The Defence of the Realm

by Christopher Andrew

  GARBO and his case officer, Tomás Harris, made such a successful partnership partly because of their shared sense of the absurd. Like ULTRA, the Double-Cross System was one of the best-kept secrets in British history. Even Churchill was not told until the spring of 1943. While Bletchley welcomed the Prime Minister’s personal interest in its work, the DG, Sir David Petrie, preferred to keep Churchill at arm’s length for fear that he might try to interfere. At some point, possibly after the war, King George VI was also told. Masterman was informed that his classified report on the Double-Cross System, written in the summer of 1945, was still in the King’s private despatch case at the time of his death seven years later.8

  Guy Liddell was well aware, as head of counter-espionage during the Second World War, that post-war Soviet intelligence would prove a more difficult target than the wartime Abwehr. He wrote in November 1942: ‘There is no doubt that the Russians are far better in the matter of espionage than any country in the world. I am perfectly certain that they are well bedded down here and that we should be making more investigations. They will be a great source of trouble to us when the war is over.’9 Liddell did not suspect, however, that the areas where Soviet intelligence was ‘well bedded down’ included the intelligence community. With very few exceptions (of whom Hollis was the most notable), Anthony Blunt was popular with staff at all levels of the Security Service. Liddell was so impressed by Guy Burgess, who was recruited by Blunt as an MI5 agent, that he would have liked him to become an officer. As a wartime SIS officer, Kim Philby too successfully ingratiated himself with MI5 both by the use of his considerable personal charm and by claiming to have told the head of SIS Section V, Felix Cowgill, that he had been ‘quite wrong’ to withhold some Abwehr decrypts from MI5. When Philby, by then head of Section IX, was posted abroad late in 1946, Liddell was ‘profoundly sorry’ to see him go.

  But if MI5 misjudged the Five, so did the Centre. Though the Soviet Union had an unequalled ability before and during the Second World War to attract to its intelligence services well-educated ideological Western agents seduced by the myth-image of Stalin’s Russia as the world’s first worker-peasant state, it was not, as Liddell believed, ‘far better in the matter of espionage than any country in the world’. Soviet agent-running during and immediately after the Second World War was less sophisticated than it later became. Much of what the Five achieved was in spite, rather than because, of their handling by the Centre. The KGB later concluded that they were the ablest group of foreign agents in its history. During the Second World War, however, the Five increasingly fell victim to the paranoid tendencies of the Stalinist intelligence system. In October 1943 the Centre informed its London residency that it was now clear that all along the Five had been double agents, working on the instructions of SIS and MI5. There were few more farcical moments in the history of Soviet intelligence than the Centre’s decision to despatch to London an eight-man surveillance team, none of whom spoke English, to trail the Five and other supposedly bogus agents in the hope of discovering their meetings with their non-existent MI5 case officers. Perhaps to compensate for the failure of this impossible mission the team misidentified some of the visitors to the Soviet embassy in London as suspected MI5 agents provocateurs. The Five were not officially absolved of the charge of being British deception agents until after the D-Day landings. MI5’s subsequent failure to recognize the gap between the outstanding achievements of the Five and the sometimes dismal quality of the Centre’s management of them hampered its investigation of the case, which was not fully resolved until almost half a century after Philby’s recruitment.

  Many Labour MPs elected in the landslide victory of 1945, who were unaware of the still classified triumphs of the Double-Cross System, viewed MI5 with suspicions dating back to the Zinoviev letter of 1924, which they blamed for the fall of Ramsay MacDonald’s first Labour government. Once the former chief constable Sir Percy Sillitoe became DG, however, Clement Attlee placed more confidence in MI5 than in some of his cabinet ministers, whom he excluded from the decision to build a British atomic bomb on the grounds that they ‘were not fit to be trusted with secrets of this kind’. It was Attlee who began the tradition that after every general election MI5 informs the incoming prime minister if there is evidence that anyone nominated for ministerial office is a security risk. Unlike subsequent prime ministers, he also asked to be informed of any sign of subversion among ministers’ families. Attlee had more frequent personal meetings with Sillitoe than any other prime minister had with the DG during MI5’s first hundred years. Sillitoe was instructed to inform Attlee whenever the Service had information that any MP of whatever party was ‘a proven member of a subversive organisation’. Though very little record survives of matters discussed when Sillitoe called at Number Ten, in 1947 Attlee told the DG he had no doubt that the Labour MP Stephen Swingler ‘was a C.P. member’. Attlee almost certainly also mentioned other ‘crypto-Communists’ on Labour benches. Morgan Phillips, general secretary of the Labour Party during the Attlee government, kept a ‘Lost Sheep’ file on pro-Soviet MPs such as Swingler.10 In 1961, after Labour had been out of power for ten years, Attlee’s successor as Party leader, Hugh Gaitskell, following discussion with his closest associates, decided to give MI5 a list of sixteen Labour MPs who, they believed, ‘were in effect members of the CPGB . . . or men under Communist Party direction’, as well as the names of nine ‘possible’ crypto-Communists. The DDG, Graham Mitchell, declined even to discuss the list, on the grounds that MI5 records ‘could be used only in the interests of the security of the realm as a whole’ – and not to assist any political party.11 The Service has maintained that position ever since.

  The onset of the Cold War began the most dangerous period in the history of the United Kingdom. When the Second World War began, no intelligence community could have foreseen that it would end with the dawn of the nuclear age. ‘A single demand of you, comrades!’ Stalin told a secret meeting in the Kremlin after the US Army Air Forces dropped an atomic bomb on Hiroshima in August 1945. ‘Provide us with atomic weapons in the shortest possible time. You know that Hiroshima has shaken the whole world.’12

  The discovery in September 1949 that the Soviet Union had successfully tested an atomic bomb, about two years earlier than expected, shocked many British intelligence officers. The chairman of the JIC, William Hayter, was so taken aback that, before announcing the news, he cleared the room of secretaries and, despite the fact that all of the committee had an obligation to keep secrets, asked any member who doubted whether he could keep this particular secret to leave as well. The sense of shock was amplified by the almost simultaneous discovery that the plans of the first US atomic bomb, tested in the New Mexico desert less than a month before Hiroshima, had been betrayed to Soviet intelligence by Klaus Fuchs. The atom spies provided much of the motivation for the Attlee government’s introduction of the Purge Procedure, designed to prevent Communists and the few remaining Fascists from gaining access to classified material. The Security Service was well aware of the implications for civil liberties and anxious not to gain a reputation as ‘black reactionaries’ out to ‘victimise unfortunate Civil Servants’. In the event the Purge Procedure seems to have led to few injustices. Between 1947 and 1956, US purges led to the sacking of 2,700 federal employees and the resignation of another 12,000. In Britain from 1948 to 1954 there were 124 dismissals (a total which probably included resignations and transfers to other jobs).

  British fears of thermonuclear war were never greater than in the five years before the Cuban Missile Crisis. A government White Paper publicly admitted for the first time in 1957 that there was ‘no means of providing adequate protection for the people of this country against the consequences of an attack by nuclear weapons’. A schoolgirl interrupted her diary, which she acknowledged was mainly devoted to ‘boys, boys, boys’, to ask the question: ‘I wonder if World War III is on the way, it certainly seems like it, doesn’t it? The future is like a great gloomy cloud looming
ahead that will swallow us up.’13

  Like the schoolgirl, MI5’s top management felt forced to peer into the nuclear abyss. It decided, without informing most staff, that in a nuclear war ‘it was no good envisaging an organised Head Office existing anywhere; indeed there would be nothing to do.’ Once officially informed that ‘we were not to plan for any long-term war’, the Service abandoned most of the plans it had made for wartime internment of the Soviet Union’s leading supporters. For the first time, the Security Service was left with no major role in the preparations for war set out in the government’s War Book.14 As DG at the time of the 1962 missile crisis, Sir Roger Hollis knew that if the Third World War began he was likely to end his days with the Prime Minister, the War Cabinet and senior defence and intelligence staff in the doomsday bunker in the Cotswolds which had been chosen as the wartime seat (and probable grave) of the British government.

  Though the prospect of a nuclear Armageddon began to recede after the peaceful settlement of the missile crisis, the 1960s were on the whole a depressing decade for MI5 senior management. Both Mitchell and Hollis suffered the unparalleled humiliation of being investigated on suspicion of being Soviet agents. The Security Service meanwhile found it increasingly difficult to cope with the steady increase in the size of the KGB and GRU residencies which, as it later acknowledged, ‘threatened to swamp our then meagre resources’. A long campaign by the Service in Whitehall ended in October 1971 with Operation FOOT: the unprecedented expulsion of 105 Soviet intelligence officers. Over the previous year, the long-drawn-out nightmare of the investigation of Mitchell and Hollis had ended with both being found innocent of charges for which there had never been any serious evidence. FOOT had an extraordinary international impact, enhancing the Service’s prestige with foreign friends in several continents. The Ghanaian intelligence liaison officer, for example, was reported to be ‘clearly delighted’ and ‘remarked that this was good ammunition to use to persuade Cabinet Ministers that the threat from Russian espionage and its scale were serious’.15

  FOOT marked a major turning point in MI5’s counter-espionage operations. The expulsions, combined with the ceiling placed on Soviet officials in London and a policy of refusing visas to known hostile intelligence officers, turned the United Kingdom, for the first time, into a hard target for Soviet intelligence: a considerable achievement which, as in the First World War, was too elusive to be adequately identified by ‘performance indicators’. For the remainder of the Cold War, Soviet Bloc intelligence agencies diverted a significant amount of their operations against UK targets to third countries, where the local security services were judged to be less effective.

  For most of the Cold War, however, there was much about the role of intelligence in the Soviet system which MI5, like the rest of Western intelligence, did not fully understand. Part of the problem was the lack of a reliable account of the long-term development of Soviet intelligence. The founding father of US intelligence analysis, Sherman Kent, had complained in 1955 that intelligence was the only profession which lacked a serious literature: ‘From my point of view this is a matter of greatest importance. As long as this discipline lacks a literature, its methods, its vocabulary, its body of doctrine, and even its fundamental theory run the risk of never reaching full maturity.’16 Practising economists and politicians, among others, are rightly critical of the remoteness of some academic research from the real world in which they operate. But economics without economic history and politics without political history, without, in other words, a dependable record of past experience, would be what Kent forecast intelligence would remain without a serious intelligence literature – immature disciplines.17

  The lack of a reliable history of Soviet intelligence made it much easier for Peter Wright to weave his vast conspiracy theories of Soviet strategic deception of Western intelligence. Soviet intelligence, he claimed, had already run such a deception a generation earlier in the TRUST operation of the 1920s in which a bogus anti-Bolshevik monarchist underground run by the OGPU had lured the former SIS agent Sidney Reilly across the Russian border to interrogation and execution, as well as deceiving a series of Western intelligence officers. Reilly was no longer (if he ever had been) the ‘master spy’ portrayed by some of his admirers. One of his secretaries complained that he sometimes became delusional: ‘Once he thought he was Jesus Christ.’ TRUST was a well-executed operation against less than first-rate opponents.18

  Wright, however, transformed it into a strategic deception comparable in importance to the Double-Cross System before the Normandy landings. ‘The Trust’, he claimed, ‘persuaded the British not to attack the Soviet Government because it would be done by internal forces,’ organized by the supposed monarchist underground.19 A reliable history of the interwar duel between Soviet and British intelligence would have exposed this claim as nonsense, but at the time no such history was available. Both ULTRA and the Double-Cross System were still classified top secret and it was not difficult for the unscrupulous Wright to assert that he had privileged access to more major secrets and to fend off challenges with the argument, ‘If you knew what I know . . .’20 The TRUST precedent, as falsified by Wright, gave an element of plausibility (albeit inadequate) to his argument that ‘in 1963, there was no doubt that the Soviets had the necessary conditions to begin a major disinformation exercise.’21 In reality, the ‘necessary conditions’ did not exist.

  The lack of a long-term perspective on the relationship between intelligence and policy in the Soviet Union did some damage to even the bestbalanced British intelligence assessments during much of the Cold War. The long history of autocratic rulers reveals the almost invariable requirement to tell the ruler what the ruler wishes to hear. Western intelligence analysts who worked to a shorter timescale tended to underestimate the degree to which that requirement degraded the intelligence supplied to the Soviet leadership, and thus to misunderstand the gulf which often separated frequently impressive Soviet intelligence collection from the dismal level of intelligence analysis. Though the gulf was greatest under Stalin, twenty years after his death political correctness remained a key constituent of intelligence reports to Leonid Brezhnev. According to Vadim Kirpichenko, first deputy head of the FCD (KGB foreign intelligence), anything which might ‘upset Leonid Ilyich’ was removed from the reports. When Soviet policy suffered setbacks, analysts knew they were on safe ground if they blamed them on imperialist conspiracies. Western intelligence analysts underestimated the role of conspiracy theory in Soviet intelligence assessment.22

  It simply did not occur to MI5 officers (or, so far as is known, to those of any other Western intelligence agency) to suspect at any stage during the investigation of the Cambridge Five that the Five’s recurrent failure to report what the Centre wished to hear might have caused them to be classed in the middle of the Second World War as a British deception operation – or that, forty years later, the Centre might believe President Reagan, supported by his British allies, to be planning a nuclear first strike against the Soviet Union. The intelligence provided by Oleg Gordievsky on Operation RYAN in 1982 came as a complete surprise on both sides of the Atlantic.23

  Lack of a long-term perspective also hampered the Security Service’s early response to the Northern Ireland Troubles. No file has been found in Service archives for the 1970s which makes any reference to the experience of British intelligence in Ireland in the period between the Easter Rising of 1916 and the founding of the Free State in 1922. Had the intelligence community and the security forces been aware at the beginning of the Troubles of the problems caused half a century earlier by the lack of co-ordination between the military, the police and the metropolitan intelligence agencies,24 similar kinds of confusion would have been less likely to recur. For historical reasons which went back to the Fenian ‘Dynamite War’ of the 1880s, intelligence confusion during the Troubles extended to the British mainland. The Security Service did not gain the lead intelligence role against Irish Republican terrorism in Britai
n until 1992 – despite the fact it already had the lead role against PIRA on the continent and the lead role in Britain against all other terrorists, including even Loyalist paramilitaries from Northern Ireland.

  The beginning of the Troubles coincided with the end of most Security Service involvement in the Empire and Commonwealth. ‘Overseas Service’, wrote Anthony Simkins, then B1 (personnel), in 1954, ‘brings good young officers on very fast indeed’ – so fast, in his view, that there was ‘a risk that they become a bit swollen headed in the process’.25 For a quarter of a century after the Second World War, MI5 officers and many other staff spent, on average, a quarter to a third of their careers on overseas postings. As a result the Service acquired more expertise on, inter alia, Anglophone Africa, India, South-East Asia and the West Indies than it yet possessed on Northern Ireland. The only occasion on which the Service ever admitted that it had gone ‘a little outside the strict terms’ of its Charter (the 1952 Maxwell Fyfe Directive) was in the surveillance of colonial delegations during independence negotiations in the early 1960s. The Home Secretary, Rab Butler, immediately condoned the breach because of the ‘great value to the government negotiators’ of the intelligence obtained.26 Following the transfer of most of the Service’s Commonwealth responsibilities to SIS, MI5 officers were, for some years, more reluctant to accept postings to Belfast than they had been to Nairobi or Kingston, Jamaica. After the introduction of direct rule in Northern Ireland in 1972, the Service was asked to fill the new post of director and co-ordinator of intelligence in Belfast, but, because no suitable MI5 officer could be found, the first DCI was an outsider. After an urgent request for more intelligence at the beginning of direct rule from the newly created Northern Ireland Office, an Irish Joint Section was established by MI5 and SIS with jointly staffed offices in Belfast and London. Because of MI5’s lack of Northern Ireland expertise, SIS was initially the senior partner. By the end of the decade, however, MI5 had come to terms with its new role; the Belfast station was wholly funded and mainly staffed by the Security Service. The IJS was wound up in 1984.27