Skip Navigation

Executive Summary And Principal Conclusions

Executive Summary and Principal Conclusions

1. Patrick Finucane, a practising lawyer, was murdered in his home in North Belfast on the evening of Sunday 12 February 1989. The attack was carried out by gunmen from the loyalist paramilitary group, the Ulster Defence Association (UDA), as he sat down for dinner with his wife, Geraldine, and their three young children. Geraldine Finucane was injured in the attack.

2. I was appointed by the Secretary of State for Northern Ireland on 12 October 2011 to conduct an independent Review into the question of State involvement in the murder. On that day the Prime Minister, the Rt Hon David Cameron MP, made the following comments in the House of Commons:

"I profoundly believe that the right thing for the Finucane family, for Northern Ireland and for everyone in the United Kingdom is … for the British Government to do the really important thing, which is to open up and tell the truth about what happened 22 years ago. Frank acknowledgement of what went wrong, an apology for what happened - that is what is required."

3. The Prime Minister's comments were followed by an Oral Statement by the then Secretary of State for Northern Ireland, the Rt Hon Owen Paterson MP, who said:

"The Government accept the clear conclusions of Lord Stevens and Judge Cory that there was collusion. I want to reiterate the Government's apology in the House today. The Government are deeply sorry for what happened. Despite the clear conclusions of previous investigations and reports, there is still only limited information in the public domain. That is why my right honourable friend the Prime Minister and I have committed to establishing a further process to ensure that the truth is revealed."

4. I was provided with the following Terms of Reference for my Review:

"Drawing from the extensive investigations that have already taken place, to produce a full public account of any involvement by the Army, the Royal Ulster Constabulary, the Security Service or other UK Government body in the murder of Patrick Finucane.

The Review will have full access to the Stevens archive and all Government papers, including any Ministry of Defence, Security Service, Home Office, Cabinet Office or Northern Ireland Office files that you believe are relevant. The account should be provided to the Secretary of State for Northern Ireland by December 2012, for the purpose of its publication."

The work of my independent Review

5. The references in the Secretary of State's statement to "the clear conclusions of Lord Stevens and Judge Cory" are to the separate investigations touching on Patrick Finucane's murder which Lord (formerly Sir John) Stevens, the former Metropolitan Police Commissioner, and former Justice Peter Cory (of the Supreme Court of Canada) had previously undertaken. Whilst I have drawn on their investigations and findings, I decided to approach my task by looking at the evidence afresh, without feeling bound by their earlier findings - particularly as Justice Cory expressed his own to be "provisional".

6. Sir John Stevens carried out a series of criminal investigations into alleged collusion between the security forces and loyalist paramilitaries in Northern Ireland. His third investigation (Stevens III) was commissioned in 1999 and focused specifically on the murder of Patrick Finucane. As a result of the three investigations carried out by Sir John, I am the beneficiary of some 12,000 witness statements, 32,000 documents and, in all, over a million pages of material.

7. However, although the volume of material already collated by Sir John Stevens was enormous, I decided at the outset of my Review that it was important to conduct a far more wide-ranging process than a straightforward examination of the available evidence gathered by the criminal investigations. I have, therefore, sought and received new documentary material from all the organisations cited in my Terms of Reference and a number of Government Departments. That material has included new and significant information that was not available to Sir John Stevens or Justice Cory.

8. A further aspect of my Review has involved engaging with key individuals who could assist me in producing a full public account. Amongst others, I met with individuals who had served in the Army, the RUC and the Security Service, and questioned them about these matters. I also received a series of written submissions. I have engaged in this process both in order to receive new information that could assist me in my work, and to provide individuals and organisations with an opportunity to make representations about issues on which they have been the subject of criticism.

9. I also came to the conclusion that it was insufficient for my Report to involve merely an account of my findings. In view of the background to this case, and the intense controversy it has aroused, I announced on 10 November 2011 that I would be declassifying and publishing documents, including original intelligence material, alongside my Report. As part of this process, I identified the key documents relating directly to the murder of Patrick Finucane, and have published them in Volume II of this Report.

10. In view of the serious obstruction of previous criminal investigations examined in this Report, it is important to acknowledge that all relevant Government Departments and Agencies co-operated fully and openly with my Review. Although I had no statutory powers of compulsion, I was given access to all the evidence that I sought, including highly sensitive intelligence files. I should specifically acknowledge the assistance provided by the Ministry of Defence, the Security Service and the Police Service of Northern Ireland, all of which held a large quantity of relevant material. The assistance and co-operation provided by these organisations was exemplary.

11. I should also record that retired senior intelligence officers, and the former BBC journalist John Ware, also engaged extensively with the work of my Review and provided me with important evidence and insights. Their assistance was provided voluntarily and was extremely valuable in enabling me to produce this Report.

12. In this Executive Summary and Principal Conclusions, I deal in outline form with the key facts and findings in this case. However, my conclusions have been based on a detailed examination of the evidence. This Summary must, therefore, be considered in conjunction with my full Report, in which I set out the evidence that I have considered and my reasons for reaching certain conclusions.

The context to the murder of Patrick Finucane

13. In his statement to Parliament on 12 October 2011, the Secretary of State for Northern Ireland expanded on my remit by explaining that:

"[The Government] accepting collusion is not sufficient in itself. The public now need to know the extent and nature of that collusion."

14. In order to examine the extent and nature of what the Government accepts to be collusion, I have undertaken an exhaustive examination of the context to the murder of Patrick Finucane in 1989.

15. My Report deals at length with the historical and security background in Northern Ireland in the late 1980s. In view of the gravity of my findings, it is essential to highlight in this Summary some of the key contextual issues.

16. By 1989 Northern Ireland had experienced over two decades of sustained and often brutal conflict. Although the Government viewed the paramilitary violence endemic during the Troubles as a form of terrorism, it essentially treated both republican and loyalist terrorists as criminals who were to be brought to justice within the judicial system in place in Northern Ireland.

17. The security forces and intelligence agencies faced an extraordinarily difficult task. Many RUC and Army personnel were murdered throughout the course of the Troubles as they sought to carry out their duties to the public and the State. During the late 1980s there was an upsurge in Irish Republican Army (IRA) violence following the importation of arms from Libya and an increasingly militant loyalist backlash following the 1985 Anglo-Irish Agreement, which was seen by many loyalists as a betrayal by the UK Government.

18. A significant part of my Report focuses on the organisations involved in intelligence-gathering in Northern Ireland. I believe that the intelligence-led security response to the Troubles did play a significant role in constraining all terrorist organisations, to the extent that they were forced to realise that their aims were not achievable by violence.

19. In the context of this Report, it is important to acknowledge, in particular, that the work of the Royal Ulster Constabulary's Special Branch (RUC SB) and the Security Service had a significant impact in thwarting and constraining loyalist terrorist groups. Many intelligence-led operations against republican paramilitary groups were also notably successful during this period.

20. In order to understand how an event such as the murder of Patrick Finucane could have occurred, it is important to consider a number of other key contextual themes which I have examined, as follows:

(i) the lack of any adequate framework or guidance for the handling of agents in Northern Ireland;

(ii) the actions of the Army agent Brian Nelson during the period 1987-89 and the accountability of agencies of the State in his case;

(iii) the responsibilities of the RUC, and in particular the handling of threat intelligence during the period; and

(iv) the flow of information from members of the security forces to loyalist paramilitaries during the late 1980s.

I summarise my findings on each of these thematic issues below.

The lack of an adequate framework for agent-handling

21. Intelligence gained from human agents is, clearly, a potent weapon for the State in countering terrorism of the kind that prevailed during the Troubles. Nowhere was the need for a proper legal framework for agent-handling thrown into sharper focus than in Northern Ireland. I have accordingly considered at length the accusations that have been made against successive governments that they each failed to provide a proper and lawful regime for the conduct of agent-handling operations.

22. The submissions made to my Review by all former intelligence officers stressed that an agent could only provide the most valuable, and potentially life-saving, intelligence if they were infiltrated into the heart of a terrorist group. It followed that agents who were so infiltrated would, in order to maintain their cover, be required of necessity to engage in criminal conspiracies with their terrorist associates (whilst, in theory, seeking to help the security forces to frustrate the realisation of these plans).

23. In my view, the running of effective agents in Northern Ireland was such a fraught and difficult task that it manifestly required the support of a clear legal and policy framework. I have established, though, that there was no adequate framework in Northern Ireland in the late 1980s. Accordingly, each of the three agencies running agents - the RUC SB, the Army's Force Research Unit (FRU) and the Security Service - operated under their own separate regimes. The result was that: the RUC SB had no workable guidelines; the FRU were subject to Directives and Instructions that were contradictory; and the Security Service received no effective external guidance to make clear the extent to which their agents could be permitted to engage in criminality in order to gather intelligence.

24. It was apparent that successive Governments knew that agents were being run by the intelligence agencies in Northern Ireland without recourse to any effective guidance or a proper legal framework. I found that repeated attempts were made by senior RUC, Security Service and (latterly) Army officers to raise this very issue with Government Ministers at Cabinet level. Yet it was not until 1993 that some Cabinet Ministers belatedly came to support the creation of a legislative framework. Even then, it was not until seven years later, when the Regulation of Investigatory Powers Act 2000 (RIPA) was passed, that any description of a statutory regime was created.

25. The practical implications of the failure to provide policy direction on agent-handling were significant. It meant that agent-handlers and their superiors were expected to gather intelligence without clear guidance as to the extent to which their agents could become involved in criminal activity in order to achieve this objective. Intelligence officers were, in effect, being asked to perform a task that, in some cases, could not be achieved effectively in ways that were lawful. It is my view that those charged with upholding the law should never be put in the position of potentially having to break the law in order to discharge their official duties.

26. My overall conclusion is that there was a wilful and abject failure by successive Governments to provide the clear policy and legal framework necessary for agent-handling operations to take place effectively and within the law.

The actions of Brian Nelson from 1987 to 1989 and the accountability of State agencies in his case

27. Brian Nelson was recruited and handled by the FRU, a covert section of the Army which ran agents in Northern Ireland. Nelson's case provides perhaps the most striking illustration of the failures evident in the handling of some agents in Northern Ireland at the time. Serious concerns over the nature of the FRU's handling of him were raised by the then Attorney General, Sir Patrick Mayhew QC, in 1991. At his trial the following year, Nelson pleaded guilty to a number of serious criminal offences, including five conspiracies to murder.

28. Nelson had previously been sentenced to seven years' imprisonment for offences relating to the kidnapping and torture of a partially sighted man. He was released from prison in 1977, and was first recruited and run as a FRU agent from 1984 to 1985. During this period he played a pivotal role in the targeting and attempted murder of a Sinn Féin Councillor. Despite this background, and after living for a time in West Germany, Nelson was re-recruited by the FRU in 1987 and persuaded to move back to Northern Ireland. There he was tasked with re-infiltrating the UDA with a view to becoming their 'Intelligence Officer'. Although Nelson achieved this objective rapidly, there were clear warning signs that should have raised serious questions about his suitability to be employed as an agent of the State.

29. The subsequent running of Brian Nelson as an Army agent is a matter which has caused me grave concern. The evidence suggests that following his re-recruitment he played some part in at least four murders and ten attempted murders. I am also satisfied that, with the knowledge and acquiescence of his FRU handlers, Nelson extensively updated and disseminated targeting material to other loyalist paramilitaries which they subsequently used in their efforts to carry out terrorist attacks. My overall assessment of Nelson is that he was motivated by a desire to see what he perceived to be 'legitimate' republican targets killed, and that his actions materially increased the targeting capacity of the UDA and thereby furthered their murderous objectives.

30. In the light of the contradictory explanations that have been offered over the intervening years, the accountability of the relevant State organisations for this situation is a complex matter that has required rigorous analysis of the evidence. It is clear that Nelson was re-infiltrated into the UDA and tasked to focus its targeting on what the FRU's Commanding Officer referred to as 'PIRA [Provisional IRA] activists'. The stated rationale for this tasking was that such targets would prove more difficult for the UDA to attack as it would take time to locate them, thus making it easier for the security forces to take the necessary counter-measures in order to save lives.

31. The evidence, however, demonstrates that such counter-measures were only very rarely taken in response to intelligence provided by Nelson. Indeed, I have found just three cases in which the security forces took action on the intelligence that he provided to seek to frustrate UDA attacks. In each instance there are indications in the intelligence documents to suggest that, in the absence of specific considerations relating to these cases, the attacks might otherwise have proceeded.

32. I am satisfied that Brian Nelson's desire to see republicans attacked was clearly apparent to the FRU throughout this period. I have also come to the view that, given the nature of the tasking he received from the FRU, Nelson was unlikely to have believed there was any distinction between his objective of targeting supposedly 'legitimate' republican targets on the one hand, and the objectives of the FRU on the other.

33. In this context, I was particularly concerned by the fact that, on occasions, Nelson's FRU handlers provided him with information that was subsequently used for targeting purposes. These actions are, in my view, indicative of handlers in some instances deliberately facilitating Nelson's targeting of PIRA members.

34. The reaction of the FRU to Brian Nelson's involvement in UDA attacks and murders was wholly unacceptable. One example was after the murder of Terence McDaid in May 1988 (a case of mistaken identity), when his handler sought to reassure a troubled Nelson by telling him that the victim was a member of PIRA. Another was after the UDA murdered James Pratt Craig (a notorious loyalist believed by the UDA to have links with PIRA) in October 1988. Nelson was dismayed that his handlers had not celebrated the occasion with a drink. Despite initially expressing disapproval, one of Nelson's handlers later recorded that, had they known the details of his role in inciting the attack, a drink might have been appropriate.

35. However, I have concluded that accountability for what went wrong in the Nelson case did not rest solely with the FRU and their Commanding Officer. Procedural provisions were in place within the Army chain of command which should have enabled senior officers to supervise such cases, but these were not applied in relation to Brian Nelson. There was, therefore, a signal failure by the Army to ensure adequate supervision in this case.

36. I have also considered the position of the Security Service in relation to the Nelson case. It is clear that, on the two occasions when the Service became actively involved in discussions about the exploitation of his intelligence, their intervention was aimed at disrupting the UDA's plans. Following the efforts to frustrate the UDA's plan in May 1987, in which Nelson was deeply involved, to kill the then President of Sinn Féin, Gerry Adams, a senior Security Service officer sent a perceptive telegram. He warned that British Intelligence and the Government could face accusations of conspiracy to murder if such an attack was to be repeated and Nelson's involvement was to get into the public domain.

37. By the summer of 1988 the Security Service were aware that Nelson was motivated by a desire to see the UDA carry out attacks against 'justifiable' targets. Although this did serve to discourage the Service from seeking to run Nelson as their own agent, it failed to prompt any effort on their part to provide a greater degree of guidance to the FRU in handling the case. Whilst I agree with Justice Cory's view that there is no evidence to demonstrate collusive acts on the part of the Security Service in relation to the Nelson case, I have reached the conclusion that the Service failed to carry out their advisory and co-ordinating duties adequately in relation to Nelson and the FRU.

38. In terms of accountability, however, the most serious issue of all related to the failure of the RUC SB to respond to Nelson's intelligence. The RUC at this time enjoyed a primacy over other agencies in the exploitation of intelligence from all sources to protect individuals under threat.

39. On this specific issue I have found what I have termed to be a fundamental 'accountability gap'. This accountability gap was created by the separate positions taken up by the FRU and the RUC SB in their attempts to explain why intelligence was not acted on. The FRU maintained that all intelligence relating to threats to life reported by Nelson was passed to the RUC SB, who were expected to act on it. The RUC SB, on the other hand, have insisted that the FRU did not provide them with the necessary information to enable them to prevent attacks. In any event, some RUC SB officers implied that Nelson was an ineffective and low-level agent whose intelligence could, therefore, be ignored.

40. Neither position, in my view, is sustainable on the available evidence. The overall pattern of the passage of intelligence corresponds far more closely with the position put forward by the FRU than with that of the RUC SB. In almost all of the relevant murders or attempted murders that I have reviewed, it was clear that the FRU passed intelligence to the RUC SB prior to the attack indicating that the individual concerned was under threat. Nevertheless, I have also concluded that the FRU should have been aware that the RUC SB were taking no action on the bulk of the intelligence being supplied to them. Taken as a whole, an extraordinary state of affairs was created in which both the Army and the RUC SB had prior notice of a series of planned UDA assassinations, yet nothing was done by the RUC to seek to prevent these attacks.

The responsibilities of the RUC and their handling of threat intelligence

41. References in my Report to 'threat intelligence' refer to information received by the authorities indicating that the life of an individual was under threat. In view of the RUC's failure to take action as a result of specific intelligence provided by Brian Nelson, I have examined more generally the approach that the RUC took towards threat intelligence.

42. During the period 1987-89, the RUC SB's Threat Book for the Greater Belfast area recorded 730 instances of republican paramilitary threats to targeted individuals, with just 36 recorded instances of threats by loyalists. During this period, republicans were responsible for 55% of the murders in the Belfast area and loyalists for 45% of the murders, although I recognise that, in practice, levels of violence from republicans were proportionately higher than this statistic alone conveys. Nevertheless, I am satisfied that there was a seriously disproportionate focus by the RUC on acting upon threat intelligence that related to individuals who were being targeted by republican paramilitary groups.

43. This pattern was not, in my view, driven by an inherently sectarian bias. It needs to be understood with reference to two key contextual considerations. The first is that the RUC SB were averse to providing warnings to those (from any community) who were considered to be 'untrustworthy' and who might, therefore, have decided to publicise the fact that there was a threat to their life. If this happened, the RUC SB feared that their intelligence 'source' would be endangered. By contrast, if members of the security forces were, for example, targeted by PIRA, the RUC SB may have trusted them to receive warnings without publicising them.

44. My extensive research into the contemporary material leads me to the view that, when certain individuals were targeted, the reaction of the RUC SB was also influenced to a significant extent by whether or not the individual under threat could be 'traced' as a paramilitary on either side of the sectarian divide. This theme is evident in documents from throughout the period, but is perhaps most graphically demonstrated by the handling of earlier threat intelligence reports relating to another solicitor, Oliver Kelly, whom the RUC believed had links to paramilitaries. It was clear to me that steps were often not taken to secure the protection of those who were considered to be (as referred to in one intelligence document) "a thorn in the side" of the security forces during this period of the Troubles.

The flow of information from members of the security forces to loyalists

45. I also considered the scale and nature of the assistance being provided to the UDA by members of the security forces during the late 1980s. It is clear that there were extensive 'leaks' of security force information to the UDA and other loyalist paramilitary groups.

46. Many stalwart individuals served in the security forces during this time and my conclusion should not be taken to impugn the reputation of the majority of RUC and Ulster Defence Regiment (UDR) officers, who served with distinction during what was an extraordinarily violent period. Nevertheless, it is clear that some individuals within those organisations provided assistance to loyalist paramilitaries in instances where they shared a common desire to see republican paramilitaries killed. Such leaks were not institutional nor systemic, though they could certainly be described as widespread.

47. Briefing provided to Government Ministers during this period tended to suggest that such leaks related only to a small number of 'rogue' individuals who provided loyalists with 'low-level' montage photographs held by the security forces. However, those briefings were inconsistent with the evidence that a very large volume of information was being passed by some members of the security forces to loyalist paramilitaries, including reported leaks of highly sensitive information.

48. I have examined a sample of intelligence, graded as reliable, relating mainly to security force leaks to the UDA in the Greater Belfast area. This showed that between January 1987 and September 1989 there were 270 separate instances of leaks. The so-called 'intelligence dump' that was developed by Brian Nelson, together with the reporting that he provided to his FRU handlers, demonstrated that a very large quantity of information of security force origin was provided to the UDA, and that the UDA cultivated and continued to maintain a number of security force 'contacts'.

49. In 1985 the Security Service assessed that 85% of the UDA's 'intelligence' originated from sources within the security forces. I am satisfied that this proportion would have remained largely unchanged by February 1989, the time of Patrick Finucane's murder. During this period, the UDA were heavily reliant on the flow of security force leaks to enable them to identify republican targets. This meant that many UDA attacks could be traced back to assistance initially provided by one of their security force contacts. Although some limited action was taken by the authorities to combat these leaks, my overall view is that, prior to the commencement of the Stevens I Investigation in September 1989, such efforts were inadequate given the scale of the problem.

The murder of Patrick Finucane

50. In producing this full public account, I cannot subscribe to the view expressed in some quarters that the murder of Patrick Finucane can be explained solely by reference to the actions of agents of the State. It is clear that Mr Finucane was the victim of a particularly violent UDA gang which played a central part in planning and executing his brutal murder. Nevertheless, this does not detract in any way from the key roles that known agents of the State played in it.

51. In that regard, I have concluded that two agents who were at the time in the pay of agencies of the State were involved in Patrick Finucane's murder, together with another who was to become an agent of the State after his involvement in that murder became known to the agency that later employed him.

The handling of previous threats to the life of Patrick Finucane

52. Before analysing the murder of Patrick Finucane, I decided that it would be instructive to consider two previous UDA conspiracies to murder him and the responses of the Security Service and the RUC SB to these plans.

53. During the first such conspiracy, in 1981, the Security Service and Secret Intelligence Service's Irish Joint Section (IJS) received intelligence indicating that the UDA had gathered information on Patrick Finucane and may have been planning to attack him. They shared that intelligence with the Head of the RUC SB, who assessed the threat as "very real and imminent". IJS and RUC SB officers discussed a number of different potential responses to this threat. Startlingly, the discussions even included a proposal, initially made by a source, to carry out a 'mock attack' on Patrick Finucane's home in order to cause him to flee; this proposal, however, was ruled out as an extremely dangerous course of action.

54. In the event, notwithstanding the apparent seriousness of the threat to Patrick Finucane's life, the decision was taken by the RUC SB, supported by the IJS, to take no action to warn or otherwise protect him because to do so could compromise an agent from whom the intelligence derived. In view of the imminence of the threat, this approach was, in my view, wholly inconsistent with a State's obligations under Article 2 of the European Convention on Human Rights (ECHR) to take appropriate steps to protect the lives of its citizens.

55. Some years later, in 1985, the Security Service again received intelligence indicating that Patrick Finucane was considered to be a "priority" target by a senior UDA figure. This information was shared with the RUC SB, but there is no evidence that any action was taken to warn or otherwise seek to protect him.

56. In terms of causation, the handling of these two previous threats to Patrick Finucane's life cannot be said to have led to his murder in 1989. However, the response of the intelligence agencies upon learning of these threats does raise serious concerns. Primary responsibility for these failings must lie with the RUC SB, though I am also critical of the role played by the Security Service, which supported the RUC SB's decision to take no action in 1981, and appear to have made no attempt to prompt them into taking any action in 1985.

The theft of a UDR gun in 1987

57. The first identifiable action by an employee of the State that contributed directly to the murder of Patrick Finucane came in August 1987. A Colour Sergeant of the UDR stole weapons from Palace Barracks, County Down, and sold them to Kenneth Barrett, a UDA member who in 2004 pleaded guilty to Patrick Finucane's murder. The weapons stolen included a 9mm Browning pistol that was subsequently used in the murder. The Colour Sergeant was himself convicted in 1988 for his role in the theft.

The failure to take action against the West Belfast UDA

58. The intensive work on this case carried out by the Stevens III Investigation has meant that the other key UDA suspects involved in Patrick Finucane's murder have been identified. Regrettably, however, given the passage of time, there has been insufficient admissible evidence to prosecute any of this group of individuals.

59. Nevertheless, I was able to trace the involvement of that particular gang through a series of murders and attacks by the West Belfast UDA during the year prior to Patrick Finucane's murder. I have seen the significant amount of intelligence that the RUC SB received at the time linking those individuals to the attacks, at least some of which was passed to the RUC Criminal Investigation Department (CID). Whilst the intelligence was not of the kind that would have been admissible in evidence for the purpose of bringing criminal charges, in my view it could have enabled the police to identify suspects and thereby develop evidential leads. Arrests could also have been highly effective in disrupting the plans of those paramilitaries to carry out imminent attacks. Indeed, the gathering of intelligence to effect arrests was cited by the Commanding Officer of the FRU as one of the key reasons for running Brian Nelson as an agent in the heart of the UDA.

60. The evidence clearly shows that the RUC had failed to take action against this gang in relation to a series of attacks they carried out over the year preceding Patrick Finucane's murder. For example, none of the four individuals reliably linked to the murder of Terence McDaid in May 1988 was arrested in connection with that attack. Similarly, none of the six who were linked to the Gerard Slane's murder in September 1988 was arrested in connection with the killing. Indeed, two of the key ringleaders of this group were not arrested at all during the period between May 1987 (the point at which Nelson returned to Belfast) and February 1989, despite Nelson having provided extensive intelligence about their terrorist activities.

61. Subject to what I say below, the record of the RUC in arresting and bringing charges against loyalist terrorists across Northern Ireland as a whole during this period withstands scrutiny and was, indeed, disproportionately better than their record against republican terrorists. Nevertheless, their response to the West Belfast UDA during the period 1987-89 does stand out in stark contrast to that general pattern. It is clear to me that such action as the RUC took to investigate and disrupt the gang who went on to murder Patrick Finucane was grossly inadequate.

Security Service propaganda initiatives

62. I have examined in depth the Security Service propaganda initiatives that were conducted in Northern Ireland during the 1980s. Those initiatives were conceived to rebut republican propaganda and also had as objectives the exposure and 'unnerving' of republican paramilitary figures. I have established that, although the focus of the initiatives was on republican paramilitaries, Patrick Finucane came to be included in them prior to his murder.

63. Given the background to these initiatives, and in the sort of circumstances which then prevailed in Northern Ireland, the use of such propaganda could, unless it was very carefully controlled, have had manifestly undesirable results. Intelligence documents show that senior officers of the Security Service later - rightly, in my view - terminated the initiatives after recognising that they had been on "dangerous ground". It is a matter of significant concern to me that no political clearance was sought or obtained for the Service's involvement in these initiatives.

64. The effect of the propaganda relating to Patrick Finucane was to identify him with the activities of his clients. In my view this propaganda breached basic principles of the criminal justice system, namely, that lawyers should be allowed to carry out their duties to their clients without intimidation, and without the inference being made that a lawyer supports their client's objectives.

65. Although I am satisfied that there was no intention that the propaganda initiatives should incite loyalists to attack Patrick Finucane, they could undoubtedly have served to further legitimise him as a potential target for loyalist paramilitaries. The fact that the propaganda could have had such an effect was, in my view, a consequence that should have been foreseen by the Security Service at the time.

Comments made by Douglas Hogg MP

66. I have also examined the comments made by the then Parliamentary Under Secretary at the Home Office, Douglas Hogg MP, just four weeks prior to Patrick Finucane's murder. During a House of Commons debate on 17 January 1989, Mr Hogg stated that there were a number of solicitors in Northern Ireland who were "unduly sympathetic to the cause of the IRA". Seamus Mallon, a Social Democratic and Labour Party (SDLP) MP, responded to these comments during the debate by stating that, as a result of the Minister's statement, lawyers in Northern Ireland would become "targets for assassins' bullets".

67. Douglas Hogg's comments followed an oral briefing provided to him by the Chief Constable and other senior officers of the RUC in Belfast on 24 November 1988. At that briefing, Mr Hogg was told that some solicitors were "effectively in the pockets of terrorists". On 13 January 1989, Mr Hogg was provided with 'profiles' of Patrick Finucane and Oliver Kelly produced by the RUC. The material provided to Mr Hogg did not, in my view, substantiate a claim that either of those solicitors was 'effectively in the pockets of terrorists'. Having reviewed the background to this issue, I am satisfied that the RUC should have known that Mr Hogg intended to make the thrust of their briefing public in some form. I concur with the conclusion reached by Sir John Stevens in 2003 that the Minister was thereby compromised.

68. It is clear to me that Mr Hogg had no knowledge whatsoever of the threat to defence solicitors in Northern Ireland prior to making his statement. Further, I am satisfied that there is no basis for any claim that he intended his comments to provide a form of political encouragement for an attack on any solicitor. It is, however, clear that in the febrile and violent context of Belfast in 1989 loyalist paramilitaries did indeed pick up on the comments that the Minister made in Parliament. There are, therefore, grounds for believing that those comments did, albeit unwittingly on Mr Hogg's part, increase the vulnerability of defence solicitors such as Patrick Finucane practising in Northern Ireland at the time.

The handling of the 1988 threat intelligence relating to Patrick Finucane

69. The Security Service received intelligence in December 1988 indicating that a meeting of UDA military commanders was going to discuss plans to kill three solicitors. Although he was not referred to by name, it was clear, in my view, that one of the solicitors under threat at that time was Patrick Finucane.

70. Had the Security Service properly pursued this threat intelligence with the RUC SB, I believe that there would have been a documentary record of such action having been taken. However, no such record exists. The handling of this threat intelligence by the Security Service has caused me particular concern in view of their propaganda initiatives that I have referred to above. Having become involved in those initiatives, which painted Patrick Finucane as a PIRA figure, there was, in my view, an even greater obligation on the Service to take appropriate action on any threat intelligence that they subsequently received.

The alleged involvement of RUC officers in the murder

71. I have considered a number of allegations relating to the possible involvement of RUC officers in encouraging and facilitating the murder of Patrick Finucane.

72. Allegations have been made in the reports of a number of non-governmental organisations that RUC officers issued threats in relation to Patrick Finucane when speaking privately to some of his clients accused of terrorist offences. However, in view of the uncorroborated nature of these allegations and the questionable reliability of the individuals who reported the threats, I do not feel that the evidence on this issue substantiates these specific allegations. I have also come to the view that, even if such threats had been made, they were unlikely to have caused Patrick Finucane's murder since it is inconceivable that his clients would have communicated these threats to loyalist paramilitaries.

73. The critical issue, in my view, was to determine whether RUC officers had been involved in inciting loyalists in custody to attack Patrick Finucane. Allegations that RUC officers had incited loyalists in this manner were first expressed privately by the Ambassador of the Government of Ireland to the Cabinet Secretary on 13 February 1989, the day after Patrick Finucane's murder.

74. Having considered the evidence on this issue closely, I believe, on the balance of probabilities, that an RUC officer or officers did propose Patrick Finucane (along with at least one other man) as a UDA target when speaking to a loyalist paramilitary, who I refer to throughout my Report as L/03, in RUC Castlereagh on either 8 or 9 December 1988. In the absence of any video or audio recording, or direct admissions from those involved, the evidence is not capable of indicating which specific officer or officers may have been responsible for this incitement.

75. I have also considered the allegations made by Kenneth Barrett who, as I have described, subsequently pleaded guilty to his involvement in Patrick Finucane's murder. Barrett maintained that an RUC officer provided the UDA with information about Patrick Finucane which encouraged the group to attack him. As regards the broad allegations that Barrett made to RUC officers who were seeking to recruit him as an agent in October 1991 - namely, that the UDA received 'intelligence' about Patrick Finucane from a police source - I am persuaded that they are essentially accurate. However, although Barrett was intimately involved in the murder of Mr Finucane, I came to the view that he did not have the detailed level of knowledge about the identity of the UDA's police 'contact' that, when speaking to BBC Panorama journalists many years later, he was to claim to have had.

76. I also examined the role of the then 'Chairman' of the UDA, Thomas 'Tucker' Lyttle, and the nature of his relationship with RUC officers. I am satisfied that Lyttle was aware of the UDA's plan to murder Patrick Finucane, although he may not have been involved in the operational planning of the attack.

77. Lyttle alleged to the BBC journalist John Ware that he had discussed the targeting of Patrick Finucane with an RUC SB officer, who had in effect provided tacit approval for the UDA to carry out the attack. I was unable to adequately test the veracity of this allegation as Lyttle is now deceased. I do not, however, discount the possibility that such a discussion took place. A wider examination of Lyttle's role did cause me serious concerns about the nature of his RUC 'contacts' and access to sensitive information. I believe that an RUC SB 'contact' did inform Lyttle of the identity of a loyalist who was passing information to PIRA. This resulted in the UDA kidnapping, interrogating and expelling that person from Northern Ireland in November 1988.

78. I believe that on 16 February 1989, four days after Patrick Finucane's murder, Lyttle gained access to sensitive RUC intelligence about PIRA's Belfast Brigade, including the terrorist activity of Patrick Finucane's brother, Seamus Finucane. The overall impression I formed was that the nature of Lyttle's contact with some RUC officers provided him with an entirely improper degree of protection and assistance in his role as UDA Chairman and the so-called 'Brigadier' for the West Belfast UDA during this period.

The involvement of State agents in the murder

79. A central feature of my Review was to establish the level and degree of involvement that agents of the State had in Patrick Finucane's murder. I examined the role that the FRU's agent, Brian Nelson, had in the murder. It is clear to me that, during the course of Sir John Stevens' previous investigations, Nelson had always significantly downplayed the true extent of his involvement in the conspiracy.

80. By his own admission, Brian Nelson was asked by one of the key loyalist figures behind the murder plot, whom I shall refer to as L/28, to target Patrick Finucane. Although Nelson claimed not to have subsequently conducted that requested targeting, I am satisfied that he did.

81. Nelson produced and stored in his intelligence dump what were termed 'Personality cards' ('P cards') on individuals about whom he was gathering targeting information. The 'P cards' were then used by the UDA for the purpose of attacking those targets. I believe it is likely that Nelson produced a 'P card' relating to Patrick Finucane which he may have disseminated himself to other loyalist paramilitaries, although it is also possible that they could have taken it directly from his intelligence dump.

82. I also believe that, prior to the murder, Nelson carried out a reconnaissance ('recce'), on the Finucanes' home in North Belfast. Such 'recces' were carried out in order to gather information which would subsequently assist a 'hit team' to attack an individual.

83. I am sure that Nelson passed a photograph of Patrick Finucane to L/28 and Kenneth Barrett on Tuesday 7 February 1989, five days before the murder took place. It was a published photograph depicting Patrick Finucane with a client, Patrick McGeown, outside a court. Nelson subsequently claimed that he was duped into handing over the photograph, but I do not accept that suggestion. I believe that he consciously handed it over in the knowledge that the UDA would use it to assist their targeting of Patrick Finucane. Considering all the evidence regarding his involvement cumulatively, I am satisfied that Nelson played an important part in the murder.

84. Given that Nelson was a FRU agent, I also considered in detail whether he informed his FRU handlers about the role he played in targeting Patrick Finucane. The evidence on this question is complex but, nevertheless, I am satisfied that it points towards a clear conclusion that Nelson did not inform his handlers about his involvement in that respect. It follows, therefore, that in my view the FRU did not have foreknowledge of the conspiracy within the UDA to murder Patrick Finucane.

85. It is important to position these findings in the context of my wider examination of Brian Nelson's role. In many cases it would not be reasonable to hold a State agency accountable for the actions of its agent in circumstances where the agent had concealed their activities from their handlers or were otherwise acting without authority. In Nelson's case, however, I considered that there were a number of important factors that must be taken into account.

86. Amongst those factors, it was clear to the FRU from at least June 1988 onwards that Nelson was prepared to withhold information from his handlers if he felt that he was carrying out 'justifiable' targeting. I also took into account the broader pattern of his conduct prior to February 1989 and the fact that the FRU re-infiltrated him into the UDA in circumstances in which he might otherwise not have re-involved himself in terrorist activity.

87. These considerations all lead me to the conclusion that, in being tasked by the FRU to target 'PIRA activists' for the UDA, Nelson would, to all intents and purposes, properly be considered to be acting in a position equivalent to an employee of the Ministry of Defence. It follows, therefore, that the Army must bear a degree of responsibility for Brian Nelson's targeting activity during 1987-89, including that of Patrick Finucane. This must be so irrespective of the nature of the information that he failed to impart to his FRU handlers in that case and some others.

The involvement of the RUC SB agent William Stobie in the murder

88. William Stobie was a loyalist paramilitary who was recruited as an agent by the RUC SB in February 1988. My Report outlines a number of criticisms relating to his recruitment and handling. Following his recruitment, Stobie's possible involvement in a murder committed by the UDA in November 1987 ceased to be pursued by the authorities. Although he was subsequently handled and paid by the RUC SB, Stobie often appears to have received and distributed weapons for use in UDA attacks without reference to his handlers. Apart from a single instance, I have seen no evidence to indicate that the RUC SB exploited the intelligence that he provided to them prior to February 1989 to frustrate loyalist terrorist activity.

89. Prior to Patrick Finucane's murder, William Stobie was holding two 9mm Browning pistols for the UDA in addition to other weaponry. On 6 February 1989 he handed over a Heckler & Koch pistol to L/03. Later that evening he was instructed by another loyalist, L/20, to provide a 9mm Browning pistol for use in an imminent attack on a "top PIRA man". I am satisfied that Stobie handed over the 9mm Browning to a UDA hit team on Sunday 12 February 1989 and that this gun was subsequently used to murder Patrick Finucane. I do not, however, believe that Stobie himself knew that Patrick Finucane was to be the target of the attack.

90. I am satisfied that it should have been clear to the RUC SB from the threat intelligence that Stobie provided to them that the UDA were about to mount an imminent attack and that L/20 was a key figure in this plot. Despite the range of options that would have been available to the RUC SB to disrupt the planned attack - as discussed in more detail in the Report - it is clear that they took no action whatsoever to act on the threat intelligence.

91. It is possible that Stobie, as he has claimed in some of his accounts, informed his handlers on 12 February 1989 - shortly before the murder took place - that he had handed over the weapon to the hit team. The evidence on this issue is inconclusive, but I did reach the conclusion that, from the evening of 9 February 1989, it was entirely foreseeable by the RUC SB that Stobie would shortly hand over a 9mm Browning pistol for use in an imminent UDA attack. They were also aware of the identity of a key figure in the operation, the UDA Commander L/20. In this regard I concur with Sir John Stevens' view that proper exploitation of William Stobie's intelligence prior to the attack could have prevented the murder of Patrick Finucane on 12 February.

The investigation into the murder

92. The European Court of Human Rights found in 2003 that the UK Government had breached its procedural obligations under Article 2 of the ECHR by failing to carry out an adequate official investigation into the murder. Additional material that was not available to the Court further highlights the importance of this finding. There was, for example, no investigative follow-up action in response to two credible intelligence reports that RUC officers had been congratulating loyalists held in custody for the UDA having carried out the murder.

93. The significance of the RUC SB's failure to act on William Stobie's intelligence provided prior to the murder becomes the greater when considered in conjunction with the information that he supplied after it took place. I am satisfied that he told his handlers that he had been asked to collect a 9mm Browning pistol from L/20 on 15 February and that he did subsequently collect it. This clearly provided the RUC SB with a potential opportunity to recover the weapon that was probably one of those used to murder Patrick Finucane and to arrest one of the UDA ringleaders responsible for the attack. The RUC SB never revealed this critical information regarding the probable murder weapon to the RUC CID murder investigation team.

94. It is clear that highly relevant intelligence was withheld from the CID murder investigation team, though it must also be acknowledged that the CID team did not exploit some of the intelligence that it had. This undoubtedly had a significant impact in preventing attempts to bring Patrick Finucane's murderers to justice. Key UDA suspects such as L/20 and L/28 were not investigated or arrested until the Stevens III Investigation in 1999, over ten years after the murder. The failure of the RUC to ensure an adequate investigation into the murder of Patrick Finucane is particularly significant when considered alongside the wider inadequacy of the action taken against the West Belfast UDA prior to the murder.

The recruitment of Kenneth Barrett in 1991

95. Kenneth Barrett has admitted to being the man responsible for driving two UDA gunmen to Patrick Finucane's home on 12 February 1989. He pleaded guilty to his part in the murder in 2004. He had been linked to it by reliable intelligence that the RUC SB had received as early as 16 February 1989, but he was not arrested until the Stevens III Investigation began ten years later in 1999.

96. At a meeting on 3 October 1991, Barrett told two RUC CID officers and an RUC SB officer that he had been involved in the murder. However, he added a caveat to this admission by stating that he was talking 'hypothetically'.

97. It is clear that, following this 'admission', the RUC CID officers wanted to pursue Barrett in order to build an evidence-based case to charge him with Patrick Finucane's murder. However, the decision of the RUC SB, taken at Superintendent level, was to recruit Barrett as an agent. The murder investigation against him was effectively then dropped. Despite being so deeply implicated in the murder, Barrett became a paid agent of the State, working for the RUC SB from that point.

98. The 'admission' confession that Barrett had made to the three RUC officers on 3 October 1991 was recorded covertly on a tape held by the RUC SB. This tape recording could have represented strong evidence of Barrett's involvement in the murder, subject to a judge at a trial using his discretion to exclude the evidence in the event of a successful argument being raised over its legal admissibility.

99. By the time of the Stevens III Investigation, however, this tape containing the 'admission' had disappeared. It transpired that the tape provided to the investigators was of a subsequent conversation in which Barrett did not repeat his previous admission to involvement in Patrick Finucane's murder. I am satisfied that the disappearance of the original tape with Barrett's 'admission' to the murder was another deliberate act designed to obstruct the investigation into the murder of Patrick Finucane.

The obstruction of the Stevens I Investigation

100. The approach of the Army and the RUC SB to the Stevens I Investigation into security force 'leaks' is particularly instructive in illustrating the attitude taken towards tackling serious allegations of collusion.

101. Both the Army and the RUC SB consciously failed to provide Sir John Stevens with important material relevant to his criminal investigation. I do not accept the Army's position that, in failing to provide information to the Stevens I Investigation, it was acting in accordance with instructions issued by the RUC Chief Constable to the General Officer Commanding in Northern Ireland that the Army was to deny Stevens access to any intelligence information. The Army, in my view, clearly had its own agenda in seeking to protect its agent, Brian Nelson. This protection even extended to advising Nelson on how to resist police interrogation in the event that he was arrested by the Stevens team.

102. In my view, the fact that senior Army officers deliberately lied to criminal investigators by informing them that they did not run agents in Northern Ireland was an attempt to deflect the Stevens Investigation from learning of the existence of Brian Nelson. Indeed, the very existence of the FRU was hidden from Sir John Stevens until he decided to arrest Brian Nelson. The evening before his impending arrest, Nelson fled to the mainland. When a new date was decided upon to effect his arrest, an unexplained fire broke out at the Stevens team's headquarters.

103. It is, however, also clear that the RUC SB too were responsible for seriously obstructing the investigation. They withheld significant quantities of information, including Army and Security Service material that was in their possession. There is also evidence to suggest that the RUC SB sought to direct the Stevens I Investigation towards examining security force 'leaks' from the UDR and concealed information indicating that a similarly large number of leaks had emanated from RUC sources. Although the RUC SB were aware that the FRU had possession of Brian Nelson's intelligence dump from September 1989, SB officers were later uniformly to claim to have had no knowledge whatsoever that Nelson or the FRU possessed such material.

104. This extensive obstruction resulted in an extraordinary situation in which important evidence in a major criminal investigation remained concealed in an Army office for nearly four months.

The prosecution of Brian Nelson

105. I have also considered in detail the response of Government Ministers and the RUC to the proposed prosecution of Brian Nelson. In Volume II of this Report I have released an extensive set of Cabinet-level correspondence relating to this issue.

106. The Director of Public Prosecutions for Northern Ireland (DPP(NI)) and the Attorney General, Sir Patrick Mayhew, deserve significant credit for withstanding considerable political pressure designed to ensure that Brian Nelson was not prosecuted. That pressure arose during the course of the Shawcross exercise that was conducted in relation to Brian Nelson's case, under the established convention whereby Government Ministers can draw to the attention of the Attorney General any public interest considerations which may arise in relation to possible prosecution proceedings. The Secretary of State for Defence, the Secretary of State for Northern Ireland and senior Government officials all argued strongly that the prosecution would not be in the public interest. It is clear that, had their views prevailed, an agent of the State who had actively promoted and facilitated a series of terrorist attacks would have escaped justice.

107. It is clear to me that the Army and Ministry of Defence (MoD) officials provided the Secretary of State for Defence with highly misleading and, in parts, factually inaccurate advice about the FRU's handling of Brian Nelson. As a direct result of this advice, the submissions sent by the Secretary of State for Defence to the Attorney General, and by extension to the DPP(NI), were seriously misleading. When the inaccurate information that the Army and the MoD had supplied was exposed by the Attorney General, the MoD nevertheless refused to re-examine its position and failed, despite the wishes of the then Secretary of State for Defence, to ensure that an investigation was subsequently carried out as to why Nelson's intelligence had not been used to prevent UDA attacks. The documentary evidence also shows that senior RUC officers provided contradictory and, in parts, highly misleading submissions to the DPP(NI).

Allegations that Government Ministers sanctioned collusive activity

108. As mentioned above and elsewhere in my Report, I am critical of the failure by successive Governments to put in place an infrastructure underpinning the conduct of intelligence agents and handlers, and the manner in which Ministers made representations contending that Brian Nelson should not be prosecuted. However, I should note that I have found no evidence whatsoever to suggest that any Government Minister had foreknowledge of Patrick Finucane's murder, nor that they were they subsequently informed of any intelligence that any agency of the State had received about the threat to his life. Similarly, there is no evidence at all that Government Ministers had any knowledge at the material time of Brian Nelson's targeting activity, nor that they in any way encouraged or directed any form of collusive activity with the UDA.

Lessons for the future

109. It was not part of my remit to make recommendations to the UK Government. However, given the nature of my Report, I have reflected on the key lessons that can be learned for the future.

110. In many respects Northern Ireland has changed dramatically since 1989. The violent, vengeful context of the late 1980s in Belfast is, hopefully, part of the past and never to return.

111. The machinery of the State that was activated to deal with the Troubles has now changed significantly. Many of the organisations mentioned in my Report - the RUC SB, the FRU, the UDR - no longer exist. The British Army no longer patrols the streets of Northern Ireland. Policing and justice powers have been devolved to the power-sharing Northern Ireland Executive. There is no suggestion that lawyers in Northern Ireland today are intimidated by the authorities or provided with inadequate protection by the State.

112. There are, nevertheless, 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 over 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.

Overall assessment

114. In establishing this Review, the Government accepted that there had been collusion in the murder of Patrick Finucane, and indeed apologised for this. In analysing what is meant by collusion, I preferred to adopt the narrower definition used by Lord MacLean in the Billy Wright Inquiry Report, rather than the one adopted by Justice Cory in his Collusion Inquiry Report. Nevertheless, even by reference to that narrower definition, it is clear for the reasons I outline in this Report that the threshold for a finding of collusion is met in this case.

115. Overall, I am left in significant doubt as to whether Patrick Finucane would have been murdered by the UDA in February 1989 had it not been for the different strands of involvement by elements of the State. The significance is not so much, as Sir John Stevens concluded in 2003, that the murder could have been prevented, though I entirely concur with this finding. The real importance, in my view, is that a series of positive actions by employees of the State actively furthered and facilitated his murder and that, in the aftermath of the murder, there was a relentless attempt to defeat the ends of justice.

116. My Review of the evidence relating to Patrick Finucane's case has left me in no doubt that agents of the State were involved in carrying out serious violations of human rights up to and including murder. However, despite the different strands of involvement by elements of the State, I am satisfied that they were not linked to an over-arching State conspiracy to murder Patrick Finucane. Nevertheless, each of the facets of the collusion that were manifest in his case - the passage of information from members of the security forces to the UDA, the failure to act on threat intelligence, the participation of State agents in the murder and the subsequent failure to investigate and arrest key members of the West Belfast UDA - can each be explained by the wider thematic issues which I have examined as part of this Review.

117. In spite of the gravity of my findings, I must also stress that it would be a serious mistake for this Report to be used to promote or reinforce a particular narrative of any of the groups involved in the Troubles in Northern Ireland. My remit has, by its nature, involved only an examination of the actions of the British State and its agents, and loyalist terrorist organisations. I have no doubt, however, that PIRA was the single greatest source of violence during this period and that a holistic account of events of the late 1980s in Northern Ireland would reveal the full calculating brutality of that terrorist group. The abiding impression of this period in Northern Ireland must be of an extremely dark and violent time in which a lawyer could so callously and tragically be murdered as a result of discharging his professional legal duties.