Читать книгу GCHQ - Richard Aldrich - Страница 19

…with ‘Sigint’ locking onto targets with pinpoint accuracy, our military ached to have a go.

Tim Hardy, Special Branch, Sarawak, April 1964¹

In the 1950s, GCHQ’s top priorities were warning of an impending war with Russia, and gathering intelligence on Moscow’s growing nuclear arsenal. However, on a day-to-day basis, the Middle East, Africa and Asia were the regions where sigint made a tangible difference. Since the end of the Second World War, Britain had been involved in a prolonged ‘escape from empire’, retreating from her colonies and hoping to replace them with a vibrant Commonwealth of newly independent states. The reality was more complex, since many of these countries contained elements that were keen to evict the British faster than they wished to go. Some hosted guerrilla groups sympathetic to Moscow, others were divided communities that faced a troubled journey towards independence. The result was that Britain was involved in an endless litany of small wars that stretched from the dusty deserts of Yemen to the steamy jungles of Borneo. Because these were often guerrilla wars, finding the enemy could be the main challenge, and here sigint was in its element. Moreover, right across Asia and Africa, cyphers were less secure than those of countries like Russia, so GCHQ could also read plenty of high-grade diplomatic traffic.

Although sigint helped to smooth the end of Britain’s empire, GCHQ itself did not always want empire to come to an end. Because the 1950s and 1960s were an era when a great deal of communications was sent over long distances using high-frequency radio, GCHQ depended on the remnants of empire to provide a global network of ground stations to collect these signals. Indeed, Britain’s imperial real estate was one of the key contributions to UKUSA, and was of particular assistance to the United States. Accordingly, in many colonies there were defence and intelligence bases that Britain wished to retain, prompting officials to drag their feet over independence. Elsewhere, the British attempted to persuade post-independence governments to permit some bases to remain.²

Throughout the 1950s Britain fought one of the most protracted colonial struggles of the post-war era, the Malayan Emergency. The enemy were a hardened band of Communist guerrillas who had been Britain’s uneasy allies against the Japanese during the war. The military forces of the Malayan Communist Party, or ‘MCP’, led by Ching Peng, operated from refuges in the dense jungle. Britain did not initially recognise the seriousness of the Emergency in Malaya, allowing it to get out of hand. However, in October 1951 the MCP succeeded in assassinating Sir Henry Gurney, the British High Commissioner. Thereafter, striking back at the guerrillas and eliminating Ching Peng became a near-obsession for the security authorities in London. When Oliver Lyttelton, the Colonial Secretary, returned to London to report on Gurney’s assassination he promised the Cabinet that he would form special teams ‘aimed at certain individuals’. These were effectively killer squads, and he gave a firm assurance that they would ‘hunt down individual men from Communist higher formations through their families, properties, sweethearts etc.’.³

Locating the guerrilla headquarters in Malaya was easier said than done. In 1950 a sigint-equipped Lancaster from the RAF’s 192 Squadron was sent out to help in the hunt for the insurgents by tracking their radio communications. Later, undercover agents planted batteries with excessively high power on the guerrillas to damage their radios. When they were repaired, the workshops the guerrillas used were bribed to secretly modify the sets to give out a stronger signal. This gave the opportunity for sigint to achieve a direction-finding fix on the main guerrilla bases. Bombers from the RAF and the Royal Australian Air Force were standing by, and lightning raids were carried out on the deemed location of the signals. Avro Lincoln bombers dropped thousands of tons of bombs into the dense jungle at likely guerrilla locations. Their pilots were always impressed by the resilience of the jungle: their largest bombs vanished into the triple-canopied green foliage below them, and from the aircraft little impact was visible. It is not known how successful these operations were, but Ching Peng, the most important prize, certainly eluded them.⁴

In January 1952, Sir Gerald Templer arrived as the new High Commissioner in Malaya. Templer possessed the authority and charisma necessary to create a unified government machine and to implement an effective counter-insurgency strategy. Although famed for his emphasis on ‘hearts and minds’, he also sorted out intelligence, creating a coherent structure in which the army, the police and the civil authorities were forced to share intelligence. All this was done with his customary fiery language – he was quite incapable of uttering a sentence without a cussword in it.⁵

Despite Templer’s forceful direction, intelligence did not improve overnight. An important intelligence issue that was never quite resolved was the question of who was actually behind the insurgency. The Colonial Office and the Special Branch officers of the Malayan Police preferred to interpret the Emergency as a wicked plot initiated by Stalin or else Mao, while the British diplomats tended to see it more as a local anticolonial uprising. During the mid-1950s GCHQ began to intercept what it believed to be wireless traffic between the MCP guerrilla leadership and the Chinese Communist Party in Peking. The Special Branch presented this intelligence to senior British officials in Kuala Lumpur with some delight as evidence of its theory of external direction, but only in a summarised form. Diplomats in Kuala Lumpur were sceptical, and asked to see the full transcripts of the transmissions. A major altercation followed, with the diplomats accusing the Special Branch of bending the evidence, while the policemen accused the diplomats of a lack of trust. The issue of exactly how close the MCP was to Peking was never resolved.⁶

GCHQ’s most important outpost in Asia was Hong Kong. China was the venue of one of Britain’s early Cold War code-breaking triumphs. Between March 1943 and July 1947 GCHQ was able to read the high-grade Russian cypher traffic passing between Moscow and its mission at the headquarters of Mao Tse-tung’s People’s Liberation Army in Yunnan. This was a highly secret programme, and GCHQ only began passing material to the Americans in March 1946. The decision not to share until this point may have reflected anxieties about the strong differences within the American administration about China policy, but it is noticeable that the spring of 1946 also marks the advent of the revised BRUSA agreement.⁷ Exactly how this breakthrough was achieved when many other Russian high-grade cypher systems remained immune to attack is still a mystery. However, SIS had placed a rather eccentric officer called Michael Lindsay at Mao’s headquarters in Yunnan, where he was assisting the Chinese Communist communications team as their ‘principal radio adviser’. This may eventually prove to be part of the story.⁸

The British colony of Hong Kong was of special value to the United States. This reflected the fact that, after the end of the Chinese Civil War that brought Mao Tse-tung to power in 1949, the United States did not even have an embassy in mainland China. ‘Hong Kong became an American watchtower on China,’ recalls Jack Smith, who looked after the Far East in the CIA’s Office of National Estimates.⁹ GCHQ joined with the Americans and the equivalent Australian organisation, Defence Signals Branch, to develop the facilities in Hong Kong. Washington received the full intercept output of Hong Kong, but with the onset of the Korean War demands for intelligence went up sharply, and Washington considered that combined US–UK intercept facilities in the Far East were ‘far short of requirements’.¹⁰

In July 1952 the US Communications Intelligence Board persuaded its British opposite numbers of the ‘urgent need’ to send an additional eight-hundred-strong US Air Force sigint unit to Hong Kong to join the hard-pressed British and Australians. However, this was vetoed by the Governor of Hong Kong, Sir Alexander Grantham, who detested the way in which his territory had become host to a myriad of espionage activities.¹¹ Once the Chinese had intervened in the Korean War, an attack on Hong Kong by China was always a possibility. Therefore GCHQ negotiated emergency facilities at Okinawa in Japan for the British and Australian sigint personnel working there.¹²

Even in 1955, the United States was still negotiating for new sites in Asia. Sigint sites were not small or discreet, often requiring vast acres of wireless masts known as ‘aerial farms’ to capture signals of interest. In Taiwan, American officials had run into trouble securing a 335-acre site near Nan-Szu-Pu airfield where they had plans to locate hundreds of personnel from the Army Security Agency.¹³ With repeated clashes between the United States and Communist China over the Taiwan Straits in the late 1950s, the British government reviewed the future of Hong Kong, which seemed exposed, and pondered the short-term value of the continued British presence in the colony. Much turned on the mysteries of the UKUSA alliance, the Anglo–American–Commonwealth sigint pact of cooperation, since Hong Kong hosted British, Australian and American eavesdroppers.¹⁴ Alongside the GCHQ activities there were also vast British and American programmes in Hong Kong for running agents and interviewing defectors from mainland China. During the 1950s and 1960s, both the State Department and the Pentagon considered Hong Kong to be the single most important British overseas territory from the point of view of intelligence-gathering.¹⁵

In order to stimulate more defectors from China to Hong Kong, Britain launched ‘Operation Debenture’ in 1954. This was a covert radio project and constituted ‘the first UK operations of any magnitude for the penetration of Mainland China’. The aim was to provide an undercover broadcasting station that would increase the desire for contacts with the West amongst the Chinese middle classes, and increase defections across the border into Hong Kong. The emphasis was on the ‘purely “intelligence” angle’, and the defectors were needed because SIS human agent coverage of China was weak. The original intention had been to place this ‘black station’ in Hong Kong, but it was eventually located in Singapore, hidden at one of the military bases.¹⁶

The main GCHQ sigint stations in Hong Kong were on the coast at Little Sai Wan and the curiously-named outpost known as ‘Batty’s Belvedere’. The contribution of Australia’s Defence Signals Branch was important, since Australia had identified China as its top sigint target, followed by Indonesia and then Vietnam.¹⁷ During the late 1950s the commander of the sigint station was an Australian called Ken Sly, and originally it was staffed by airmen from the RAF’s 367 Signals Unit.¹⁸ A constant flow of National Servicemen had learnt Chinese at RAF Wythall near Birmingham and later at RAF North Luffenham in Leicestershire, but by 1957 the increasing use of civilians with qualifications in the language was reducing this considerable training requirement. There was also a separate cohort of Vietnamese linguists.¹⁹ Civilianisation brought unexpected security problems, since civilians could not be used for some of the menial duties carried out by service personnel. GCHQ tried to address this problem by employing deaf and dumb locals in the more sensitive locations on the sites.²⁰

Ken Sly was well aware of the attentions of Chinese intelligence. One of the locally employed Chinese, Wal Bin Chang, showed a propensity for taking photographs of groups on social occasions, and ‘also took care to photograph each one of us separately’. Moreover, he tended to volunteer for extra duties at unsociable hours. He was eventually captured on the border trying to cross over into Communist China with a number of documents, including a description of the personal habits of every NCO and officer at the base. He had been entertaining some of them in ‘girlie bars’, and admitted that he had persuaded one of the officers to sleep with his wife, adding: ‘In this way I will be able to obtain much more information of value to our side.’ The officer in question was swiftly discharged. Military staff at overseas listening stations working for GCHQ were a continual target for this sort of honey-trap.²¹ Ken Sly was eventually replaced by a civilian with the rank of Senior Linguist Officer, and moved on to serve in Australia and then with GCHQ at Cheltenham.²²

In both Hong Kong and Cyprus, the British were experimenting with intelligence-gathering radar. At Hong Kong the main site was located three thousand feet up the precipitous cliffs of Tai Mo Shan in the New Territories. Operated jointly by the RAF’s 117 Signals Unit and the Australians, it peered out into Chinese airspace, and its main purpose was ‘to provide intelligence information for the UK, USA and Australia’.²³ Western aircraft regularly intruded over the border to generate an elint response from Chinese defences.²⁴ The site was constructed with great difficulty in 1957 and was operated continuously into the 1980s. By a heroic effort, cranes and lorries had moved materials up to the summit by means of what was little more than a jeep track. During construction a ten-ton crane had been lost over the edge, but fortunately the RAF driver leapt clear before the vehicle disappeared over the cliff. Later, the RAF Regiment, known as the ‘Rock Apes’, who guarded the base, lost two Land Rovers over the cliff. This prompted a local humorist to erect a sign at the base of the uphill trail that warned: ‘Beware of Falling Rocks’.²⁵

GCHQ does not seem to have broken much high-grade Chinese traffic; nevertheless, there were intelligence success stories. One of the most important was the prediction of the detonation of China’s first nuclear weapon in 1964. Like all such programmes, China’s efforts to acquire a nuclear weapon required a vast technical and industrial effort, therefore imagery from overflights together with relatively low-level signals gave a good indication of progress. Archie Potts, the UK’s Deputy Director of Atomic Energy Intelligence, noted that for about five years the British had been aware of an important secret programme controlled by ‘a special ministry’. Plant construction had begun in 1958, with an elaborate effort to produce uranium ore. The Chinese had also ceased their public complaints about superpowers with nuclear weapons. All this prefaced China’s first nuclear test.²⁶

Although NSA viewed Hong Kong as Britain’s single most valuable overseas sigint station, GCHQ placed more emphasis on the Middle East. Immediately after the war, Britain had numerous interception stations. The most important was at Heliopolis in Egypt, which boasted many civilian operators and took in much of the region’s diplomatic traffic. The Army ran a large intercept station at Sarafand in Palestine, while the RAF ran a similar installation at RAF Habbaniya in Iraq. There were undercover listening stations buried within embassies and consulates in countries such as Turkey. By the 1950s Britain had also developed covert sites in northern Iran that were focused on Russia. However, the British Empire in the Middle East consisted of very few formal colonies and had long been an agglomeration of mandates, shaky treaty relationships and uncertain base rights granted by royalist regimes. Egypt, which had achieved independence in 1935, was especially anxious to divest itself of the disfiguring presence of British bases. Accordingly, British sigint gradually fell back towards its last proper colonial foothold in the region, the island of Cyprus.

Cyprus was increasingly the home for every kind of secret radio activity in the Middle East. This included not only Britain’s sigint assets but also the monitoring sites of the CIA’s Foreign Broadcast Information Service, which listened in to news broadcasts around the world. In addition, Cyprus offered a safe haven for Britain’s overt and covert propaganda broadcasting in the region. This mushroomed during the premiership of Anthony Eden, who nurtured a special hatred of Egypt’s nationalist leader General Gamal Abdel Nasser, whom he viewed as a dangerous dictator. Eden urged a reduction of British radio propaganda directed at the Soviets in favour of targeting Nasser.²⁷ As early as 1954 he insisted that a new broadcasting station in Aden covering Iraq and Syria was to receive ‘first priority’, since Nasser’s radio station, The Voice of Egypt, was busily pouring out its own vitriolic message.²⁸ Britain’s main radio weapon against Nasser was the SIS-owned station in Cyprus, Sharq el-Adna. ‘Sharq’ had originated as a wartime British propaganda radio station that had been taken over by SIS in 1948, and been evacuated from Palestine to the safety of Cyprus. It was soon thought to be the most popular station in the region.²⁹ SIS was working with John Rennie, the head of Britain’s Information Research Department, to accelerate four other radio projects in the Middle East, including a secretive ‘black station’ that was being developed at two other sites on Cyprus with a transmitter that could reach as far as Aden.³⁰

On 29 October 1956 Eden launched ‘Operation Musketeer’, a surprise attack to capture the Suez Canal, which Nasser had recently nationalised. Sigint and radio warfare had an important part to play. Arrangements were made for the force commanders to receive a range of key intelligence materials from national sources, including photo-reconnaissance cover and ‘all CX [SIS] reports on Egypt’, as well as material from ‘special sources’, a somewhat coy cover name for sigint. GCHQ attached liaison officers to the main Army, Navy and RAF commanders, and detailed instructions were generated to provide cover for the ‘protection of SIGINT material’.³¹ Most of the sigint coverage came from 2 Wireless Regiment at Ayios Nikolaos near Famagusta in eastern Cyprus, with additional help from listeners at Dingli on Malta. While the coverage was good, the radio channels available to push this material forward to field commanders were often choked. In addition, a small tactical ‘Y’ intercept unit was being prepared to accompany the land force from Cyprus to the landings in Egypt, and was eventually based at Port Said.³²

The British not only had to hide the invasion preparations from the Egyptians, but also from the Americans. Britain had engaged in an elaborate plot with the French and the Israelis which hid the real reasons for the intervention by presenting it as the arrival of a so-called ‘peace-keeping’ force for the disputed Suez Canal Zone. Eisenhower and his Secretary of State John Foster Dulles were astonished by Anglo–French–Israeli collusion over Suez. In the autumn of 1956 Washington’s eyes were elsewhere, distracted by the uprising in Hungary, while in the Middle East its focus was on the possible breakup of Jordan and the likelihood of Israeli and Arab attempts to divide the spoils. American U-2 flights out of Turkey detected an Israeli mobilisation, but this was interpreted by some as part of Israeli ambitions on the West Bank. Allen Dulles, the Director of CIA, was tracking reports of an imminent coup in Syria.

Nevertheless, the ability of the British to hide ‘Operation Musketeer’ from NSA raises some interesting questions. What were American sigint liaison officers doing? During the Suez invasion there was a US Sixth Fleet exercise off Crete, yet American Naval intelligence conceded frankly that it had ‘no warning of British intentions’.³³ Much of the story can be explained by NSA’s obsessive focus on Russia, with the vast majority of its assets in locations such as Turkey looking northwards to the missile-testing stations of the Caucasus. Meanwhile NSA depended on GCHQ for much of its coverage of the Middle East. Moreover, the crisis occurred just as the American code-breakers were moving to their new building at Fort Meade. The failure to spot the Suez Crisis had a significant effect on NSA, triggering a post-mortem and the creation of new divisions based on country or geographical lines.³⁴

The British deliberately blanked their American allies. In a neat piece of choreography, the British Ambassador to Washington was replaced at this moment, with the new man being sent across the Atlantic by passenger liner. He was thus in mid-ocean when the Suez Crisis broke, and could not be accused of having deceived the Americans. In Tel Aviv, the British and French Military Attachés were told to give their American counterpart a wide berth.³⁵ However, the American Military Attaché realised something was up when his civilian driver, a reservist in the Israeli Army who had only one arm, one leg and was blind in one eye, was suddenly recalled to duty. His American employer deduced – quite correctly – that if his driver was being mobilised it could only mean one thing: imminent war.³⁶

The sharpest Americans knew something was afoot. On 12 September 1956 Robert Amory, Deputy Director for Intelligence at the CIA, set up a highly secret joint group from the CIA, NSA, the State Department and military intelligence to watch the Middle East round the clock.³⁷ Its main source of information was an expansion of the U-2 spy plane operations from Wiesbaden covering the Middle East. The CIA’s own U-2 official history claims that this allowed them to predict the attack on Egypt three days before it took place.³⁸ This is probably an exaggeration: the U-2 evidence of growing forces on the ground was not precise enough to make such a forecast. Allen Dulles, the Director of the CIA, told Eisenhower he believed the Israelis were about to attack Jordan. Eisenhower attached special significance to NSA reports of an increase in signals traffic between Tel Aviv and Paris.³⁹ Almost certainly from sigint, the Americans had also picked up news of a secret meeting between the British Foreign Secretary, Selwyn Lloyd, and the French in Paris on or about 15 October. This was the very sensitive meeting that sealed the deal over the Suez invasion. Allen Dulles recalls: ‘I remember I had a long talk with Foster [Dulles] about what this might mean in view of the fact that we were not otherwise informed about it.’⁴⁰ But Eisenhower personally dismissed the significance of the military build-up on Cyprus, refusing to believe that Britain would be ‘stupid enough to be dragged into this’. Remarkably, six weeks after the invasion of Suez, many in the CIA were still uncertain whether the British had colluded directly with the Israelis.⁴¹ Both NSA and the CIA had also failed to predict the Russian invasion of Hungary, so 1956 was not their best year.⁴²

Deliberate American pressure on the pound eventually forced Britain’s ignominious withdrawal from Suez, and contributed to Eden’s sudden resignation in January 1957. Eden’s foreign policy may have failed, but the intelligence support he received had been excellent. In the wake of Suez, Selwyn Lloyd wrote to Eric Jones, the Director of GCHQ, congratulating him on the torrents of Middle East intelligence that sigint had provided during the crisis, particularly after the seizure of the canal. ‘I have observed the volume of material which has been produced by G.C.H.Q. relating to all the countries in the Middle East area,’ he wrote, suggesting that the traffic of many countries was being read, and added: ‘I am writing to let you know how valuable we have found this material and how much I appreciate the hard work and skill involved in its production.’ Jones passed on these congratulations to units such as the Army’s 2 Wireless Regiment on Cyprus and the RAF’s 192 Squadron.⁴³ There had also been shipborne signals interception by the Royal Navy. The RAF airborne signals element was especially important during the invasion. The ageing RB-29 Washingtons had been despatched from Watton to map the characteristics of Egyptian anti-aircraft defence. This included the habit of shutting down air-defence radar routinely just after midday – a priceless piece of information.⁴⁴

At a higher level, GCHQ read much of Cairo’s diplomatic traffic with key embassies in the region during the mid-1950s, such as those in Amman and Damascus.⁴⁵ It also read traffic with Egypt’s London Embassy.⁴⁶ No less importantly, GCHQ stepped up its watch on the Soviets. On 15 November 1956, Britain’s leaders were reassured that there was ‘still no evidence from signals intelligence sources of any large-scale Soviet preparations to intervene by force in the Middle East’.⁴⁷ However, there had been problems. Some of the newly civilianised sigint sites had complained about working round the clock during the crisis, causing managers to wonder about the wisdom of non-military intercept operations.⁴⁸

Despite GCHQ’s operational success, the Suez Crisis left a problematic legacy. It led directly to the eviction of GCHQ from some of its more valuable real estate in the Middle East and the Indian Ocean. In December 1956 GCHQ was just opening a large and well-equipped secret sigint station covering the Indian Ocean at Perkar on Ceylon, which had been constructed at a cost of close to £2 million. The Ceylonese government had wanted to free up access to the old sigint site at HMS Anderson for redevelopment. The purpose of the GCHQ site at Perkar was hidden from the Ceylonese, requiring the British to generate a cover story. Much debate had taken place in London over whether to let the Ceylonese Prime Minister, Solomon Bandaranaike, in on the real function of the station. GCHQ decided against candour, fearing ‘leakage’.⁴⁹ British officials had always been convinced that ‘the real purpose could be easily disguised’.⁵⁰

Endless effort had gone into the Perkar site. By 1955 it had been upgraded to monitor signals traffic from ‘all bearings’, and boasted a vast aerial farm that covered more than four hundred acres.⁵¹ Yet the Suez operation effectively destroyed this expensive new facility almost as soon as it was completed. The Ceylonese were incensed at Eden’s imperial escapade, and believed the British had refuelled ships in Ceylon en route to the invasion of Egypt. They now demanded a schedule for the removal of all foreign bases, without exception. The Treasury was aghast, stating that even a brief visit to Ceylon ‘brings home the complexity of these installations’ and ‘their vital importance’. Officials came up with the preposterous idea of using service personnel in civilian clothes in the hope of assuaging the Ceylonese.⁵² Bandaranaike stamped his foot, insisting that all the British, however attired, had to go. A compromise was agreed: ‘The GCHQ station can be given up entirely, but we should like to keep it in operation for five years.’ Ultimately, Britain had lost the best site in the Indian Ocean.⁵³

GCHQ felt the reverberations of Suez elsewhere. In Iraq, Britain enjoyed a good relationship with the ruler King Faisal. As a result, the British had been allowed to retain a number of bases. One of these was RAF Habbaniya, not far from Baghdad. Superficially this looked like so many military aerodromes in the Middle East, but in fact it housed 123 Signals Squadron, later 276 Signals Squadron, which ran a large sigint monitoring station. Airborne sigint flights from Habbaniya crossed into Iran, and then loitered over the Caspian Sea. However, as a result of Suez, Faisal’s political situation deteriorated rapidly, with uprisings in the cities of Najaf and Hayy. Iraq’s membership of the Baghdad Pact, a British-managed military alliance, only exacerbated popular hatred of the regime. Then, in the summer of 1958, Faisal’s ally, King Hussein of Jordan, asked for military assistance during a growing crisis in the Lebanon. The Iraqi Army put together an expeditionary force, but in the early hours of 14 July 1958 the assembled column turned against its own supreme commander, marched right into Baghdad and carried out a coup. Revolutionary officers arrived at the Royal Palace at 8 o’clock in the morning and ordered the King, his immediate family and his personal servants into the courtyard. They were politely asked to turn away from their captors, whereupon they were machine-gunned. Most died instantly, but Faisal survived a few hours. Fortunately, GCHQ intercepts of Egyptian diplomatic traffic gave precise information about Nasser’s parallel plots against the King of neighbouring Jordan a few days later, prompting timely British support for the beleaguered monarch.⁵⁴

However, Britain’s time in Iraq was now up, and the final departure from RAF Habbaniya was anything but orderly. The vast base had quickly been occupied by the Iraqi Fourth Armoured Division, and the British had even been denied access to their own signals installations and aerial farms. Most of the RAF’s 276 Signals Unit were evacuated to temporary tented accommodation on Cyprus, where they continued their interception work amid terrible conditions. Three hundred personnel remained at Habbaniya, presiding over the residual technical facilities and stores. They were continually provoked by Iraqi forces, and it was not unusual for them to ‘end up in the Iraqi guard room’. Although much of the radio equipment had been removed, the remnants included specialist signals vehicles, machine tools and fuel, together with the entire contents of a nearby RAF hospital.⁵⁵