GCHQ

by Richard Aldrich

  Another massive achievement was that Bletchley Park and its diplomatic equivalent at Berkeley Street in London were properly ‘integrated’, mixing up staff from the three armed services and civilians. This was immediately obvious to any visitor from the curious blend of uniform and civilian dress, often in exotic combinations. Occasionally a visiting Admiral or General would fulminate to see members of his service dressed in colourful pullovers, and demand that they return to full uniform. However, the top brass on day trips from Whitehall were little more than a temporary nuisance. During the 1940s a sigint service which mixed up civilians and personnel from the armed services was quite remarkable. It would take the Americans until the early 1950s to achieve an integrated organisation that mirrored Bletchley. In Nazi Germany as Calvocoressi recalls, the situation had been even worse, for there ‘six or seven different cryptographic establishments fought each other almost as venomously as they fought the enemy’.45

  In the social anthropology of intelligence, sigint was emerging as the dominant tribe. ‘The Ultra community at BP saw itself as – perhaps was – an elite within an elite,’ recalled one code-breaker. Material gathered by other kinds of intelligence agencies was merely ‘Top Secret’, but sigint material was compartmentalised as ‘Top Secret-Ultra’. The ability to impose draconian security on its product would be a hallmark of a fully-fledged sigint organisation, and dominated its relations with its friends and allies in the code-breaking world. This security obsession also extended to people. The security rule at Bletchley Park was ‘Once in, never out.’ In other words, once people had worked in sigint, there was a reluctance to allow them to move to other areas of war work, and they were effectively ‘captive’ for the duration of the war.

  Dominance was partly about size. By the end of the war, over ten thousand people were labouring under Bletchley’s direction. The expanded bombe effort alone led to the creation of five further outstations as far away as Stanmore and Eastcote on the outskirts of London. Working alongside GC&CS were the listening units of the armed forces, known as the Y services. Although these fed high-grade material to Bletchley Park, they also worked on low-grade material for their own purposes. Often considered ‘poor relations’, they derived their intelligence either from listening in to low-level tactical communications that were not encrypted, including clear voice traffic, or by simply analysing the flow of traffic. Analysing the patterns of radio traffic, including volume and direction, even without breaking the codes, could reveal a great deal of information about the enemy, and GC&CS worked closely with the armed services to develop what were known as the ‘Y stations’. Bill Millward, who continued to serve long after the war, recalls that Bletchley Park’s relationship to the Y services was to become ‘a sort of university of signals intelligence, developing techniques which all might share’.46 The Y services had been largely responsible for deducing the enemy ‘order of battle’, the structure, strength and location of the units of the German armed forces. The Navy ran intercept sites at Scarborough and Winchester. The Army ran a site at Fort Bridgelands near Chatham, and later opened a station at Beaumanor Hall near Loughborough in Leicestershire. The RAF were located at Cheadle in Cheshire, and developed a large new site at Chicksands near Baldock in Bedfordshire. Many of these locations would continue as sigint sites after August 1945.47 All of them were symptomatic of an industrial revolution in secret intelligence: both Bletchley Park and the outstations operated like factories, with three gruelling shifts each day.

  At a deeper level, there had also been a social revolution in British intelligence. Brilliant individuals who only a year before had been members of international chess teams or wrestling with obscure mathematical problems in Cambridge colleges, were now focused on intelligence. Remorselessly logical, they could see that Bletchley Park was the intelligence machine of the future. Moreover, they were outsiders, with no sense of bureaucratic anxiety and no fear of the ‘Establishment’. They fearlessly articulated what to them was self-evident. GC&CS, once a small school of code-breakers working in the service of SIS, had now vastly outgrown its parent organisation. Gladwyn Jebb, one of a number of rising British diplomats who were temporarily attached to intelligence duties during the war, noticed this dramatic change. The organisations like Bletchley Park had been forced to recruit widely from industry and the universities to fill their ranks, so they had forward-looking staff who brought with them modern organisational techniques.48 Jebb complained that SIS had ‘too much of what I would call the “false beard” mentality…more especially amongst those who have been in the show for a very long time’. The world had moved on, he argued: ‘The idea of a deeply mysterious “Master Spy”, sitting in some unknown office and directing an army of anonymous agents, is as outdated as it is romantic.’49

  The Americans had also opened the eyes of GC&CS to what was technologically possible. Although hobbled by the bitter Army–Navy divide, Washington nevertheless threw vast scientific resources at sigint. On their visits across the Atlantic, a core of determined individuals from Bletchley were able to glimpse what the future might hold. In 1944 a small group of talented British code-breakers began the long-range planning that would turn wartime Bletchley Park – with its chess players and crossword puzzlers – into Britain’s premier post-war secret service, with a strong sense of identity, a large budget and predatory designs on other agencies. Three key figures were instrumental in this: Gordon Welchman, the man behind Bletchley Park’s intelligence processing centre; Harry Hinsley, who would serve as the ‘sherpa’ for the Anglo–American–Commonwealth sigint summits after 1945; and Edward Crankshaw, who had handled wartime sigint discussions with the Soviets. Hugh Foss joined them on his return from a posting in Washington.50

  On 15 September 1944, only weeks after the liberation of Paris, this planning group began to consider GC&CS’s post-war future. It was led by Gordon Welchman, who was Assistant Director for Mechanisation, and had also been responsible for Hut Six, where Enigma was broken. Some of the exciting ideas the group advanced for the future of GC&CS grew largely out of the Hut Six experience. It called for a more centralised ‘Foreign Intelligence Office’ as part of a coherent national intelligence organisation, and for a comprehensive body dealing with all forms of sigint, together with a modern signals security organisation with the latest communications engineering. This, the group believed, could become a truly modern ‘Intelligence Centre’ governing all types of interception activities.51

  Welchman’s group was tough-minded. There were, it argued, few people in GC&CS with real ability in general planning and strategic coordination. They observed, ‘it would be difficult to count as many as a dozen’. This talent should not be wasted on the final year of the war against Japan. Instead, as soon as the war in Europe was over, ‘as many as possible of the few potential planners should be set to work in the direction of our three immediate objectives, instead of devoting more of their time to Japanese problems’. GC&CS should not lose touch with developments in the field of Japanese sigint problems, since there were interesting things to learn in this sphere. However, it should merely extract technical benefits from the Japanese War, rather than expend resources upon it. British commanders in Burma, like Field Marshal Bill Slim, realised that they were now a low priority for the intelligence services, and complained bitterly about it.52

  GC&CS realised that speed was of the essence. It was ‘imperative to make an approach to the present Prime Minister at the earliest possible moment’. Any successor to Churchill, it reasoned, however sympathetic, could not have a real appreciation of ‘the fruits of intelligence in this war’, or Churchill’s keen appreciation of the importance of tight security. In Churchill it had a heavyweight advocate, and it feared a return to the pre-war situation of under-recognition of what sigint could achieve; even now, the true scale of its wartime output was known to only a very few in high places. Moreover, the really talented sigint planners were newcomers, and would soon be recalled to their pre-war occupations unl
ess some positive action was taken to retain them. Quite simply, this came down to cash. GC&CS had to have the status to secure ‘a sufficiently liberal supply of money to enable it to attract men of first rate ability’, particularly engineers and electronics experts. It was also aware that it would have to give equal weight to all types of intelligence about foreign countries, ‘including scientific, commercial and economic matters’. This was a tacit reference to the targeting of friendly states.53

  In January 1945, the torch of post-war planning passed to William F. Clarke. Clarke, who had served continuously in code-breaking from 1916, warned that the ‘enormous power wielded by the Treasury’ might soon be brought to bear on GC&CS. As had happened in 1919, work on military cyphers might cease in favour of concentration on diplomatic material only. This, he insisted, could be ‘disastrous’, because the resulting damage to ongoing cryptographic research might mean that in the event of a sudden future conflict, enemy military traffic would prove inaccessible. Even more problematic was the challenge of building up the prestige of GC&CS. Its very secrecy was its worst enemy, ensuring that many in elevated government circles did not know its true value. There was also the ‘potential danger’ of a Labour government coming to power, since the interwar Labour government had found many aspects of the secret state to be repellent.

  Clarke also paused to consider the emerging United Nations. Allowing himself some momentary Utopian thoughts, he observed that if the new organisation took the step of abolishing all code and cypher communications, this action ‘would contribute more to a permanent peace than any other’. However, he conceded that this ‘is probably the counsel of perfection’, and was highly improbable. Instead, he predicted that energetic code-making and code-breaking would persist into the post-war world. On the matter of who would control the British code-breakers, he felt that in the past neither the Admiralty nor the Foreign Office had been satisfactory. The current system of control by SIS also brought with it ‘certain disadvantages’. Clarke vigorously asserted that GC&CS should break free, not only of SIS but also of the Foreign Office. Instead it should be a separate organisation under either the Chiefs of Staff or the Cabinet Office, and should be regarded as a wholly separate third secret service.54

  As late as October 1944, some senior figures at Bletchley were still arguing for re-absorption by SIS. John Tiltman, the Soviet specialist, argued that the code-breakers should be ‘closely fused with S.I.S. under the Director General [Sir Stewart Menzies] as the one and only Intelligence producing service’.55 However, the stock of Menzies was continuing to fall among senior figures in Whitehall. In January 1945, the Chairman of the JIC, Victor Cavendish Bentinck, concocted his own influential vision of ‘the intelligence machine’. He suggested that GC&CS should remain under the overall direction of ‘C’, but at the same time it would be a separate organisation and ‘not a part of SIS’. It would boast its own budget alongside the other secret services as part of the Secret Vote, Britain’s quaintly titled intelligence budget.56 It was thus Commander Edward Travis, not Menzies, who determined the final shape of GC&CS shortly after VJ-Day. Although peace had arrived, Travis’s mind was already focused on possible future conflict with the Soviets. Recalling the earliest days of the last war, he observed, ‘When information was most urgently required, very little was forthcoming.’ The next war was likely to be of shorter duration, with little time for mobilisation. In such a conflict the British would have to fight with what they had. It was essential that continuity be maintained, and that rapid expansion was possible on the eve of war.

  Exactly when the post-war term ‘GCHQ’ came into common usage is a matter of dispute. It was first used as a cover name to confuse workmen dropping off furniture at the Bletchley Park site as early as the end of 1939.57 By 1946, although technically still merely a cover name, it was used more and more widely to denote Britain’s code-breakers. Travis decided that the new post-war GCHQ would be divided into five groups run by his key subordinates.58 To cover its multifarious tasks, he hoped to have a thousand civilians plus a hundred military staff at a new sigint centre located somewhere near to the policy-makers in London. By contrast, the outlying Y stations would be manned by about five thousand additional personnel, of whom only a few would be civilians. GCHQ’s own core staff fell rapidly from an end-of-war strength of 8,902 to a projected 1,010 for 1946.59 Despite the dramatic drop in numbers, Travis concluded that the post-war deal he had struck with the Treasury was ‘on the whole most satisfactory’. For him it was about quality rather than quantity. A few days before Christmas 1945 he explained: ‘The war proved beyond doubt that the more difficult aspects of our work call for staff of the highest calibre, the successes by the Professors and Dons among our temporary staff, especially perhaps the high grade mathematicians, put that beyond doubt.’ He wanted suitable conditions with which to attract these sorts of people, although he knew this would be difficult.60 Captain Edmund Wilson, Travis’s Principal Establishment Officer, echoed this view, arguing that of the 260 officers to be kept on in their post-war establishment, some two hundred of them must have not only initiative but also ‘first class brains’.61

  Where would GCHQ’s new centre be? What it craved was a site in central London, next to the policy-makers, but even with the post-war demobilisation of many government departments, nothing suitable could be found. The solution was what John Betjeman would immortalise as ‘Metroland’. GCHQ moved to the outer fringes of north-west London, close to Harrow and Pinner. The precise location was Eastcote, which had been used as a wartime outstation of Bletchley Park. It was also close to Dollis Hill, where the laboratories of the Post Office Research Department had built the remarkable ‘Colossus’ computer. Together with Stanmore, Eastcote was one of two large out-stations built in 1943 to accommodate the ever-expanding number of bombes that were being used to cope with the flood of Enigma traffic. However, while it provided reasonable single-storey buildings that were superior to the huts of Bletchley, the overall site was regarded as cramped and unattractive. In June 1946, William Bodsworth, a British code-breaker, returned from a period in America to the cold and rain of an English summer to take over GCHQ’s Soviet section. He found his first sight of Eastcote ‘frankly shattering’. Expecting ‘a nice old country house’, instead he found it to be ‘more cheerless than any of the temporary buildings I have seen in this racket either here or abroad’.62

  Those who were leaving Bletchley for good and returning to civilian occupations were given the security warning of their lives. Edward Travis issued a ‘Special Order’ to everyone in GCHQ. He began by thanking them all for their admirable achievements and the substantial contribution they had made to the winning of the war. He then moved quickly on to the matter of maintaining secrecy, even after the end of hostilities. ‘At some future time we may be called upon again to use the same methods. It is therefore as vital as ever not to relax from the high standards of security that we have hitherto maintained. The temptation to “own up” to our friends and families as to what our war work has been is a very real and natural one. It must be resisted absolutely.’63 However, in the Far East, the secret of ‘Magic’, the breaking of Japanese diplomatic codes, was already out. When Bruce Keith, commander of the vast British sigint station located at HMS Anderson in Ceylon, tried to outline Travis’s tight security measures, some of his subordinates openly laughed at him and observed that ‘the Americans had spilled the beans in the paper the other day’.64

  The move from Bletchley to Eastcote was undertaken during early 1946 in four main parties. The first was the priority group, and included the Soviet and East European Division; the last arrived in April 1946.65 Staff turning up in leafy Pinner in search of lodgings were allowed to refer to their place of work as ‘GCHQ’, but they were told firmly that any reference to ‘signals intelligence’ was forbidden.66 Between 1945 and 1948 the term ‘GCHQ’ was used interchangeably with both ‘London Signals Intelligence Centre’ and ‘Station X’.67

  Blet
chley Park was now an empty shell in the Bedfordshire countryside. Barbara Abernethy, who had worked as Denniston’s personal assistant, recalls: ‘We just closed down the huts, put all the files away and sent them down to Eastcote. I was the last person left at Bletchley Park. I locked the gate and took the key down to Eastcote. That was it.’68 Much of the machinery was broken up, including examples of the mighty ‘Colossus’ computational machine. However, Professor Max Newman, who had been central to its development, managed to secure two ‘Colossus’ machines for his new computing department at Manchester University. These were transported by the Ministry of War Transport at the price of thirty-four shillings a ton. Newman offered to send a junior university lecturer down ‘to sit on the van’ to make sure that the precious machines were not damaged in transit.69 In fact, this was not quite the end of Bletchley Park’s active life in sigint, since GCHQ continued to use it for training courses as late as the 1960s.

  The intention behind GCHQ’s post-war move to London was to service the centres of power in British government. Accordingly, in the autumn of 1945 Travis took the opportunity to look at how the sigint product – the ‘blue jackets’ or ‘BJs’– circulated around Whitehall. The Foreign Office was a big customer, receiving three sets of BJs daily. One set stayed with Ernest Bevin, the new Foreign Secretary and his war-weary Permanent Under-Secretary, Cadogan, ‘for their immediate information’. Another went to the Services Liaison Department, which worked closely with the JIC. The third went to the main departments. Virtually everyone in the operational core of the Foreign Office habitually saw BJs, but they were always kept separate from other documents in special boxes which were locked up overnight.70