Intelligence Post WWII Part I

Argentinian Invasion of the Falklands

Military operations have changed greatly since the end of the Second World War, most of all because the development of nuclear weapons has effectively prevented the major states from fighting the sort of full-scale struggles for decision which are the subject of this book. Big wars are now too dangerous for big countries to fight. That does not mean that the world has become a safer place for the common man. On the contrary. It is estimated that armed conflict since 1945 has killed fifty million people, as many as died in the Second World War. Most of the victims, however, have perished in small-scale, random struggles, many scarcely to be dignified even by the name of civil war. In the last fifty years it is not the methods or weapons of 1939—45 that have harvested the major proportion of violent deaths – aerial bombardment or battles between great tank armies or the relentless grind of infantry attrition – but skirmish and all too often massacre with cheap small arms.

Even in such few major wars as have been fought, there have been few large-scale conventional battles and their number has tended to decline over time. Thus, while the Korean war of 1950–3 was almost exclusively a conflict of infantry and tank armies, and the Arab–Israeli wars of 1956–73 likewise, the biggest war of all, in Vietnam, was a protracted counter-insurgency struggle, marked by the clash of armies scarcely at all. Though the Iran–Iraq war of 1980–8 saw much heavy fighting, Iran’s lack of heavy equipment and use of under age conscripts in suicide attacks made it an unequal contest bearing little resemblance to other wars of the twentieth century. In 1991 Iraq was forced to abandon its illegal occupation of Kuwait as a result of defeat in one major tank battle; but its army, more concerned to surrender than to stand its ground, cannot really be said to have given battle at all. The same can be said of its performance in the 2nd Gulf War of 2003, in which intelligence played an important role in the targeting early on of the Iraq leadership.

That episode apart, the post-war military record yields few examples of outcomes being influenced by operational intelligence of the sort assessed in the previous chapters. Intelligence services have never been busier than they are in the nuclear world and consume more money than has ever before been spent. By far the greater proportion both of effort and funds is devoted, however, to early warning and to listening, continuous processes, intended to sustain security, not to achieve success in specific or short-term circumstances. The elaborate infrastructure of early warning – radar stations, underwater sensors, space satellite systems, radio interception towers – is enormously expensive to build, maintain and operate and so are its mobile auxiliaries, particularly airborne surveillance squadrons. The intelligence material thus collected, categorised by professionals as sigint (signals intelligence), overlapping with comint (communications intelligence) and elint (electronic intelligence), requires processing and interpretation by thousands of analysts and computer technicians. What they do and what they achieve is rarely published. The public anyhow seems indifferent to what is unquestionably the most significant sector of contemporary intelligence activity. Understandably, the complexities of intelligence technique must baffle even highly educated laymen. Only the most specialist of experts can hope to comprehend what intelligence agencies now do. It is possible, with application, for the interested general reader to follow descriptions of how the Enigma machine worked and of how the problems it presented to cryptanalysts were overcome. Modern ciphers, created through the application of enormous prime numbers to language, belong in the realm of the highest mathematics and are alleged to defy attack even by the most powerful computers yet built.

It is not surprising, therefore, that the intelligence world attracts attention only when there is a breach of security, typically in recent years by the ‘defection in place’ of an intelligence operative who yields to greed or lust or exhibits defects of character not identified at the time of recruitment. There has been a steady trickle of such scandals, long post-dating the sensational unmasking of the ‘Cambridge’ spies in Britain and affecting the American and Soviet services which were presumed to have been warned against such occurrences in their own ranks by the ‘Third’ and ‘Fifth’ Man episodes.

Public interest is also engaged by accounts of the effect of human intelligence, humint, on recent or current military operations, where such effect can be shown. Humint has unquestionably played a major part in Israel’s successful efforts to hold at bay its Arab neighbours in four major wars, much minor conflict and its continuous struggle for security, for the ingathering of Jews from neighbouring lands allowed its intelligence services to recruit patriotic operatives who spoke Arabic bilingually and were able to pass as natives in their countries of former residence. It is understandable that the successes of Israeli humint remain almost completely secret. During the Vietnam War the American CIA conducted a large-scale campaign of destabilisation against the Viet Cong, largely by the targeted assassination of Viet Cong leaders in the South Vietnamese villages. Operation Phoenix remains unacknowledged; the Vietnam War was eventually lost; it would nevertheless be illuminating to know what effect Phoenix had on its conduct.

The only conventional military conflict of recent times for which a reasonably complete picture of the influence of intelligence on operations is available in all or most of its complexity – signit, elint, comint, humint and photographic or imaging intelligence – is the Falklands War of 1982, between Britain and Argentina. Rights of sovereignty over the Atlantic islands of the Falklands or Malvinas, which include such Antarctic outliers as South Georgia, Graham Land and the South Shetland, Orkney and Sandwich groups, has been disputed between Britain and Argentina since the nineteenth century. The small Falklands population is exclusively British (the other territories are effectively uninhabited) but it is a universal and deeply held belief in Argentina that the lands are theirs. Argentina has a troubled political history. Once a country of great wealth, which attracted to it over the last century large numbers of immigrants, including poor Italians seeking a better life outside Europe and an English minority who came to supply its commercial and professional class, Argentina suffered serious economic decline in the mid-twentieth century. Discontent brought to power a populist Peronist regime, so called after Colonel Juan Peron, its leader. Peronist mismanagement provoked a military coup in the 1970s. When the military junta itself became unpopular, it decided to restore its fortunes by reviving the claim to the Falklands. Recovering the Malvinas was a cause around which all Argentinians could unite.

Britain was long used to Argentina’s Falklands demands. It did not take their revival in 1981–2 very seriously. Negotiations proceeded at the United Nations in New York: they were not marked by urgency and the British found the Argentinians in reasonable mood. Unknown to Britain, however, the junta, led by General Leopoldo Galtieri, had already decided to mount an invasion at latest by the October of 1982, when it was calculated that the only Royal Naval ship on station, the ice patrol vessel Endurance, long scheduled for retirement, would have been withdrawn. As late as March 1982, no military preparations had been made and no diplomatic crisis appeared to impend. Then what seems a chance factor altered the tempo. An Argentinian scrap reclamation party arrived at Leith in South Georgia, the Falklands dependency, declaring it was there to dismantle an old whaling station. The scrap men raised the Argentinian flag but failed to seek permission for their work from the local station of the British Antarctic Survey, the government authority. When visited, they hauled down the flag but did not regularise their presence. Constantino Davidoff, their leader, denied then and afterwards that he was sponsored by the Argentinian navy but he is believed to have had a meeting with naval officers before landing. Once he was ashore, the British Foreign Office felt it had to act; the Ministry of Defence was more reluctant, since it regarded operations 8,000 miles from home as beyond its capabilities. Under Foreign Office pressure, a case was made to the Prime Minister, Mrs Margaret Thatcher, who ordered Endurance, with a party of marines from Port Stanley, the Falklands capital, to sail for South Georgia and to await orders.

The unexpected despatch of Endurance perturbed the junta. If the scrap men were removed, Argentinian prestige would be damaged; but the presence of Endurance challenged it to military action, which it did not plan to take for several months. The Argentinians havered, first sending a naval ship to take off most of the scrap men, then sending another with a party of Argentinian marines to ‘protect’ those left. It was the turn of the British government to dither. It sought guidance from its own and the American intelligence services as to what Argentina intended. The signs were unclear. Budgetary economies had run down the Secret Intelligence Service (MI6) station in Buenos Aires; what signal information could be supplied by Government Communications Headquarters (GCHQ), by the American Central Intelligence Agency (CIA) and by its sister signals organisation, the National Security Agency (NSA), did not clarify the picture. The British agencies enjoyed a warm and co-operative relationship with the American, based on much exchange of mutually useful material; but the CIA depended on MI6 for human intelligence, while both GCHQ and the NSA were confused by the volume of radio traffic suddenly generated in the South Atlantic by Argentinian but also Chilean vessels; the two navies were conducting a large-scale but routine exercise.

Britain fell into a week-long bout of indecision; it had decided it could not tolerate any further Argentinian intervention in the affairs of its South Atlantic dependencies; but it shrank from any overt measure that would provoke Argentina to action. Eventually, the decision was taken out of its hands. On 26 March, the junta, under pressure from street demonstrations against its economic austerity programme, but even more fearful of public reaction if it appeared to back down before British diplomatic protest over the South Georgia affair, decided to advance the timetable for its invasion of the Falklands and launch the operation at once.

The Falklands were effectively undefended. Of their population of 1,800, 120 of the men belonged to the Falklands Islands Defence Force, but they were untrained and equipped only with small arms. An official British military presence was provided by Naval Party 8901, a detachment of forty Royal Marines; their number had recently been doubled by the arrival of their reliefs. Apart from Endurance, currently in Antarctica, there were no naval ships in the Southern Hemisphere. The Argentine armada, which began to land at dawn on 2 April, could not therefore be repelled, though it was briefly opposed. Naval Party 8901, depleted by the despatch of twelve men to reinforce South Georgia, was ordered by the governor, Sir Rex Hunt, who had been warned by London that an invasion force was at sea, to guard the airfield and the harbour. When an advance party of 150 Argentinian commandos landed, they were engaged and, in a firefight around Government House, two were killed. It was clear to Sir Rex Hunt, however, that resistance was hopeless and, after two hours, he ordered surrender. Soon afterwards the vanguard of 12,000 Argentinian troops began to land, while the Argentinian air force took control of the airfield.

The news caused an immediate and major political crisis in London. The second of April was a Friday; an emergency session of parliament, which never sits at the weekend, was called for the following day. The consensus at Westminster was that, if the government could not demonstrate its willingness and ability to confront the Argentinians, it would have to resign. Fortunately for Mrs Thatcher, a woman of iron will but untried powers of decision, she had already instituted precautionary measures. Alerted by the enormous volume of radio traffic generated by Argentinian preparations, she had ordered a submarine to sail for the South Atlantic on the previous Monday, 29 March. Much more important, indeed, as was to prove critically for the whole Falklands saga, she had on Wednesday evening ordered that a naval and military task force should be assembled to depart at once for the South Atlantic. Her desire to recapture the Falklands was never in doubt; the impetus to the decision was supplied by the arrival in her room in the House of Commons when she was consulting her ministers of the First Sea Lord, Admiral Sir Henry Leach, who gave it as his professional opinion that Britain had the power to mount such an operation and that the navy could set out by the coming weekend. He also assured the Prime Minister of victory. On return to his office he sent a signal, ‘The task force is to be made ready and sailed.’

Its first elements departed on Monday 5 April, while its military complement was hastily assembled in Britain to follow. Three submarines, two nuclear-powered, one diesel, formed the spearhead; there were to follow, over the course of the weeks to come, 2 aircraft carriers, embarking 20 Harrier aircraft and 23 helicopters, 23 destroyers and frigates, 2 amphibious ships, 6 landing ships, 75 transports, ranging in size from large passenger liners to trawlers, and 21 tankers. The majority of the transports and tankers were ‘taken up from trade’, chartered or requisitioned, that is, from the merchant service.

The troops to be embarked would eventually comprise the whole of 3 Commando Brigade (40, 42, and 45 Commando, Royal Marines, 29 Commando Regiment Royal Artillery and 59 Commando Squadron Royal Engineers), attached to which were 2nd and 3rd Battalions The Parachute Regiment, two troops of light armoured vehicles of the Blues and Royals, thirteen air defence troops, the commando logistic regiment and the brigade’s helicopter squadron. There was also a large complement of Special Forces, including three sections of the Special Boat Squadron (SBS) and two squadrons of the Special Air Service (SAS). To follow later was 5 Infantry Brigade (2nd Scots Guards, 1st Welsh Guards and 1st/7th Gurkha Rifles) with some artillery and helicopters. The Royal Air Force deployed elements of seventeen squadrons, flying fighters, bombers, helicopters, reconnaissance aircraft and air refuelling tankers.

Refuelling, in the air and at sea, was an essential requirement, for the task force was to operate without a land base nearer than Ascension Island in the middle of the Atlantic. Until the airfield at Port Stanley could be recaptured, air refuelling was less vital, for long flights over the ocean could not be numerous. All fuel, and other supplies to the warships, however, had to be transferred ship-to-ship while under way.

The assembly of the task force was a race against time, not only because of the need to confront the Argentinians with an armed response as rapidly as possible but also because of the season; the onset of the South Atlantic winter at the end of June would bring sub-Arctic weather necessitating withdrawal from the region. Everything, from completing dockyard maintenance to supplying the soldiers with warm clothing, had to be done at the highest speed; at the outset it seemed that many requirements could not be met.

It was not only the pace of material preparation that had to be forced; so too did that of planning and intelligence gathering. The two were intimately connected and interdependent. Britain had no base in the region and no allies. Chile, long on bad terms with its Argentinian neighbour, was disposed to be helpful but could not risk openly siding with Britain; most other South American countries supported Argentina’s claim to the Falklands, if only out of regional solidarity. How was the campaign to be fought? Clearly there must be an amphibious landing but it would have to be launched from the task force’s ships, not from land. That required the navy to close up to the islands, at least while the troops got ashore, but also to remain nearby during daylight so that the carrier aircraft could provide support. Worryingly the islands, though 400 miles from the nearest stretch of Argentinian coast, were just not far enough offshore to lie outside the range of the enemy’s land-based aircraft. The troops, once landed, would be vulnerable to air attack. Far more worryingly, the warships and transports would also be at risk, except when at night they could stand off to the east into the broad expanse of the ocean.

How serious was the risk? That proved, both at the outset of the campaign and during its development, an embarrassingly difficult question to answer. No one in Britain really knew; no one, indeed, knew anything much that was useful about Argentina’s armed forces. For reasons of economy, the Secret Intelligence Service (MI6) had closed down all but one of its stations in South America; that remaining was in Buenos Aires but its chief was too overworked to collect anything but political intelligence. The service attachés, navy, army and air, were supposed to report on their Argentinian opposite numbers; but in recent years they were more often required to act as salesmen for the British defence industries, so the excuse went afterwards; in practice, attaché appointments were final postings at the end of a middling officer’s career, a farewell present for an unexceptionable life. This was not particular to Argentina but the general rule; only those officers posted to the Soviet Union had the duty of acquiring intelligence and were fitted by ability and training to do so.

Yet the collection of pertinent information in any reasonably open society, which Argentina was, is not difficult and need not conflict with diplomatic propriety. Readily available service magazines contain valuable snippets of information which, if collated, quickly yield an order of battle; so do local newspapers, from stories about local men in uniform and the social affairs of locally stationed units. Service histories are also fruitful sources; units tend to occupy the same barracks for decades. Armies, and navies, are relatively unchanging organisations and, to anyone who takes the trouble to form a picture of their organisation, rarely conceal secrets about their location, strength or function requiring specialised intelligence scrutiny to uncover.

The archives of the Defence Intelligence Service in London ought, in short, to have contained copious and detailed reports on the Argentinian navy, army and air force in April 1982. They did not. The cupboard was almost bare. The officers of the task force have in consequence left a record of a shaming and hurried search in public libraries for such standard works as Jane’s Fighting Ships and the Institute for Strategic Studies’ Military Balance. Little was to be found. The Military Balance allots no more than two or three pages to a country the size of Argentina; Jane’s Fighting Ships is largely a photographic album. Moreover, as the most important of Argentina’s warships, the carrier Veinticinco de Mayo, was the ex-British HMS Venerable, venerable indeed since launched in 1943, and three of its largest destroyers were British-built or designed, Jane’s could tell little the British did not know already. The marines and soldiers scanning the Military Balance must have been even more disheartened. It lists the barest information of numbers of units and quantities of equipment and those in separate sections; no picture of units’ capabilities is discernible, therefore, while units are not named nor are their peacetime locations specified. That omission may have been seriously misleading in the frenzied days of early April 1982. The Argentinian army’s three best formations were the VI, VIII and XI Mountain Brigades (Peron, incidentally, was a mountain infantry officer), which, by reason of their training and familiarity with cold climate, seemed the obvious choice for Falklands duty. Because of the junta’s fear that Chile might profit from their commitment to the Falklands to strengthen its position in the disputed Cape Horn region, however, it had left the mountain brigades in their peacetime stations and decided to employ lower-grade formations drawn from the warm borders of Uruguay. GCHQ is known to have been intercepting the mountain brigades’ radio traffic, confirming that they were still located in the far south even as the invasion fleet put to sea. The task force officers, apparently dependent wholly on scantily published information about the location and capability of their potential opponents, did not even know that.

The navy was quite as badly informed. Admiral Sandy Woodward, commanding the warships and transports aboard the old carrier Hermes, had a general picture of the risk he faced. It consisted of three elements: attack by land-based Argentinian aircraft, some of which were equipped to launch Exocet, the French-supplied sea-skimming missile (also aboard some of Woodward’s ships), which was difficult to distract by electronic counter-measure and deadly if it struck home; the Argentinian surface fleet, known from radio intercepts to be at sea and organised in two groups formed respectively around the Veinticinco de Mayo and the ex-American heavy cruiser Belgrano, apparently deployed to mount a pincer movement; and Argentinian submarines. The diesel-propelled submarines were known to be difficult to detect but, it was believed, could be held at bay by the British nuclear submarines in the area; the surface fleet had been warned not to enter an ‘exclusion zone’ proclaimed around the islands by Britain and would be attacked if it did (it did not but was attacked anyhow, by HM Submarine Conqueror, and Belgrano sunk); it was hoped to overcome the Exocet threat by positioning destroyers and frigates as radar pickets between the islands and Argentina to provide early warning and to distract any missiles that got through by firing ‘chaff’, which simulated a larger target than the threatened ship.

In practice the two Argentinian diesel submarines did not manage to attack the task force; the surface fleet, partially incapacitated by equipment failure aboard the Veinticinco de Mayo, turned back from the exclusion zone and returned to port after the sinking of the Belgrano. The Exocet aircraft, by contrast, inflicted heavy damage on the task force and, with others delivering more conventional ordnance, came close to achieving a naval victory that would have secured the Falklands and humiliated Britain for decades to come.

The Argentinian air-launched Exocet, a modified version of the maritime model, known as the AM-39, was mounted on a Super Etendard aircraft, supplied by France, like the missile itself. The British believed correctly that Argentina had only five AM-39s, but wrongly that it had only one Super Etendard; the right number was five. As important as the aircraft–missile combination was the maritime reconnaissance aircraft that alerted the Super Etendards at their Rio Grande base to the presence of the task force within attack range. An antiquated American aeroplane, the SP-2H Neptune, it possessed the capability to linger beyond the horizon formed by the earth’s curvature but to keep the British under radar surveillance by bobbing up over it at regular intervals. The Super Etendards, when vectored towards the target, flew at sea level, beneath British radar, until close enough for the Exocet to strike. The pilots needed to gain altitude only once or twice, and then briefly, for their own radars to acquire their targets and automatically programme the missiles to depart in the correct direction. Once launched the Exocet maintained height just above sea level by an on-board altimeter and finally homed on the target ship down the beam of its own radar.

Admiral Woodward and his staff had been wrongly informed that the Super Etendards’ range was only 425 miles, too short to reach the task force east of the islands. In fact, by refuelling from one of Argentina’s two KC-130 tankers, they could achieve launch positions. On 4 May, two days after the sinking of the Belgrano, two Super Etendards, flying from Rio Grande, approached the task force; their directing Neptune had been spotted by British radar but was thought to be searching for Belgrano survivors. Glasgow and Coventry, deployed as radar pickets west of the task force, caught echoes of the attacking aircraft as they rose above the horizon to correct their final approach paths. The British ships fired chaff and both Exocets, travelling only six feet above the sea, were deflected by their own course-corrections. Sheffield, twenty miles distant, was currently transmitting on its radio link to satellite, which prevented its hearing the warnings transmitted by its sister ships or operating its own radar. Its crew were therefore oblivious of impending danger and neither fired chaff nor manoeuvred. She was hit in the forward engine room by one of the Exocets which, though its warhead failed to explode, started a fire that eventually forced her abandonment, after heavy loss of life.

The manifestation of the Exocet threat was to exert a decisive effect both on the management of the campaign and on the intelligence effort that underlay it. Admiral Woodward at once withdrew the task force far to the east of the islands, where it was to remain until the landings began on 21 May. At the same time the Northwood joint services headquarters, from which Operation Corporate, as the campaign was code-named, was directed, began a frenzied search for means to improve intelligence collection and to strike directly at the Argentinian air menace. Of signal intelligence there was no shortage; the Argentinian army, navy and air force generated a large volume of traffic, which was intercepted not only by GCHQ, through its intercept station at Two Boats on Ascension Island, ostensibly a branch of the Cable and Wireless Company, but by the NSA, the American intelligence community having decided to lend its British partners full support at this time of need, and by a New Zealand intercept station at Waiouru. The United States was also generous with satellite intelligence. The National Reconnaissance Office (NRO) had three systems in operation that could together provide electronic and imaging data, White Cloud, KH-8 and KH-11; it could also offer data from occasional overflights by the SR-71 high-altitude reconnaissance aircraft.

The limitation on the usefulness of overhead surveillance was, first, its intermittence – White Cloud made only two passes a day – but, second, that by the time it became available, the damage had been done. Overhead surveillance could have warned of the Argentinian invasion fleet setting sail, in time for the British government to have issued an ultimatum; once the fleet had arrived, it could supply little further information that was useful.

It was, among other factors, for that reason that the Northwood headquarters decided, after the shock of the first Exocet attack, to move from passive to active counter-intelligence methods. Since traditional means of warning – including satellite intelligence – had failed to avert the threat, the Ministry of Defence would be ordered to mount operations to eliminate the risk at source. Britain’s special forces would be committed to find and destroy the Exocet units in their home bases.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.