GCHQ

In the summer of 1949 the US Navy picked its latest submarines for this mission, the USS Cochino and the USS Tusk. They had been built at the end of the war, and in 1948 they were modified to bring them up to U-boat standards and fitted with the latest snorkels. This allowed them to run submerged for long periods on diesel power, venting their exhaust to the surface. They had also been streamlined and fitted with the fastest available propulsion systems. The specialist elint equipment for capturing missile control signals, or ‘telemetry’, was installed by British sigint technicians at Portsmouth. The submarine chosen for fitting was the Cochino, under Commander Rafael Benitez. The name ‘Cochino’ was supposed to denote a species of trigger fish, but in Spanish it simply meant ‘The Pig’. Preparations were masterminded by Harris M. Austin from the US Naval Security Group and civilian sigint engineers. Additional aerials, known as ‘ears’, were fitted to the tailfin. The elaborate listening technology required the drilling of small holes for wires in the submarine’s pressure hull, which weakened it and did not best please the crew. Trials were held along the British coast in July 1949. In August the Cochino, escorted by three other submarines, including the Tusk, headed for Arctic waters. In the Barents Sea, the Cochino separated and sat off the coast hoping to collect the high-frequency signals that indicated a missile test, but found nothing of great interest. After a few days of lurking, it headed back to a rendezvous with the Tusk.[25]

However, disaster now struck. Four hundred miles north of the Arctic Circle the Cochino ran into a severe storm. Water poured in through a malfunctioning snorkel and a serious battery fire developed, burning for fourteen hours and producing large volumes of dangerous hydrogen. Some of the crew battled the fire using breathing equipment, but after a series of explosions they staggered back and admitted defeat. One crew member recalls: ‘They formed a grotesque aspect with their faces and hair burned. The skin was falling from their hands.’[26] Commander Benitez and his seventy-eight crew decided to abandon ship just after midnight. Despite a rescue by the Tusk, seven men were lost to the stormy seas off the Norwegian coast. Six of these fatalities were brave rescuers from the Tusk who were equipped with faulty survival suits, while the seventh was civilian signals intelligence expert Robert W. Philo from the Cochino. Commander Benitez was the last man to make the treacherous crossing – effected by a swaying plank – between the two vessels. By the time the Tusk pulled clear the Cochino was already half-submerged. ‘With a final burst of spray she disappeared from sight,’ plunging into 950 feet of water. The Tusk took the casualties, many with severe burns, to Hammerfest in Norway, from where they were flown to London.[27] These were the first casualties in one of the most secretive and dangerous areas of Cold War signals intelligence activity. However, London and Washington were not deterred. By the early 1950s, British and American sigint submarines were regular visitors to the headquarters of the Soviet fleet.[28]

Elint flights over the open sea were also sensitive and risky. On 8 April 1950 a US Navy elint aircraft, a PB4Y-2 Privateer, launched from Bremerhaven in northern Germany, was shot down while trying to identify new Soviet missile bases along the Baltic coast. The crew of four, who had named their aircraft the Turbulent Turtle, all perished. The Soviets later salvaged the Privateer’s elint equipment from the waters of the Baltic, and were in no doubt about the nature of the mission. Further missions were postponed.[29] Within a month of the shootdown of the Privateer, General Omar Bradley, then Chairman of the US Joint Chiefs of Staff, set out the case for resuming the flights, insisting that the intelligence they gathered was of the ‘utmost importance’. President Truman finally agreed to a resumption when told that US aircraft close to Soviet-controlled territory would be armed ‘and instructed to shoot in self-defense’. Truman minuted, ‘Good sense, it seems to me.’ The President’s green light was received on 6 June 1950, but after the outbreak of the Korean War later that month the flights were suspended for another few weeks due to ‘current hyper-tension and fear of further shoot-downs’. By the end of 1950, regular operations with RB-50Gs, ‘special mission’ elint aircraft adapted from an upgraded Superfortress bomber, were operating out of RAF Lakenheath airbase in East Anglia.[30]

Norway was an early partner in all types of sigint operations. In 1952, Rear Admiral Anthony Buzzard, Britain’s Director of Naval Intelligence, paid a secretive visit to Norway. His requests were so sensitive that only handwritten notes were taken at the meetings. Buzzard asked for permission to launch special reconnaissance flights from Norway into Russian airspace using the RAF’s new Canberra aircraft, and also to run elint flights conducted within Norwegian airspace. However, the fate of the USS Cochino and then of the American Privateer had alerted the Norwegians to elint operations as a potential flashpoint, and they were cautious. Only permission for the latter flights within Norwegian airspace was granted.[31] By the mid-1950s the Norwegian Defence Intelligence Staff was beginning to experiment with the use of commercial trawlers as platforms for intelligence-gathering in the Barents Sea. Initially these were used for photographic reconnaissance, but they were gradually expanded to involve sigint monitoring. A ‘cover’ shipping company, Egerfangst, was established to run these operations, and its first vessel, the Eger, was in operation by 1956 using equipment supplied by the American NSA.[32]

Although the most important sigint collected at short range came through perilous operations by air and sea, the British and Americans also boasted vast armies of land-based listeners crouching over their radio sets in wooden huts, often in inhospitable locations. Tactical sigint in peacetime presented a problem, since there was not much for Y service sigint to listen to. Yet on the first day of any future war with Russia – and war, if it came, was expected to come suddenly – the RAF would be required to reconstitute its vast legions of secret listeners.[33] In the event, the three services kept a large inter-service intercept formation in place, using personnel who were doing National Service. During peacetime they were lent to GCHQ, and spent much of their time collecting a wide range of signals, including diplomatic and commercial traffic. One RAF sigint officer observed, ‘Our only function is to receive the stuff in its cryptic form – a purely mechanical process – and pass it on to the body whose job it is to break it down.’[34] By 1950 a system was in place whereby all those beginning National Service were asked if they would volunteer to learn Russian. The huge numbers of personnel who were trained up guaranteed a vast pool of tactical sigint operators who could be recalled on the eve of war, although GCHQ worried about how to hide the scale of the operation. This had profound consequences for the balance of power within post-war British sigint. It ensured that while the overall British sigint programme was coordinated by GCHQ, it was in fact provided by a complex alliance of GCHQ and the three armed services. This secret pact suited everyone – except for the Treasury, which struggled to track sigint spending, hidden as it was under a welter of misleading headings and cover organisations.[35]

The outbreak of the Korean War on 25 June 1950 triggered a massive expansion of the riskier and more dangerous short-range operations to collect all types of sigint, including elint. Tensions were high, because the strategic planners believed that the outbreak of global war was not far away. Hitherto the Americans had been dependent on the British to cover much of north-west Europe, but now Americans began to arrive in numbers and their listening stations sprouted all over Britain, often disguised as RAF stations. In 1952 the 47th Radio Squadron of the US Air Force Security Service arrived at Kirknewton airbase in Scotland, from where it could monitor shipping off the Kola Peninsula. In October that year Squadron Leader J.R. Mitchell became the first dedicated ‘Liaison Officer for GCHQ’ on elint in Washington.[36]

The accelerated pace of operations paid dividends. By 1952 elint experts in London and Washington had achieved a comprehensive picture of the Soviet Air Force. This had not required the breaking of Soviet codes. Instead, most of it was achieved through a mixture of elint or direction-finding, which simply meant using triangulation to locate specific Soviet units. GCHQ also listened in on clear voice traffic used by Soviet air-defence controllers giving instructions to fighters. There were large gaps in both Soviet air warning and coastal radar, which were mapped carefully. Anti-aircraft radar around Moscow was examined with special attention. Elint experts had been able to follow air-defence exercises in which the Soviets had used tiny strips of aluminium foil dropped from aircraft, known as ‘chaff’, as a radio countermeasure to fool radar operators into thinking large numbers of aircraft were airborne. Some of the more sophisticated Soviet work was thought to have been carried out by German experts captured by the Russians after 1945. Korea itself had proved to be a bonanza, with new Soviet radio equipment being captured, including direction-finding equipment which showed ‘marked improvement in design and construction’.[37]

Success in the exciting new field of elint offset some recent disappointments. GCHQ and its American partners had not yet recovered the medium-grade Soviet cyphers lost during the infamous ‘Black Friday’ of October 1948. They had not detected the advent of the first Soviet atom bomb, nor had they anticipated the outbreak of the Korean War. However, the elint effort against the Soviet Air Force, which also involved direction-finding and traffic analysis, was one of the key areas in which GCHQ could claim outstanding achievement in the first post-war decade – and it was sustained.

There was an especially secret reason why GCHQ and NSA examined the operational anatomy of the Soviet ‘nuclear bear’ so minutely. During the early 1950s, target intelligence officers in London and Washington had been busy exchanging sensitive data on ‘the mission of blunting the Russian atomic offensive’. This meant planning early counter-force attacks against Soviet nuclear forces, especially bombers, in the hope of destroying them on the ground in Eastern Europe before they could be used in a future war. GCHQ had given particular attention to this matter because of the vulnerability of Britain, and the Americans were impressed by the progress London had made on it. GCHQ and the RAF’s secret units had amassed ‘a significant amount of evaluated intelligence, particularly in the special intelligence field, which would be of the greatest value’ if war broke out.[38]

American officers considered that ‘vigorous efforts should be taken immediately to ensure rapid development of a joint research program to insure maximum exploitation of the British resources’. In short, it was not just the raw elint that the British had collected, but their sophisticated analysis of it that allowed it to be turned into high-quality finished intelligence, a legacy of the skills garnered during Bletchley Park’s Hut Three operation. Most of the airfields and the operational procedures for the Soviet Union’s nuclear air force in the European theatre had been mapped by 1952.[39] In June of that year a team from the US Air Force Security Service centre at Brooks Field, led by Major Hill, visited GCHQ and one of its outstations at Knockholt in Kent to further converge their activities in this area. Hill also wanted to discuss the creation of new ‘ground-based electronic intercept stations’ in Europe. Korea had greatly accelerated preparations for a ‘hot war’, and GCHQ’s elint success on Soviet air defences helped it to justify budget increases.[40]

Throughout 1951 and 1952, global war often seemed imminent. Communist China had entered the Korean War in 1951, and numerous Soviet advisers were busy assisting the North Korean forces. Soviet and American pilots were actually fighting each other in the skies of East Asia. Although the public were never told, sigint made this fact clear to the secret listeners. In this increasingly fevered atmosphere, improved intelligence was given a high priority. On 22 January 1952 the British Chiefs of Staff met the Permanent Under-Secretary of the Foreign Office to review plans for accelerating intelligence. GCHQ was given a large tranche of new money over five years under the heading ‘Methods to Improve’. Its extensive shopping list included larger computers and ‘high speed analytical equipment’ for renewed attacks on high-grade Soviet communications. These were given the highest priority, and government research and supply elements were instructed accordingly. GCHQ and the Admiralty were beginning a new programme to build better receivers for ground-based and seaborne ‘Technical Search Operations’ which were critical to elint. Again, much of this was about targeting, and the Chiefs of Staff continually reiterated the ‘very great importance’ of speeding up technical development in these areas.[41] By November 1952, British defence chiefs wanted increased expenditure on intelligence, and were unanimous that in the short term the emphasis should be on sigint.[42]

Late 1952 was an exciting time for GCHQ. Equipped with a larger budget, staff had begun to move to their new headquarters at the twin sites in Cheltenham. It also had a new Director, Wing Commander Eric Jones. Given the rise of airborne sigint, it was appropriate that an RAF officer should have succeeded Edward Travis, who had been increasingly ill during the late 1940s with lumbago.[43] Jones was a Bletchley Park veteran who had proved himself while in charge of the critically important Hut Three. Bill Millward, another long-serving GCHQ veteran, recalls that at first glance ‘his qualifications for the post were not apparent’. He had spent the 1930s as a cloth merchant in Macclesfield. However, he was a quick learner, a natural diplomat and a man of obvious principle.[44] He had proved an excellent liaison officer in Washington towards the end of the war, and after the ‘happy outcome’ of the BRUSA conference of 1946 he had decided to ‘stay in the racket’.[45] This was good for GCHQ, since Jones proved to be a leader who inspired instinctive trust. In late 1952 he was busy filling the three hundred extra staff posts recently authorised. GCHQ had already proposed an additional increment of a further 366 staff, and was going from strength to strength.[46]

In the late summer and the autumn of 1952, senior GCHQ officers like John Somerville had been pressing defence scientists and intelligence chiefs to join them in planning the future expansion of airborne sigint.[47] Good results were being obtained, and GCHQ was making the most of the facilities at its disposal, but there were just not enough ‘ferret’ flights, and the elint-collection effort needed ‘more equipment, personnel, aircraft and ships’.[48] The RAF was using its three RB-29 Washingtons, which were big enough for the increasingly complex equipment required.[49] It was a Washington that had captured the first recordings of ‘Scan Odd’, the new airborne radar which equipped Soviet fighters. However, the RAF’s Washingtons were slow and vulnerable compared to the upgraded American variant, which could reach 400 mph, as fast as a wartime fighter.[50] British elint specialists longed to enter the jet age with the military version of the de Havilland Comet, but there seemed to be no hope of getting this desirable aircraft. Part of the problem was that 192 Squadron was not seen as a front-line fighting unit, tended to be overlooked and was continually moved between commands over the next two decades.[51] There was also an acute shortage of staff at GCHQ qualified in the rarefied field of elint analysis, which was made worse by the upheaval of the move to Cheltenham.[52]

The required extra momentum came from the Americans. Washington pressed the British for more spending and more effort at a major US/UK elint conference in December 1952.[53] Accordingly, in 1953, Britain’s elint specialists acquired more small specialist Canberra jet aircraft, and a year later, after much discussion, the Treasury reluctantly approved the purchase of three much-prized de Havilland Comet C2s, a modified version of Britain’s first jet airliner. Until now, all of the sigint aircraft had been rough-and-ready adaptations. The Comet C2 was Britain’s first dedicated airborne sigint platform designed from scratch. In the spring of 1957 the first C2 arrived at CSE Watton and was placed in the hands of SRIF, the RAF’s secret team of sigint engineers, who were based in No.3 Hangar. Their task was to cram a whole mini-sigint ground station into the cramped interior of an airliner. For George Baillie, the Principal Scientific Officer at Watton, this was the most complex task his team would ever undertake. Equipping the three Comets was extremely expensive and time-consuming. It was therefore with complete horror that they discovered a fire in No.4 Hangar in the early hours of the morning of 3 June 1959. One plane was completely destroyed, leaving CSE with only two Comets to fulfil the many missions requested by GCHQ. The Treasury boggled at the cost of the unscheduled replacement of the third aircraft, and it was two years before GCHQ’s Deputy Director Joe Hooper persuaded it to find the money.[54]

In America, the new NSA was now responsible for communications intelligence or ‘comint’, the most important area of signals collection and code-breaking. However, battles over elint stretched on into the 1960s. Even ten years after the creation of NSA, American officials envied the more centralised British model, which placed GCHQ in charge of both fields of sigint activity.[55] In 1953 the British approached the Americans to suggest a combined organisation for planning electronic warfare and radio countermeasures. The problem for the Americans was that they would first have to settle their bitter inter-service disputes, which was proving near impossible.[56] Notwithstanding this, the two nations managed to get together for a major US/UK Electronic Warfare Conference every two years.[57] The GCHQ approach made sense, because it was becoming harder to distinguish between signals that carried communications and other types of electronic signals. The Americans noted wistfully that this ‘has been recognised by the British, who have placed all electronic search and reconnaissance under the control of COMINT authorities’ – in other words, under GCHQ. In the US, control over these matters remained fiercely contested as late as the Vietnam War.[58]

By 1953, elint was considered so valuable that more aggressive British operations were being authorised. These included clandestine operations on the ground. GCHQ had manufactured a new short-range elint reception and analysis kit, code-named ‘Deaf Aid’, that looked like a suitcase. SIS was sending agents into the Eastern Bloc with these sets – the results are still classified. The Joint Intelligence Committee decided that the equipment would also be secretly deployed in diplomatic posts inside the Eastern Bloc alongside the covert comint stations operated by the Diplomatic Wireless Service.[59] By February 1954 experimental versions were also being deployed on the ground by the British Military Mission, or ‘Brixmis’, effectively a team of roving military attachés in East Germany.[60] GCHQ was settled into its new accommodation by January 1954, allowing it to work with the Ministry of Defence’s Directorate of Scientific Intelligence on the much-needed expansion of T Division, the analytical unit at Cheltenham which made sense of elint once it was collected.[61]

All these new intelligence activities meant risk. In October 1952 Earl Alexander, the Defence Secretary, and Anthony Eden, the Foreign Secretary, had been chewing over the delicate question of who should approve the expanding programme of ‘ferret’ flights and perilous submarine missions around the edge of the Soviet Union. They agreed that the buck should be passed upwards to Downing Street, and a process developed whereby a list of proposed secret sigint missions was regularly sent to Prime Minister Winston Churchill. Rather disingenuously, Alexander told Churchill that Britain had in the past, in cooperation with the Americans, ‘carried out one or two flights with special aircraft near Soviet territory with the intention of “sniffing” at Russian transmissions’. In reality, a veritable fleet of elint aircraft had been buzzing around the perimeter of the Soviet Union for more than five years. Churchill was assured that they kept at least thirty miles from the Soviet coastline. Alexander continued:

The aircraft are almost certain to be picked up by Russian radar. In fact we shall be disappointed if they are not, for that is the whole object of the operations. But since the flights will take place at the darkest period of the month, and the Russians do not (as far as we know) possess airborne radar, the risks of actual interception are small.

In other words, if the intelligence operations against Soviet air defences were to succeed, they had to actually create alerts and prompt the Soviets to launch their fighters, as it was precisely these procedures that GCHQ wanted to listen in to. Alexander added that a key purpose in undertaking these operations was ‘making our own contribution to the Anglo–American intelligence pool from which we will expect valuable returns in kind’.[62] What would happen if a secret British spy plane crashed inside the Soviet Union, or if a submarine was caught inside a Soviet harbour? So far the GCHQ’s clandestine collection programmes had been remarkably free of incidents, but this enviable record was not to last for much longer.[63]