Burke Trend and Dick White worked well together, but they were confronted by many challenges. By the late 1960s British intelligence was facing a climate of increasing exposure and revelation, and for some years the press had considered espionage and security to be fair game for stories. Now, with journalists like Chapman Pincher on the prowl and the additional excitement caused by the publication of Kim Philby’s memoirs in 1968, the floodgates were opening. Society was becoming less deferential, and newspapers were increasingly reluctant to accept an official definition of what constituted the national interest. Harold Wilson’s antagonistic relationship with the press, and his endless talk of conspiracies, made things much worse.49 Indeed, it was amazing that GCHQ had not been mentioned publicly during the course of the D-Notice affair. NSA was not so lucky: the previous year a series of articles on sigint in the New York Times had revealed short-range collection from the American Embassy in Moscow, as well as satellite interception of the car phones with which Politburo-level Zil limousines were equipped as they sped around Moscow.50
Above all, the Vietnam War served as a focus for dissent on both sides of the Atlantic, and crystallised the sense of generational change. Everywhere established authority and conventional attitudes were being challenged. The spring of 1968 was marked by violent anti-war protests on the streets in Washington, London and Paris. In the early summer of that year, Joseph Wenger, one of NSA’s most senior code-breakers, confided in a close friend and fellow naval officer that his children did not respect his service to his country, and had complained that he was engaged in a ‘shady business’. This had clearly upset him. His friend remarked that Sir Dick White had recently received exactly the same reaction from his own children when his position as Chief of SIS was revealed by the press. White’s rejoinder was a good one. He reportedly asked his children, ‘Do you consider me to be any less good a husband and father and good citizen?’ Upon reflection, his children admitted that they ‘guessed not’. Nevertheless, the spirit of the age was presenting new challenges for intelligence officers everywhere.51
13
Intelligence for Doomsday
In March to August 1968 the JIC consistently took the view that the USSR was unlikely to invade Czechoslovakia.
Douglas Nicoll, Internal Report on the JIC and Warning of Aggression1
During the early hours of 21 August 1968, Soviet forces invaded Czechoslovakia. Accompanied by troops from three other Warsaw Pact countries, their purpose was to crush the ‘Prague Spring’, a liberalising tendency within the Eastern Bloc which Moscow considered dangerous. Czech youths threw petrol bombs at Soviet tanks as they moved through the streets of Prague. Buses were set ablaze, and protesters finally massed in Wenceslas Square. Elsewhere, students actually climbed onto Soviet tanks to argue with their drivers. However, the Czech Army stayed in their barracks, and there was no serious fighting. Within a day, Alexander Dubček, the reformist leader, and his colleagues were on their way to a KGB compound in the Carpathian Mountains. Moscow rationalised its move on the basis of the ‘Brezhnev Doctrine’, which asserted the right to prevent any member state from leaving the Warsaw Pact. The Red Army, which had not hitherto been stationed on Czech soil, arrived in force and remained as uninvited guests all the way through to 1991.2
The best possible sources on Soviet intentions towards Czechoslovakia during the summer of 1968 were the ultra-secret short-range sigint stations hidden in the British and American Embassies in Moscow. These were able to scoop up a variety of interesting telephone calls, including those made from the cars of the senior Soviet leadership. Code-named ‘Gamma-Guppy’, this material was given a very limited circulation. However, the Soviets were aware of these operations, and were careful about what they said on the phone. In mid-August, Marshal Andrei Grechko, the Chief of Warsaw Pact armed forces, had been flying to various Eastern Bloc capitals to assess opinion on an invasion. On his return, he called Leonid Brezhnev, the Soviet Premier, from Moscow Airport. However, he spoke in a prearranged code, and no one in Western intelligence knew if this cryptic conversation meant the invasion was on or off.3
For the British, the Soviet invasion was an intelligence disaster of the first order. Dick White, who had just been appointed as Britain’s first Cabinet Office Intelligence Coordinator, was horrified by the failure of the JIC to offer any warning that it was imminent. The Defence Secretary Denis Healey and the Chief of the Defence Staff also subjected the recently created Defence Intelligence Staff, with its specific warning role with regard to Soviet military actions, to severe criticism. Dick White set up a small committee to look into the workings of the DIS, and indeed for a while ‘its very future seemed fraught with peril’.4 Conclusions were soon reached. ‘Dick White asserted that in 1968 the DIS had been correct in their forecast of the Czech invasion by the Soviets, whereas the FCO [Foreign and Commonwealth Office] and the Friends [SIS] had been wrong.’ Sigint had not provided a clear warning, and had been unable to distinguish invasion preparations from routine Warsaw Pact military exercises. By contrast, defence sources had the real information, and so the DIS had come to the right conclusion. However, as White put it, the DIS’s ‘cutting edge’ had not been sharp enough to persuade the JIC.5
Although sigint had not offered a definitive warning, Britain had excellent assets on the ground. A team of British military liaison officers known as ‘Brixmis’ who were stationed in East Germany under a long-standing agreement with the Soviets had been watching the Red Army as it mobilised for a ‘major exercise’. The classic question these observers confronted was whether this was an exercise, a bluff, or a real invasion. The fabulous anthropological knowledge that Brixmis had developed over twenty years of watching its subject provided the key warning indicator. It had observed the Red Army going out on exercise countless times, and knew how it behaved. A routine exercise provided an opportunity to leave broken or faulty vehicles in barracks to be worked on by engineers. What marked this out as the ‘real thing’ was that the Soviets took everything with them. Vehicles were piled high with personal effects, showing that the troops were not expecting to come back for some time. Unserviceable vehicles were also taken. Indeed, there were so many of these that the Soviets ran out of towing chains, and made makeshift ones by plaiting fence chains together. This was a small detail, but to the expert watchers of Brixmis it spoke volumes. The Soviets were definitely up to something.6
Why did the JIC get the invasion of Czechoslovakia so badly wrong, despite such excellent indicators? Quite simply because its chairman, a rather lofty diplomat called Denis Greenhill, refused to accept the evidence that was staring him in the face. He insisted that the Soviet mobilisation was only an attempt to apply psychological pressure, and argued that if he was in the Soviets’ shoes he would think the wave of international criticism an invasion would provoke too high a price to pay. In other words, he thought like a British decision-maker, not a Russian one – a classic example of a basic analytical mistake called ‘mirrorimaging’. The CIA made much the same mistake, and only the West Germans accurately predicted the invasion. After the inquiry, Dick White decided to beef up the Defence Intelligence Staff. His chosen instrument was the redoubtable Admiral Louis Le Bailly, whom he made Director General of Intelligence, or ‘DGI’, at the Ministry of Defence. The DGI became Deputy Chair of the Joint Intelligence Committee, sitting alongside the senior military intelligence officer, who was already a member of the committee. The DGI was also boosted with his own personal staff. Thereafter, the diplomats and the military fought for control of Britain’s central intelligence machinery.7
In the late 1960s the West had launched a charm offensive towards the Eastern Bloc called ‘détente’, which aimed to encourage precisely the liberal tendencies that Brezhnev had now cracked down on. The invasion of Czechoslovakia in August 1968 did not derail Western efforts to pursue détente. However, it did raise awkward questions. NATO forces had presumed a good degree of warning of any Soviet attack on the West, but would they a
ctually receive this warning if a Third World War broke out? The invasion also prompted British commanders in Germany to think harder about intelligence in the first anxious hours of a confrontation between East and West. Although Brixmis provided a fabulous source of both operational and technical intelligence, it was a peacetime mission and was expected to be rounded up before any military action took place. Thereafter, commanders would need a reliable source of intelligence. Their primary need was to track the movement of the main Soviet thrusts, together with reinforcements mustering anything up to three hundred miles to the rear. In a future war, British commanders would hope to disrupt the Warsaw Pact’s emerging battle plan and destroy its momentum. The most demanding task would be surveillance well behind the enemy front line.
One might have expected senior British commanders to have turned to sigint and GCHQ. Instead, during the 1960s those in Germany seem to have rejected the wonders of electronic monitoring in favour of the least technical option, human reconnaissance from ‘stay-behind patrols’. This was often referred to in the local parlance of hardy special forces soldiers as ‘the Mk.1. Eyeball’. From the onset of any future war, intelligence inside the Soviet-occupied areas would have been provided by dedicated stay-behind parties from NATO special forces. Prevailing doctrine suggested that these special forces had several deep-penetration roles in wartime. These included the collection of intelligence by active or passive methods, offensive operations by small parties, cooperation with partisans or guerrillas, and assistance to escapers such as downed pilots. However, in the context of a global war, intelligence was deemed to be the predominant task of SAS-type units. The only sabotage-type activities that were considered important were efforts to destroy enemy nuclear weapons or missile sites.8 This emphasis on intelligence reflected a growing anxiety that Warsaw Pact forces might move too fast to allow sigint or air reconnaissance to provide effective targeting intelligence for NATO artillery, and eventually tactical nuclear strikes. Everyone knew that NATO commanders would press for early deployment of nuclear weapons, for fear that Warsaw Pact units would deliberately move close to their opponents, making the use of tactical nuclear weapons increasingly difficult.9
In the 1960s the British Army on the Rhine developed a secret new force for this important intelligence role. This involved adding a Special Reconnaissance Squadron from the Royal Armoured Corps to strengthen 23 SAS Regiment. During an initial alert, the Special Reconnaissance Squadron was expected to hold the fort until the arrival of 23 SAS, which would be flown in from Britain. These units were on short readiness times: their unofficial motto was ‘Wait and Fly – Dig and Die’.10 After the arrival of 23 SAS the two formations were to operate as a single unit, giving priority to sightings of ‘nuclear units, formation HQs, armour, and bridging and ferrying equipment’. Their main task would be to provide the target intelligence for battlefield missile systems and heavy artillery.
These special units were based at Padeborn in central Germany, and were equipped with high-frequency Morse to provide long range and, hopefully, continued communications. The expected onrush of Warsaw Pact tanks meant there was no need for them to practise skills to penetrate the enemy front line. Instead, the drill was to move forward quickly, by any available soft transport, such as three-ton trucks. Special forces would eventually meet the units tasked as the delaying force, and as these elements withdrew, the special forces would stay behind. Preparations for such operations had become quite elaborate by the late 1960s, with pre-identified hides offering good fields of observation over likely routes of advance for the Red Army, and some pre-positioned SAS stocks which had been buried underground. Much of this activity was focused on what commanders referred to as the ‘demolition belt’. These were ‘killing zones’, some way east of the Rhine, where it was hoped that bottlenecks would occur amongst aggressor forces twenty-four hours or so after the Warsaw Pact forces had attacked.11
Some special forces units were equipped with Atomic Demolition Munitions (ADMs) which it was hoped would slow the advance of Soviet armour. The US special atomic demolition munitions programme appears to have been code-named ‘Green Light’, and was active from the mid-1960s to the mid-1980s. In 1960 Sergeant Major Joe Garner of US Army Special Forces was probably the first person to make a parachute jump with a small atomic weapon, which was strapped to his body during a field exercise.12 Although ADMs posed serious problems of contamination because they would have generated a great deal of fallout, they were nevertheless popular with military planners, since they were considered to be more defensive and less escalatory than other types of tactical nuclear weapons.13 By 1971, Britain appears to have been moving ahead with its own ADM programme, designated ‘Project Clipeus’; however, its full extent is still shrouded in secrecy. Clipeus was a James Bond-type suitcase nuclear weapon, which even now seems to have more in common with spy fiction than reality.14
Depending on a few soldiers for the targeting of nuclear artillery and key battlefield missile systems such as ‘Corporal’ and ‘Honest John’ during the first few hours of war was an obvious weakness in British plans, and there were increasing concerns about the potential vulnerability of such troops. An extensive programme of research was carried out during 23 SAS’s annual autumn training exercise held in Germany, code-named ‘Badger’s Lair’. Eight SAS teams were deployed on the Soltau training area, while British Army signals teams played the role of the KGB and conducted elaborate tests to investigate their vulnerability to Soviet intercept and direction-finding procedures. RAF units tested concealment procedures by overflying the SAS hides with thermal-imaging cameras and monochrome photography. Vulnerability to searches by dog patrols was also tested. To the dismay of the SAS, of the thirty-nine hides they created, all but two were found within the first six hours. The patrols and dogs proved highly effective, but even more remarkable was the success of electronic warfare sensor vehicles. To the surprise of the research teams, these intercepted not only Morse code but also burst-encrypted traffic, produced by a special radio that stored a message and then sent its entire contents in less than a second, to try to defeat enemy listeners. Direction-finding bearings were achieved at ranges of up to twelve kilometres, and accurate bearings using triangulation between three vehicles at ranges of five kilometres.15 The ability to track burst transmissions was especially alarming, and prompted a decision to develop better communications equipment for stay-behind parties.16
This heavy dependence on SAS-type activities raises profound questions about sigint. After all, it had been a key source of real-time operational intelligence during the Second World War, and as we have seen, had continued to be important in small conflicts such as the Confrontation in Borneo. British commanders in Germany enjoyed their own significant sigint and electronic warfare capability, which was directed by an Intercept Control Centre. In time of war there would have been some additional sigint from airborne collection, albeit these planes’ main role would have been to support the British V-bombers. The backbone of this was 225 Signals Squadron, which was tasked to support British forces in Germany. These units had provided an invaluable intelligence contribution in peacetime through their work on the Soviet order of battle. Over the years, operators had learned the radio signatures of individual Soviet units, allowing them to learn a great deal about each distinct formation. However, these tactical sigint units themselves felt that their very success in peacetime had led to a dangerous overestimation of their likely contribution in any future war. They observed:
In a war of limited duration in NW Europe, if the standard of security in WP [Warsaw Pact] communication links is good, the timely intelligence and useful steerage that 225 can provide will be very small…The problem…is not so much one of equipment quantities but rather the difficulty of conducting EW [Electronic Warfare] in a highly mobile tactical environment…The wartime limitations of 225 Signal Squadron are not widely known; as a result the squadron’s capabilities are overrated.
Electronic warfare was
one of the few planned intelligence sources for determining the location of enemy headquarters. Yet sigint specialists feared that the Soviets might be routinely ‘remoting’ the radios associated with their major headquarters, at a distance of perhaps two kilometres, in which case the effectiveness of any direction-finding efforts would be drastically reduced. Their only hope was that under the stress of war, Warsaw Pact communications security might lapse, but this was by no means certain. Indeed, some predicted that for the first twenty-four hours the enemy might advance on predesignated lines and keep near radio silence.17
Senior British officers clearly hoped that their tactical sigint organisation would provide some information on enemy deployment patterns as well as intentions in the first few hours of war. They also desired information on the ‘location of enemy headquarters and missile launching sites’. However, in reality only the smaller missile-launching sites were likely to be within intercept range, while the more important SCUD missile sites would be outside the typical operating range of tactical sigint, which was forty to fifty kilometres. British commanders were intrigued by the American decision to introduce airborne tactical sigint systems on light aircraft, but noted that these would have to fly far from the battle area in order to survive.18 In fact, this American system – known as ‘Guardrail’ – proved highly effective.19 The deployment of these small sigint aircraft reflected good previous experiences with the U-2s which had undertaken regular perimeter sigint flights around the Eastern Bloc.20
GCHQ Page 27