Bit by bit, with deep and understandable reluctance, the CIA has been opening its archives. Only fragments of its view of the Falklands War are currently available, but they are still worth close study.
Intelligence material often plays a large part in the formulation of foreign policy, sometimes a decisive one, but it is almost always held back far longer than other government information, creating a structural weakness in our understanding of the recent past.
the cia's declassified files: the crest database
CIA Director's angry memo to almost everyone re: the leaks that caused Argentina to change its cypher, 9 Jun 1982
Around 40 CIA documents on the war have been released through the agency's "CIA Records Search Tool" database (CREST). These are items more than 25 years old that the agency has been obliged to release by a Clinton-era Executive Order (EO 12958/13256), rather than responses to Freedom of Information Act requests, which probably gives this mini-collection a little more coherence and completeness than it might otherwise have. One of the great problems with relying on FOI to produce historical information is that the selection of things released is dependent on the interests of the persons or organisations who made the original request, and their interests probably won't be the same as yours, indeed are likely to be different - and might be completely crazy. You generally won't know why one document has been released and not another, since the terms of the original requests are not themselves released alongside each document.
Even so, there is much about this CIA collection that is obscure. While you can sometimes work out in Presidential Library files what part of the whole has been released and what not, and roughly why, in the case of these CIA documents that is all but impossible. What we are looking at here might be a fifth of the whole, or a fiftieth. It might contain much of what they knew about the Falklands, or the merest fraction. It might be representative, it might be completely the reverse. One obvious element is definitely missing: there is no signals intelligence here. But we know that the US had broken Argentinian military codes, until leaks in the US press caused the Argentinians to change their cypher, of and it would be a surprise if it wasn't reading diplomatic traffic as well.
Some of the CREST documents are so heavily redacted that they resemble a kind of two-dimensional Swiss cheese. Redacting becomes so compulsive an activity that even unclassified documents get redacted sometimes. Generally though the Falklands documents have been released without too much of this kind of semi-shredding. Unhelpfully, you need to visit in person to use this database, which is stored at National Archives II in College Park, Maryland, and print out what you find. They give you free paper at least. The team at margaretthatcher.org has done the work and now uploaded the Falklands material. There is more useful information available in the library section of the CIA's excellent website, including a number of additional Falklands documents, and we have added in a handful also from the official-sounding but resolutely independent "National Security Archive", based at George Washington University in Washington D.C.
This CIA material includes twice-daily situation reports at the very beginning of the war, some satellite image interpretations, as well as high-level internal papers updating the Director and Deputy Director. It includes too a number of the "National Intelligence Daily" (NID) intelligence summaries circulated only to senior officials in the White House and Cabinet Members, the CIA's 'newspaper', printed in tabloid form and some "Monthly Warning Assessments" (MWA).
Given the fragmentary picture these documents provide of the agency's knowledge and outlook, conclusions are best presented as a series of points rather than a connected story, with suitable warning that they are incomplete and that possession of the full record would require them to be amended.
The earliest document is a memo from CIA Director William Casey to Secretary of State Al Haig, copied to the White House, providing a "Quick Intelligence Assessment on Falkland Affairs (April 2, 1982)". This helpfully reminds one how useless intelligence reports can sometimes be. Forced to make snap judgments without much hard information, the interpretation inevitably tended towards the bland. The British, Casey solemnly suggested, would be hoping to use the UN Security Council "to put Argentina in the dock as an 'aggressor'". He explained that "British economic options are limited" and suspected that the Argentinians would be likely reinforce the islands before the British Task Force arrived. And so on: there are pages of this. Any averagely well-informed reader would have been tempted to skip large chunks of it in the hope of finding something seriously useful further on. More interesting was the point that many of the ships in the Argentinian fleet were 40 years old - i.e, dated from WW2 - and would need to be withdrawn as soon as possible from the wintry southern seas. "We also do not believe they would fare well in a full-scale naval engagement with the British". How Argentina's actions had been received in Latin America is nowhere touched on in the document, but must surely be in the redacted sections. The memo wraps up with Cold War aspects - Cuba likely to make trouble if it thought Argentina might win, though only rhetorically, the Soviets likely to supply surveillance information on the British Task Force to the Argentinians as part of a general strategy of ingratiating themselves with Buenos Aires, but no more. Almost the most telling point in the memo is the reminder that the post-Afghanistan US grain embargo on the Soviet Union, 1980-81, had created a big new trade between Argentina and the USSR, the former supplying no less than 10 per cent of the latter's imports, though this was hardly classified information. The CIA was wrong, however, in thinking that Moscow would work with Argentina to block British attempts to persuade the UN Security Council to condemn the invasion by Resolution, "threatening to veto it". This might have seemed a safe bet, but in fact there was no Soviet veto, to Argentinian disappointment and general surprise, UNSCR 502 passing easily on 5 April. This early success anchored the British diplomatic position from the beginning of the conflict, because almost no one could be found to defend Argentina's first use of force.
The Argentinians understood, of course, the centrality of the Cold War to US perceptions and were themselves playing the Soviet card with Washington from the beginning of the war - warning the US that the Soviets were watching and waiting (CIA sitrep, 3 Apr). Later they threatened to turn to them for arms supplies (MWA, 28 May). Of course, this was a familiar move in third world diplomacy. It is not clear how seriously the CIA took it in early April, but by the time the junta faced defeat - indeed "military humiliation", in the agency's eyes (MWA, 26 May) - they were watching closely. On 28 May Deputy Director for Intelligence Robert Gates told the CIA Director, William Casey, that a special memo had been prepared for the White House on Cuban assistance to Argentina and Casey himself was specially briefed on Soviet responses prior to a meeting with Haig. The movement of the Soviet ELINT satellite into orbit above the South Atlantic is noted in the 30 Apr 1982 NSC meeting on the previous page. But anxieties settled after the war. An Interagency survey of the "Prospects for Soviet Arms Deliveries to Argentina" in July 1983 concluded that: "Only as a last resort would the Argentines move to major purchases that would involve a long time arms relationship with the Soviets".
The CIA debated the possibility that the Galtieri junta might be replaced by a "highly nationalistic military regime" (MWA, 30 Apr) - as if it wasn't one, or not much of one. One has the sense that Galtieri was their preferred strongman, a caudillo in the making maybe, and that change was thought unlikely to be an improvement from a US perspective.
apr - jun 1982: the military campaign
The agency saw major difficulties for a British assault on the islands and thought at one point we were underestimating Argentine forces (NID 19 May), but consistently, and from early on, it believed a British victory was 'likely' (eg, MWA 30 Apr). Several times they revert to Casey's initial point that much of the Argentine navy was clapped out.
British attacks on the mainland of Argentina were thought a real possibility, though one that diminished significantly once troops were landed safely on the islands (which happened on 21 May). It was believed that British submarines might attack the Argentine navy, particularly its carrier (the former HMS Venerable), even within the country's coastal waters (ie, 12 mile limit), if there were big losses during landing of forces. If the troopship Canberra was hit, the agency believed landings would have to be aborted.
After the landings the agency judged (NID 24 May) the British were planning a speedy campaign and were willing to tolerate high casualties to wrap things up quickly, warning that "serious reverses" would create big political pressures at home and perhaps open the way to a Pym government. If on the other hand the campaign was quickly successful, "chances would increase dramatically for an election as early as next spring, with a major Tory victory likely". That Goose Green would be attacked ahead of Port Stanley, rather than bypassed, was correctly predicted.
There were believed to be 17 US citizens on the Falklands at the outbreak of the war. Efforts were made through the Argentine occupying forces to persuade them to leave before the Total Exclusion Zone was declared (29 Apr - TEZ was the following day).
Judging from the junta's efforts to acquire new equipment and supplies even as its forces faced defeat on the Falklands, the CIA thought it was planning to fight a campaign of long-term resistance (MWA 28 May) and that the Argentine government regarded its Falklands garrison as 'expendable' (Watch Committee 26 May).
There is limited military analysis in the CREST documents and there is a dearth of material from the Department of Defense on the Falklands War. The CIA predicted an outright British assault on Port Stanley led by Scorpion light tanks, but of course the Argentinians surrendered following a series of small battles in high ground outside the town so no assault on the town itself was needed. The analysts noted the effectiveness of Argentine air force attacks after the landing at San Carlos, but were unaware apparently of the failure of some of their bombs to explode. They doubted the Argentines could keep up their rate of air attacks, and questioned also whether Britain was being truthful about the number of Harriers it had lost (NID 29 May). The role of EXOCET missiles was discussed, but a redaction obtrudes, perhaps touching on Argentine efforts to buy more of these effective missiles.
The CIA Director, William Casey, sent a sharp memorandum to Cabinet Officers and a wide range of agency heads (but not the White House) regarding "Unauthorized Disclosures on the Falklands Situation" (9 Jun). He warned of the "seriousness of leaks compromising sensitive collection sources and methods" and urged investigations to determine whether leaks should be referred to the Attorney-General - i.e., considered for criminal prosecution. The 30 Apr meeting above noted that press stories had caused the Argentines to realise the US was reading their military codes and to change their cipher, locking the US out (for a time at least). ABC Nightline had broken the story on 14 Apr, but the 15 Apr New York Times article mentioned on the previous page pops up in the CREST database when you search for 'Falklands': plainly, it drew the agency's special attention.
The US had no satellite over the South Atlantic at the outbreak of the war and took some time to reposition one. Observation conditions were less than perfect, with the southern hemisphere moving into winter, so the CIA analyses of 28 May of images over Port Stanley and southern Argentina may be the earliest they got. Even then they note problems with cloud cover. These photographs showed increased defensive preparations around Port Stanley, while a series of images of mainland military installations in north and south of the country showed the Argentine navy safely tied up in port, aircraft carrier, submarines and all. Given that signals intelligence was available, apparently in some abundance, it is unlikely on this showing that satellite intelligence played anything like as significant a role in the war. Many, of course, thought otherwise at the time. A State Department telegram of 4 May shows that the Embassy in Buenos Aires was unable to kill the story that a US satellite had given the British the position of the General Belgrano, whereas the British Official History of the Falklands War, by Lawrence Freedman (vol.2, p285) credibly explains that it was an intercepted and decrypted signal from Admiral Lombardo late on 1 May that showed the British that the Argentine navy was planning a coordinated strike against the Task Force the following day, information that led directly to the decision to sink the ship, which had first been found when the British nuclear submarine south of the islands got a sonar trace of the Argentine tanker sent to refuel it. One cannot safely assume that the signals intelligence came from the US side to the British either: we had our own formidable listening and decrypting capability, and it was only the day before the crucial signal that Inman was telling his colleagures that the US had been locked out.
News of the Argentine surrender was circulated as a "CIA late item" in the National Intelligence Daily deriving the information straight from MT's office (NID 15 Jun).
jun 1982 - jul 1983: after effects - damage to U.S. Interests in south america judged 'manageable'
Almost immediately the war was over Casey requested briefing on the impact of the war on global high technology transfers (Gates memo, 18 Jun).
The agency pointed out internal divisions in the Argentinian army resulting from recriminations over the war and saw in them a potent threat to the successor regime of General Reynaldo Bignone (NID 29 Jul). Returning soldiers from the garrison were expected to have unhappy stories to tell, with possibly radicalising results.
Long-term fallout from the war in the form of anti-US sentiment in Latin America was a genuine concern for the US, but swiftly diminished. This was apparent even at the most difficult moment, days after the Argentine surrender on 14 June.
- The CIA prepared a lengthy intelligence appreciation for the President and other senior figures on 18 June, "Short-Term Prospects for Central America". This commented: "The circumstances of the Falklands war will doubtless reduce the readiness of some Latin American states to support US initiatives concerning Central America. The direct adverse impact in Central America, however, is not likely to be substantial".
- The State Department took a longer look later in the year, as part of an inter-agency review of "US Policy Towards South America in the Wake of the Falklands Crisis". This study concluded: "Aside from Argentina, crisis-induced damage to U.S.-Latin American relations has varied widely from country to country and appears manageable on most matters". Actions to "restore and assert US influence in South America" included a resumption of arms sales to Argentina (and Chile) by end 1982 - much upsetting MT, but the President was firm - and a decision to maintain "its diplomatic position on the fundamental Falklands issues as it was before the crisis: U.S. neutrality on the question of sovereignty over the islands and support for negotiations or other peaceful efforts to resolve this dispute". An important distinction was drawn between US opposition to Argentine first use of force, "which was widely accepted in Latin America", a region of many territorial disputes and small countries scared of bigger ones, and US support for the British military campaign "which was just as widely resented". "Fortunately for us, Argentina's reputation for arrogance, and the collapse of its forces on the ground, helped to cushion reactions". Argentina's nuclear weapons programme was the region's most sophisticated and attracted a wary intelligence eye.
- Amongst the handful of Pentagon documents on the war is a characteristically clear-eyed assessment of the "Military Lessons from the Falklands" from Caspar Weinberger, written for the President on 19 July. "In the final analysis, the battle for the Falklands appears to have been a closer call than many of us would believe. The British won primarily because their forces, inferior in numbers at first, were superior in training, leadership and equipment". "But luck also played a significant part". Unlike the CIA he was well-aware that Argentinian bombs had failed to explode.
- Assessment of the war was an early task at the NSC for Marine Colonel Oliver North, who served there from 1981-86. This role perhaps explains an event a few years later, on the eve of MT's visit to Camp David in December 1984, when the then National Security Adviser, Bud McFarlane, suggested to the President that he sound her out on "steps they could take to assist the Nicaraguan resistance. Through intermediaries we have been advised that the Chilean government is prepared to provide up to 48 BLOWPIPE surface-to-air missiles to the freedom fighters". These British missiles had been "staged in Chile during the Falklands War".
- The probability of a long-term Argentinian tilt towards Moscow was judged low, as already noted above from the Interagency Study of 1 July 1983.
The CIA's daily 'newspaper', for the most senior US officials