The Defence of the Realm

by Christopher Andrew

  The point made by Dr Evatt was in fact unanswerable on the basis of the cover story. He suggested that it would be a reasonable Russian cover to attribute a leakage to Australia if the Russians had in fact a well-placed agent in some other part of the Empire, and that if the defector had accepted in good faith the attribution of the leak to Australia, no amount of examination by the U.K. authorities could go further than establish that the defector genuinely believed what he had been told . . . There was a very considerable risk that Dr Evatt and his colleagues would regard us either as fools for failing to see his point, or as knaves for knowing a great deal more than we were prepared to tell them, and it is by no means certain that they do not have suspicions in regard to the latter alternative.16

  Since the cover story had failed to convince Chifley and his ministers, the only solution appeared to be to reveal the true source. On Attlee’s instructions Sillitoe visited Washington to seek American consent to do so. With the support of the Secretary of State, George Marshall, he eventually gained US agreement to telling Chifley, Evatt and Dedman, in the strictest confidence, of the SIGINT evidence.17

  By the time this permission was obtained, however, United States alarm at the state of Australian security had led it to place an embargo on the transfer to Australia of all classified information – a decision communicated to the Australian ambassador in Washington ‘suddenly and without any reason given’.18 When Hollis paid a further visit to Australia in August and September 1948, accompanied by Robert Hemblys-Scales of B2, he found the Defence Ministry and service chiefs deeply concerned by the embargo and anxious to reform Australian security – a ‘most refreshing change’ from what Hollis regarded as the ‘lethargy’ of most of the Canberra government, Evatt and Chifley in particular. The source of much of the ‘lethargy’, he believed, was Longfield Lloyd, head of the Commonwealth Investigation Service (CIS), Australia’s post-war security service:

  He . . . claims already to know and understand the workings of the M.G.B. [forerunner of the KGB] and Russian espionage methods generally and to be able to cope with them adequately. In fact he demonstrably knows nothing of these matters and has almost no resources to cope with them . . . I think we must assume that the ostrich-like attitude towards security of Evatt and the P.M. probably comes in part from such assurances from Lloyd. I am quite convinced that the C.I.S. with its existing staff and Lloyd at its head has no possibility of ever becoming an effective counter-espionage service.19

  Once briefed on the existence of the Canberra decrypts, Chifley and Dedman, who between them determined Australian security policy, dropped their earlier objections to the reform of the CIS and set out to mend their fences with Britain and the United States. The impact of the Soviet decrypts was, almost certainly, greater on Dedman than on Chifley. The Defence Department had suspected since late 1944 that there had been leakages of classified information from External Affairs to the Soviet Union. The post-war Australian SIGINT agency, the Defence Signals Bureau, headed by a British GCHQ officer, came under the authority of the Defence Department, whose powerful secretary, Sir Frederick Shedden, jealously protected its control of SIGINT. The able but morbidly distrustful Evatt was a controversial figure even in External Affairs and was loathed by Shedden.20 Though Hollis had been authorized to reveal VENONA to Evatt as well as Chifley and Dedman, the External Affairs Minister was away during his visit and there is no evidence in Security Service files that he was ever briefed.21

  On 17 September, shortly before Hollis returned to London, the Australian Defence Committee formally acknowledged that the CIS did not have the required ‘counter-espionage capacity’ and recommended the establishment of a new security organization. Three days later, Chifley approved the founding of an agency ‘similar to MI5’ and said he would write to Attlee asking him to station a Security Service officer in Australia to ‘provide advice’ to it. When Hollis left Australia, Hemblys-Scales remained to give advice on both the new organization and the Australian spy-hunt. Though lacking physical presence and described by an Australian colleague as ‘skinnyish’ with a ‘pimply, pasty face’, Hemblys-Scales played a major role in the reform of Australian security.22 So strong was Security Service influence on the setting up of the Australian Security Intelligence Organization (ASIO) that Shedden referred to it in a secret memorandum of December 1948 as a ‘proposed MI5 section’.23

  Hollis’s reason for returning to London in September 1948 was to take part in a secret Commonwealth Security Conference, largely prompted by what he called ‘the dangers to which bad security can expose us’, as demonstrated by the Australian experience.24 Hollis reported afterwards to Sillitoe that the dominions* had little grasp of counter-espionage and ‘look to the Security Service for guidance and leadership’:

  In our discussions on counter-espionage at the Conference, we put forward the proposal that it would be redundant for each constituent member of the Commonwealth security services to undertake the basic study of the Russian Intelligence Service which is now carried on in B2b, and that on this matter B2b should act as a coordinating point. Canada was unable to agree that B2b in London should have a formal directing function in Canadian counter-espionage work, though the other dominions raised no objection to this direction. In order to meet the Canadians, the final recommendation was drafted to read as follows:-

  There should be the fullest exchange among these security services of information on counter-espionage and upon the organisation, personnel, methods and targets of all hostile intelligence services.

  There was general agreement that London would be in possession of very much more information than the other parts of the Commonwealth and that the information which we put out would in fact fulfil a directing function.

  . . . The security services of the Dominions do not have access to much of the material which is available to us, nor are most of them in close enough contact with the United States services to share the product of their work. It is therefore only through London that the whole available supply of intelligence information on the Russian Intelligence Service can be made available to the Dominions.25

  Early in 1949 Hollis returned to Australia to give detailed advice on the charter, organization and senior personnel of ASIO, which was formally established on 16 March by a directive from Chifley to its first head, Mr Justice Reed. Hollis was accompanied by Courtenay Young, who was to replace Hemblys-Scales as the MI5 representative and to become the first security liaison officer (SLO) with ASIO. For its first fifteen months almost all ASIO’s energies were devoted to following up leads from VENONA provided by the Security Service. Young kept the decrypts themselves in a huge Chubb safe to which no one in ASIO had access.26 When passing on information to ASIO, he guaranteed its authenticity but concealed the SIGINT source. In consultation with Courtenay Young and with the approval of the Prime Minister (initially Chifley; from December 1949, Sir Robert Menzies), ASIO instituted telechecks on the main suspects identified by VENONA.27 Director B (Dick White) noted with evident satisfaction after a visit to Australia in November 1949: ‘ASIO never embark upon a new line of enquiry without first consulting the SLO. On his side Mr Courtenay Young has displayed great patience and ability and I feel sure that he is personally liked and respected by all members of the ASIO.’28 On becoming the second head of ASIO in 1950, Charles Spry, previously Director of Military Intelligence, was indoctrinated into VENONA (of which he had previously been unaware) by Courtenay Young.29 Derek Hamblen, Young’s successor as SLO in 1951, had a major row during his first week in Australia as a result of Spry’s insecure handling of a decrypt. They subsequently became firm friends and Hamblen was given an office next to Spry’s in ASIO’s Melbourne headquarters, where they spent many hours discussing the VENONA leads.30

  Though there was no Australian system of Home Office Warrants, approval for telechecks on suspects named by the SLO was ‘sought and granted orally’. One of the first telephones to be tapped was that of Fyodor Andreyevich Nosov, a TASS journali
st and Soviet intelligence co-optee, who under the codename TEKHNIK was identified by the VENONA decrypts as the main point of contact with the Australian spy-ring. Though ASIO also obtained authority to bug Nosov’s flat, it lacked MI5’s experience in planting microphones. The team which bored a hole in Nosov’s ceiling from the flat above to insert a microphone were horrified to discover not merely that the microphone was easily visible in the ceiling but that plaster disturbed by the drilling had fallen on the floor. Having persuaded the caretaker of the apartment block to let them into Nosov’s flat, they removed the microphone and cleared up the mess, but had no doubt that Nosov would realize that an attempt had been made to bug his flat.31

  Apart from Nosov, the main initial targets of the MI5–ASIO investigation were two Australian diplomats identified by VENONA decrypts as ideological Soviet agents: Ian Milner (codenamed BUR), who had joined the Communist Party while a Rhodes Scholar at Oxford in 1934 and had been seconded to the UN secretariat in New York in December 1946; and Jim Hill (TOURIST), brother of a leading Communist lawyer, who was posted as first secretary to the Australian high commission in London early in 1950 so that he could be kept under surveillance by the Security Service.32 In May 1950 Hill was visited in his Aldwych office by MI5’s leading interrogator, Jim Skardon, who a few months earlier had coaxed a confession out of the atom spy Klaus Fuchs. Though plainly shocked, Hill protested his innocence. Guy Liddell noted in his diary: ‘There is no doubt in our minds that Hill is a guilty party. He telephoned to his wife from a callbox immediately after the first interview, but said no more than that something extremely serious had happened in regard to something that he had done.’33 Having heard that Hill had been questioned, Ian Milner fled to Prague, where he spent the rest of his career teaching English literature at Charles University. Though never prosecuted, Hill returned to Australia and left External Affairs early in 1951.34

  VENONA revealed that, in one important respect, Soviet intelligence operations in Australia differed from those in most Western countries. Virtually all Soviet espionage was organized by a leading Australian Communist Party official, Wally Clayton (codenamed KLOD), described by one member of his spy-ring in External Affairs as ‘a shadowy figure’ who ‘wouldn’t look at me when I was reporting’. A fellow Communist who lent Clayton his flat for meetings with his agents said later that ‘the mysterious quality in his work . . . was mirrored in his face, which nearly always wore a furtive expression, although Clayton was not unlikeable.’ Clayton ceased being a Party functionary in 1951, probably as the result of the compromise of his spy-ring.35 Though he refused to confess to espionage and there was no evidence capable of convicting him (since SIGINT was regarded as too secret to use), the KGB was believed to be ‘disturbed that so important and (to them) potentially dangerous a witness should remain at large’. Plans appear to have been made later for him to defect to Moscow, but Clayton’s passport was revoked before he could leave the country.36 Thereafter ASIO showed only intermittent interest in him.

  In the autumn of 1948 Arthur Martin (B2b), a former wartime army signals officer, and others in the small circle of Security Service officers indoctrinated into the VENONA decrypts had begun to realize that the American ASA must be decrypting Soviet wartime intelligence traffic with residencies in the United States as well as in Australia.37 ASA remained unwilling to discuss American VENONA with GCHQ until early in 1949, when decrypted telegrams exchanged between the Centre and its residencies in Washington and New York during the final year of the war revealed that there had been a Soviet spy within the British embassy who could not be identified without British assistance. Once informed of the existence of the decrypts early in 1949, Liddell sought the help of Sir Edward Travis, the Director of GCHQ, in persuading the United States Communications Intelligence Board (USCIB), which was responsible for co-ordinating the US SIGINT effort (a mission in which it had limited success), to disclose how many more telegrams dealing with Soviet intelligence operations in the United States had been decrypted and to make them available in London. In addition, he asked Travis to arrange with USCIB for the SLO in Washington, Dick Thistlethwaite, to be given access to all available decrypts and authorized to discuss what action was to be taken on them with ‘appropriate’ American colleagues.38

  Travis, however, discovered to his surprise that even USCIB had not been informed of the American VENONA.39 He then approached General Carter W. Clarke, the head of ASA, with whom he had worked closely on ULTRA during the war. Clarke appeared surprised that the GCHQ representative in Washington was not already seeing all available VENONA decrypts and gave instructions that he was to have full access. It was also agreed that the FBI should give Thistlethwaite and the SIS liaison officer in Washington all the intelligence derived from VENONA on Soviet espionage in the United States.40 The case officer with whom they dealt was Robert Lamphere, the FBI’s chief liaison with ASA (and its successor AFSA) on VENONA investigations. Thistlethwaite found him ‘exceptionally friendly and able’.41 He and his SIS colleague did, however, agree that they should not receive ‘details of purely United States domestic’ interest.42 Geoffrey Patterson, who succeeded Thistlethwaite as SLO in June, later reported to Sillitoe: ‘The raw [VENONA] material, when processed, is liable to produce most startling information about American domestic affairs, and I cannot see the authorities here allowing it to reach the hands of another country, however friendly that country may be.’43

  Despite this restriction, it was an extraordinary comment on the closeness of the transatlantic Special Relationship that ASA, the FBI and the Armed Forces Security Agency (AFSA), which took over responsibility for VENONA production on its foundation in July 1949, should have been prepared to share with all three British intelligence agencies intelligence affecting US interests which they were not prepared to communicate to the CIA and most of the American intelligence community. After his meeting with Clarke, Travis telegraphed to Sillitoe and Menzies:

  Only very few of American Military Intelligence [of which ASA was a part], A.S.A. and F.B.I. are in [the VENONA] picture, and nobody in any other Department. Imperative, therefore, no mention this line of information be made to any American even though he be indoctrinated, e.g., United States personnel at G.C.H.Q., members C.I.A., F.B.I. representative in U.K. etc.44

  Thistlethwaite telegraphed a further warning that even Rear Admiral Inglis, head of USCIB and Director of US Naval Intelligence, then visiting Britain and the continent, knew nothing of the American VENONA decrypts: ‘We are therefore not at liberty to discuss it with him. Grateful if you would ensure “C” also appreciates this.’45

  On becoming first head of AFSA in July 1949, Rear Admiral Earl F. Stone, also a member of USCIB, appears to have been seriously put out to discover that he had previously been kept in ignorance of VENONA. He was also – to quote a doubtless euphemistic FBI memorandum – ‘very much disturbed’ to discover that, though ASA had called in the Bureau in an attempt to identify the codenames of the Soviet agents mentioned in the VENONA decrypts, it had failed to inform either the President or the Director of Central Intelligence (DCI). Stone insisted that ASA do so promptly. Carter Clarke ‘vehemently disagreed’. After what was probably a blazing row, Stone and Clarke took their dispute to the Chairman of the Joint Chiefs of Staff, General Omar Nelson Bradley.46

  Truman was a strong supporter of Bradley, and had enthusiastically supported his appointment as army chief of staff in succession to Eisenhower a year earlier.47 In 1949, however, Bradley was guilty of an extraordinary act of insubordination to the Commander in Chief, siding with Clarke in his dispute with Stone over whether to reveal VENONA to Truman and the CIA. Bradley announced that if, in his opinion, the contents of the decrypts ever warranted it he ‘would personally assume the responsibility of advising the President or anyone else in authority’.48 Clarke assured Patterson that, even if Bradley did decide to pass on VENONA material to Truman or the National Security Council, he would do so in a ‘suitably dressed up form’ which
would conceal its SIGINT origin.49 Bradley was even more anxious to keep the VENONA secret from the CIA – and in particular from its head, Rear Admiral Roscoe H. Hillenkoetter, DCI from 1947 to 1950. Clarke doubtless reflected Bradley’s views as well as his own when he told Patterson that he had ‘very good reason for suspecting that it is not safe to pass such secret material to C.I.A.’. That ‘very good reason’ was, almost certainly, the evidence contained in the decrypts that OSS had been heavily penetrated during the war and the understandable (though mistaken) fear that its peacetime successor, the CIA, was also penetrated.50 The Director of the FBI, J. Edgar Hoover, was equally suspicious of CIA security and viewed the Agency as an upstart rival.51 Hillenkoetter was frequently Truman’s first caller of the day, bringing with him the intelligence briefing which later became known as the President’s Daily Brief. If Truman were informed of VENONA, he would be likely to raise the subject with his DCI. Only by keeping the secret from the President, therefore, could Bradley and Hoover be sure that it was kept from the DCI.52

  VENONA revealed much less about NKVD/NKGB intelligence operations in the United Kingdom than in the United States, chiefly because far fewer wartime messages were intercepted between Moscow and London than between Moscow and the US. Even before the Soviet entry into the war, the Foreign Office had agreed that the Soviet embassy in London could communicate with Moscow by radio on set frequencies. These radio messages were initially intercepted and recorded in the hope that they could eventually be decrypted, but interception (save for that of GRU traffic, which continued until April 1942) ceased in August 194153 because of the need to concentrate resources on the production of ULTRA intelligence based on the decryption of Enigma and other high-grade enemy ciphers. Interception of Soviet traffic did not resume until June 1945. In the United States, by contrast, there was no agreement allowing the Soviet embassy and consulates to communicate with Moscow by radio. Instead, Soviet messages were written out for transmission by cable companies, which, in accordance with wartime censorship laws, supplied copies to the US authorities.54