This is an article which I wrote back in 2007. I am adding it to my blog for those who are interested.



CIA HeadquartersIn June the U.S. Central Intelligence Agency (CIA) declassified hundreds of once-secret documents. The documents, released the CIA website, in response to a 1992 Freedom of Information Act request, were once called the “family jewels” by its employees. Most, of course, relate to the Soviet Union, China and Cuba – familiar American concerns. But among the documents now available are a number relating to South Africa. (See

Three documents provide an insight into the American intelligence community’s understanding of the situation in South Africa in the late 1980’s and early 1990’s. Although parts are still blacked out, they make fascinating reading.


  • “Africa Review; Special Issue: South Africa entering the 1990’s,” dated 20/1/1989. Secret
  • “South Africa: Weathering the Storm,” dated 25/6/1992. Confidential
  • “Prospects for the South African Transition,” dated 1/1/1994. Secret


This article is an attempt to highlight some of the key issues raised by the reports. It is not a definitive or academic review of the documents.


  1. The view in 1989


The overall impression one gets from reading this comprehensive (70 page) document is that the CIA assumed that white rule would continue for the immediate future. There was also no suggestion of a change in attitude at the heart of government. It opens with the following statement:


“The ruling National Party (NP) under President P.W. Both has weathered more than four years of unprecedented domestic and international pressure….Despite toying occasionally with ideas such as federalism or Swiss-style cantons, the NP is recent years has strayed little from its strategy crafted during the 1970’s. Elements of the strategy include a tough security posture, modest reforms of social and economic apartheid, limited black power-sharing at the local and regional levels, a national advisory body to discuss constitutional change with blacks acceptable to the government, privatization and deregulation to boost economic performance and a mixture of coercion and diplomacy in the region….Pretoria appears unwilling to press ahead rapidly on even its very timid reform agenda prior to national parliamentary elections expected later this year.” (page 1)



While it is, of course, easy to be wise after the event, it is difficult to believe that this assessment could be made by one of the world’s best resourced intelligence agencies just a year before the momentous announcement by President F.W. de Klerk on 2 February 1990 that the African National Congress and Pan Africanist Congress would be unbanned and Nelson Mandela released. It would seem that the CIA had little idea of the depth of the contacts between the ANC and the government, with meetings with Mandela at least as early as 1985. Indeed, a section marked “Outlook” begins with this prognosis:


“We believe that the NP will probably muddle through at least the next four years in power, but without achieving any lasting solutions to its electoral, black, economic or regional challenges. Although the seriousness of events or crises facing the government may at times appear – or be portrayed by its foes to the left and right – as regime threatening, Pretoria probably will be able to at least cope with the challenges.” (5)



It goes on to report that the National Party does see a need for change, but not anytime soon.


“Despite its strategy for the next few years, we believe that even the NP accepts the inevitability of fundamental political change in South Africa. Pretoria’s policies are aimed at delaying that day as long as possible. The current leadership probably believes its security forces can guarantee continued white privilege well into the next century.” (6)



Indeed, the CIA appears to see the threat from the far right Conservative Party (CP) is at least as serious for the government as the opposition of the ANC. The CP with its promise of returning the country to the days of strict apartheid before the reforms brought in following the Durban strikes of 1973 and the Soweto uprising of 1976, were doing well at the polls. Overall, the CIA’s assessment is that the party will not outs the National Party. “In our judgement it is almost certain that the NP will retain control of parliament in the next general election, if it is held later this year, but the Conservatives appear poised for their strongest performance to date.” (8)


ANC – majority rule “not around the corner”


The Americans accept that despite a fierce crackdown on internal black opposition to the government from the United Democratic Front and the unions – Cosatu – brought in during the state of emergency, these movements have not been defeated. Although more than 30,000 activists had been detained they were still able to operate. Strategies of co-option are seen to have failed. Institutions, from the homelands to township government had, the report concluded, been rejected by black people.


The CIA’s view of the ANC is not entirely clear, since the first page of their analysis is blacked out. However, their take on the movement’s goals and objectives is legible. “The ANC’s short-term objectives are fairly clear and its two-track military-political strategy reflects a realisation that majority rule is not around the corner.” (22) On the military front the CIA says what it calls “…the impressive bombing campaign last year – designed to discredit the government’s nationwide elections – suggests its internal network is alive and operational.” (22)


At the same time the Agency notes what it calls “one negative outcome of this shift towards greater operational freedom at the local level” – namely the “…controversial attacks on civilian targets.” (22) This, the CIA maintains has produced what it calls a “heated debate among senior ANC leaders.”


“The ANC political leadership has been loathe to endorse any premeditated campaign against civilians, especially whites, because such attacks risk Western condemnation and even harsher South African retaliation. ANC military officials, on the other hand, believe that the group must increase urban attacks to undermine white morale and thereby force Pretoria to negotiate. The ANC senior leadership will allow periodic increases in military activity inside South Africa to placate restive insurgents and preclude a split between the political and military wings over strategy. Nonetheless, it almost certainly will take steps to reign in the military wing if it starts operating too independently.” (22)



The document also looks at the prospects for Mandela’s release, noting discussions on the subject within the National Party. But it concludes that this is not likely in the near future.


“Despite wanting to resolve the issue, Pretoria is unlikely to risk alienating conservative whites by releasing Mandela prior to national parliamentary elections expected to be called later this year. Even over the longer term, Pretoria is unlikely to release Mandela unless it is convinced it can contain any resulting black moves towards political mobilization.” (26)




The assessment of the CIA is clear. Government repression will continue, and will continue to successfully contain the ANC and black opposition in general. Contacts between the National Party and the ANC are noted, but not given a great deal of weight.


The document contains a well researched case study of what it calls “Rugby Talks with the ANC” – the meetings in Harare between Louis Luyt, chairman of the South African Rugby Board, and ANC officials in October 1988. These paved the way for talks between Danie Craven of the South African Rugby Union and ANC secretary general, Alfred Nzo, Director of Publicity, Thabo Mbeki and Cultural Secretary, Barbara Masakella.


Although the meeting is reported to have gone “quite well” the CIA believes it led to a backlash, with both parties coming under a “storm of criticism” from their constituents. The report concludes:


“Despite the difficulty of reaching compromise on tangential issues in the absence of serious negotiations on fundamental political questions, the rugby talks probably were useful. For its part, the ANC sees progress wherever it can engage in dialogue with Afrikaners. The rugby meeting and other ANC-Afrikaner talks provide each side an opportunity to put forward its case to the other side, and may encourage some white South Africans to begin seriously questioning South Africa’s options.” (29-30)


The reality was very different and fundamental change was already on the cards.


The National Party leadership had already made up its mind that there was no future in its past strategy. These meetings, which the CIA correctly reported on, were paving the way for the talks which led to President de Klerk’s announcement – just over a year after the report was published – of the unbanning of the movements and the freeing of Mandela. It would seem from this report that American intelligence network had failed to penetrate either the ANC or the National Party. Nor were their analysts capable of reading the runes. In this they were, of course, in good company, since very few analysts really understood what was going on. But the CIA, with its huge resources, might have been expected to do better.


  1. The view in 1992



By the time this document was being prepared talks had taken place at Codesa (Convention for a democratic South Africa) and broken up in recriminations. The ANC had walked out, accusing the government of having a hand in the upsurge of what was termed ‘black-on-black’ violence, mainly between supporters of the ANC and Inkatha.


The CIA’s estimation was not to blame the government, but their conclusion is interesting.


“We do not believe or have evidence to support allegations that President de Klerk is directing efforts to foment violence and thereby weaken the ANC. On the contrary, his efforts to purge the security forces, particularly the police, of righwingers have been so pronounced that the local media have openly speculated on the possibility of security elements mounting a coup. Nonetheless, a growing body of credible reporting indicates that rogue rightwing sympathizers within the security forces are providing support and assistance, including arms, to any forces that are instigating violence.” (1)



In the original document a question mark has been written in the margin at the point where the de Klerk’s purge of the security forces is mentioned. Who wrote this is not indicated.


The Agency’s assessment is that the ANC walkout is not too serious, as both sides needed each other.


“A lengthy disruption of talks would stretch out the transition process and make it even more violent. The ANC and its allies, however, are no match for white-controlled security forces, while the NP cannot make a new system work without the cooperation of the ANC and its allies. Consequently, a halt to the talks probably will be temporary as the principal actors try to restart the process and avoid unprecedented bloodshed and irreparable damage to the economy.

Our judgement that the transition program will stay on track is based more on the previous behaviour of the key player and the assumption that they fully appreciate the disastrous consequences of a breakdown that on any current compelling evidence….

The United States and other foreign governments can play an important role in keeping reform and transition on track. Pretoria wants political support and economic investment. The ANC views its international standing as a key strength. Both sides know that an abandonment of the present course would undermine the favourable stance that Washington and other governments have adopted.” (v)


Beyond this, the only really interesting elements of this short document are the beliefs that “…a final constitution may still be years away…” and an assessment that although the ANC’s alliance with the SACP would survive in the short term it might come under stress in the years ahead.


“…differences over future economic policy; ANC attempts to broaden its base of support among relatively conservative voters, including Indian and mixed-race Coloureds; and competition for the urban black vote in the runup to national elections will ultimately strain the alliance, perhaps to the breaking point.” (5)


  1. The view in 1994

The report opens with the arresting subheading: “ANC Headed for Big Electoral Win” which more or less sums up the CIA reading of the situation. “An electoral loss by the ANC – estimated one in 20 chance – would push South Africa to the brink off anarchy.”



Its assessment is that the ANC will win the election hands down.


“We judge that the ANC will win the election with a margin of victory wide enough to ensure that it will control the interim government. The ANC might even win more than two-thirds of the seats in the new national legislature, which would allow it to monopolise the drafting of the final constitution.” (4)


Coming just ahead of South Africa’s first fully democratic elections in April 2004 the CIA believes there is really only one major hurdle to be overcome – the threat from the far-right. After noting that violence was on the increase it arrived at this conclusion:


“Coup plotting by rightwing elements within the security forces is under way. A coup would be difficult to coordinate and seems unlikely to occur under current conditions, but a climate more conducive to a coup could develop quickly – and even a failed coup attempt could disrupts the electoral schedule.” (3)


If there was such a coup it would provoke a violent reaction. “The scenario could range from the current rate of bloodshed and fighting to a high-intensity conflict in which a central government, vastly superior in military resources got involved…..A civil war would impose costs on neighbouring countries in cross-border refugee populations, and disruption in the flows of vital commodities from and through South Africa.” (5)


Once elected the CIA believes the ANC will have a “nine in 10 chance” of surviving for the remainder of the year, but would come under concerted pressure from the far right.


“The most pressing business of a new ANC government will be to defuse hostile black conservatives and white rightists. It is unclear whether white extremists of the IFP will pose the greater threat. The two forces might act in tandem to undermine an ANC government.” (7)


The Agency points out that whites are routinely armed and that most had some form of military training. They were dangerous and the CIA predicts that “Militant Afrikaners will resort to terrorism, among other tactics, to harass an ANC-led government and to try to gain greater autonomy.” (7) Given the hold of far right groups on the civil service and industry the Agency regards them as a serious threat.


“No matter which path the AVF chooses, one of its members, the neo-Nazi Afrikaner Resistance Movement – which has over 50,000 supporters and several thousand well-trained paramilitary troops – and some 130 other white extremist groups will plot assassinations and conduct bombings.” (8)


Given these threats there is some mulling over of what the US might do. Its ability to affect major developments, the CIA believes, is “limited.” Gone is the optimism of the previous report that South Africans will listen to outside advice. And should the worst come to the worst even military action under the UN is seen as improbable.


“ While the top parties welcome international election observers, they do not favour using a UN peacekeeping forces to safeguard the election. If key parties, most notably the ANC, suddenly changed their stance on this issue, we judge that the costs and risks of such a ‘short fuse’ mission would be prohibitively high. A large-scale UN peacekeeping mission almost certainly could not be pulled together in time and, even if it could, would not guarantee that the election would be free, fair or peaceful. Moreover, the mission of such a force would quickly become controversial if the security situation began to deteriorate significantly.” (12)


In summary, the CIA believed South Africa was, for better or worse, on its own. In this, at least, they were correct.




These three documents – none of which record any directions to agents or prescriptions for US government action – reveal an intelligence agency that is both more reasonable and less informed than some would have expected. These are not the conclusions of a ferociously right-wing organisation, attempting to intervene in world affairs at the drop of a hat. Rather, they are sober, cautious assessments. Some of the conclusions were wide of the mark, but they were not unreasonable. All one can say is that it might have been expected that they would have been better informed than they were in practice. Perhaps we over-estimates the abilities of governments and their intelligence services to know what is taking place.