PRIME MINISTER

CHAPTER 1 - Introduction

1.1       On 8 August 1992, Michael John Smith was arrested by the police. On 12 August 1992, he was brought before the magistrates having been charged with four offences under the Official Secrets Act 1911:

(a)                communicating material to another for a purpose prejudicial to the safety or interests of the State, between 1 January 1990 and 1 January 1991 (Section 1(1)(c));

(b)                communicating material to another for a purpose prejudicial to the safety or interests of the State, between 1 January 1991 and 1 May 1992 (Section 1(1)(c));

(c)                making a sketch or note for a purpose prejudicial to the safety or interests of the State, between 30 April 1992 and 8 August 1992 (Section 1(1)(b)); and

(d)                obtaining or collecting material for a purpose prejudicial to the safety or interests of the State, between 30 April 1992 and 8 August 1992 (Section 1(1)(c)).

1.2       Smith was convicted on the first, second and fourth counts at the Central Criminal Court on 18 November 1993 and sentenced to 25 years imprisonment. He was acquitted on the third charge.

1.3       Smith applied to the Court of Appeal (Criminal Division) for leave to appeal against conviction and sentence. On 8th June 1995 his appeal against conviction was dismissed. His appeal against sentence was allowed, and the sentence was reduced from 25 years to 20 years.

1.4       By your letter of 10 January 1994, you asked the Security Commission to investigate and report on the case with the following terms of reference:

To investigate the circumstances in which breaches of security have or may have occurred arising out of the case of Michael John Smith, who was convicted on 18 November 1993 of offences under section 1 of the Official Secrets Act 1911; and to advise in the light of that investigation whether any change in security arrangements is necessary or desirable.

1.5       In our report, below, we outline the significant features of the case. Our main conclusions are set out in chapters 8 to 10 and are summarised in chapter 11.

Conduct of the inquiry

1.6       The Commission first met on 22 February 1994 and on a further six occasions. We completed and submitted our report in August 1994, together with our recommendations. The report could not, for obvious reasons, be published so long as Smiths appeal was outstanding.

1.7       Since completing our original report, certain further matters have come to light as a result of Smiths appeal. In particular, it has emerged that some of the material supplied by MoD on which we based our damage assessment, was seriously incorrect. Furthermore, one of the documents found in Smiths possession on his arrest had been wrongly classified. We met again to discuss this matter on 3 July 1995. We deal with these matters separately in Annex A.

1.8       A particular difficulty in this inquiry was the fact that much of the evidence related to events which happened more than fifteen years ago, long before the period covered by the trial. It was not therefore always possible or practicable to interview people who had been directly involved. Our enquiries therefore concentrated primarily on evidence from those in the Security Service, in the Ministry of Defence and in the companies for which Smith worked who are currently responsible for the relevant areas. We were however able to benefit from the evidence of some people who had been responsible at the time for security in the relevant companies. The Commission is grateful for the assistance it has received from all these witnesses. They are listed at Annex B.

1.9       This was the first case to be referred to the Security Commission involving someone in industry. It became apparent at the start of the inquiry that the main issues at stake were: (1) the initial identification of Smith as a communist; (ii) the delay in removing him from classified work once he had been so identified; (iii) the decision taken some eight years later to grant him access to classified information again; and (iv) the time taken to identify him as the person responsible for passing sensitive information to the Soviet Union. In all of this, the Commission has paid particular attention to the lines of communication between the Security Service, the Ministry of Defence and the companies concerned. In addition, the Commission has noted the extent to which both the security environment and security procedures have evolved since the events which form the main focus of this inquiry took place. We have attempted to reflect this appropriately in our conclusions.

1.10      The Commission was supported throughout the inquiry by the Secretariat, Mr R D J Wright and Mr M J Sterling of the Cabinet Office. Mr Wright has now been succeeded by Mr J K Barron. We are greatly indebted to them all for their assistance.

CHAPTER 2 - Michael John Smith

2.1       Michael John Smith was born on 22 September 1948. Between 1960 and 1967 he attended Ockendon Courts County Secondary Modern School, Essex, where he obtained nine O levels and four A levels. Smith then studied at the University of Surrey for four years and he gained a BSc Honours Degree in Electronic and Electrical Engineering in June 1971. Smith worked as a trainee assistant electronic engineer at W & T Avery Ltd, Smethwick, Warley from November 1971. In April 1972, Smith moved to work as a junior engineer for Rediffusion in Chessington.

2.2       In July 1976, Smith started work as a test engineer in the Quality Assurance department of EMI Defence Electronics at Feltham, Middlesex. In this job he needed a security clearance to allow him access to material classified up to SECRET. Accordingly, before his appointment, EMI applied for a Normal Vetting (NV) clearance for Smith. This was granted, giving Smith access to information classified CONFIDENTIAL and SECRET but not TOP SECRET, and between July 1976 and May 1978 Smith was employed by EMI on a wide range of work which brought him into contact with both classified and unclassified material.

2.3       In May 1978, Smiths security clearance was revoked. To ensure he had no further access to classified information, Smith was moved to work on projects wholly outside the defence field at EMI Medical at Hayes. Following an internal reorganisation, Smith was made redundant by EMI on 30 September 1985.

2.4       In December 1985, Smith began work as a Senior Quality Assurance engineer and manager with GEC at the Hirst Research Centre (HRC). This work required him to have access to CONFIDENTIAL material and on 9 July 1986 he was granted a partial clearance which would enable him to have such access. In July 1992, he was made redundant by GEC due to the abolition of his post and the general reduction in the numbers of staff employed at HRC.

2.5       He was arrested on suspicion of offences under the Official Secrets Act on 8 August 1992.

CHAPTER 3 - The Assessment of Smith

Action by the Security Service

3.1       Michael John Smith first came to the notice of the Security Service in November 1971 when a Michael Smith living in Birmingham applied to join the Communist Party of Great Britain (CPGB). Efforts were made at that stage by the Security Service and the police, at the formers request, to identify Smith but without result.

3.2       In January 1973, a Michael John Smith, now known to be the same person, separately came to the notice of the Security Service as having attended a District Congress of the CPGB in Surrey. The report gave a Chessington address for him and his occupation as an engineer. No connection was made with the Smith in Birmingham because of the common surname and the different address. The report was wrongly put away on an existing file for another Michael John Smith, a fully identified member of the CPGB living in Surrey, unconnected with the subject of this inquiry. Although the existence of two CPGB members in Surrey called Michael John Smith may seem an improbable coincidence, we have confirmed that this was in fact the case.

3.3       The Security Service received further reports on a Mike Smith, M Smith or Michael Smith as being active in the Surrey CPGB between 1973 and 1975. At that time, there were 25,000-30,000 members of the CPGB and the aim was to resolve as many new identifications as possible within the Security Service before approaching the police for help. The Security Service accordingly undertook its own inquiries for three years before seeking help from the Metropolitan Police. Nothing in these routine enquiries pointed either to the same man or to the Michael Smith from Birmingham.

3.4       In October 1976, the Security Service wrote to the Metropolitan Police asking for their help in completing the identification of Michael John Smith. On 29 March 1977, an interim report was received giving full particulars on the Michael J Smith known to be active in the Surrey District of the CPGB. This report indicated that he was working for EMI, a list X company (see Annex C). This report was confirmed in May 1977. A permanent Security Service record on Smith was made up by 24 June 1977 which drew together all the previous material on Smith, including the papers which had originally been misfiled.

3.5       On 4 July 1977, F Branch of the Security Service (responsible for counter-subversion) drew Smith's subversive record to the attention of C2, the section responsible for advising List X firms on security matters. Although on 24 August 1977 the C2 Adviser visited EMI on a routine visit, no mention was made of Smith. On 7 February 1978, the C2 Adviser visited EMI and discussed the Smith case with the firm.

3.6       The period from 1971 to 1978 is the first critical period in the case. In 1971, the Security Service had information that a Michael Smith in Birmingham was a communist. By March 1977 it had linked intelligence on the Smith active in the Surrey CPGB between 1973 and 1976 with information that he was working at EMI. This information was confirmed in May 1977. By July 1977, three months after the initial identification, this information had reached the C2 Adviser within the Security Service whose responsibility it should have been to raise the case with EMI. However, this was not done until February 1978. No clear explanation has emerged for this delay. The Security Service has been unable to help us from its records and the C2 Adviser concerned is no longer alive.

Action with EMI

3.7       In February 1978, once it was recognised that action needed to be taken with EMI to resolve Smiths position, the Security Manager at EMI Feltham was asked to prepare a profile of Smith as a matter of urgency. The Security Service took the view that given Smith's connection with the Communist Party he should be given a "B" vetting assessment, i.e. that he should not be given access to information classified CONFIDENTIAL or above unless there were overriding departmental reasons. Between February and May 1978 there were a number of discussions between the Security Service, the Ministry of Defence and EMI about Smith. A decision was taken to move Smith from classified work.

3.8       Details of the vetting assessment process are set out at Annex D. At that time, the NV procedure was covert. Employees were not normally aware that they had been subject to Security Service and criminal record checks and it was regarded as important that the NV process remained covert both in its own right and to protect sources of information. Accordingly, it was considered important that Smith should not be alerted to the questions being raised about his security clearance, and particular emphasis was placed on finding a way of moving him away from classified work without arousing his suspicions. In May 1978 it became possible to move him on promotion to EMI Medical at Hayes, and thereby to remove the need for him to have access to classified material.

3.9       In June 1978, the Security Service had obtained from the police conclusive proof that the Smith identified in Birmingham in 1971 was in fact the same person as the Smith in Surrey working for EMI. This served to confirm his earlier activities as a CPGB member in Birmingham.

3.10      Between May 1978 and November 1979, Smith tried to get back onto classified work at EMI. He had not been told directly why he was not being allowed to move back into defence-related work. He assumed, perhaps not surprisingly given what he knew about his own activities, that he was being baulked on security grounds. He therefore asked to see the EMI Security Manager. Following discussions with the Security Service and MoD, it was agreed that the security manager should interview Smith to see if he would volunteer information on his previous political background. At an interview on 12 November 1979, despite being given ample opportunity to make a clean breast of his political past, Smith claimed that there was nothing in his background which would preclude his getting a security clearance. He volunteered a political leaning to the Liberal party.

3.11      We now know that Smith met his handlers in Vienna in August 1979 where he was subjected to, and passed, a lie detector test. He did not tell the KGB until August or September 1979 that he had lost his clearance. It is not entirely clear whether thereafter Smith was being urged by the KGB to attempt to regain his security clearance or whether he was doing so at his own initiative. From the KGBs point of view, there may have been advantage in Smith making a fuss about his clearance, whatever the outcome. At best, MoD would allow Smith access again to classified information. At worse, it would give the KGB some idea as to the strength of the case against Smith. Nonetheless, these efforts could have backfired. If the Security Service had found out at that stage about Smith's espionage activities, as distinct from just his political activities, he would almost certainly have been arrested much earlier, or the Security Service would at least have taken steps to restrict the damage he could do. There were risks, therefore in Smith's attempts to regain his clearance. In fact, his efforts to do so served to reduce, rather than to increase, any suspicions about his activities, the Security Service later reaching the view that if he really was a spy he was unlikely to risk drawing attention to himself.

3.12      During the interview at EMI Smith sought the opportunity to discuss his position with MoD. In further discussions between the Security Service, MoD and EMI, it was agreed that there was little alternative but to agree to give Smith the further interview he had sought in order to resolve the questions about his political affiliations and activities one way or another. In preparation for this interview Smith completed a standard security questionnaire and in response to the question asking whether he had been a member of or in sympathy with any communist organisation answered no. He did, however, give details of others (a former teacher of his and other acquaintances) whom he knew to espouse communist or Trotskyist beliefs.

3.13      On 10 June 1980 Smith was interviewed by a Security Service officer acting under MoD cover. Smith was directly challenged about his affiliations to the CPGB and the Young Communist League (YCL). Initially, he denied these connections. Eventually, however, after being confronted with some of the Security Services information, Smith admitted his involvement with the YCL between 1971 and 1976. He claimed to have resigned in 1976 because he was disillusioned with the party. He said he bitterly regretted his past involvement. In the opinion of the interviewing officer, it was impossible to tell whether Smith was concealing anything further. It was recommended, however, that as Smith was a proven and persistent liar, it would be unwise to grant security clearance and that he should not be allowed access to information classified CONFIDENTIAL and above. It was agreed that Smiths security clearance should be reviewed after two years.

3.14      In November 1980, Smith wrote to the Director of Security (Procurement Executive) in MoD complaining about the denial of his security clearance. In his allegations, Smith included information about breaches of security at the EMI site at Feltham which could only have been obtained from people working there. At MoDs instigation, EMI undertook an investigation into these claims and submitted a report on their conclusions to MoD. Although the scope of the investigation was constrained by the perceived need not to alert Smiths contacts at EMI, it concluded that there was no foundation for Smiths claims. Immediate steps were taken, however, to constrain his further access to the Feltham site.

3.15      Around this time it appears Smith also considered raising his case with the Prime Minister. A draft of a letter to the Prime Minister, never apparently sent, was found in Smiths home following his arrest.

3.16      On 24 June 1982 the Security Service reviewed Smiths security clearance. They wrote to MoD stating that as Smith had not come to adverse notice since 1976, they would be prepared to make a C assessment - i.e. that Smith should not necessarily be debarred from access to information classified CONFIDENTIAL or above, but that the department might wish to make other arrangements, especially if access to delicate material was involved. This would in principle have allowed Smith to have access to material classified CONFIDENTIAL and some SECRET but not TOP SECRET material.

3.17      The change in the level of assessment was based solely on the six years that had elapsed since Smith had been active in the CPGB. The Security Service qualified this relaxation heavily, however, by advising that, despite the relaxation from a B to a C assessment, the MoD should have regard to the fact that Smith had lied and concealed the truth in completing his security questionnaire. There was accordingly a presumption of unfitness for clearance under the criteria for assessing trustworthiness outlined in the Manual of Personnel Security Measures, i.e. that the individual had deliberately omitted significant information from, or falsified ... a Security Questionnaire; and has shown himself ... to be dishonest, untrustworthy. The Security Service said that MoD might feel able to discount this if sufficient positive information was received on Smiths character with particular regard to his reliability and discretion. In the event, MoD continued to deny Smith access to classified information.

3.18      From information learned subsequently, it is now known that Smith was stood down by the KGB in 1983 or 1984.

3.19      Smith was made redundant by EMI on 30 September 1985. During September 1985, Smith made applications to work for Marconi Defence Systems Ltd and Gresham Lion plc. In both cases security questionnaires were completed and the Security Service repeated their earlier assessment of Smith, i.e. that he remained clear of communist activity but that in view of his past untrustworthiness the heavily qualified C assessment would have to be maintained. They advised that the subject need not necessarily be debarred from access to information classified CONFIDENTIAL or above but that MoD (as the contracting department) might wish to make other arrangements if delicate material was involved. MoD continued to deny clearance.

3.20      In December 1985 Smith joined GEC Hirst Research Centre as a senior quality assurance engineer and signed a copy of the Official Secrets Act declaration. GEC submitted a security questionnaire to the Security Service on 10 January 1986. The Security Service advised MoD that they held to their earlier assessment.

3.21      GEC indicated to MoD that they wanted to employ Smith on work classified up to CONFIDENTIAL. They indicated, however, that they would be able to keep his access down to a minimum. The Security Service visited GEC on 21 May 1986 to assess what level of access to classified information Smith would need. On the basis of that visit and the latest vetting advice given in September 1985, MoD agreed that Smith could have access to classified information not exceeding CONFIDENTIAL, subject to his needing access for his work and to periodic review of this access by the Security Service Adviser. MoD wrote to GEC on 9 July 1986 confirming that Smith could be allowed access to classified information up to the level of CONFIDENTIAL, subject to the need to know.

CHAPTER 4 Smiths Espionage Activities

4.1       Smith was recruited as a spy by Viktor Oshchenko, who befriended him at a union meeting in 1975. At Oshchenkos bidding, Smith severed his links with his communist associates and sought a new job with a view to his playing a more active open role in the political or trade union spheres. At this stage, the KGBs objective appeared to be to enable him to establish himself in positions where he could influence developments in favour of the Soviet Unions aims and report back inside information of potential value to the KGB. With the KGBs support, Smith applied for a job with EMI. To his case officers surprise, he got the job. Smith eventually became a fully recruited agent and passed information to the Russians certainly until Oshchenko left the UK in late September 1979, and possibly thereafter. Smith was reactivated by the KGB in 1990.

Espionage at EMI

4.2       An intelligence lead which reached the investigating section in the autumn of 1984 indicated that the Russians were in possession of information relating to a radio (sic) fuse for a Western atomic bomb. Analysis of the source material indicated that the compromise had occurred via the KGB Residency in London, probably in late 1978 or early 1979.

4.3       The nature of the lead was such that, initially, it was impossible to be certain whether the fuse in question was of British or American origin. Initially, the US authorities were adamant that the material that had been compromised was theirs. After much debate, it was concluded in 1987 that the compromised information in fact related to the British XN-715 radar fuse for the British free nuclear bomb. This had been developed principally by Thorn EMI in conjunction with a number of MoD research establishments.

4.4       The Security Service investigation of the lead concentrated initially on attempting to identify documents relating to the project which might have been compromised. This was not straightforward in view of the technical nature of the material, the substantial number of papers in circulation and the ambiguity of the available intelligence. In order to allow coherent analysis, a full document audit of the relevant technical assessments and specifications was undertaken. Research produced a list of something in excess of 150 possible candidates, at that stage including Smith, who had access to relevant documents. Further refinement, completed by 1989, produced a short-list of six documents with an authorised distribution, not including Smith, of 36 individuals.

4.5       By December 1991, a clear picture had emerged of the XN-715 project, the people involved and document security in the MoD branches, RAE Farnborough and Thorn EMI, A short-list of four potential suspects had been drawn up of whom Smith was one. In the context of the investigation, Smith was initially given little weight as a candidate because he had been transferred from the project before the date on which the material was believed to have been passed to the KGB. However, in March 1992, a defector examined the lead material and concluded that although the lead was dated February 1979, it could include material offered up to six months before this date. In March 1992, therefore, Smith became the most likely culprit. His involvement was finally confirmed conclusively by the defection in July 1992 of Viktor Oshchenko who had recruited and run Smith up to September 1979.

Espionage at GEC

4.6       Despite the fact that the most serious of Smith's known espionage activities occurred whilst he was working for EMI, as outlined in Chapter 6, his trial was confined to charges under the Official Secrets Act relating to the documents taken from GEC in his possession at the time of his arrest. The case for the prosecution was that Smith was an agent for the Russian Intelligence Services and that between 1990 and 1992 he had communicated to his handlers material intended to be of use to the Russians.

4.7       It was assumed that the documents in his possession at the time of his arrest were collected for his last delivery before he was made redundant by HRC. There is no way of knowing what other information he may have passed since his reactivation by the KGB in September 1990.

CHAPTER 5 - The Intelligence Operation and Smiths Arrest

5.1       On 25 July 1992, Viktor Oshchenko, a senior officer of the Russian Intelligence Service stationed in Paris, defected to the west and was granted political asylum in the United Kingdom. He confirmed that he had recruited and run Smith as a KGB agent in the period 1975-79. It was decided that Smith should be arrested as quickly as possible in order to prevent him covering his tracks once Oshchenkos defection became public knowledge.

5.2       On the morning of Saturday 8 August 1992, a member of the Security Service telephoned Smith at his home in Kingston-upon-Thames. The officer introduced himself as George, a colleague of Smiths old friend Viktor, and explained that it was very urgent that they should meet. He arranged to call Smith at a telephone kiosk near his home.

5.3       Shortly afterwards, Smith left the house and walked to the telephone kiosk, under surveillance by the Metropolitan Police. Due to a breakdown in communications, Smith did not receive the planned telephone call and he returned home. He was then arrested by plain clothes officers for offences under the Official Secrets Act.

5.4       Over the next three days, while Smith was being interviewed at Paddington Green Police Station, his house and car were searched. Three envelopes were found in a bedroom, two of which contained £1,000 each in £50 notes. Analysis by the Special Branch of Smith's financial affairs revealed unexplained income of over £20,000 in the period from October 1990 to June 1992. The third envelope contained a letter dated 24 September 1990 signed Williams which served to recommission Smith as a Soviet agent. It also included four sheets of paper in Smiths handwriting apparently referring to clandestine meetings. One of the notes was apparently a shopping list of projects he was targeting including the micronvalve project at HRC. Expert evidence was given at the trial by the Security Service and Oleg Gordievsky, a former KGB officer, to the effect that the notes were typical of the Russian Intelligence Services tradecraft.

5.5       In the boot of Smiths car, police found a plastic bag full of documents and some components. Amongst the documents were handwritten notes headed Micromachining Project, Micron-Valve Project and other subjects. Smith would have no need of such documents in the course of his normal employment. In addition, Smith had also removed a number of documents from HRC without permission. These included specifications for devices employed in the Rapier missile system and in smart weapons, radars and electronic warfare systems, and a document now known to have been wrongly classified (see Annex A) relating to another device in a current weapon system, to which Smith should not have had any access and which it appears he had stolen from the filing cabinet of a former colleague in his room.

CHAPTER 6 - Damage Assessment

6.1       In relation to the XN-715 fuse for the WE177 free-fall bomb, there is no evidence that the former Soviet Union took any operational advantage of this information, although as the result of Smiths spying it had the technological capability to do so. The damage from Smiths compromise was therefore potentially extremely high.

6.2       The material known to have been obtained by Smith during his time at GEC which led to the charges and his conviction under the Official Secrets Act, was a mixed bag. Some was already in the public domain and some was of value more for its commercial than for its military potential. But a number of documents contained more sensitive material, relating to weapons systems. The potential damage to the UK overall including the Rapier is assessed as considerable. In the case of the other current weapon system, the detailed information contained in the document which should have been classified SECRET would have enabled an intelligent enemy to deduce operating parameters which would have allowed counter-measures to be developed. The potential damage to the national interest in the case of that weapons system is assessed as serious.

CHAPTER 7 - Changes Made by The Security Service and Ministry of Defence Since the 1970s

7.1       The inquiry by the Security Commission had to be conducted in the context that the events in question took place some years ago. Much has changed since then. For example, a revised policy on vetting was introduced in November 1990, and a further revision was announced by the Prime Minister in a statement to the House of Commons on 15 December 1994. The main change in relation to the Smith case was that Normal Vetting, which had been a covert process in 1976, was superseded by an overt system of vetting in 1990.

7.2       In addition, the Security Commission has been called on to investigate a number of security breaches since that time and its recommendations have led to significant improvements in the area of vetting and more widely.

7.3       The Smith case also needs to be viewed in the light of the prevailing circumstances of the late 1970s. At that time, much of the Security Services effort was directed against the threat from subversive elements in the UK. The threat from subversive organisations has decreased to the point where it is assessed by the Security Service as low. The CPGB no longer exists. The main surviving organisation in the UK is the Communist Party of Britain (CPB) which is assessed to be only about 1100 strong, compared to 25,000 to 30,000 in the CPGB in the early 1970s. The reduction in the threat has been such that it has been agreed interdepartmentally that the investigation of subversive organisations can be reduced accordingly.

7.4       The system for prioritising the actions of C2 officers in dealings with List X firms has also been changed, although the essential features, until recently supplemented (see paragraph 8.8), have remained much the same.

7.5       Since the 1970s, the Security Service has made a number of changes to the organisation and procedures within F Branch in relation to the identification of individual subversives. These have included the computerisation of traces. In addition, the practice whereby F Branch was used as a training ground for new recruits has now ceased.

CHAPTER 8 - The Granting and Withdrawal of Smiths Security Clearance

8.1       The significant features of this case fall into three main episodes. First, there is the time from Smiths original coming to attention in 1971, through his initial clearance in 1976 to the withdrawal of his clearance in 1978. Second, there is the period leading up to the partial reinstatement of his clearance in 1986. Third, there is the period between learning in 1984 that highly sensitive information on a western nuclear fuse was in Soviet hands by 1979 and the linking of this compromise to Smith in 1992.

8.2       This and the following chapters examine in more detail the lessons that can be learned for the future from each of these episodes.

The granting and withdrawal of Smiths clearance

8.3       In identifying people who may pose a security risk, the task is to match information on people who are likely to pose risks with information on people who have access to sensitive government information. Where there is such a match, action needs to be taken.

8.4       It is clear from the evidence we have received that, in the 1970s, the number of those identified as posing a potential threat was very high. In particular, the CPGB and other organisations avowedly committed to undermining Parliamentary democracy numbered many thousands of members and associates. This is no longer the case, and this change alone has made the Services task in monitoring potential subversives significantly more manageable.

8.5       Evidence from the Smith case, however, indicates a number of lessons that can still usefully be learned. First, the matching of traces to people and people to traces should always be a two-way process. In this case, the Service had held information since January 1973 that a Michael John Smith was a member of the CPGB in Surrey. Attempts were made thereafter to identify this individual more conclusively. Separately, in 1976 EMI sought approval for a Michael John Smith living in Chessington, Surrey to be given access to information classified up to SECRET. We accept that the commonness of the name was a complicating feature in this case. Nonetheless, we find it surprising that EMIs application never triggered a connection with the information from 1973. The explanation we have been given is that Smith did not at this stage have a personal file. Because of the volume of cases, a personal file would only have been opened after the identification of an individual had been firmly established. There was no system for checking against temporary files. The matching of traces to people was thus incomplete.

8.6       We are glad to be assured that the matching process is now complete and comprehensive. Temporary files would now be checked in similar circumstances. This together with the introduction of computer systems and the reduction in the number of organisations and individuals presenting an ideologically motivated threat to the interests of the UK, reduces significantly the likelihood that, when the Service is in possession of both the elements of a match as they were in this case, the match would fail to be made. We recommend that the Security Service keeps its organisation and internal lines of communication under continuous review to ensure that this remains the case.

8.7       Second, there was an unwarranted delay between the matching of the information on Smiths access to sensitive material with the information on his background and the taking of any follow-up action with EMI to resolve this. As already indicated, this delay was particularly damaging to the interests of the UK: it was while Smith continued to have access until May 1978 that he acquired the especially sensitive material which was later passed to the Soviet Union.

8.8       The Security Service accepts that there is no satisfactory explanation for this delay in communicating with the company other than a failure on the part of the individual concerned and on the part of the Services monitoring systems. The Security Service outlined to the Commission the systems in place to ensure that C2 Security Advisers do not overlook outstanding action with List X firms. Matters of security interest that come to light are normally raised with firms immediately by letter or telephone. Only trivial matters are left to be raised at visits. Before a visit the Adviser calls up the file for that firm, which should contain requests for any action by the Adviser. In any contact with a List X firm about an individual, records of that contact are placed on both the firms and the individual's files. The Service accepts these safeguards failed on this occasion and it has taken steps following earlier discussion with the Commission to strengthen its arrangements for prompting essential follow-up action on the part of desk officers. We welcome this response on the part of the Security Service and have no further recommendations to make in this respect.

CHAPTER 9 - The Partial Reinstatement of Clearance in 1986

9.1       From 1978 when his clearance was withdrawn, Smith made repeated attempts to have it reinstated. These were partially successful by 1982. At that stage, the Security Service revised their assessment of Smith. In line with their standard practice, the Service concluded that since Smiths last known communist associations were six years old, and since he had not come to notice in the meantime, there was no objection on association grounds to his having access to information classified up to CONFIDENTIAL if he needed such access for his work. On character grounds, however, the Service continued to have reservations. It accordingly urged the clearance authority (in this case MoD) to weigh carefully in the balance the fact that Smith was a proven liar before granting him any access to classified information.

9.2       MoD did so. Smith was accordingly refused clearance consistently until 1986. In December 1985, however, having been made redundant by EMI, Smith joined GEC. In January 1986, some weeks after he had joined the company, GEC sought clearance for Smith. MoD checked with the company on the degree of access Smith would need and was told that access up to CONFIDENTIAL would be essential for the job for which he had been recruited but that this access would be fairly limited. MoD at this stage then decided to grant him access up to CONFIDENTIAL, on a need to know basis and subject to regular review by the Security Controller. However, MoD told the Company only:

that we are prepared to accept that Mr Smith may be allowed access to classified material up to the level of CONFIDENTIAL, subject to the need to know.

The Company observed this express constraint, as far as we can tell, as scrupulously as it could. However, MoD failed to inform anyone in writing that Smiths access should be subject to regular review. This failure meant that the Company never put in hand such arrangements. Nor did the Security Service do anything to monitor how Smith was doing in his regular dealings with the Company thereafter.

9.3       A number of lessons arise from this phase of the case. First, we are concerned about the grounds on which the Security Services recommendation on Smiths clearance was relaxed. Clearly, the Service is entitled to adopt guidelines in performing its functions and advising on clearances generally. But guidelines should not be applied inflexibly. We consider that the way in which the guidelines were applied in this case, on the basis of which the Security Service concluded that objections to Smiths clearance could be withdrawn solely because he had left the CPGB six years before, was too rigid and mechanistic. We understand that changes have already been made to these guidelines and that further discretion will be built into the decision-making process in the future. We welcome this.

9.4       We are also concerned about the way in which MoD in the event agreed to Smith having access to CONFIDENTIAL information. There is no evidence of there being any system for obtaining an up-to-date assessment of his integrity and reliability, for example by seeking and reviewing performance reports from EMI. And despite the fact that the original firm evidence of his lack of integrity had arisen from the interviews he had been given by the EMI security manager and by the Security Service in 1979 and 1980, there was no attempt to reinterview him to make a further assessment of his reliability. We find this surprising.

9.5       We deliberated whether we should recommend that individuals who are proven liars should never again be granted security clearance. There are strong arguments in favour of such a rule. On balance, however, we consider that such an approach would suffer from the same defect that we have just criticised, namely that it would be too rigid. It would not allow proper account to be taken of mitigating circumstances in genuinely difficult cases. It would also, we believe, tip the scales too far against fairness to individuals in favour of national security.

9.6       Instead we propose that lies in the filling in of security forms or in the context of security interviews should be regarded as creating a very strong presumption against the granting or reinstatement of a security clearance. This should be made abundantly clear to candidates for clearance from the outset so that they can decide for themselves whether or not to proceed with seeking a clearance, recognising that lying may have very long term implications. So we recommend.

9.7       More generally, we recommend that wherever after ordinary inquiries significant and otherwise unresolvable doubts remain about an individuals SECRET or TOP SECRET clearance on character grounds, those doubts should always be explored in an interview with the subject before a decision is made to grant access. It will not always be the case that character doubts will be resolved by such interviews. An interview in this instance, given Smiths apparent proficiency and fluency as a liar, may not have produced a different outcome. But we believe that in many if not most cases the judgement to be reached on any clearance, whichever way it may go, will be strengthened by the conduct of a face to face interview. This will of course be easier now than it was under a covert vetting system in the 1970s and 80s insofar as the vetting system itself is now wholly overt. We expect our recommendation to apply equally to initial clearances, to reviews and to appeals for reinstatement following a refusal or withdrawal of clearance and to apply to people employed directly in Government departments and the Services or by contractors in industry.

9.8       We have also looked carefully at the lines of communication between the Security Service, the contracting department (in this case MoD) and the contractor in respect of clearances for staff in industry. These involve the Security Service recommending a level of clearance on the basis of the information available to it; the contracting department determining a level of clearance on the basis of the Security Services (C3 Section - responsible for vetting) advice and any further information available to it; the contractor applying the outcome; and the Security Service (C2 Section) then remaining in touch with the contractor both to advise it on ways of achieving the outcome and to ensure that it complies with any requirements laid down by the contracting department (including the general rules laid down in Security in Industry).

9.9       This is a long chain of communication, and potentially confusing. We do not consider there is anything necessarily wrong with this, however, as long as everyone is clear as to their own role and as long as communications are themselves abundantly clear.

9.10      The Smith case highlights two possible sources of confusion which we consider need to be addressed. First, the definitions of the different levels of assessment are unnecessarily complicated. They also make it too easily for responsibilities between the Security Service and the contracting department to be muddled. We recommend that these definitions be modified so that they convey unambiguously the thrust of the Security Services advice. The Security Service should also ensure that, if they are aware that a third party holds relevant information on a subjects character and conduct, departments attention is drawn to this. They should make it clear that it is for the employing or contracting department to obtain this information and to determine what level of access is granted in the light of this and any other information available to it.

9.11      Second, and closely related to this, we consider it essential that, in industry, the contracting department works jointly with the contractor to review the character and conduct of individuals. The contracting department cannot, clearly, reach such an assessment without the help of the individuals employer; nor can the assessment be left solely to the employer. We therefore recommend that decisions on the clearance of individuals in List X companies are taken by the contracting department only after appropriate consultation with the employing Company.

9.12      In these instances, it will be for the employing Company to apply any constraints on clearances. To do so effectively, they need to be entirely clear as to what is expected of them. We therefore recommend that the Security Service, in consultation with contracting departments, reviews with industry how the levels of access granted to individuals are described to ensure that they are capable of effective implementation; if they are not, the Security Service should agree with representatives of industry alternative definitions which can be given proper effect in the manufacturing and service sectors.

9.13      We also consider a further defect in the system lay in the failure to arrange for the monitoring of Smiths behaviour after his clearance was partially reinstated. We do not, as we make clear earlier, hold the Company responsible for this. In future, however, we recommend that when a partial or provisional clearance has been given to an individual, or a where clearance has been given with significant doubts attached to it, the contracting department should inform the Security Service. The Security Service should put in hand arrangements to ensure that the behaviour and performance of such individuals is explicitly reviewed with the employing Company on every visit thereafter until such time as the individual comes off the watch list. This implies that the Security Service will need to take the contractors Security Controller (or, sometimes the Board member with responsibility for security) more closely into its confidence in such cases. We consider that, subject to the protection of sources, this would generally be beneficial rather than threatening to overall security.

CHAPTER 10 - The Linking of The Compromise To Smith

10.1      The first indication that a major compromise had taken place in 1978/79 was received in 1984. This was only partial and ambiguous. It was such as to allow for a number of different interpretations as to what it was that had been compromised, and even whether the information had originated from the US - as assumed at first by the US authorities - or the UK. The initial investigation of this intelligence therefore went down what can now be seen to be a number of blind alleys. This is not unusual in this sort of work, all of which requires painstaking and detailed analysis.

10.2      The field of those who were known to have had relevant access to the project information was very large, in excess of 150 people. Smith was amongst these and his candidacy was specifically considered in 1987 and again at frequent intervals during 1988.

10.3      By June 1989, the field of investigation had narrowed significantly. The subject of the compromise appeared clear. The investigation had focused on six documents and a possible 36 people amongst those who had had access to them. Smith was not amongst these. The Service considered it highly unlikely that he was the source, since the original intelligence suggested that the leak had taken place after he had been denied access to the material in question. It had never before been known for an agent to hold information for any length of time before passing it on.

10.4      We understand and appreciate how the Security Service may have reached this conclusion (although we comment later on the consequences of the Service accepting as readily as it did the validity of its own judgements in this and other areas). We are, nonetheless, critical of the approach adopted by the Security Service in such a serious case in two respects.

10.5      First, of the 150 candidates, Smith was the only one who is known to have come to notice on security grounds. Whilst it is always dangerous to jump too early to obvious but not necessarily correct solutions, we remain surprised that the timing of the leak was seen as a factor weighing more heavily in Smiths favour than his security background weighed against him.

10.6      Second, despite the fact that Smith remained a candidate, however loosely, for what was a very serious security breach, the Security Service took no action to restrict his continuing access to classified information. They could, for example, have approached GEC to see if some further limitation could have been placed on his access to classified information or at least to ensure that his behaviour was closely and effectively monitored. If this had been done, Smith might well have been prevented from passing over information from GEC to the Soviet Union between his reactivation by the KGB in September 1990 and his arrest in 1992.

10.7      Certainly, by the time the list of candidates had narrowed to four, including Smith, in December 1991, the Security Service would appear to have had sufficient grounds to take action to limit any further damage. We were told that the Security Service did consider taking such action but decided not to do so. They took the view that it could not have been done without alerting Smith to the fact that he was under suspicion. This would have enabled him to cover his tracks and so escape prosecution. It might also have put sources at risk.

10.8      Clearly the handling of such cases must always be a matter of judgement. But we are bound to question whether the judgement in this case was correct. The result in practice was that the investigation of the 1978/79 leak and the prosecution of the culprit was afforded a higher priority than preventing any further risk to national security.

CHAPTER 11 - Summary and Conclusions

11.1      Although Smith was convicted on the evidence of his espionage whilst employed at GEC, it is likely that he in fact inflicted even greater damage to UK interests during the earlier period when he was still employed at EMI and it is that period that has inevitably attracted much of our attention in both our investigation and this report. Regrettably, our inquiry has shown that the handling of Smith from a security viewpoint and the investigation which eventually linked Smith to the compromise of the XN-715 fuse was characterised by a number of delays and errors, and by judgements which, although founded on experience, proved to be incorrect.

11.2      In particular, the evidence we have received has indicated that, in addition to the commonness of his name, a number of other features of the case came paradoxically to work in Smiths favour. First, he had been a member of the CPGB. The Security Service knew that, since the 1960s, the KGB had as a matter of policy avoided recruiting Communist Party members as agents. They therefore judged it unlikely that Smith was a spy. Second, Smith had left the CPGB. Although membership of the CPGB was a barrier to the granting of a security clearance, the severance of his links meant that he was again able to be granted a partial security clearance. Third, he had attempted to have his clearance reinstated. This ran counter to previous experience which suggested that genuine agents would do anything to avoid drawing attention to themselves. Fourth, he had held on to valuable information before passing it to his handlers in the KGB. Past experience suggested that agents would always pass on information quickly so as to avoid being caught red-handed.

11.3      We should stress that we do not question the value of guidelines in general or the need to learn from experience. Judgements based on experience have an important part to play in securing the effectiveness of the Services operations. We question only the extent to which they governed decision-making in this case. We note in this connection the dictum that in espionage, the only rule is that there are no rules. Smith was successful because, temporarily at least, the Service forgot this.

11.4      Finally, we would make some general comments on the vetting clearance process as a whole. As we have noted earlier, the granting or withholding of clearances is always a matter of judgement. In difficult cases, a balance has to be achieved between the need adequately to protect sensitive material and information and the need to be fair to individuals. If clearances are withdrawn or withheld, people can lose their jobs or suffer curtailment of their career aspirations. It is to the credit of government departments that they try to achieve a proper balance in these respects, rather than merely adopting a safety first attitude, possibly to the detriment of individuals. Regrettably, however, on this occasion MoD decided to reverse its original judgement and so granted him a limited clearance.

11.5      We note generally that a number of complicating factors arose in this case because the vetting system at the time was covert. Now that it is overt, we consider it important that departments and agencies use this greater openness constructively to allow them to be more robust in, and more directly accountable for, their judgements on vetting. To be so, current appeal arrangements may need to be strengthened to give individuals both in the Civil Service and in industry the opportunity to present their case against, and departments the requirement to defend, vetting decisions which affect them adversely. Had such an arrangement been in place at the time, we believe MoD might more readily have been able to resist the granting of a clearance to Smith, even though it might have resulted in his losing his job with GEC. As a QA engineer, Smith had many employment options other than those involving classified information. It would have been better if he had been forced to look outside the defence field for work.

11.6      Summary of the Security Commissions recommendations

(a)          The Security Service should continue to keep its organisation and internal lines of communication under continuous review (paragraph 8.6).

(b)          Lies in the filling in of security forms or in the context of security interviews should be regarded as creating a very strong presumption against the granting or reinstatement of a security clearance. This should be made abundantly clear to candidates for clearance from the outset so that they can decide for themselves whether or not to proceed with seeking a clearance, recognising that lying may have very long term implications (paragraph 9.6).

(c)          Wherever after ordinary inquiries significant and otherwise unresolvable doubts remain about an individuals SECRET or TOP SECRET clearance on character grounds, those doubts should always be explored in an interview with the subject before a decision is made to grant access. We expect our recommendation to apply equally to initial clearances, to reviews and to appeals for reinstatement following a refusal or withdrawal of clearance (paragraph 9.7).

(d)          The definitions of different levels of assessment should be modified so that they convey unambiguously the thrust of the Security Service's advice. The Service should also ensure that, if they are aware that a third party holds relevant information on a subjects character and conduct, departments attention is drawn to this. They should make it clear that it is for the employing or contracting department to obtain this information and to determine what level of access is granted in the light of this and any other information available to it (paragraph 9.10).

(e)          Decisions on the clearance of individuals in List X companies should be taken by the contracting department only after appropriate consultation with the employing company (paragraph 9.11).

(f)          The Security Service, in consultation with contracting departments, should review with industry how the levels of access granted to individuals are described to ensure that they are capable of effective implementation; if they are not, the Security Service should agree with representatives of industry alternative definitions which can be given proper effect in the manufacturing and service sectors (paragraph 9.12).

(g)          When a partial or provisional clearance has been given to an individual, or where a clearance has been given with significant doubts attached to it, the contracting department should inform the Security Service. The Security Service should put in hand arrangements to ensure that the behaviour and performance of such individuals is explicitly reviewed with the employing Company on every visit thereafter until such time as the individual comes off the watch list (paragraph 9.13).

(h)          Judgements based on experience have an important part to play in securing the effectiveness of the Security Services operations. But they should not be given excessive credence (paragraph 11.3).

(i)           Appeal arrangements may need to be strengthened to give individuals both in the Civil Service and industry the opportunity to present their case against, and departments the requirement to defend, vetting decisions which affect them adversely (paragraph 11.5).

LLOYD OF BERWICK

(The Rt Hon The Lord Lloyd of Berwick)

JOHN BLELLOCH

(Sir John Blelloch KCB)

DEREK BOORMAN

(Lieutenant-General Sir Derek Boorman KCB)

TOMBS OF BRAILES

(The Rt Hon The Lord Tombs of Brailes)

July 1995

ANNEX A - MoD Damage Assessment

1.         At our second meeting on 1 March 1994, we asked MoD to supply us with an up-to date assessment of the damage resulting from Smiths activities at EMI between July 1976 and May 1978 and at GEC between December 1985 and the date of Smiths arrest. Mod prepared two documents dated 8 March 1994. The first dealt with the compromise of the XN-715 fuse during the earlier period. The damage was assessed as potentially extremely serious. We have no reason to change our views as set out in Chapter 6 of our report.

2.         The second document dealt with the later period during Smiths employment with GEC. This was the period covered by the evidence at Smiths trial. We were advised that in terms of national security, Smiths activities had caused some damage to the UK, but not serious damage. We had no reason to doubt the accuracy of this advice. Accordingly, the second paragraph of Chapter 6 read as follows:

6.2       The material known to have been obtained by Smith during his time at GEC which lead to the charges and his conviction under the Official Secrets Act, was a mixed bag. Some was already in the public domain and some was of value more for its commercial than for its military potential. The formal assessment is that this information may have caused some, but not serious, damage to the UKs national security.

3.         In May 1995, in the course of preparing for Smiths appeal, a discrepancy between the damage assessment report dated 8 March 1994 and the case presented on behalf of the Crown at Smiths trial became apparent. MoD was asked to prepare a further report to explain the discrepancy. This was done and submitted to the Security Commission. It now appears that one of the documents found in Smiths possession at the time of his arrest, about which there was a good deal of evidence given in camera at the trial, was potentially of great value to an intelligent enemy in connection with one of the UKs current weapons system. It is MoDs view that the document in question should have been classified as SECRET. If it had been properly classified as SECRET, Smith should not have had access to it.

4.         We have taken further evidence from the MoD about the preparation of the original damage assessment report. The document which should have been classified as SECRET was among the documents considered when the assessment was being prepared, But the importance of the document was not recognised by the individual who prepared the damage assessment report, although it had become apparent to those who gave evidence on behalf of the Crown at the trial. The individual did not discuss his report before it was submitted to the Security Commission with those who would, or should, at once have seen that it was incorrect. MoD have accepted that this was a serious error on their part, and have apologised to the Commission.

5.         As for the wrong classification of the document, we have heard the following explanation: at the time the document was created it was not specifically linked to a particular weapons system. It was classified in accordance with the Security Aspects Letter attached to the relevant contract, but this itself was couched in very broad terms. The MoD view is that this indicated that the document should have been classified at least CONFIDENTIAL. We recognise the difficulties involved in assigning the appropriate protective marking to technical documents at the early stage of a project. We also recognise that in some cases information which may be of no great importance in itself only becomes important when aggregated with other information. We do not wish to encourage over-classification. Nevertheless the classification of the document in question was clearly wrong. The difficulties to which we have referred above make it all the more essential that, so far as possible, vetting assessments should be correct, and that the measures which we have recommended in Chapter 9, paragraphs 10-13 should be carried through.

ANNEX B - List of Witnesses

Security Service

Mrs S Rimington           Director General

Members of the Security Service

Ministry of Defence

Mr A G Rucker                        AUS (Security)

Mr A P Stringer                       Director MoD Security

Mr J F MacCulloch       Deputy Director of MoD Security (Scientific and Technical)

Dr D I Weatherley       Head of Technology, Weapons System Sector, Defence Research Agency

Other witnesses

Mr E G S Ley              Group Security Manager, Thorn EMI

Dr S Cundy                  Director, Hirst Research Laboratory

Mr D Barlow               Quality Assurance Manager, GEC

Mr H Coppinger           ex-Security Manager, Thorn EMI

Mr J S McMichael        ex-Security Controller, Thorn EMI

ANNEX C - The List X System

1.         The List X System has for many years been administered by the Security Service on behalf of contracting departments. It was developed originally to cater for the large number of contracts classified CONFIDENTIAL or above by the Ministry of Defence, though a number of other departments have since taken advantage of the system. Companies currently on List X include not only the leading defence contractors, but also main and sub-contractors in the fields of information technology, research and development, reprographics, data processing, technical writing and translation, and the transmission of classified information and its archive. The list also includes companies offering professional services (e.g. management consultants, lawyers, architects, accountants).

2.         Within the C2 (now D1X) Branch of the Security Service officers (known as Advisers) are allocated List X firms and are the initial point of contact for any matters relating to security. Firms are categorised according to the quantity of classified information they hold, and each adviser is allocated a mixture of high and low priority firms.

3.         Companies on List X were issued with guidance in the form of Security in Industry - now replaced by the Manual of Protective Security. This is kept up-to-date by periodic revisions and by the issue of interim guidance based on Physical and/or Personnel Security Notices issued by the Cabinet Office. Companies on List X are also sent security education material and their staff have the opportunity of participating in courses run by the Security Service on a range of protective security matters.

ANNEX D - The Vetting Assessment Process

In 1976 the Security Service used the following definitions in making assessments in normal vetting cases:

A        The Security Service advises that the candidate should not have access to information classified CONFIDENTIAL or above.

B        The Security Service advises that the candidate should not have access to information classified CONFIDENTIAL or above unless you [the department] decide that there are overriding departmental/service considerations.

C        The Security Service advises that the above information should not necessarily debar the candidate from access to information classified CONFIDENTIAL or above but you [the department] may prefer to make other arrangements especially if access to particularly delicate information is involved.