commissioner; to establish a procedure for the investigation by a Tribunal or, in some cases, by the Commissioner of complaints about the Service; and for connected purposes.”

The provisions of the 1989 Act, in its original form, are relatively brief. They comprise just seven sections, albeit with schedules. Sections 1 and 2 provide as follows:

“1.

(1) There shall continue to be a Security Service (in this Act referred to as “the Service”) under the authority of the Secretary of State.

(2)

The function of the Service shall be the protection of national security and, in particular, its protection against threats from espionage, terrorism and sabotage, from the activities of agents of foreign powers and from actions intended to overthrow or undermine parliamentary democracy by political, industrial or violent means.

(3)

It shall also be the function of the Service to safeguard the economic well-being of the United Kingdom against threats posed by the actions or intentions of persons outside the British Islands.

2.

(1) The operations of the Service shall continue to be under the control of a Director-General appointed by the Secretary of State.

(2)

The Director-General shall be responsible for the efficiency of the Service and it shall be his duty to ensure—

(a)

that there are arrangements for securing that no information is obtained by the Service except so far as necessary for the proper discharge of its functions or disclosed by it except so far as necessary for that purpose or for the purpose of preventing or detecting serious crime; and

(b)

that the Service does not take any action to further the interests of any political party.

3.

The arrangements mentioned in subsection (2) (a) above shall be such as to ensure that information in the possession of the Service is not disclosed for use in determining whether a person should be employed, or continue to be employed, by any person, or in any office or capacity, except in accordance with provisions in that behalf approved by the Secretary of State.

4.

The Director-General shall make an annual report on the work of the Service to the Prime Minister and the Secretary of State and may at any time report to either of them on any matter relating to its work.”

Section 3 makes provision for warrants to be issued by the Secretary of State, authorising what otherwise would be unlawful entrance on or interference with property. It provides, in the relevant respects, as follows:

“(1)

No entry on or interference with property shall be unlawful if it is authorised by a warrant issued by the Secretary of State under this section.

2.

The Secretary of State may on an application made by the Service issue a warrant under this section authorising the taking of such action as is specified in the warrant in respect of any property so specified if the Secretary of State—

(a)

thinks it necessary for the action to be taken in order to obtain information which—

(i)

is likely to be of substantial value in assisting the Service to discharge any of its functions; and

(ii)

cannot reasonably be obtained by other means; and

(b)

is satisfied that satisfactory arrangements are in force under section 2(2)(a) above with respect to the disclosure of information obtained by virtue of this section and that the information obtained under the warrant will be subject to those arrangements.

3.

A warrant shall not be issued under this section except— (a) under the hand of the Secretary of State; or

(b)

in an urgent case where the Secretary of State has expressly authorised its issue and a statement of that fact is endorsed on it, under the hand of an official of his department of or above Grade 3.

Subsequently, the system of warrants provided in s. 3 of the 1989 Act was itself replaced by s. 5 of the Intelligence Services Act 1994 (“the 1994 Act”). The 1994 Act also put on a statutory basis the Secret Intelligence Service (that is, MI6) and the Government Communications Headquarters. The same drafting technique as used in the 1989 Act (“shall continue”) was used for this purpose in s. 1 of the 1994 Act.

The appellants placed some reliance on the terms of s. 7 of the 1994 Act. That provides, in part, as follows:

“(1)

If, apart from this section, a person would be liable in the

United Kingdom for any act done outside the British Islands, he

shall not be so liable if the act is one which is authorised to be done by virtue of an authorisation given by the Secretary of State under this section.

2.

In subsection (1) above “liable in the United Kingdom” means liable under the criminal or civil law of any part of the United Kingdom.”

We were also referred to, and again the appellants placed some reliance on, various sections of RIPA, in particular sections 26 and 27 of RIPA. (RIPA, we note, had repealed the Interception of Communications Act 1985).

“26.

Conduct to which Part II applies.

(1)

This Part applies to the following conduct—

(a)

directed surveillance;

(b)

intrusive surveillance; and

(c)

the conduct and use of covert human intelligence sources.

3.

Subject to subsection (6), surveillance is directed for the purposes of this Part if it is covert but not intrusive and is undertaken—

(a)

for the purposes of a specific investigation or a specific operation;

(b)

in such a manner as is likely to result in the obtaining of private information about a person (whether or not one specifically identified for the purposes of the investigation or operation); and

(c)

otherwise than by way of an immediate response to events or circumstances the nature of which is such that it would not be reasonably practicable for an authorisation under this Part to be sought for the carrying out of the surveillance.

4.

Subject to subsections (4) to (6), surveillance is intrusive for the purposes of this Part if, and only if, it is covert surveillance that—

(a)

is carried out in relation to anything taking place on any residential premises or in any private vehicle; and

(b)involves the presence of an individual on the premises or in the vehicle or is carried out by means of a surveillance device.

(4)

For the purposes of this Part surveillance is not intrusive to the extent that—

(a)

it is carried out by means only of a surveillance device designed or adapted principally for the purpose of providing information about the location of a vehicle; or

(b)

it is surveillance consisting in any such interception of a communication as falls within section 48(4).

(5)

For the purposes of this Part surveillance which—

(a)

is carried out by means of a surveillance device in relation to anything taking place on any residential premises or in any private vehicle, but

(b)

is carried out without that device being present on the premises or in the vehicle,

is not intrusive unless the device is such that it consistently provides information of the same quality and detail as might be expected to be obtained from a device actually present on the premises or in the vehicle.

(6)

For the purposes of this Part surveillance which—

(a)

is carried out by means of apparatus designed or adapted for the purpose of detecting the installation or use in any residential or other premises of a television receiver (within the meaning of Part 4 of the Communications Act 2003), and

(b)

is carried out from outside those premises exclusively for that purpose, is neither directed nor intrusive.

(7)

In this Part—

(a)

references to the conduct of a covert human intelligence source are references to any conduct of such a source which falls within any of paragraphs (a) to (c) of subsection (8), or is incidental to anything falling within any of those paragraphs; and

(b)

references to the use of a covert human intelligence source are references to inducing, asking or assisting a person to engage in the conduct of such a source, or to obtain information by means of the conduct of such a source.

(8)

For the purposes of this Part a person is a covert human intelligence source if—

(a)

he establishes or maintains a personal or other relationship with a person for the covert purpose of facilitating the doing of anything falling within paragraph (b) or (c);

(b)

he covertly uses such a relationship to obtain information or to provide access to any information to another person; or

(c)

he covertly discloses information obtained by the use of such a relationship, or as a consequence of the existence of such a relationship.

(9)

For the purposes of this section—

(a)

surveillance is covert if, and only if, it is carried out in a manner that is calculated to ensure that persons who are subject to the surveillance are unaware that it is or may be taking place;

(b)

a purpose is covert, in relation to the establishment or maintenance of a personal or other relationship, if and only if the relationship is conducted in a manner that is calculated to ensure that one of the parties to the relationship is unaware of the purpose; and

(c)

a relationship is used covertly, and information obtained as mentioned in subsection (8)(c) is disclosed covertly, if and only if it is used or, as the case may be, disclosed in a manner that is calculated to ensure that one of the parties to the relationship is unaware of the use or disclosure in question.

(10)

In this section “private information”, in relation to a person, includes any information relating to his private or family life.

(11)

References in this section, in relation to a vehicle, to the presence of a surveillance device in the vehicle include references to its being located on or under the vehicle and also include references to its being attached to it.

27.

Lawful surveillance etc.

(1)

Conduct to which this Part applies shall be lawful for all purposes if—

(a)

an authorisation under this Part confers an entitlement to engage in that conduct on the person whose conduct it is; and

(b)

his conduct is in accordance with the authorisation.

(2)

A person shall not be subject to any civil liability in respect of any conduct of his which—

(a)

is incidental to any conduct that is lawful by virtue of subsection (1); and

(b)

is not itself conduct an authorisation or warrant for which is capable of being granted under a relevant enactment and might reasonably have been expected to have been sought in the case in question.

(3)

The conduct that may be authorised under this Part includes conduct outside the United Kingdom.

(4)

In this section “relevant enactment” means—

(a)

an enactment contained in this Act or the Investigatory Powers Act 2016;

(b)

section 5 of the Intelligence Services Act 1994 (warrants for the intelligence services); or

(c)

an enactment contained in Part III of the Police Act 1997 (powers of the police and of officers of Revenue and Customs).”

Consequently, the scheme of authorisation envisaged by RIPA is effectively delimited to the situations specified in s. 26.

Then, by s. 29 of RIPA, there is provision (extending to England and Wales, Scotland and Northern Ireland) for the designated person to “have power to grant authorisation for the conduct or the use of a covert human intelligence source.” That section then sets out the requirements before any such authorisation may be given.

On one view, these provisions of RIPA could be said to have a potentially significant bearing on the issues thrown up in the present case. The context of RIPA at all events suggests that one particular concern was to address situations which might raise issues of potential breaches of article 8 rights under the European Convention on Human Rights (“the Convention”). In the event, it was common ground before the Tribunal, as it records in its judgment, that by reason of s. 80 of RIPA (the terms of which we do not need here to set out) the source of any power, if it exists, for the Security Service to act as it does under the Guidance is not to be found in RIPA. That was accepted by the Tribunal and has been the basis on which the appeal was argued before this court. It has therefore not been suggested that the provisions of RIPA can or do operate in any way as some form of implied repeal of the 1989 Act or of any other pre-existing power (if any) to “authorise” criminal conduct. We will, therefore, ourselves proceed on that basis.

Put shortly, the appellants’ case on the principal issue is that there was and is no legitimate source for the power asserted by the respondents; and accordingly the Guidance, and conduct undertaken purportedly pursuant to the Guidance, is ultra

Since the resolution of this issue ultimately depends on the interpretation of the 1989 Act, we enquired whether there were any Explanatory Notes or statements in the Parliamentary debates which might cast light on the matter. We were informed that there were none. Mr Jaffey did refer us to a limited selection of statements taken from Hansard in the debates on the 1994 Act. But not only does that Act postdate the 1989 Act but also we found those (selected) statements really too broad to be of any material assistance.

The Tribunal was very strongly constituted, comprising as it did Singh LJ (President), Lord Boyd of Duncansby (Vice-President), Sir Richard McLaughlin, Mr Charles Flint QC and Professor Graham Zellick QC.

The points ultimately advanced by the appellants before the Tribunal were seven in number. They were summarised by the Tribunal as follows (the Tribunal taking the Guidance as constituting the policy):

“(1)

There is no lawful basis for the policy, either in statute or at common law.

(2)

The policy amounts to an unlawful de facto power to dispense with the criminal law.

(3)

The secret nature of the policy, both in the past and now, means that it is unlawful under domestic principles of public law.

(4)

For the purposes of the ECHR, the policy was not and is not “in accordance with law”.

(5)

Any deprivation of liberty effected pursuant to a purported authorisation given under the policy violates the procedural rights under Article 5 of the ECHR.

(6)

Supervision of the operation of the policy by the Intelligence Services Commissioner (“ISC”) in the past, and now the Investigatory Powers Commissioner (“IPC”), does not satisfy the positive investigative duty imposed by Articles 2, 3 and 5 of the ECHR.

(7)

Conduct authorised under the policy in breach of Articles 2, 3, 5 and 6 of the ECHR is in breach of the negative and preventative obligations in the ECHR. It is submitted that the policy itself is unlawful to the extent that it sanctions or acquiesces in such conduct.”

The majority found in favour of the respondents on all issues. The minority (Mr Flint QC and Professor Zellick QC) disagreed on the first issue (and, in consequence, on the fourth issue). The minority agreed with the majority on all the remaining issues.

Permission to appeal was granted by the Tribunal itself in respect of the first, fourth, fifth, sixth and seventh issues. Permission to appeal was then granted by Dingemans LJ on the second issue on 11 May 2020. Permission to appeal was refused by

The judgment of the majority

In dealing with the first issue, the majority of the Tribunal noted the acceptance on behalf of the respondents that the power to do what the Security Service does arises, if at all, from the 1989 Act and nothing else. The submission of the respondents was that the power was to be found in s. 1 of the 1989 Act as a matter of necessary implication. The argument on behalf of the appellants, on the other hand, was that no such power could be implied into the 1989 Act. Express words, it was said, would have been needed to achieve the result contended for; and the proposed implication was neither necessary nor in accordance with the principle of legality. Those remain the essential positions of the parties on this appeal, on this issue.

Having outlined the arguments, the majority said this at paragraph 52 of the judgment:

“What became clear during the course of the hearing before this Tribunal is that this fundamental issue of principle does not depend upon there being a policy at all. Although these proceedings have arisen in the form of a challenge to the policy set out in the Guidelines, Mr Jaffey acknowledged at the hearing that, if his submissions are correct, the Security Service has no power to undertake the activities in question at all. Even if it had no policy on the subject, it would still lack such power. Furthermore, in answer to questions from the Tribunal, Mr Jaffey made it clear that his submissions would have the consequence, if correct, that the Security Service would not in practice be able to “run” an agent who is embedded within a proscribed organisation at all. This is because inherent in such an operation would be the fact that a proscribed organisation is banned by the criminal law of this country.”

It went on, at paragraph 54 of its judgment, as follows:

“If the Security Service runs an agent in a proscribed organisation, and in particular if it embeds that agent into the organisation, it necessarily knows that that person must be a member of an illegal organisation and he may thereby be committing a criminal offence without more. Mr Jaffey submitted that this would not prevent the Security Service from running agents in other circumstances. Nevertheless, in our view, this would strike at the core activities of the Security Service. If that is the result on the correct interpretation of the

After reviewing various legal authorities, the majority went on to accept the submissions made on behalf of the respondents on this issue. In essence, the majority gave four reasons, which we summarise as follows:

The conclusion of the majority on this issue was expressed in these terms at paragraph 67 of the judgment:

“We conclude on the first issue that the Security Service does have that power as a matter of public law. It is important to appreciate that this does not mean that it has any power to confer immunity from liability under either the criminal law or the civil law (e.g. the law of tort) on either its own officers or on agents handled by them. It does not purport to confer any such immunity and has no power to do so.”

The majority went on, at paragraph 70, to express the respectful view that the judgments of the minority “fail sufficiently to draw the distinction between a power

On the second issue, on which the Tribunal was unanimous, the essence of the reasoning was that the Guidance did not purport to confer any immunity from criminal prosecution on anyone and in fact required that to be made clear to agents. There was no de jure immunity. As to the assertions of some kind of de facto immunity, that involved speculation as to what the outcome might be in any individual case and also involved an incorrect argument that the Guidance in some way cut across the constitutional independence of the prosecuting authorities, whether in England and Wales or in Scotland or in Northern Ireland.

On the fourth issue (we need not discuss the third issue), the Guidance was, given the conclusion of the majority, in accordance with domestic law, as being authorised by the 1989 Act. To the extent that the appellants had sought to argue that there must in law be sufficient oversight and safeguards with regard to a discretionary power vested in the Executive, it was further held that the oversight powers conferred by the legislation on the Intelligence Services Commissioner, and subsequently the Investigatory Powers Commissioner, did provide sufficient oversight and safeguards. Further, the validity of any relevant policy in this respect was in principle subject to review by tribunals or courts.

The Tribunal took the remaining issues, raised by reference to the Convention, together. It firmly rejected them. It was not disputed that the Security Service was a “public authority” subject to s. 6 of the Human Rights Act 1998. But there was, it held, nothing inherent in the Guidance to create a significant risk of any breach of Article 2 or Article 3 or of any other Convention rights; and in any event the position could not be properly addressed generically but only in the light of the specific facts of an individual case. Indeed, for like reasons, the appellants did not even have standing to raise human rights arguments in reliance on the Convention in this way: none of the appellants could properly be classed as a person “who is (or would be) a victim” for the purposes of s.7 of the 1998 Act.

The judgments of the minority

Mr Flint QC and Professor Zellick QC each delivered a judgment. Professor Zellick QC expressed agreement with the reasoning of Mr Flint QC but delivered a lengthy judgment of his own, expressing his own reasoning.

In his judgment, Mr Flint QC accepted “the operational necessity for the Service to run agents who may need to participate in serious criminal activity”. He described that operational necessity as “very clear”. His concern was that if the requisite power was to be found in the 1989 Act by way of implication (as the respondents argued) then there were no limits on that power set out in the statute. He further reasoned that the express insertion of the provisions of s. 3 with regard to warrants told against the implication of the power contended for. He also considered that fundamental rights could not be overridden by general or ambiguous words. He considered that the case of Morris v Beardmore [1981] AC 446 (a case not cited by any party in argument before the Tribunal) was relevant in this respect. He considered that that case also, in particular, established the proposition that the fact that the person who exercises a power to do an act may be liable to prosecution did not make the act lawful; and that was the answer to the respondents’ arguments that the implied power did not involve a dispensation from criminal investigation and prosecution.

He further took the view that such a conclusion was consistent with the 1994 Act (and in particular the express provisions of s. 7 of that Act) and with the provisions of s. 26 and s. 27 of RIPA. He in fact considered that the Guidance (and its predecessors) involved “parallel procedures outside the scope of the restraints and safeguards provided by RIPA”.

In expressing his view that the 1989 Act provided no legal basis for the Guidance, Mr Flint QC further said:

“An implied power which authorises conduct contrary to the general criminal and civil law but leaves the person engaging in such conduct liable to criminal prosecution would be extraordinary”.

He stated at paragraph 130, by way of conclusion:

“The policy under challenge has been exercised with scrupulous care by the Security Service so as to discharge its essential functions in protecting national security, whilst giving proper regard to the human rights of persons who may be affected by the activities of agents. But I am unable to find that the 1989 Act provides any legal basis for the policy under challenge…..”

He went on to refer to the need for new legislation.

In the course of his forthright judgment, Professor Zellick QC said of agents “… I readily accept that such [criminal] participation is a necessary and inescapable feature of their use”. (He later described it as “essential.”) Nevertheless, he regarded the central propositions of the respondents’ arguments (and, by extension, the reasoning of the majority) as “untenable”. His view was that the Royal Prerogative had not, before 1989, provided the necessary power; that s. 1 of the 1989 Act was not itself the source of any of the Security Service’s powers; and in any event the general wording of the 1989 Act could not be read so as to imply such a power: “It would require language much more specific to indicate Parliament’s intention to confer a power touching, as the policy here does, the principle of legality and the Rule of Law.”

He went on to elaborate his view that s. 1 of the 1989 Act conferred no powers at all, but simply “sets out functions or purposes – it might be called an objects clause”. He said (at paragraph 153 of his judgment):

“If the argument of the Respondents as to section 1(2) is correct, MI5 would have been able to rely on it to justify the property interference authorised by section 3 if there had been no section 3. Can that possibly be correct? Where does it end? What other powers does MI5 have as a result of the section?”

He described the respondents’ arguments as “fallacious”. As to their reliance on a preexisting prerogative power he said this:

“Had there been such a prerogative power, an argument that Parliament could not have intended to abrogate it would have had some cogency, but if it was not lawful before the Act, as I judge to be the case, it could be lawful afterwards only if the Act specifically said so (as it did with section 3).” He further stated this at paragraph 161 of his judgment:

“The simple if awkward fact is that, in accordance with the

Government’s wishes, Parliament gave no attention to this aspect of MI5’s work and the Government was content for it to remain under wraps and accept whatever legal or political risks it entailed. To attribute an intention to Parliament in these circumstances is fanciful.”

He considered the contents of the de Silva Report to provide “irrefutable support” for this view.

Dealing further with the arguments on necessary implication and various of the legal authorities, he then said this at paragraph 174:

“I do not see how the Respondents’ argument satisfies the requirements set out in these authorities. Even if the power in question here does not seek to override a fundamental human right such as LPP material, it is plainly a matter of considerable legal importance such that specific language is required. The power may well be sensible and desirable, even essential, but Parliament would, I fancy, be astonished to be told that it had conferred this power in 1989.”

He went on to state that to use the doctrine of necessary implication to imply a power into a section which was “wholly silent about powers” was unprecedented. As to the arguments on immunity, he stated that that was an “entirely separate issue”. He stated (reflecting the position of Mr Flint QC) that:

“A power to condone or permit the commission of crimes cannot acquire the quality of legality because those breaking the law in compliance with the authorisation may subsequently be prosecuted.”

As to the argument based on agents necessarily working within proscribed organisations, he said that was “not relevant” to whether the power existed. He went on to say this at the end of paragraph 179 of his judgment:

“If the MI5 policy did no more than turn a blind eye to offences in relation to proscribed organisation, I would have had no hesitation in denying the Claimants the declaration sought. There is no direct impact on the legal rights of others; and it would be an exaggeration to describe such a policy as subversive of the Rule of Law.”

We have set out a summary (and we stress it is only a summary) of the respective judgments in the Tribunal as we think it indicates the nature and parameters of the competing views on the central issue of vires: as well as indicating the basis for the (unanimous) reasoning of the Tribunal on the other issues.

We turn to the first ground of appeal in short, whether the Security Service has the legal power (vires) to run agents who participate in criminality. That, as we assess matters, is really the critical issue on this appeal.

We are of the firm view the decision of the majority is correct in law. We agree with their conclusion.

Some points can, in our opinion, be made at the outset.

Proposed legislation is, as it happens, now to be found in the Covert Human Intelligence Sources (Criminal Conduct) Bill, to which we have referred above. But, as we have indicated, we do not think that the fact that there is now such a Bill can be used to determine the issue of statutory interpretation of the 1989 Act.

We should perhaps make clear that, as Mr Jaffey informed us, it is not the case that the appellants take a position on the substantive issues being addressed by the latest Bill. Their concern and aim has always, as we were told, been quite different: to ensure that such sensitive matters are exposed to proper public debate and scrutiny. Indeed, the appellants have by these proceedings achieved the publicising (albeit in redacted form) of the Guidance; and, moreover, there is now the public debate and scrutiny being accorded to the Bill as it currently makes its way through Parliament.

One thing, at least, is uncontroversial. Clearly the use of agents is not of itself in any way unlawful. But further, even where ostensible criminality on the part of an agent is involved – whether in the form of membership of a proscribed organisation or in the form of participating in other criminal activities undertaken by that organisation or both – it does not necessarily follow that the conduct of the agent, or instructing handler, actually is necessarily criminal.

This is because, for example, the necessary mental element for the postulated crime or attempted crime may well be lacking. Or there may be a potential defence: for example, self-defence (including defence of another or of property) or duress or necessity. Moreover, in the context of crimes of violence, there may, for instance, be a potentially available defence under s. 3 of the Criminal Law Act 1967. That in the relevant respects provides:

“(1)

A person may use such force as is reasonable in the circumstances in the prevention of crime, or in effecting or assisting in the lawful arrest of offenders or suspected offenders or of persons unlawfully at large.”

We were given a most interesting resumé by Mr Perry of defences which may be run by way of justification and defences which may be run by way of excuse. There is a characteristically lucid discussion of the conceptual position in Smith, Hogan and Ormerod’s Criminal Law (15^th ed.) at pages 278-280. We need not for present purposes further explore these aspects. But the reality is, and as the majority in the Tribunal noted in paragraph 64 of the judgment, that the Guidance must be taken to extend to the use of agents who participate in possible criminality. We consider this very important. And Mr Jaffey expressly accepted before us (in our view, rightly) that it would be lawful for authorisation under the Guidance to be given to cases which only may involve criminality. This, surely, must potentially be a large class of cases. At all events, on the appellants’ arguments the focus then is to be confined to cases which do, indeed, necessarily involve criminality: either as correctly so assessed at the time or (even if not so assessed at the time) if it turns out to be so in the event of a subsequent conviction.

But, that said, the further reality still has to be confronted: that there will be (and will have been) some situations where agents will commit (and will have committed) a crime. Membership of a proscribed organisation is one convenient and obvious potential example; but of course there may be others. And in such situations, the handler – no doubt in part motivated by a need for the agent’s cover to be maintained and/or the credibility of the agent to be enhanced – may have given authorisation under the Guidance for that. Mr Jaffey was at all events entitled to point out that the Guidance itself is in terms directed at participation in crime: not just possible crime.

There can be no doubt that such a situation existed before the 1989 Act. In such circumstances, we agree with the majority that such a power would, if lawful before the 1989 Act, “continue” in the language of s. 1 (1) and s. 2 (1). We also agree with the majority that the availability of such a power is consistent – and necessarily consistent – with the “efficient” running of the Security Service for the purposes of s. 2 (2): for efficiency must surely include effectiveness.

We thus cannot, with respect, accept the view of Professor Zellick QC that s. 1 of the 1989 Act is only concerned with functions and purposes and simply defines the limits or scope of the Security Service’s activities: in short, that s. 1 (2) was a limiting provision. The 1989 Act is silent about and says nothing expressly about powers . However, since the Security Service’s functions could not be achieved without its having the necessary powers, it follows by necessary implication, in our view, that the 1989 Act is confirming the continuance of the powers which the Security Service previously had, in order to fulfil the functions now specified in s. 1 (2) and (3) or (in the case of the Director-General) in s. 2. The functions of the Security Service being similar to those before the 1989 Act, by necessary implication the powers to carry out those functions continued and remained the same.

This conclusion is also consistent with authority, if authority be needed. Thus in the case of Hazell v Hammersmith and Fulham London Borough Council [1993] 2 AC 1 Lord Templeman said at p.29 E that:

“…the word “functions” embraces all the duties and powers of a local authority: the sum total of the activities Parliament has entrusted to it. Those activities are its functions ….”

It is true that he was, in that case, talking in terms of a statutory body in the context of particular statutory local government provisions. But those observations seem to us to be directly in point for present purposes. They also accord with the statements of Laws LJ in the Court of Appeal in R (A) v Director of Establishments of the Security Service [2009] EWCA Civ 24, [2010] 3 All ER 416, where at paragraph 28 of his judgment he stated:

“All the functions of the Security Service are and have been since the coming into force of the Security Service Act 1989, statutory functions”

Laws LJ plainly contemplated that the Security Service (and Director-General) had powers incidental to those functions.

We also, with respect, do not agree with the suggestion made by Professor Zellick QC, essentially basing himself on the discussion in the de Silva report, that Parliament gave “no attention to this aspect of MI5’s work” and that “to attribute an intention to Parliament in these circumstances is fanciful”. We take the contrary view. It is not at all fanciful to accept that Parliament indeed intended the Security Service to retain, subject to the rule of law, its (operationally necessary) powers in this regard which it was exercising – as it assuredly was – before 1989; and it is most surprising to suggest, if that is the suggestion, that Parliament was intending to remove those (operationally necessary) powers from the Security Service, still less without express words.

On behalf of the respondents, Sir James accepted before us, as he had below, that the respondents’ case rested on implication into the 1989 Act. He also accepted, as he had below, that the relevant test was one of necessity.

We were referred to a number of authorities in this respect; but will ourselves refer to just two.

Thus in R (Morgan Grenfell & Co Ltd) v Special Commissioner of Income Tax [2002] UKHL 21, [2003] 1 AC 563 the issue was whether a statute by implication could compel disclosure of documents covered by legal professional privilege. It was held that it could not. A “fundamental right” such as legal professional privilege could only be overridden by express words or by necessary implication. Lord Hobhouse in the course of his concurring speech said this at paragraph 45:

“It is accepted that the statute does not contain any express words that abrogate the taxpayer's common law right to rely upon legal professional privilege. The question therefore becomes whether there is a necessary implication to that effect. A necessary implication is not the same as a reasonable implication as was pointed out by Lord Hutton in B (A Minor) v Director of Public Prosecutions [2000] 2 AC 428, 481 . A necessary implication is one which necessarily follows from the express provisions of the statute construed in their context. It distinguishes between what it would have been sensible or reasonable for Parliament to have included or what Parliament would, if it had thought about it, probably have included and what it is clear that the express language of the statute shows that the statute must have included. A necessary implication is a matter of express language and logic not interpretation.”

The last sentence of that paragraph perhaps is rather hard to follow. The rest is clear.

This approach was adopted, and in one important respect expanded, by the Supreme Court in R (Black) v Justice Secretary [2017] UKSC 81, [2018] AC 215. At paragraph 36 of her judgment, with which the other Justices of the Supreme Court agreed, Lady Hale stated, uncontroversially, that the goal of all statutory interpretation was to discover the intention of the legislation. She went on:

Having regard both to the context and to the purpose of the 1989 Act, and having regard to the operational necessity for agents to participate in criminality, it must follow, in our judgment, as a matter of necessary implication that the Security Service under the 1989 Act was intended to retain the (essential) power to instruct agents to participate in criminality, whether potential or actual. The key question then would be whether that could be so if the instruction could not lawfully be given. Certainly, as we accept, if to do so before 1989 was ultra vires then nothing in the provisions of the 1989 Act itself changed that thereafter. It was that particular point that caused Mr Flint QC (entirely) and Professor Zellick QC (primarily) to disagree with the majority.

Here, the respondents rely on a power which they say was, prior to 1989, available under the Royal Prerogative (the Security Service being an emanation of the Crown for this purpose). They also emphasise that a “power” is to be distinguished from an “immunity”, citing (as was cited below) Hohfeld’s Fundamental Legal Conceptions as applied in Judicial Reasoning (1919). Hohfeld suggested in this respect that a power is to be regarded as the legal ability to do something. But, Mr Jaffey then objects, there is no legal ability of the Security Service to authorise the committal of a crime (or, indeed, any otherwise unlawful act). There could, he said, be no such power under the Royal Prerogative; and, he said, the 1989 Act does not itself expressly confer such a power. He goes on to submit that that is reinforced by the provisions of s. 3 of the 1989 Act. For that has expressly provided that no entry into or interference with property shall be unlawful if authorised by warrant issued by the Secretary of State. Thus, it was accepted by Parliament that such invasions of property needed statutorily sanctioned authorisation. But, his argument goes on, that was not done in the case of authorisation of criminal participation.

In common with the majority of the Tribunal, we do not accept these arguments.

It seems to us, as it seemed to the majority, that it is critical for these purposes to distinguish carefully (and as Hohfeld explains) between a power and an immunity. We agree with the majority of the Tribunal that the judgments of the minority fail sufficiently to draw this fundamental distinction.

We also think it critical to bear in mind that the Guidance in effect connotes a twostage process. The first is directed at what the handler is empowered to do. If a handler acts entirely in accordance with the Guidance then, in his or her capacity as an officer of the Security Service, he or she is empowered, as part of his or her duties, to run an agent in this way. The second stage is then directed at what the handler may instruct the agent to do. It is at this particular stage that the word “authorise” needs to be treated with caution, in our opinion. This is because the word “authorise” can frequently be taken as connoting the conferring of legitimacy on conduct which otherwise would not be legitimate. But that is not so in the present context. It is not so because the Guidance is specific that such “authorisation” has no legal effect and confers no immunity from prosecution.

It is evident that the 1989 Act does not itself purport in any way to grant general immunity in respect of participation in criminality. Since, as we have emphasised, the Guidance itself expressly states that an authorisation of an agent participating in criminality has no legal effect and does not confer, either on the agent or on those involved in the authorisation process, any immunity from prosecution, it follows that, in our view, it cannot properly be said that the 1989 Act or the Guidance seek to place the Security Service and its officers and agents above the criminal (or other) law. It cannot be said that they involve a suspension of or dispensation with the rule of law. Rather, as we see it, they are an endorsement of it. And it further follows that, since there has been no suspension of or dispensation with the law, Mr Jaffey’s rather hopeful reliance on the Bill of Rights 1689 also falls away.

This also, we consider, is the explanation for and meaning of s. 3 of the 1989 Act. Where the provisions of that section are complied with, what would otherwise be an unlawful breaking in or interference with property are declared by statute not to be unlawful. Immunity from a charge or claim in such a situation is thereby statutorily conferred. But it does not follow at all from that that exercises of other powers by the Security Service in a context where criminality or other unlawfulness may be involved is prohibited. On the contrary, all those (necessary) powers continue: but in circumstances where the 1989 Act has not conferred immunity. The same can be said for provisions such as s. 7 of the 1994 Act (to the extent that subsequent legislation can be relied upon at all in construing the 1989 Act). Those are to be regarded as provisions governing situations where Parliament has elected to grant immunity. But in the exercise of powers otherwise falling within s. 1 of the 1989 Act no immunity is conferred.

As to the position before 1989, we do not in any event accept the premise of the judgments of the minority that the commission of an unlawful (and a fortiori criminal) act, even where liability to sanction in accordance with law in respect of such an act is retained, can never be within the ambit of the Royal Prerogative.

This is not the time or place for a lengthy exegesis on so difficult, elusive and abstruse a concept as the Royal Prerogative. But it may at least be noted that in the famous case of Attorney-General v De Keyser’s Royal Hotel Ltd [1920] AC 508, it was accepted that what would be an unlawful trespass could be justified as an exercise of the Royal Prerogative in the case of necessity, in circumstances of defence of the realm in time of war. The issue then was, in that case, whether there was an obligation to compensate the land-owner in circumstances of there being relevant concurrent legislation.

Burmah Oil Co. Ltd v Lord Advocate [1965] AC 75 again involved a claimed exercise of the Royal Prerogative in time of war. In that case, oil refineries and pipelines belonging to the claimants had been deliberately destroyed, without their consent, by the military forces in Burma during the Second World War in accordance with Government Policy, in order to deny such resources to the advancing enemy. The issue, again, was whether compensation was payable. It is quite true that the case was so decided in the context of the acts being done in time of war; moreover the central issue (as in De Keyser’s Hotel) was whether compensation was payable. But it is to be

National security threats are, no doubt, not to be equated precisely with threats arising in time of war: although the carnage and destruction when such threats to national security achieve actuality – as evidenced during the time of “the Troubles” in Northern Ireland or by more recent outrages such as have occurred in London and Manchester – at least are strongly illustrative of parallels. Certainly, as we see it, national security threats can, by their very nature, pose a threat to the stability of the State as well as a threat to life. Such considerations, indeed, find reflection in the language of s.1 (2) of the 1989 Act. They also find reflection in the very fact that the Security Service was established in the first place by reference to the Royal Prerogative.

A broad approach in this regard is, we consider, to be found in the decision of the Court of Appeal in R v Secretary of State for the Home Department, ex parte Northumbria Police Authority [1989] QB 26. In that case, a Home Office policy permitted chief officers of the police, in certain circumstances, to obtain plastic baton rounds and CS gas canisters from a central store, without obtaining prior approval of the relevant police authority. It was argued that no such power was conferred by the Police Act 1964. The Court of Appeal rejected that. But it was also held by the Court of Appeal that, under the Royal Prerogative, the Secretary of State had authority at all times (not just in times of war or actual emergency) to maintain the Queen’s peace and to keep law and order, unless to do so was contrary to statute; and, furthermore, could act pre-emptively in doing so.

Thus Croom-Johnson LJ, after citation of various authorities, said this at p.44B:

“…I have no doubt that the Crown does have a prerogative power to keep the peace, which is bound up with the undoubted right to see that crime is prevented and justice administered.” Purchas LJ said this at p. 53 D-F:

“It is well established that the courts will intervene to prevent executive action under prerogative powers in violation of property or other rights of the individual where this is inconsistent with statutory provisions providing for the same executive action. Where the executive action is directed towards the benefit or protection of the individual, it is unlikely that its use will attract the intervention of the courts. In my judgment, before the courts will hold that such executive action is contrary to legislation, express and unequivocal terms must be found in the statute which deprive the individual from receiving the benefit or protection intended by the exercise of prerogative power.”

He went on to say at p. 53 H:

“The prerogative power is to do all that is reasonably necessary to keep the Queen’s peace.”

Nourse LJ made statements to similar effect. He referred to a duty “to protect the lives and property of the King’s subjects”. He went on to say at p.58A:

“There is no historical or other basis for denying to the war prerogative a sister prerogative of keeping the peace within the realm.”

In the present case, the functions of the Security Service and Director-General, as stated in s. 1 and s. 2 of the 1989 Act, are all essentially geared to maintaining peace and stability within the realm. Protection against terrorism and sabotage are expressly included as part of those functions. That was also the position before 1989; and the operational necessity of running criminally participating agents unquestionably had been occurring over many years prior to 1989. As we read the 1989 Act, there is nothing in that Act which prohibits such an activity or policy. Nor was our attention drawn to any statutory measure in existence prior to 1989 bearing on that activity or policy such that it would be incompatible with any statutory provisions, in the sense indicated by Purchas LJ in the Northumbria Police Authority case (or, indeed, in the sense indicated in the De Keyser’s Hotel and Burmah Oil cases).

All this, in our opinion, further points strongly to the Security Service having, and always having had, the power, by its officers, to run agents who participate in criminality, whether possible or actual, in order to fulfil its function to protect the public: provided that there is no immunity from criminal sanction.

Moreover, the Guidance in no way issues a command as to what handlers are required to instruct (or agents are required to do). These are indeed “Guidelines”, which give a discretion to individuals: albeit a discretion to be exercised within the stated parameters. Furthermore, the Guidance itself in terms, by paragraph 8, stipulates that authorisation may only be given where the authorising officer is satisfied that the potential harm to the public interest from the criminal activity is outweighed by the benefit to the public interest derived from the anticipated information the agent may provide and that the benefit is proportionate to the activity in question. Thus, contrary to the submission of Mr Jaffey, there is indeed a limit to what criminality may be authorised, having regard to the necessity, public interest and proportionality requirements stipulated first by the Maxwell Fyfe Directive and then (in more detail) by the Guidance itself. There is also the very real practical limit on such decision making arising from the known risk of prosecution and from which, as the Guidance makes absolutely clear, there is no immunity, whether for handler or agent.

The conclusion which we reach is, we consider, reinforced by three other matters.

First, we think that the case of Buckoke v Greater London Council [1971] 1 Ch 655 provides a powerful analogy to the present case. We accept that the context of that case was different from the present. But at all events the background there also involved the potential protection of life and property. The position was this. Under the statutory provisions then in force it was an offence if any driver failed to stop at traffic lights when showing red. No exception was, at the time, made for emergency workers. If, however, drivers of Fire Brigade vehicles complied with the letter of the law vital minutes could be lost in an emergency. Guidance in the form of a Brigade Order by the Chief Officer issued to drivers, while stressing that they were subject to the law, in effect tacitly accepted that to perform their urgent duties they might have to cross through a red light. The matter was left to the decision of the individual drivers. The validity of the relevant Brigade Order was challenged. It was objected that it was in substance an order by employer to employee telling him how to break the law if he decides to break it, and in effect was an encouragement to break the law. But both at first instance and on appeal the validity of the Brigade Order was upheld: albeit the reasoning of the judges varied somewhat.

In this context, the judges accepted that a breach of the law would have occurred and that prosecution could result (though emphasising that an exercise of discretion ordinarily not to prosecute should be expected). Lord Denning MR in terms said (at p.699E) that if a driver had made clear that he was not going to pass through a red light except when there was no risk of collision and after taking due precautions, then an order to crewmen to travel with such a driver was a lawful order. As for the Brigade Order itself, Buckley LJ, for example, said (at p. 679B):

“The Order does not confer any discretion on drivers to break the law: it limits that discretion which they individually exercise.”

Thus it is noticeable that so far from it being held that the Chief Officer had no power (vires) to issue the Brigade Order, rather the Brigade Order was in effect commended. We consider that similar observations can be made about the Guidance in the present case. It acknowledges the obligation of individuals to comply with the law and acknowledges the risk of prosecution if that is not done; but, amongst other things, it gives guidance as to the public interest and proportionality considerations which must be taken into account before any instruction is given by an individual officer. We, in fact, think it most likely that those drafting the Guidance – and the Guidance clearly must have had legal input – would have had well in mind the corresponding factors which informed the approach taken in Buckoke.

Second, and as noted by the Tribunal itself, Sir John Donaldson MR in the Court of Appeal in the Spycatcher litigation, Attorney-General v Guardian Newspapers Ltd [1990] 1 AC 109 (litigation which preceded, and provided context for, the 1989 Act), in the course of his judgment said at p.189-190:

“It would be a sad day for democracy and the rule of law if the service were ever to be considered to be above or exempt from the law of the land. And it is not. At any time any member of the service who breaks the law is liable to be prosecuted. But there is a need for some discretion and common sense. Let us suppose that the service has information which suggests that a spy may be operating from particular premises. It needs to have confirmation. It may well consider that, if he proves to be a spy, the interests of the nation are better served by letting him continue with his activities under surveillance and in ignorance that he has been detected rather than by arresting him. What is the service expected to do? A secret search of the premises is the obvious answer. Is this really "wrongdoing?

Let us test it in a mundane context known to us all. Prior to the passing of section 79 of the Road Traffic Regulation Act 1967, fire engines and ambulances, unlike police vehicles, had no exemption from the speed limits. Their drivers hurrying to an emergency broke the law. So far as I am aware that is still the position in relation to crossing traffic lights which are showing red and driving on the wrong side of the road to bypass a traffic jam. The responsible authorities in a very proper exercise of discretion simply do not prosecute them.

Even in the context of the work of the Security Service which, I must stress, is the defence of the realm, there must be stringent limits to what breaches of the law can be considered excusable. Thus I cannot conceive of physical violence ever coming within this category. Or physical restraint, other than in the powers of arrest enjoyed by every citizen or under the authority of a lawful warrant of arrest. But covert invasions of privacy, which I think is what Mr. Wright means by “burglary,” may in some circumstances be a different matter.

It may be that the time has come when Parliament should regularise the position of the service. It is certainly a tenable view. The alternative view, which is equally tenable, is that the public interest is better served by leaving the members of the service liable to prosecution for any breach of the law at the instance of a private individual or of a public prosecuting authority, but may expect that prosecuting authorities will exercise a wise discretion and that in an appropriate case the Attorney-General would enter a nolle prosequi, justifying his action to Parliament if necessary. In so acting, the AttorneyGeneral is not acting as a political minister or as a colleague of ministers. He acts personally and in a quasi-judicial capacity as representing the Crown (see article entitled "How the security services are bound by the rule of law" by Lord Hailsham in "The Independent," 3 February 1988). It is not for me to form or express any view on which is the most appropriate course to adopt in the interests of the security of the nation and the maintenance of the rule of law. However that problem is resolved, it is absurd to contend that any breach of the law, whatever its character, will constitute such "wrongdoing" as to deprive the service of the secrecy without which it cannot possibly operate.”

We appreciate that the minority in the Tribunal evidently considered untenable what the Master of the Rolls plainly had considered to be “equally tenable”. But we respectfully consider those observations of the Master of the Rolls (albeit this was a dissenting judgment in the Court of Appeal) to be valid. In any event, it clearly would have come as a total surprise to the Master of the Rolls to be told that the Security Service did not even have the vires to permit, through its officers, agents to participate in criminality.

Third, in placing emphasis on the principle of legality Mr Jaffey drew attention to what Lord Hoffmann had said in R v Secretary of State for the Home Department, ex parte Simms [2000] 2 AC 115 at p. 131 E-G:

“Parliamentary sovereignty means that Parliament can, if it chooses, legislate contrary to fundamental principles of human rights. The Human Rights Act 1998 will not detract from this power. The constraints upon its exercise by Parliament are ultimately political, not legal. But the principle of legality means that Parliament must squarely confront what it is doing and accept the political cost. Fundamental rights cannot be overridden by general or ambiguous words. This is because there is too great a risk that the full implications of their unqualified meaning may have passed unnoticed in the democratic process. In the absence of express language or necessary implication to the contrary, the courts therefore presume that even the most general words were intended to be subject to the basic rights of the individual.”

But, in the present context, we do not think that this general statement of principle really advances matters. Although the law ordinarily does not tolerate an “ends justify the means” approach, reliance on the principle of legality remains perhaps a rather paradoxical argument for the appellants in this case to deploy, given that the activities of the agents are being “authorised” precisely with a view to preventing the taking of innocent life and to inhibiting the activities of those having no regard whatsoever to any principle of legality. But in any event, even though Mr Jaffey postulated some potential extreme instances, it is very difficult to see how “fundamental rights” will necessarily be “overridden” if the 1989 Act is to be interpreted as permitting the continuation of the “authorisation” of undercover agents to participate in criminality in the sense which we have explained, namely without granting immunity from criminal or civil sanction.

Finally, in dealing with this ground of appeal, we should refer to the decision of the House of Lords in Morris v Beardmore [1981] AC 446, on which Mr Flint QC, one of the minority in the Tribunal, had particularly relied.

That case concerned the requirement of the provision of a specimen for breath testing, under s. 8 of the Road Traffic Act 1972. In that case, a driver had left the scene of collision and had driven home. The police arrived at his house one and a half hours later. He refused to let them come up to his bedroom and required them to leave. Instead of leaving, they went to his bedroom and then arrested him for refusing to undergo a breath test. A subsequent charge was brought against him.

It was held by the House of Lords that the requirement to take the breath test had been made unlawfully, as the police officers were trespassers in the driver’s home at the time. It was stated that the interpretative presumption was that, in the absence of express provision to the contrary, Parliament did not intend to authorise tortious conduct. Further (and this was the point on which Mr Flint QC placed particular emphasis), it was held that it was no answer to say that a civil claim could still lie against the police for trespass.

With respect, we do not think that decision casts any real light on the present case. Morris v Beardmore involved a question of interpretation of the Road Traffic Act 1972, where the context and purpose are wholly different from the present. Moreover, the sanctity of a person’s own home was there much emphasised: it is noticeable, for example, that the House of Lords in terms left open the position where the police entered without licence the property of a third party in order to administer the breathtest (“hedge-hopping”).

We therefore conclude that before 1989 the Security Service had the vires to run agents who participate in criminality in order to protect national security, in the sense which we have described; and that power continued under the 1989 Act, on its true interpretation. Given that there was and is no immunity from prosecution, such a conclusion does not place the Security Service above the law. We therefore endorse the conclusion of the majority of the Tribunal; and in particular we endorse their emphasis on the important differentiation between a power and an immunity. As to the view of Mr Flint QC, that an implied power which authorises conduct contrary to the general criminal law but leaves the person engaging in such conduct liable to a criminal prosecution would be “extraordinary”, we take an entirely contrary view. Set in the context of the essential functions of the Security Service, as they always have been, it would, as we see it, be extraordinary if it were otherwise.

Having reached that conclusion, we need express no opinion on the potential implications for the police, and their use of undercover officers and agents, if the appellants’ arguments were well-founded. This is another point which had troubled the majority (see paragraph 66 of the judgment) and which they were inclined to consider lent further support to their conclusion. These are deep waters, as cases such as Yip-Chiu Cheung v The Queen [1995] 1 AC 111 and R v Loosely [2001] UKHL 53, [2001] 1 WLR 2060 illustrate. We prefer, for present purposes, to express no views on that particular issue.

In summary, we therefore reject the first ground of appeal. We agree with the majority of the Tribunal that Parliament did not intend in enacting the 1989 Act to bring an end to an essential part of the Security Service’s core activities. As Sir

We turn to the second ground of appeal. We can take that altogether more shortly:

The issue here is whether the Guidance creates a de facto immunity from prosecution. In agreement with the unanimous view of the Tribunal, we see insuperable objections to the arguments of the appellants.

Self-evidently, the appellants are not in a position to argue that the Guidance purports to grant any immunity: to the contrary, the Guidance emphasises, as we have said, the requirements of the law and the risk of prosecution. For the reasons given above, it therefore cannot be said that the Security Service has purported to dispense with the law, contrary to the Bill of Rights. There is, de jure, no immunity from prosecution.

The present situation demonstrably is wholly different from the case of The King v The London County Council [1931] 2KB 21, to which Mr Jaffey referred us. In that case, there was at that time a longstanding statutory prohibition on opening premises for public entertainment, for an admission fee, on Sundays. Nevertheless, the County Council in that case had granted a cinema company a licence to do so, on condition of payment of a sum to charity. In short, the Council purported to sanction a breach of the law in return for payment. Unsurprisingly, it was held that the County Council had no authority to dispense with observance of a statute. But that is a situation far removed from the present case, in circumstances where neither s. 1 of the 1989 Act nor the Guidance itself purports to confer, or does confer, any immunity from prosecution at all.

In fact, we found it very difficult to ascertain just what was the basis for the very generalised assertions of de facto immunity.

Certainly Mr Jaffey struggled, unsuccessfully, to identify any legal obligation on the Security Service even to notify the prosecuting authorities where potential participation in criminality was proposed to be authorised. Nor, in any event, is it at all clear how breach of such notification duty (if any) could create a de facto immunity, where no immunity otherwise exists.

We of course accept that the constitutional position of prosecutors is designed to be, and is, independent of the Executive. The constitutional arrangements in this regard vary somewhat between England and Wales, Scotland and Northern Ireland: but in general terms the overall position in this context is broadly the same. In terms of England and Wales, all that Mr Jaffey could do was point to the Directory of Civil

Mr Jaffey could do no better with regard to Scotland. He had rather more material available to him with regard to Northern Ireland. For by s. 5 of the Criminal Law (Northern Ireland) Act 1967 it is, in effect, made a duty of every person who knows or believes that an arrestable offence has been committed and that he has information which may materially assist in the apprehension, prosecution or conviction of any such person, to give such information to a constable; and “if, without reasonable excuse, he fails to do so” he is guilty of an offence. So even here there is an exception for reasonable excuse.

We overall agree with the Tribunal’s disposal of this issue. It said this at paragraph 79 of its judgment:

“There are several difficulties in the way of that submission. The first is that it presupposes what the outcome would be after the event in a criminal court on the individual facts of a particular case. As we have already mentioned, that is not usually the function of a civil court or tribunal. Certainly it is not usually their function to give a categorical view on matters of criminal law in advance. Secondly, it may well be, depending on the facts of a particular case, that a person will have “reasonable excuse” where they rely on the fact that conduct was authorised in accordance with the policy of the Security Service. Thirdly, there might well be a need to interpret section 5 of the 1967 Act consistently with the duties which fall upon the Director-General in section 2(2) of the 1989 Act, in particular his duty not to disclose certain information except so far as necessary for the purpose of discharging the functions of the Security Service or for the purpose of preventing or detecting serious crime.”

In such circumstances, we also cannot agree with the appellants’ argument that the Security Service by its Guidance is undermining the independence of prosecution (or police) authorities. On the contrary, the Guidance respects it. What it does do is to indicate what the Security Service would intend to say by way of making representations to the prosecution authorities when those authorities are considering, where they are required to do so, whether a prosecution would be in the public interest.

Further, if the Security Service were required to give notification where the potential crime is inchoate and before achievement of the actual intended act then it would potentially be put in an invidious and fraught position: first, because (to repeat) what will be authorised will in very many situations have been assessed only as carrying the risk of being criminal; second, because there is then the inevitable enhancement of the potential for security leaks and for the compromising of the agent. And the prosecuting authorities would themselves also potentially be in an invidious and fraught position: if only because it is a general principle that prosecuting authorities themselves cannot, in advance of knowing all the facts, grant a “proleptic grant of immunity from prosecution”: see R (Pretty) v Director of Public Prosecutions [2001]

Moreover, where a crime has in fact been committed and identified, and where there is then an investigation or prosecution, the evidence is that the Security Service will then provide materials to the prosecution authorities which are relevant to investigation and prosecution. The undisputed evidence generally was that the Security Service works closely with the police in counter-terrorism operations. The evidence also reveals that there is, for example, a Memorandum of Understanding between the Security Service, the police and the Counter Terrorism Division of the Crown Prosecution Service. We were told that there were corresponding protocols in Scotland and Northern Ireland.

We do not propose to say more on this ground of appeal. We reject it.

Given the conclusions this court has reached as expressed above, the Guidance is “in accordance with” the domestic law.

It was submitted, nevertheless, on behalf of the appellants that interference by a public authority with the rights of an individual must be sufficiently amenable to judicial oversight so as to provide appropriate safeguards against the risk of abuse of power by the public authority.

In our judgment, and in agreement with the (unanimous) decision of the Tribunal on this point, this ground of appeal is devoid of any real substance.

The 1989 Act had conferred some important oversight functions on the Commissioner (required by s. 4 (1) to be a person who held, or previously had held, high judicial office). That oversight has since been continued, and expanded, with regard to the Investigatory Powers Commissioner (where there continues to be the same requirement for holding or having held high judicial office).

It is true that the 1989 Act did not confer such oversight over all of the Security Service’s functions; nor did the various directions of the Prime Minister so require (or permit). But the challenge in this case has been to the lawfulness of the Guidance. That, ultimately, is a matter not for the Commissioner but for the Tribunal, a superior court of record, and the appellate courts – indeed, the very fact of these present proceedings indicates the availability of such oversight.

As to the suggestion that there was no oversight conferred with regard to prosecutorial decisions in this context, we see no reason why there should have been. Prosecutorial decisions are a matter solely for the relevant prosecution or police authorities, who will have their own applicable policies. Moreover, decisions by those authorities as to whether or not to prosecute in any given case are in principle capable, in appropriate circumstances, of being subject to judicial review.

The complaint about the Guidance being kept secret has been overtaken by events. We also note that permission to appeal was refused on the third ground then being sought to be advanced.

These grounds all involve allegations of potential breach of the Convention. Sir James expressly accepted before us, as he had below, that the Security Service was to be taken to be a public authority required to comply with the obligations of the Convention. Nevertheless, in our judgment, these claims by the appellants share a fatal initial weakness. That is that the appellants had and have no standing to pursue them.

In this respect, s. 7 (1) (3) and (7) of the Human Rights Act 1998 provide as follows:

“(1)

A person who claims that a public authority has acted (or proposes to act) in a way which is made unlawful by section 6(1) may—

(a)

bring proceedings against the authority under this Act in the appropriate court or tribunal, or

(b)

rely on the Convention right or rights concerned in any legal proceedings,

but only if he is (or would be) a victim of the unlawful act.

The obvious difficulty for the appellants is that in no way can they be said to be “a victim” of any unlawful act. At most they, as non-governmental organisations, seek to represent a viewpoint that might be a viewpoint that could be advanced by others who were or would be such victims.

This is contrary to s. 7 and to the general approach taken in the European Court of Human Rights by reference to article 34.

Thus in Zakharov v Russia (2016) 63 EHRR 17 this was said at paragraph 164 of the judgment:

“The Court has consistently held in its case-law that the Convention does not provide for the institution of an actio popularis and that its task is not normally to review the relevant law and practice in abstracto , but to determine whether the manner in which they were applied to, or affected, the applicant gave rise to a violation of the Convention. Accordingly, in order to be able to lodge an application in accordance with art.34, an individual must be able to show that he or she was “directly affected” by the measure complained of. This is indispensable for putting the protection mechanism of the Convention into motion, although this criterion is not to be applied in a rigid, mechanical and inflexible way throughout the proceedings.”

Mr Jaffey nevertheless sought to rely on a potential exception to that general approach, illustrated in the secret surveillance case of Klass v Germany (1979-1980) 2 EHRR 214: a case discussed in detail in Zakharov itself. In fact, having discussed that case and other authorities, the Court in Zakharov went on to say, in the context of secret surveillance cases, at paragraph 171:

“In such cases the individual may claim to be a victim of a violation occasioned by the mere existence of secret measures or of legislation permitting secret measures only if he is able to show that, due to his personal situation (emphasis added), he is potentially at risk of being subjected to such measures.”

In any event, Klass (as was Zakharov) was a case where the claim was brought by individuals, claiming to be victims of a violation of article 8 but without being in a position to prove that secret surveillance had in fact been applied to such individual(s). That demonstrably is not the position here. The claims here are not brought by individuals; are not confined to issues of secret surveillance; and are put on a far wider and more generalised basis than in Klass.

This is no mere procedural technicality. The factual situations that can arise in this context will be myriad. It would be an impossible task, and at all events most unwise, for a court to make generalised pronouncements in this sort of context which are divorced from an identified and specific factual framework. So to hold, moreover, does not preclude an actual victim in an appropriate case from coming forward and advancing such a claim, depending on the actual factual circumstances that have occurred. Indeed, the (exceptional) case of Gäfgen v Germany (2011) 52 EHRR 1 is an illustration of how critical can be the need to establish the relevant facts before making pronouncements on claims based, for example, on articles such as article 2 or article 3 of the Convention. That is also mirrored by the approach generally taken by the domestic courts: as illustrated by, for example, the decisions in Pretty and Nicklinson (cited above).

For this reason alone, we reject these grounds. We consider that the Tribunal was justified in holding that the appellants lacked sufficient standing for the purposes of s. 7 of the 1998 Act. In such circumstances, we need ourselves express no (necessarily obiter) views on the substance of the points sought to be raised on these grounds: on which, we note, the Tribunal had also, and in any event, unanimously found against the appellants.

As we have said, the members of this court were, in advance of this appeal, requested to read and did read a quantity of closed materials and the closed judgment of the Tribunal. We do not consider that any outline, in open, of the closed judgment is appropriate. We simply state here that we approve the closed judgment; and that there is, in our opinion, nothing in the closed materials which undermines the conclusions which we in any event reach. On the contrary, they only serve to confirm these conclusions. Accordingly, no further closed hearing is called for.

We conclude that the appeal fails on all grounds advanced. We uphold the conclusion of the majority of the Tribunal on the first ground and the unanimous conclusion of the Tribunal on the other grounds. We therefore dismiss this appeal. It has been accepted that there is to be no order as to costs.