The most important British VENONA breakthrough was the decryption in whole or in part of all but one of the radio messages sent by the Centre to London from 15 to 21 September 1945.55 The intercepts revealed the existence of an inner ring of British agents referred to by the Centre as ‘the valuable agent network’, which had access to British intelligence. Some years later, following lengthy analysis of collateral evidence, the codenames for three members of the network were identified as those of Philby, Burgess and Blunt. At the time, however, clues to the identities of the agents were sparse. By the mid-1950s Burgess was the only one of the three whose codename had been correctly identified in the decrypts.56 Even in the mid-1970s, twelve of the twenty-four codenames in the September 1945 decrypts had still not been identified.57 The unidentified included JACK and ROSA, an apparently important agent couple of whom no significant identifying details were given but who were, at various times, wrongly suspected of being Herbert and Jenifer Hart or Victor and Tessa Rothschild; and a high-level atom spy, KVANT (‘QUANTUM’), who it was later thought might be Bruno Pontecorvo, who defected to Russia in 1950.58 The importance of the intercepts during this single week in September 1945 provided graphic evidence of the serious intelligence loss sustained as a result of the non-interception of Soviet traffic for most of the previous four years. With the exception of that week, only five other telegrams exchanged between the Centre and the London residency were (partially) decrypted. The lack of British VENONA covering the most successful period of Soviet intelligence collection in Britain in the history of the Soviet Union was to prove an enormous handicap to post-war British counter-espionage. Had British VENONA been on the scale of its US counterpart, a number of the Security Service’s most important counter-espionage investigations – chief among them the Cambridge Five – would undoubtedly have been both more successful and more rapidly concluded. The first two important British Soviet agents discovered as a result of VENONA, Klaus Fuchs and Donald Maclean, were both identified not from messages intercepted in Britain but from decrypts of traffic between the Centre and its residencies in Washington and New York relating to their time in the United States.
Access to VENONA was tightly restricted. A list dated 2 October 1952, five years after the Security Service’s first involvement in it, records that only nine officers had unrestricted access to the decrypts; another nineteen had restricted access or were aware of their existence.59 Secretaries were not allowed to open envelopes containing VENONA material.60 As a result of Soviet penetrations of both AFSA and SIS, however, these security precautions were fatally undermined. In 1950 AFSA was shocked to discover that one of its employees, William Weisband, had been a Soviet agent ever since he joined the wartime army SIGINT agency in 1942.61 The son of Russian immigrants to the United States, Weisband was employed as a Russian linguist and roamed around first ASA, then AFSA, on the pretext of looking for projects where his linguistic skills could be of assistance. Cecil Phillips, one of the cryptanalysts who worked on VENONA, remembers Weisband as ‘very gregarious and very nosy’:
He would come around and ask questions about what you were doing . . . He was never aggressive. If you said, as I often did, ‘Nothing important,’ or ‘I’m doing something as dull as hell,’ he would wander off . . . I never heard him offer a political thought. He was around everywhere all the time. He cultivated the senior officers.
Meredith Gardner recalls Weisband looking over his shoulder at a critical moment in the project late in 1946, just as he was producing one of the first important decrypts – a telegram from the New York residency to Moscow of December 1944 listing some of the scientists who were developing the atomic bomb.62 It seems to have been over a year, however, before Weisband was able to pass on this dramatic news to Moscow. After the defection of the American KGB courier Elizabeth Bentley in 1945, the Centre ordered contact with Weisband to be broken as a security measure. Contact was not resumed until 1947.63 It was probably as a result of Weisband’s warning that the Soviet reuse of one-time pads ceased in 1948.64
By the time Weisband was arrested in 1950, the Centre had an even better-informed source on the progress being made by Meredith and his colleagues. In September 1949, a month before he arrived in Washington as the SIS liaison officer, Kim Philby was indoctrinated into VENONA.65 Soon afterwards he reported to Moscow that the atom spy successively codenamed REST and CHARLES, referred to in a number of the decrypts, had been identified as Klaus Fuchs – thus enabling Moscow to warn those of its American agents who had dealt with Fuchs that they might have to flee through Mexico. Among those who made their escape were Morris and Lona Cohen, who later reappeared in Britain using the aliases Peter and Helen Kroger and were convicted of espionage in 1961.66 After his death in 1995, Morris Cohen was posthumously declared a Hero of the Russian Federation by President Boris Yeltsin.
Soon after his arrival in Washington, Philby was taken to AFSA by the departing SIS liaison officer, Peter Dwyer, and introduced to Gardner. On this, as on a number of previous visits, Dwyer provided information which helped Gardner fill in gaps in one of the decrypts. Almost half a century later, Gardner still vividly recalled the meeting:
. . . I was very much pleased [with progress on the decrypts] and so was Dwyer, of course. Philby was looking on with no doubt rapt attention but he never said a word, never a word. And that was the last I saw of him. Philby was supposed to continue these visits, but helping me was the last thing he wanted to do.67
Despite discontinuing meetings with Gardner, however, Philby contrived to increase his access to VENONA decrypts. His anxiety to do so grew in June 1950 after the first partial decryption of telegrams to the London residency from Moscow in September 1945 referred to the existence of a ‘valuable’ British agent network, of which, as he was well aware, he had been a prominent member.68 The Security Service SLO, Geoffrey Patterson, wrote to the DG on 18 July 1950:
Philby has signalled ‘C’ to suggest that an extra copy of any material which G.C.H.Q. send to [their Washington liaison] should be enclosed for Philby and myself. At the moment [their liaison] receives only one copy which he, of course, shows to us, but he has no time in which to sit down and copy it for us. If Philby and I can have our own copy it will give us more time for studying it before we approach the F.B.I. on the subject.69
Armed with his own copies of VENONA decrypts, Philby was able to pass them on to Moscow.70 At the time he was the only person in either London or Washington able to identify a particularly important Soviet spy in Britain, codenamed STANLEY in the September 1945 telegrams from Moscow to London,71 as himself.
Thanks to Weisband and Philby, the VENONA secret was communicated to Moscow well before it reached either the President of the United States or the CIA. From 1950 onwards, following Weisband’s arrest,72 Clarke, Bradley and Hoover must have been aware that the secret they had kept from Truman and the Agency was known to Stalin and the Centre. Though American and British security procedures both failed miserably to prevent Moscow learning of the VENONA decrypts, however, they were remarkably successful in keeping the secret within the United States and Britain.
Despite the fact that VENONA probably identified more Soviet agents than any other Western intelligence operation of the Cold War, it was too secret to be mentioned in American or British courts, even in camera, and therefore led to very few convictions. It was possible to mount successful prosecutions against Soviet agents identified in the decrypts only if they could be persuaded to confess or if alternative evidence could be discovered from non-SIGINT sources. Among the few who admitted their guilt when confronted by the evidence against them (which they had no idea was based on SIGINT) were two of the atom spies at Los Alamos: the Germanborn British physicist Klaus Fuchs and the American Technical Sergeant David Greenglass. Greenglass also implicated his wife’s brother-in-law, Julius Rosenberg, whom VENONA had identified as the organizer – with some assistance from his wife Ethel – of a highly successful Soviet spyring in New York producing a wide
range of scientific and technological intelligence. When the Rosenbergs’ trial opened in 1951, Greenglass was the chief witness for the prosecution. Unlike Fuchs, who was sentenced to fourteen years’ imprisonment in 1950, the Rosenbergs were sentenced to death, though the execution was not carried out until 1953.73 They were the only Soviet spies executed during the Cold War.
When the flow of VENONA decrypts reduced to a trickle during the early 1950s, however, the Security Service senior management largely ceased to pay detailed attention to them. In October 1955 an important clue to Philby in a newly decrypted message from Moscow to the London residency of September 1945 was missed.74 The value of VENONA as a counter-espionage tool was diminished, sometimes seriously, by the extreme secrecy with which it was handled. As Peter de Wesselow, who had the main day-to-day responsibility in MI5 for handling VENONA, later recalled, it ‘was considered as of exceptional delicacy’ and the decrypted messages from Moscow to London were ‘ten times more delicate than the rest’.75 The Director of GCHQ, Group Captain E. M. Jones, made life more difficult for his cryptanalysts by denying them the real names and biographical details of the British citizens whom de Wesselow and others in MI5 suspected of being referred to by codenames in the VENONA traffic.76 In April 1956 both the DG, Sir Dick White, and the future DDG Graham Mitchell, then Director D (counter-espionage), unaware of the important clue to Philby which had been overlooked only six months earlier, mistakenly concluded that VENONA was no longer ‘worth the effort’. Though GCHQ continued to work on VENONA, the Security Service ‘virtually abandoned it’ for the next five years.77
Service interest in VENONA revived in the early 1960s as the result of the acquisition of new intelligence from the KGB defector Anatoli Golitsyn and from Swedish intelligence. Though Golitsyn had little precise intelligence on the so-called ‘Ring of Five’, he identified it as the ‘valuable agent network’ mentioned in decrypts of September 1945.78 Following requests during 1960, the Swedes supplied copies of wartime GRU telegrams exchanged between Moscow and the Stockholm residency, some of which were discovered to have employed the same one-time pads used in hitherto unbroken GRU traffic with London.79 One hundred and seventy-eight GRU messages from the period March 1940 to April 1942 were successfully decrypted in whole or part. After a gap of almost three and a half years, another 110 decrypts were produced for the period from September 1945 to March 1947.80 The main discovery from this new VENONA source was the existence of a wartime GRU agent network in Britain codenamed the ‘X Group’, which was active by, if not before, 1940. The identity of the leader of the Group, or at least its chief contact with the GRU London residency, codenamed INTELLIGENTSIA, was revealed in a decrypted telegram to Moscow on 25 July 1940 from his case officer as one of the CPGB’s wealthiest and most aristocratic members, educated at Westminster School and King’s College, Cambridge:
[The Honourable] Ivor Montagu, brother of Lord [Ewen] Montagu, the well known local communist, journalist and lecturer. He has [several words not decrypted] contacts through his influential relations. He reported that he had been detailed to organise work with me, but he had not yet obtained a single contact.
All that Ewen Montagu reported at this meeting was general political gossip. His GRU controller made clear his dissatisfaction in a telegram to Moscow on 16 August:
INTELLIGENTSIA has not yet found the people in the Military Finance Department. He has been given the address of one officer but has not found him yet . . . I have taken the liberty of pointing out to the X Group that we need a man of different calibre and one who is bolder than INTELLIGENTSIA.
Montagu did, however, provide classified reports from his friend, the scientist J. B. S. Haldane, educated at Eton and New College, Oxford, and, like Montagu, a Communist with aristocratic connections, whose war service for the Admiralty included work at the Royal Navy’s secret underwater research establishment near Gosport.81 Most other members of the X Group proved harder to identify. BARON, who was a prolific source of intelligence on German forces and troop movements in Czechoslovakia, was thought likely to be a member of the intelligence service of the Czechoslovak government in exile in London.82 There was speculation that BOB, another member of the X Group, was the future trade union leader Jack Jones, though a report of 1969 concluded that there were ‘few pointers to the identity of “BOB” and the most that can be said is that Jones cannot be eliminated as a candidate.’83 A decade after work on the GRU decrypts began, the Security Service’s VENONA experts were still uncertain what the X Group’s precise function had been:
We have not established what the X Group represents. It is not the Communist Party as such, but it is probably some fraction or undercover group of the C.P. Moscow obviously visualised it as a source of military intelligence but it is difficult to trace the connection between Ivor Montagu (whose interests were largely in Film Production, Jewish affairs, International Table Tennis etc.), a Colonel in the R[oyal] A[rtillery], a girl in a Government Department and NOBILITY, a journalist.84
The Service devoted no significant further resources to unravelling either the connection between Montagu and the rest of the X Group or the identity of NOBILITY, which remains unknown.85
* The dominions in 1948 were Australia, Canada, Ceylon/Sri Lanka, India, New Zealand, Pakistan, South Africa. Though technically a dominion until 1949, Ireland had long ceased to be an active member of the Commonwealth.
4
Vetting, Atom Spies and Protective Security
The most unwelcome increase in the Security Service’s responsibilities during the early Cold War was in vetting. Even at the end of the Second World War there was little sign of what was to come. The future DG, Roger Hollis, noted in February 1945, ‘The Civil Service has in the past shown an extreme and understandable reluctance to have its intake vetted by us.’1 Though most government departments consulted the Service about the employment of temporary personnel on secret work, they rarely did so about established staff. In at least one case, a Whitehall failure to consult Service records led to the appointment of a Communist as private secretary to a cabinet minister.2 Even when a Communist was discovered in a sensitive post, there were no grounds for dismissal. ‘All that a Department could do was to transfer him to non-secret or less secret work, if it could do this without rousing the man’s suspicions.’3
The Attlee government was initially reluctant to grapple with the problem of keeping Communists away from classified information for fear of being accused of witch-hunts by the Labour left. It was gradually spurred into action by the sensational revelations of Soviet espionage which began with the arrest of Alan Nunn May in March 1946, followed a few months later by the report of the Canadian Royal Commission on Igor Gouzenko’s disclosures of Soviet spy-rings in Canada.4 In May 1947 the newly founded Cabinet Committee on Subversive Activities (GEN 183) concluded that ‘what was done in Canada might be attempted with comparable if not equal success in any other democratic country, including our own’, and that the May case showed ‘the existence of a Soviet espionage machine in this country’:
The ideology of the Communist involves, at the least, a divided loyalty, which might in certain contingencies become active disloyalty; the Canadian case has amply demonstrated the reality of this danger. This is not to say that all Communists would be prepared, even after long exposure to Communist indoctrination, to betray their country by consenting to work for Russian espionage agents; but there is no way of separating the sheep from the goats, at least until the damage has been done or suspicion is aroused . . . We are, therefore, forced to the conclusion that the only safe course is to decide that a member of the Communist Party is not to be employed on work where he may have access to secret information.5
Attlee brooded for some months before deciding to grasp the nettle. He minuted in December 1947: ‘We cannot afford to take risks here, and the general public will support us. Fellow travellers may protest, but we should face up to this. Action should be taken in regard to Fascists as well as Comm
unists although the former are feeble.’6 Feeble though the tattered remnants of British Fascism were, they proved of some use for public relations purposes in enabling the government to claim that it was protecting the state against extremists of both left and right. In March 1948, following cabinet discussion, the Prime Minister announced in the House of Commons the introduction of what became known as the ‘Purge Procedure’ excluding both Communists and Fascists from work ‘vital to the Security of the State’. The Communist MP Willie Gallacher interjected defiantly, ‘So raise the scarlet banner high!’
The Security Service was unenthusiastic. It would have preferred a more systematic use of the existing informal vetting system which, it believed, could ‘produce as good security’ as the Purge Procedure. The Service was predictably anxious that a more public vetting system might prejudice the secrecy of its sources, especially in the Communist Party.7 But there was also a deep sense of grievance at the way in which Labour ministers, embarrassed by the unpopularity among many of their own supporters of the extension of the vetting system, seemed reluctant to take full political responsibility, preferring to let the opprobrium fall as far as possible on MI5. On 25 March 1948 Herbert Morrison, the Lord President, told Guy Liddell, the DDG, ‘I hope you chaps will be very careful in [vetting] all these Civil Servant cases.’ Though Morrison’s tone seems to have been light-hearted, Liddell gave an indignant reply:
The Defence of the Realm Page 51