Volume TWO Chapter TWO

The State outside South Africa between 1960 and 1990

 

· INTRODUCTION

The security forces will hammer them, wherever they find them. What I am saying is the policy of the government. We will not sit here with hands folded waiting for them to cross the borders … we shall settle the hash of those terrorists, their fellow-travellers and those who help them. (Minister of Defence, General Magnus Malan, Parliament, 4 February 1986.)

1 The Promotion of National Unity and Reconciliation Act (the Act) charges the Commission with investigating and documenting gross human rights abuses committed "within or outside" South Africa in the period 1960–94. This chapter focuses on the "outside", specifically the Southern African region and Western Europe. Evidence has been gathered of violations committed by South African security forces or their agents and/or surrogates in nine regional states – Botswana, Lesotho, Swaziland, Angola, Mozambique, Zimbabwe, Zambia, Tanzania, and the Seychelles – and in Western Europe – in the United Kingdom, France, Belgium, the Netherlands and Scandinavia.

2 The primary focus will be on killings and attempted killings (including targeted assassinations, cross-border raids and large-scale massacres such as that at Kassinga in Angola in 1978) and on abductions and infrastructural sabotage.

3 Another area of focus will be acts which, though they may not in and of themselves have constituted gross violations, were violations of state sovereignty and international law and invariably led to or created the conditions for the perpetration of gross violations of human rights. The reference here is to the wars in Angola and South West Africa (now Namibia), South African Police (SAP) operations in Rhodesia, surrogate-force campaigns in Angola, Mozambique, Lesotho, Zimbabwe and Zambia, and the attempted coup in the Seychelles.

4 Over three decades, the South African government’s involvement in the region expanded from occasional cross-border interventions in the 1960s to a situation in the 1980s where the South African Defence Force (SADF) was involved in various levels of warfare in six Southern African states, while covert units conducted attacks particularly in Botswana, Lesotho and Swaziland (the BLS states). Additionally, in the early 1980s, South African security and intelligence operatives attempted to overthrow the Seychelles government and co-funded a mercenary force of Presidential Guards in the Comores, which became the de facto ruling authority of that territory.

5 This involvement in the region led to the conclusion that the majority of the victims of the South African government’s attempts to maintain itself in power were outside of South Africa. Tens of thousands of people in the region died as a direct or indirect result of the South African government’s aggressive intent towards its neighbours. The lives and livelihoods of hundreds of thousands of others were disrupted by the systematic targeting of infrastructure in some of the poorest nations in Africa.

6 The South African government’s security strategy was shaped by the doctrines of pre-emptive interventionism and counter-revolutionary warfare. By the 1980s, the region had become an arena of cold war confrontation. For the leadership of the government and the SADF, the war in Angola and the other conflicts across the region were good and just wars, part of the West’s resistance to a perceived Soviet global offensive.

7 It is the Commission’s view that the destruction wrought on the region by South Africa’s counter-revolutionary war, particularly in Angola and Mozambique, was disproportionate to the threat posed by their post-independence governments and the fact that they played host to groups engaged in armed conflict with the South African government. At the time of their independence in 1975, Angola and Mozambique were severely underdeveloped and posed no credible military threat to the Republic of South Africa. Centuries of colonial exploitation had left them with a legacy of poverty and without the skills to build and manage a modern economy.

8 The Commission is, therefore, of the view that factors of race and racism should not be dismissed when attempting to explain South Africa’s conduct in the region. It finds it difficult to believe, for example, that Koevoet would have been allowed to operate on a bounty basis, or that the SADF would have killed over 600 people, many of them children and women, in the Kassinga camp in Angola, had their targets been white. From the evidence before the Commission, it appears that, while some acts of regional destabilisation may have been a defence against Communism, the purpose of the war was also to preserve white minority rule in South Africa and was, therefore, a race war.

9 The perpetration of gross human rights violations outside South Africa will be discussed through an examination of the following types of security operations:

a conventional warfare;

b police and military counter-insurgency operations;

c surrogate-force insurgency operations;

d police and military cross-border operations, including special operations of a sensitive nature or by the Civilian Co-operation Bureau (CCB);

e unconventional military operations.

 

· CONVENTIONAL WARFARE

The war in Angola

I was then in the military, you know in the paratroopers and the Special Forces, and I was decorated for a couple of operations in South West Africa. I don’t know if I must apply for amnesty for Kassinga … It was probably the most bloody exercise that we ever launched, according to me … we were parachuted into that target ... It was a terrible thing … I saw many things that happened there but I don’t want to talk about it now because I always start crying about it. It’s damaged my life. (Lieutenant Johan Frederich Verster, ex-SADF Special Forces officer, testimony to the Commission, 4 July 1997.)

It (Kassinga) was a jewel of military craftsmanship. (General J Geldenhuys, A General’s Story: From an Era of War and Peace, 1995. p. 93.)

10 The targets and objectives of the various security operations conducted in Angola by the security forces of the former South African state were two-fold.

11 First, the possibility of the Popular Movement for the Liberation of Angola (MPLA) taking power in Angola was regarded by the former government as a threat to South Africa’s security, as the MPLA was viewed as a Soviet surrogate. The introduction of Cuban forces into Angola in support of the MPLA simply confirmed that view. The South African government’s initial objective, therefore, was to prevent the MPLA from taking power at independence. When this failed, the goal became its overthrow and replacement by a ‘friendly’ anti-Communist government led by the National Union for the Total Independence of Angola (UNITA).

12 Second, the movement of the forces of the South West African People’s Organisation (SWAPO) into bases in Angola was regarded as escalating the threat to South Africa’s position in South West Africa. Aware that it would ultimately have to implement UN Resolution 435, the South African government was determined to weaken or cripple SWAPO’s military capacity in preparation for the time when SWAPO would enter the electoral stakes inside South West Africa.

13 South Africa’s active involvement in Angolan politics after the collapse of Portuguese colonial rule escalated with the SADF’s invasion of Angola, through Operation Savannah, in the second half of 1975. Though the operation was undertaken with the covert support of the US State Department, this undeclared act of war did not receive the approval of the South African cabinet. Indeed, the issue was not even raised at cabinet level until the invasion was several months old and no longer a secret.

14 The invasion was also illegal in terms of the 1957 Defence Act, which made no provision for the deployment of non-voluntary forces (conscripts) beyond South Africa’s borders. To override this legal difficulty, Parliament passed an amendment to the Act in January 1976, sanctioning the deployment of non-voluntary South African troops outside of the Republic’s borders. It was made retroactive to August 1975, the month in which Savannah was launched.

15 The Commission was not able to access any files on Operation Savannah in the SADF archives, nor did it have access to any Angolan data. There is no doubt, however, that the human and economic cost of the operation, involving over 2 000 men in mobile columns, was immense. In three to four months, the SADF swept through vast areas of central and eastern Angola, capturing numerous towns, until it was halted on the outskirts of Luanda by stiff Cuban-led resistance.

16 Civilian and military casualties on the Angolan side were considerable, given the known extent of damage to the social infrastructure. Thousands of people were displaced internally.

17 Though the SADF’s intervention failed to prevent the MPLA from taking power at independence in November 1975, and even though SADF forces were withdrawn in March 1976, South African military and political involvement in Angolan affairs continued for the next thirteen years, with human and other consequences of varying degrees of severity for all the parties involved – Angolan, South West African, South African, Cuban and Soviet. South Africa’s forces were not entirely withdrawn; the SADF created an eighteen-kilometre-wide demilitarised zone (DMZ) along a 1 000 kilometre stretch of the border, which it retained after the termination of the invasion. By creating a ‘free-fire’ zone, the SADF effected further large-scale displacement of people, this time of residents from both sides of the border.

18 Between 1976 and 1978, the SADF’s strategy in respect of Angola focused on establishing a string of bases along the border, on rearming and strengthening UNITA’s fighting capacity through the launch of Operation Silwer and on preventing SWAPO from moving south. This it did by deploying the 32 Battalion in frequent forays against SWAPO in the south of Angola.

19 Despite these efforts, by the end of 1977 the SADF believed that SWAPO had established a significant presence in the south and sought authorisation from Prime Minister Vorster for large-scale SADF operations. The concern of the SADF was shared by the administration in South West Africa, where there were plans to hold elections leading to the installation of an interim administration of which SWAPO was not to be a part. Were SWAPO to succeed in establishing a permanent foothold in the south of Angola, both the South African government and the administration in Windhoek feared that its capacity to disrupt the election would be enhanced.

Operation Reindeer: the attacks on Kassinga and Chetequera camps

20 In human rights terms, the SADF raid on Kassinga, which killed over 600 people, is possibly the single most controversial external operation of the Commission’s mandate period.

21 The SADF’s view on the situation in southern Angola was spelt out in a communication from the chief of staff operations (CSOPS) to the chief of the SADF on 27 February 19781. CSOPS argued that, since the termination of Operation Savannah, SWAPO had been successful in building up its strength in Owambo and the Eastern Caprivi, whence it was able to conduct operations in the northern areas of South West Africa. In the opinion of CSOPS, it was now imperative to deploy the full strength of the SADF’s air and land capability against SWAPO, instead of counting on a ‘hearts and minds’ programme. It was also important to counter the enemy’s propaganda by demonstrating that South Africa had so far deployed in Angola only a small part of its military capability.

22 CSOPS calculated that, by the end of 1978, SWAPO would grow from a force of 3 700 to 5 000 guerrillas, though it estimated that only 1 000 of these could be operationally deployed in the short term. The current estimate was that between 250 and 300 guerrillas were currently active in Owambo. Of SWAPO’s five principal bases in the region, CSOPS identified Kassinga as the largest, with an estimated 800 guerrillas in camp. It was, however, also the furthest from the South West African border (198 km north of the border) which made the logistics of an assault complicated.

23 In arguing for a raid, CSOPS noted that SWAPO was becoming better organised and that, as a result, the SADF’s principal operating unit inside Angola, 32 Battalion, was finding it difficult to operate. UNITA was likewise under pressure. In short, SWAPO was benefiting from the fact that South Africa was not fighting to its full potential. If operations became effective, on the other hand, SWAPO would be forced to disperse its camps into smaller components, making the organisation less effective. This would also make it more difficult for Cuban instructors to work with and easier for UNITA to operate against SWAPO.

24 On 8 March 1978, the chief of the army, Lieutenant General Viljoen, sent a communiqué (H/LEER/309) to the chief of the defence force, in which he identified the camp at Kassinga as the planning headquarters of SWAPO’s armed wing, the People’s Liberation Army of Namibia (PLAN) – subordinate only to SWAPO’s defence headquarters at Lubango. He also noted that the camp was the principal medical centre for the treatment of seriously injured guerrillas, as well as the concentration point for guerrilla recruits being dispatched to training centres in Lubango and Luanda and to operational bases in East and West Cunene. The camp also offered refresher courses in infantry warfare and mine-laying.

25 General Viljoen noted that the camp was not heavily defended and that the nearest Angolan army and Cuban forces were at Techumutete, fifteen kilometres to the south. The target lent itself to the maximum use of air power and the infliction of maximum casualties. Finally, he suggested that, given the presence in the camp of PLAN’s commander, Dimo Amaambo, important documents could be captured. Other documents in the SADF files make it clear that it was also hoped that Amaambo and other senior PLAN officials would be captured or killed.

26 Approval for the Kassinga operation – which became part of Operation Reindeer – was received in about March 1978. The original operational orders included the following priorities and instructions:

a Maximum losses were to be inflicted on the enemy but, where possible, leaders must be captured and brought out. Once the attack was completed, no prisoner-of-war was to be shot in cold blood.

b Documents as well as useful weapons were to be removed.

c Bases were to be destroyed.

d Skirmishes with Cuban and Angolan Army forces were to be avoided if at all possible.

e Photographs were to be taken after the attack to counter "enemy" allegations.

f Where possible, women and children were not to be shot.

27 The details of the plan were as follows:

phase one: from ten days prior to the operation (D-10): Low-profile coverage, with selected media references to Alpha (Kassinga);

phase two: beginning D-7: The key idea should be to create the impression of a resumption of SWAPO border violations and attacks on SADF patrols, especially against the local population. The intention would be after a relatively quiet period to refocus attention representing it as a seasonal trend. Shortly before D-1, information should be released on a SWAPO build-up. Thereafter, on D minus 1, a grave incident (real or imaginary) must take place. Either attempted assassination or cross-border attack on SADF patrol base. In the case of the latter, some casualties could be attributed to this attack;

phase three: D to D+4: Media operational:

C Army must ensure that media coverage of the operations (especially Alpha Camp) takes place. Credible coverage and immediate release are essential to counter probable hostile counter-claims of SADF operations and mass killings of civilians, especially women and children. Guidelines: Photo cover must feature: i) military features, for example,. weapons, ammunition, communications, headquarter buildings; ii) any dead must have weapons alongside them; iii) any photography of civilians must reflect humane treatment, e.g. being provided with food. Civilians should, however, be avoided altogether; iv) documents captured must feature prominently to add credibility to subsequent disclosures.

28 Accompanying this document in the file is an undated message from SWA Command to the chief of the army, which reads: "Contingency plans in progress to create own incidents that can be attributed to SWAPO should insufficient publicity or further SWAPO actions be forthcoming". The signal also emphasised the need to counter any enemy propaganda about attacking civilian targets.

29 On 25 April 1978, the SWA Command sent a signal for the attention of either General Viljoen or General Gleeson. It noted that the UN General Assembly was in session on the subject of SWA and that the debate would last until 3 May. D-day should, therefore, be delayed until the conclusion of the debate to avoid making life difficult for those countries favourable to South Africa. On 30 April, higher authority dictated a delay of at least forty-eight hours.

30 As the commanders of Reindeer waited for the go-ahead from the cabinet, a number of messages were passed on the changing situation. On 2 May 1978, the chief of the defence force signalled the chief of air staff that considerable extensions were being made to Kassinga’s defensive installations. It was also now estimated that between 1 000 and 1 500 PLAN recruits were at Kassinga, including fifty to one hundred Cubans. Permission was given for two Mirages to be added to the air component.

31 It is clear that from the SADF’s perspective, Kassinga was a military facility rather than essentially a refugee camp or refugee transit facility, as SWAPO has always claimed. The photographic evidence shown to the Commission at the SADF archives suggests a military dimension to the camp. This cannot, however, be taken as conclusive evidence that Kassinga was a military base. In the context of the ongoing war in Angola, some defensive fortification of any SWAPO facility, whether civilian or military, would have been standard practice.

32 What is evident is that human settlement in the Kassinga area had grown considerably in the period since Angola’s independence. The site was allocated to SWAPO by the Angolan government in 1976, after an appeal for help to cope with an inflow of thousands of refugees. Under SWAPO’s control, the abandoned homes in the village had been converted into offices, a kindergarten and primary school, a clinic, a sewing facility, and storage and vehicle repair workshops. New permanent structures had also been erected, plots cultivated and a set of defensive trenches dug. SWAPO had also installed two anti-aircraft guns in the centre of the village, and the camp contained a self-defence unit of approximately 300 male and female PLAN cadres.

33 The fact that Kassinga had a non-military dimension is reflected in a UNICEF report of a visit by a UNICEF delegation from its regional office in Brazzaville, published two days before the raid. Central to the report was the fact that, although well-run and well-organised, the facility was ill-equipped to cope with the rapid expansion in the size of the refugee population as a result of a steady inflow in early 1978. Kassinga was thus both a military base and a refugee camp. It housed a considerable number of combatants, including senior officers. It also housed considerable numbers of civilians. As a large facility, it was easily partitioned into military and non-military sections.

As a result of the ominous build-up of SWAPO forces in southern Angola and the extensive campaign of intimidation of the local inhabitants and the murder of political leaders in SWA, as well as the large number of border violations during the past few weeks, a limited military operation against SWAPO forces has been carried out over the border ... I trust that the limited operation will leave those who wish to threaten us under no illusions ... I have in the past while repeatedly expressed the hope that military bases will not be made available to terrorists in southern Angola, but this apparently has fallen on deaf ears. (Minister of Defence, Mr PW Botha, press release, 4 May 1978)

34 Reindeer began with an attack at 08h00 on 4 May 1978. Four Canberra jets dropped 300 Alpha bombs, followed one minute later by four Buccaneers which dropped seven 400kg fragmentation bombs. The initial target was the parade ground where between fifty to one hundred ‘enemy’ were seen immediately before the attack. At 08h04, 370 paratroopers were dropped on the camp, but many of them fell well away from the target area because of high winds. It took just over an hour for the paratroopers to group together, during which time many of Kassinga’s inhabitants were able to flee to safety. These included most of PLAN’s senior officers, including Dimo Amaambo.

35 Several hundred occupants of the camp were not so fortunate, however. By 10h30, signals from the ground were reporting heavy ‘enemy’ casualties and the capture of large quantities of weapons. At noon, the first helicopter extractions of SADF personnel began. By this time the chief of the army, Lieutenant General Viljoen, had been taken into the battlefield area aboard a Puma helicopter. General Ian Gleeson (101 Task Force), Colonel ‘Blackie’ de Swardt (SAAF) and Colonel ‘Giep’ Booysen (SA Medical Services) were in overall command of the actual operation and responsible for its planning. Fighting forces on the ground at Kassinga were led by Colonel Jan Breytenbach (32 Battalion), and Commandant Deon Ferreira, and at Chetequera by Major Frank Bestbier.

36 At 13h45 between twenty and thirty armoured cars were reported as being en route from the nearby Angolan military base at Techumutete, which also housed Cuban troops. Later, other troop carriers were spotted moving towards Kassinga. These convoys were attacked from the air and many were destroyed. According to information drawn from the Cuban archives, approximately 150 Cuban troops died in these attacks – the most serious casualty loss in their involvement in Angola.

37 By evening, the assault on Kassinga was complete and all SADF personnel, bar five missing troops, had been withdrawn from the site. The SADF’s anxiety about external reaction is reflected in a signal from the chief of the defence force, sent at 19h30, enquiring whether any women and children had been killed. This took priority over a 20h50 signal enquiring whether any Cubans had been captured. In response to the earlier query, SWA Tactical Headquarters sent out a top-secret message that night (OPS/104/04) reporting that there were many women and children at Alpha and that large numbers had been killed. Among the dead women, the message reported, many had been in uniform and many in the trenches.

38 At 22h00, SWA Tactical HQ sent a further message which read as follows (translated):

Target Alpha

Enemy losses estimated at 500 (five hundred) dead.

    1. No POWs taken because of the serious threat from the south from enemy tanks and armoured cars. Initially 75 POWs were captured and 15 ear-marked as POWs including women in uniform. Some were in civilian clothing. All looked like young recruits;
    2. Air attack caused extensive casualties and damage. All the buildings within reach were set on fire. Three 14.7 machine guns were damaged but could not be destroyed. Explosions indicated that ammunition dumps had been destroyed. A large number of documents were seized in the OC’s house. He could not be found;
    3. After the bombardment considerable numbers of the enemy were noticed in the trenches. They offered reasonably strong resistance;
    4. 14.7 machine guns could not be silenced by the ground forces, so close air support was called in. The machine guns fired on the Buccaneers at every opportunity;
    5. The enemy resisted firmly and did not run away. Some made out they were dead but were then either killed or captured.

39 The official death toll (according to an Angolan government White Paper) was 159 men, of whom only twelve were said to be soldiers, 167 women and 298 teenagers and children – a total of 624. In addition, 611 South West Africans were wounded in the attack. These were largely victims of the initial bombing attack. The dead were buried in two mass graves. Foreign journalists who saw the graves before they were covered confirmed that large numbers of the dead were women and young people wearing civilian clothes. This does not necessarily mean that they were all non-combatants. In a guerrilla camp, not all combatants would be wearing uniforms. Moreover, the figure of twelve soldiers cited by the Angolan White Paper is not credible, unless the reference to soldiers is limited to the twelve Angolan government soldiers killed during Reindeer. It is known too that PLAN forces contained women and it can be assumed that some of the women casualties were combatants.

40 The Kassinga raid formed only a part of the Reindeer operation. There were also attacks on a number of SWAPO facilities in and around Chetequera (an area known by SWAPO as "Vietnam") where over 300 South West Africans were killed and a large number captured. These prisoners – between 200 and 300 in all – were taken to the Oshakati military base where many were tortured. Nearly two years later, in January 1980, the International Red Cross reported that 118 of them were still being held at a detention centre near Mariental in the south of South West Africa (see further below).

41 Some 1 200 hundred people – South West African, Angolan, Cuban and South African – died; over 600 others, overwhelmingly South West African and Angolan, were wounded in the attacks on Kassinga and Chetequera that day. It is probable that some died later from their wounds. In addition, several hundred were captured at Chetequera. No prisoners, or perhaps at most a handful, were taken from Kassinga. Those reported in the early despatches as being held as prisoners were released when no room on the evacuation helicopters was available for them.

42 All the planning documentation, including aerial photographs, would indicate that the SADF command was convinced that Kassinga was the planning headquarters of PLAN, and thus a military target of key importance. Given the objective (expressed in the CSOPS communication to the Minister of Defence on 27 February 1978) of demonstrating South Africa’s military superiority and inflicting major damage on SWAPO’s military capacity, it would seem strange if a refugee camp had been chosen as a target. This does not, of course, obviate the possibility of mistaken target identification, though the aerial photographs and the detail of intelligence reports of what was going on at Kassinga obviously guided the SADF command to their conclusions. It is probable therefore, that the SADF could mount a plausible defence of its operation in terms of the doctrine of pre-emptive interventionism and the right of states to defend themselves.

43 This argument would, however, have to fail in this case for the simple reason that the state being defended by South Africa was South West Africa, which South Africa was occupying illegally. In terms of UN General Assembly Resolution 2145/66 and UN Security Council Resolution 385/76, South Africa’s mandate in South West Africa was revoked and its occupation deemed illegal. Hence any justification of a retaliatory operation cannot be accepted.

44 Beyond the issue of the mandate, Operation Reindeer violated international humanitarian law on other counts, one of which was the failure to take adequate steps to protect the lives of civilians. International humanitarian law stipulates that the right of parties in a conflict to adopt means of injuring the enemy is not unlimited and that a distinction must at all times be made between persons taking part in hostilities and civilians, with the latter being spared as much as possible. There is little evidence that the SADF took sufficient precautions to spare those civilians whom they knew were resident at Kassinga in large numbers.

45 The fact that the operational orders for Reindeer included the instruction that "women and children must, where possible, not be shot" is evidence of the SADF’s prior knowledge of the presence of civilians. However, this apparent intention to spare their lives was rendered meaningless by the SADF’s decision to use fragmentation bombs in the initial air assault, as such weapons kill and maim indiscriminately. Their use, therefore, in the face of the knowledge of the presence of civilians, amounts to an indiscriminate and illegitimate use of force and a violation of Protocol 1 to the Geneva Conventions of 1949. The foreseeable killing of civilians at Kassinga was therefore a breach of humanitarian law.

46 The treatment of civilians in this operation may have violated international legal covenants in another respect; in this case a provision of Protocol 11 regarding the treatment of the wounded. If press reports based on a press interview with an officer involved in the operation are accurate (see below), the SADF failed to protect and care for those wounded in the operation. According to this source, some of the wounded, irrespective of their status as combatants or civilians, men or women, were shot.

We were in enemy territory and had to get the hell out of there as quickly as possible. There were just too many wounded. We could have left them on the battlefield to die in pain and agony. We couldn’t. I was given an AK-47 and instructed to kill those who couldn’t be saved. I had to decide who was not going to live. I was the company leader, so I had to take the lead. I don’t know how many people I shot that that day. Some were conscious, some were not. We found this woman clutching her screaming baby. It was only when we saw the terrible wounds inflicted by an Air Force bomb. There was no hope for her. I had to shoot her. She looked at me. I can never describe what it did to me. It was too much. I later broke down. (Anonymous soldier, The Star, 8 May 1993.)

47 The Commission has not been able to corroborate the information contained in this quote. It did, however, speak to retired General Chris Thirion who was present at Kassinga and who stated that he was standing next to General Viljoen when he heard him give a command that no wounded were to be killed. He concedes that this does not mean that none were killed but, if so, this would have been contrary to orders. The Commission accepts, therefore, that there were no orders to kill the wounded, but is not convinced that this did not happen. The Commission also has evidence from the war in South West Africa that, on occasion, badly wounded SWAPO fighters were shot and not given medical treatment.

48 The Commission, therefore, makes no finding on the treatment of the wounded. It does, however, make a finding on the choice of weaponry used in the attack and on the question as to whether adequate care was taken to protect the lives of civilians. In making a finding, the Commission has taken cognisance of the following international legal provisions: Principles III, IV, VI of the Principles of International Law Recognised in the Charter of the Nuremberg Tribunal and in the Judgement of the Tribunal 1950:

 

THE COMMISSION FINDS THAT OPERATION REINDEER WAS A VIOLATION OF THE TERRITORIAL SOVEREIGNTY OF THE REPUBLIC OF ANGOLA AND THAT IT RESULTED IN THE COMMISSION OF GROSS HUMAN RIGHTS VIOLATIONS AGAINST THE CIVILIAN OCCUPANTS OF THE KASSINGA CAMP WHICH ENTAILED DELIBERATE PLANNING ON THE PART OF THE FOLLOWING PERSONS WHO ARE HELD ACCOUNTABLE:

• PRIME MINISTER BJ VORSTER IN HIS CAPACITY AS HEAD OF STATE (PRINCIPLE III);

• MINISTER OF DEFENCE PW BOTHA IN HIS CAPACITY AS POLITICAL HEAD OF THE SADF;

• GENERAL MAGNUS MALAN IN HIS CAPACITY AS CHIEF OF THE SADF;

• LIEUTENANTS GENERAL CONSTAND VILJOEN AND RH ROGERS IN THEIR CAPACITIES AS CHIEFS OF THE ARMY AND AIR FORCE RESPECTIVELY.

The State Security Council and Angola

49 With the accession to power of Mr PW Botha in September 1978, the war against Angola became a government priority. In March 1979, the State Security Council (SSC) adopted two strategy documents pertaining to Angola. One was a total national strategy encapsulating a long-term view; the other was a short-term strategy document. In the former, the government spelt out its goals as being to use all means – political, diplomatic, psychological, economic and military – to neutralise the Angolan government’s support for SWAPO as well as to bring down the MPLA government, if and when the assurance existed that a more friendly and stable government would replace it.

50 The short-term strategy document lists its objectives, stating that the political situation in Angola should be kept as unstable as possible, that support should be rendered to UNITA and other movements, and that clandestine operations should be launched against Angola with the aim of forcing the MPLA government into preventing SWAPO from operating in southern Angola.

51 In a section headed Opdrag en Take (Mission and Duties), four tasks are detailed:

a to subject southern Angola to a national strategy for as long as it takes to pressurise the MPLA to abandon its support for SWAPO;

b to enlist UNITA and other movements as partners ("bondgenote") against the Marxist onslaught;

c to destroy SWAPO bases in Angola through co-ordinated actions;

d to make preparations for conventional operations against Angolan, Cuban and SWAPO forces.

52 With regard to the first two tasks above, the document talks of establishing a stable anti-Communist government in the south (a UNITA government) as soon as the political situation in the area has stabilised. All tasks and strategies in the short-term plan were to be directed to this end. In other words, the immediate objective of the SADF was to partition Angola and, in effect, to bring about the secession of the south of the country.

53 The importance of UNITA and its leader, Mr Jonas Savimbi, to South African strategy at this time was stressed in a letter from the chief of the SADF, General Malan, to CSOPS (chief of staff operations) Major General Earp, dated 6 March 1979, in which he states that "Mario [SADF codename for Savimbi] se voortbestaan raak direk die toekoms van Suidelike Afrika. Hy het so belangrik geword dat ons sy veiligheid sal moet verseker". (His continued existence directly influences the future of Southern Africa. He has become so important that we will have to ensure his safety.)

54 Under economic action guidelines, the following steps were listed:

a to so disrupt the national infrastructure of Angola through clandestine operations, that an unstable situation in the country would be created;

b to disrupt, through clandestine operations, the main export harbours and railways leading to the south of Angola;

c to handle to the best advantage of South Africa all requests from the Angolan government in connection with electric power from Ruacana (the hydro-electric scheme on the Angola–Namibia border), as well as food supplies.

55 The document concludes by emphasising the urgency of the situation and the intensity of the Marxist onslaught in the region. It then goes on to argue that, owing to the abbreviated UN time-scale envisaged for South West Africa (a reference to the pending implementation of UN Resolution 435), the short-term strategy must be implemented as speedily as possible.

56 The implementation of this strategy was discussed at a series of meetings in March 1979. At a meeting between Savimbi, the chiefs of the defence force and the army and senior staff officer of special operations in the department of Military Intelligence (MI), it was agreed that UNITA’s priority would be to clear the MPLA and SWAPO out of Cuanda-Cubango province and "dele van die Cunene" (parts of the Cunene province) so that the nucleus of a UNITA government could be established and so that "die verbindingslyne tussen Angola en SWA weer daargestel kan word" (the lines of communication between Angola and SWA could be restored). A modus operandi was agreed whereby the SADF would take the initiative in Cunene province (see discussion on Operation Protea below) with support from UNITA, while the reverse would apply in Cuanda-Cubango. It was also agreed that, from this point, 32 Battalion would be semi-permanently deployed in southern Angola.

57 The post-1979 military strategy in Angola therefore took the form of a series of large-scale conventional military operations against pre-selected SWAPO and Angolan Army targets, while 32 Battalion formed the vanguard of a low-intensity counter-insurgency campaign. The third prong was Operation Silwer (see below), which provided ongoing logistical and other support to UNITA.

58 The Angolan war was an ongoing, thirteen-year-long occupation, enabling the SADF to achieve one of its aims, namely, the de facto secession of the south from central government control. This is certainly the view of former army and SADF chief, General Geldenhuys. Writing in his autobiography, he states:

A few specific cross-border operations made headlines … Each, in turn, had a positive influence on the course of the war … in the end … it was the overall effect of the almost unseen but incessant day-to-day general operations that brought us success.

59 The first big operation in this post-1979 phase of the war was launched in June 1980. Operation Sceptic began as a lightning attack on a SWAPO base complex called ‘Smokeshell’, 120 kilometres into southern Angola and developed into an extended operation which produced the first serious clashes between the SADF and the Angolan Army and mechanised elements of SWAPO. It ended by driving SWAPO from its forward bases; 380 guerrillas and seventeen members of the SADF were killed.

60 A year later, in August 1981, the SADF launched Operation Protea, the largest mechanised operation undertaken by the South African military since World War II. Protea was launched in implementation of the decision made in March 1979 to install UNITA as the de facto government of southern Angola. In the initial eighteen-day phase of the operation, the SADF occupied 50 000 square kilometres of Cunene province. Thereafter, parts of the province remained under SADF occupation until 1989, essentially as a support for UNITA which, from January 1982, took effective administrative control of most of the province.

61 The information below is drawn from a variety of sources including files in the collection "Aanvullende Dokumente" (Supplementary Documents) OD 1968 no. 20 held by the military archives.

62 According to this collection, the immediate military objectives of Operation Protea were the destruction of PLAN headquarters at Xangongo and Ngiva and the capture of the extensive caches of equipment held at Ngiva. During the initial 1981 phase of Protea, several towns in Cunene province such as Cahama, Chibemba, Xangongo, Mongua and the provincial capital Ngiva were extensively damaged by bombing and artillery attacks which also caused civilian casualties.

63 After two weeks, all the towns mentioned above, indeed most of the province, had been evacuated. According to the BBC, an estimated 160 000 Angolan civilians were rendered homeless and were forced to flee to sanctuary further north in Lubango. British press reports of the operation spoke of the SADF’s use of heavy artillery for long-range bombardment of towns and villages in combination with SAAF carpet-bombing before ground troops moved in to take control. The result in Cahama was that only four out of forty to fifty buildings in the main street were left intact. Amongst the facilities reportedly destroyed were a hospital, a dispensary and a food distribution centre.2

64 According to press reports, more than 1 200 SWAPO and Angolan Army soldiers were killed in the operation, along with several Soviet advisers. One Russian warrant officer and seventy-nine Angolans were captured. The SADF files give a casualty toll of 746 Angolan Army and eighty-one SWAPO combatants, and four Soviet advisers. The number of civilian deaths is not known to the Commission, but, given the systematic and sustained targeting of civilian centres, it must have been large. The SADF suffered fourteen dead and sixty-four wounded.

65 In terms of civilian casualties, the mass displacement of civilians and the creation of an internal refugee population, as well as the wholesale destruction of towns and socio-economic infrastructure, Operation Protea probably caused more human suffering and physical damage than any other operation in the thirteen-year-long Angola war, resulting in violations of human rights on a vast scale.

66 Protea was followed by Operations Daisy in 1981, Super and Meebos (1982), Phoenix and Askari (1983), Boswilger (1985), Modulêr and Hooper (1987–88), Packer and Displace (1988). Each of these was a smaller-scale version of Sceptic and Protea and resulted in large numbers of casualties. According to the military historian, Colonel CJ Nothling, writing in the 1989 South African Defence Review, over 8 000 ‘terrorists’ (SWAPO and Angolan Army forces) were killed in these campaigns. No figures are cited for civilian deaths. The SADF acknowledged 136 fatalities (three each in Daisy and Super, twenty-nine in Meebos, twenty-seven in Phoenix, twenty-one in Askari and fifty-three in Hooper and Modulêr).

67 The December 1983 Operation Askari was aimed at disrupting PLAN’s logistical infrastructure and its command and control systems through ground and air attacks. In its advance towards SWAPO headquarters near Cuvelai, a major battle developed between South African forces and Angolan Army units aided by two Cuban battalions. According to the SADF, this was the biggest encounter between South African and Angolan Army forces of the entire war: 324 Angolan and Cuban troops and twenty-one South Africans were killed. Another casualty was the town of Cuvelai, which was almost totally destroyed. In 1984, when there was a temporary withdrawal, Angolan authorities re-entered the ruined town and reported that facilities and most buildings had been destroyed while livestock had either been killed or taken to South West Africa. There had been extensive civilian casualties.

68 During Operation Hooper, SADF and Angolan Army forces clashed in a number of large land battles near the town of Cuito Cuanavale. The town was shelled by SADF 155mm artillery for several weeks, and largely destroyed. The SADF failed, however, to capture the town and the stalemate led eventually to negotiations and the signing of the New York Accords in December 1988. These agreements produced an SADF withdrawal from Angola (Operation Displace), the implementation of UN Resolution 435 and the independence of Namibia in March 1990.

69 The above operations all targeted either SWAPO or Angolan Army facilities. In another SADF raid, the South African Air Force (SAAF) attacked an African National Congress (ANC) camp at Nova Catengue on 14 March 1979. Essentially a transit facility, the camp housed large numbers of recent exiles (the 1976 Soweto outflow). While the facility was severely damaged in the raid, casualties amounted to three dead (including one Cuban) and fourteen wounded. Casualties were comparatively low because advance intelligence had been received and the camp had been evacuated.

The economic and human costs of the Angolan War

70 In addition to destroying Angolan towns in the south of the country, the SADF targeted economic installations in Angola, especially its petroleum facilities. These included attacks on the Luanda oil refinery on 30 November 1981, the storage tanks in the port of Namibia in June 1988 and those at Huambo in 1987. On 21 May 1985, a Recce 4 commando unit was intercepted attempting to sabotage the Cabinda Gulf Oil complex and the commander, Captain Wynand du Toit, was captured. Road bridges and the Benguela railway were also frequently sabotaged.

71 The effects of the war on Angolan civilians were devastating. UNICEF has estimated that, between 1980 and 1985, at least 100 000 Angolans died, mainly as a result of war-related famine. The cumulative effect of the battering of the economy and social infrastructure in the 1980–85 period produced an even greater escalation in the death rate after 1985. Between 1981 and 1988, again according to UNICEF, 333 000 Angolan children died of unnatural causes. The Angolan government estimated the economic cost of war damage to be US$12 billion in 1987 alone.

72 The environmental effects of the war on the south (and in the Caprivi) were devastating. Both forest lands and wildlife were destroyed. This rape of the environment was sanctioned by the SADF. In the early 1980s, covert front companies were established to facilitate trading in rare woods like teak and kiaat, and in ivory, skins and diamonds. A safari company was also set up through which the hunting of big game was regulated. Ostensibly, these activities were undertaken to raise secret funds for UNITA, but they led quickly to widespread and high-level corruption.

THE COMMISSION FINDS THAT THE SOUTH AFRICAN GOVERNMENT’S MILITARY CAMPAIGN IN ANGOLA BETWEEN 1977 AND 1988 LED TO GROSS VIOLATIONS OF HUMAN RIGHTS ON A VAST SCALE. THE COMMISSION FINDS FURTHER THAT THE CAMPAIGN CONSTITUTED A SYSTEMATIC PATTERN OF ABUSE, WHICH ENTAILED DELIBERATE PLANNING ON THE PART OF THE FORMER CABINET, THE STATE SECURITY COUNCIL AND THE LEADERSHIP OF THE SOUTH AFRICAN DEFENCE FORCE. THE COMMISSION FINDS THESE INSTITUTIONS AND THEIR MEMBERS ACCOUNTABLE FOR THE AFORESAID GROSS VIOLATIONS OF HUMAN RIGHTS.

 

· POLICE AND MILITARY COUNTER-INSURGENCY OPERATIONS IN SOUTH WEST AFRICA

The third specific incident that I remember is chasing a SWAPO unit commander or political commissar. We picked up his spoor and chased him for two days ... this was typical of the style of contacts that I was involved in. Five Casspirs, fifty men chasing one or two people running on foot. We finally did catch him, hiding in a kraal. The unit commander … lined up a bunch of Koevoet people next to the hut he was in and drove over the hut with the Casspir. Everyone then fired into the rubble ... The SWAPO commissar was pulled out of the rubble and given to me to keep alive. He had been shot in the arm and the leg and had been driven over ... because he was a commissar, he would have been carrying a handgun. John Deegan [acting unit commander] started to interrogate him while I was putting up a drip. The purpose of this interrogation was to find the handgun ... We never found the handgun because John shot him in the head out of frustration while I was still attending to him. The incident and the face of this SWAPO commissar haunted me in dreams for years. (Lance Corporal Sean Callaghan, amnesty application, December 1996.)

[The SWAPO commissar] was a veteran ... he would have been an excellent source of information but he was so fucked ... each team had an army medic and Sean started patching up this guy while I was busy interrogating him ... and he was just going "kandi shishi", even at this stage he was denying everything ... and I just started going into this uncontrollable fucking rage and he started going floppy and I remember thinking "how dare you, I’m talking to you, how dare you ignore me...why don’t you answer me" and then this is what I was told afterwards. I had my 9mm in my hand and I was just pushing my way through the team ... and apparently what happened was I started ripping … Sean had put a drip into the guy’s arm and started plugging the bullet hole to get him together ... he would have pulled through ... I ripped all the bandages, the drip off the guy, pulled out my 9mm, put the barrel between his eyes and fucking boom I executed him ... and they told me afterwards I was just screaming, I was raging ... (Warrant Officer John Deegan’s account of this incident to the Commission’s conscript hearing, Cape Town, June 1997.)

73 Three factors are central to the human rights situation in South West Africa (Namibia) in the Commission’s mandate period. The first is the fact that the South African presence in the territory was a violation of international law, and that the South African administration and its courts and security forces had no right in international law to carry out any actions affecting the South West African people. If this did not apply for the full Commission mandate period, it certainly applied from October 1966 when the UN General Assembly terminated South Africa’s mandate over South West Africa, a decision affirmed by the Security Council in 1969. In June 1971, the International Court of Justice in The Hague declared South Africa’s presence in South West Africa illegal and demanded its withdrawal. It further declared invalid all South Africa’s acts on behalf of or concerning South West Africa. The Commission’s analysis of the situation in South West Africa is informed by this position in international law, from which it follows that all security-related actions initiated by the South African and South West African administrations and their security forces were those of illegal and illegitimate authorities.

74 The second factor relates to the fact that, for twenty-three years, these authorities were engaged in a guerrilla war against an indigenous liberation movement whose armed struggle was legitimated both in terms of international law and by the overwhelming moral support of the international community.

75 The third factor relates to the sheer enormity of the topic. South Africa’s occupation of South West Africa would merit a separate truth commission of its own.

76 In the account that follows, emphasis will be laid on two particular factors – torture and extra-judicial killings. In regard to the latter, the focus will be on the bounty or ‘cash-for-corpses’ policy employed by the police counter-insurgency unit, Koevoet (crowbar). Where possible, reference will be made to those human rights violations and amnesty applications pertaining to South West Africa that were submitted to the Commission. These were, however, relatively few: only one human rights violations submission and thirteen amnesty applications were received.

Changes in human rights violations over time

77 The pattern of human rights violations in South West Africa varied over time, in accordance with the level and nature of resistance to the South African occupation. This pattern may be periodised as follows:

1960–1966

78 During this period, there was little organised resistance to South Africa’s occupation and no armed struggle. Even so, the apartheid system was enforced with even more rigidity in South West Africa than in South Africa itself and the human rights of the people of South West Africa were constantly and systematically violated, in particular through the system of contract migrant labour. During the 1960s and 1970s, up to two-thirds of South West African workers were subjected to this form of labour control and coercion. Contract workers were required to leave their families in the ‘homelands’ and to sign contracts that rendered them powerless to choose their employer or to negotiate a wage. Those who resigned from their jobs or broke their contracts were liable to deportation back to the ‘homeland’. This was a systematic violation of basic human rights which established a system of quasi-slavery. It also served to depress wages and prevent labour organisation. While modifications were made to the system after the 1971–72 contract workers’ strike, the system remained in place until 1977.

1966–1971

79 The second period falls between 1966 and 1971. In 1966, SWAPO launched an armed struggle, although only a few incidents took place in this period, mainly in the Caprivi Strip. In this period, the SAP were in direct control of the counter-insurgency effort. A number of leading members of SWAPO involved in this early phase of the armed struggle were captured and put on trial in Pretoria (see below).

1971–1974

80 The third period falls between 1971 and 1974. The Advisory Opinion of the International Court of Justice in 1971, that South Africa’s presence in South West Africa was illegal, triggered a contract workers’ strike which involved between 13 000 and 20 000 workers. Subsequent political tensions in Owamboland in the north, where most South West Africans live, resulted in considerable police activity and public floggings carried out by bantustan officials (see below).

1974–1980

81 The fourth period falls between 1974 and 1980. This was a period of militarisation. SWAPO established camps and bases in the south of Angola after Angola became independent in 1975 and began operations along the Angola–South West Africa border. In the same year, the SADF took over counter-insurgency responsibility from the SAP and established an infrastructure of bases throughout the ‘operational areas’ of Owamboland, Kavango and Caprivi. Human rights abuses by South African troops during this period escalated considerably. One consequence was a dramatic increase in the outflow of refugees, particularly from Owamboland.

1980–1988

82 The fifth period falls between 1980 and 1988. From around 1980, the nature of the war began to change. South Africa increasingly relied on Koevoet, a newly-formed special police counter-insurgency unit, which became notorious for its human rights abuses during its pursuit operations. A process of indigenising the war effort began and South West Africans were recruited and conscripted into a South West Africa Territory Force (SWATF), a largely locally-staffed military force which took on much of the burden of the war, although it remained under firm South African control at the senior officer level. A South West African Police force (SWAPOL) was established in a similar manner.

1989 onwards

83 In 1989, elections were held under UN supervision. South West Africa became independent (as Namibia) the following year and all South African police and military forces were withdrawn.

Categories of abuses

84 Human rights abuses in South West Africa fell into the following categories.

Political repression and imprisonment

85 While SWAPO was never banned in South West Africa, many of its leaders were harassed, put on trial and imprisoned, either in South Africa or in South West Africa, despite the illegality of South Africa’s presence and actions in the territory. South African legislation, including the Terrorism Act of 1967 (introduced specifically to try SWAPO leaders), the Internal Security Act (extended to South West Africa in 1976) and the Riotous Assemblies Act, were employed for these purposes.

86 Following the launch of the armed struggle in 1966, thirty-seven SWAPO activists and leaders were arrested and taken to Pretoria where they were held under the Suppression of Communism Act until the Terrorism Act was passed. This legislation was made retrospective and the SWAPO activists were then tried under its terms. All of the accused reported being tortured, mainly through beatings and electric shocks. All thirty-seven were convicted. Twenty were sentenced to life imprisonment – nine, including Mr Andimba Toivo ya Toivo, to twenty years, while the others received lesser sentences. The trial was condemned as illegal by the UN Security Council. The long-term prisoners were sent to Robben Island to join South Africa’s political prisoners.

87 After the Pretoria trial, most SWAPO political trials were held in South West Africa. However, until 1984, many of those convicted continued to be sent to Robben Island, where they were kept in the same dehumanising conditions as the South Africans, and sometimes subjected to additional abuses and beatings. By 1984, all South West African political prisoners on Robben Island had been released or transferred to gaols in South West Africa.

88 In contrast to the South African situation, no death sentence or judicial execution of a South West African was ever carried out for offences of a political nature. Death sentences were occasionally imposed but were always set aside for fear of incurring the further wrath of the international community.

89 Political repression in South West Africa was acute. The 1980 Ida Jimmy case, for example, provides a particular example of severe ill treatment. In 1980, the chairperson of SWAPO Women’s Council in South West Africa, Ms Ida Jimmy, was sentenced to seven years in prison (subsequently reduced to five) for calling for support for SWAPO guerrillas at a public meeting. Although seven months pregnant, she was kept in solitary confinement in Windhoek Central Prison. Her baby boy was taken away from her at the age eighteen months, and given to the care of relatives. The boy died a year later. Ms Jimmy was refused permission to attend the funeral.

90 As in South Africa, the authorities restricted the movements of opponents through bannings or house arrest. One such victim was the acting president of SWAPO inside South West Africa, Mr Nathaniel Maxuilili, who was banned to Kuisebmond township in Walvis Bay for a seventeen-year period between 1968 and 1985, and prohibited from speaking at meetings or from being quoted by others.

Detention and torture

91 Detention without trial was widely used by the South African authorities in South West Africa, mainly for purposes of interrogation, which almost invariably and routinely involved torture. Much of this was carried out under emergency proclamations, notably Proclamation AG26 of 1978 (which provided for indefinite detention) and AG9 of 1977, which became more and more severe and allowed police and soldiers to detain people for up to thirty days without reference to any higher authority. In 1983, the SADF revealed that it had detained a total of 2 883 people during the 1977–83 period, while the security police disclosed that they had detained 2 624 people during the same period.

92 Mass detentions in the ‘operational areas’ were common. Many detainees were held secretly and without access to lawyers or relatives for long periods, sometimes years. Such conditions provided opportunities for prolonged abuse and torture. Torture was also used as a method of intimidation by police and soldiers in the war zone, and as a way of extracting ‘operational’ information quickly. Torture methods reported in the South West African press, in affidavits by South West Africans and as a result of international human rights investigations included beatings, sleep deprivation, drowning, strangling and suffocation, suspension from ropes or poles, burnings (sometimes over open fires), electric shocks and being held against the hot exhausts of military vehicles.

93 SWAPO’s Administrative Secretary in South West Africa, Mr Axel Johannes, was a victim of repeated detentions, torture and arrests. He was detained in 1964, 1966, 1973, 1974, 1975 (twice), 1977, 1978 (twice) and 1979 (twice). During these periods, he reported being repeatedly tortured and was often held incommunicado and in solitary confinement. After his final detention in 1979, restrictions were placed on his movements and he was prohibited from leaving the township of Katutura. He went into exile in 1980.

94 Captain Pat King of the South African Security Branch was charged in 1987 with the murder of detainee Mr Johannes Kakuva, who had been killed seven years previously. The trial followed international protests as a result of an official enquiry held in 1983, which had accepted the evidence of seven men detained with Kakuva that they had been assaulted and tortured. The men said that they and Kakuva had been beaten with sticks and subjected to electric shocks on King’s orders, resulting in Kakuva’s death. The trial ended with the acquittal of Captain King.

The case of the Chetequera/Mariental detainees

An estimated 200 to 300 South West African refugees in Angola were captured by the SADF during the raid on Chetequera in 1978 and taken to Oshakati military base, where many were tortured. A year later, UN officials published the names of 130 people whom they said were still being detained, but this was met with a denial by the South African authorities. The position of these detainees was discussed at an SSC meeting on 23 April 1979 where it was noted that 119 of them were still being held at a camp (Keikanachab) near Mariental, a small town south of Windhoek. Of these, three were Angolan citizens while the rest were described as "Wambo’s uit Suidwes-Afrika" (Owambos from South West Africa). Six were said to be members of MPLA, while the rest were described as "SWAPO-lede en geharde terroriste" (SWAPO members and hardened terrorists).

The problem the SSC faced was what to do with them. They could not, in the SSC’s view, be regarded as prisoners-of-war "omdat daar nie ’n staat van oorlog bestaan nie" (because a state of war did not exist). It was also noted that they were no longer providing any useful information and could not, due to their circumstances, be used as witnesses. In short, the document concluded "hulle het geen verdere nut vir die owerhede nie" (they were of no further use to the government). Even so, the meeting decided not to release the detainees, largely for fear of the propaganda SWAPO would make out of the incident and because the six MPLA members could be used "vir moontlike onderhandelings en ander diplomatieke doeleindes" (for possible negotiations and other diplomatic purposes).

The plight of these detainees was raised by the International Red Cross in 1980 which resulted a short while later in the release of a few of the prisoners – a full two years after their forcible abduction. However, in the case of the other 124, the status quo persisted for another five and a half years. Eventually, in 1984, three South West African bishops, supported by relatives of the detainees, brought an urgent application in the Windhoek Supreme Court for the detainees to be released. The court hearing was prohibited by the South African State President PW Botha, using powers under the Defence Act (section 103 ter (4)). They were, nonetheless, released in two batches in May and October 1984. They reported having being tortured and assaulted and forced to carry out hard labour. For close on seven years, they had been denied family visits and access to lawyers.

THE ABDUCTION OF ANGOLAN CITIZENS AND SOUTH WEST AFRICAN REFUGEES FROM ANGOLAN SOIL, THEIR TORTURE AT THE OSHAKATI MILITARY BASE AND THEIR FORCIBLE DETENTION AT MARIENTAL WHERE THEY WERE DENIED BASIC RIGHTS AND FORCED TO UNDERTAKE HARD LABOUR AMOUNT, IN THE COMMISSION’S VIEW, TO ILL TREATMENT AND DEPORTATION TO SLAVE LABOUR AND, AS SUCH, TO GROSS VIOLATIONS OF HUMAN RIGHTS. FOR THIS, THE COMMISSION FINDS PRIME MINISTER BJ VORSTER, MINISTER OF DEFENCE, PW BOTHA, AND THE CHIEF OF THE SADF, GENERAL MAGNUS MALAN, TO BE RESPONSIBLE FOR THE PERPETRATION OF GROSS HUMAN RIGHTS VIOLATIONS IN THE CASE OF THE MARIENTAL DETAINEES. IT ALSO REGARDS AS INDIRECTLY RESPONSIBLE (DOLUS EVENTUALIS) ALL THOSE PRESENT AT THE SSC MEETING OF 23 APRIL 1979.

95 The extent of the torture of detainees and civilians caught up in military operations has been documented by numerous bodies:

a In 1981, a delegation from the British Council visited South West Africa and conducted extensive interviews with church and community representatives. They reported that torture and intimidation were widespread. Their documentation of more than twenty individual cases included incidents where corpses of alleged guerrillas were dragged through villages behind military vehicles.

b A 1982 visit by a delegation led by Archbishop Hurley from the Southern African Catholic Bishops’ Conference (Bishops’ Conference) detailed further cases of torture, based on interviews with 180 South West Africans. The Archbishop was subsequently charged under the Police Act for making this report and South African and South West African newspapers were threatened with prosecution if they published his statements. The charges were dropped after international protests.

c After 1982, the South West African Bar Council began to speak out about torture and assaults. It expressed shock and concern about the abuse of detainees, rapes and deaths in detention and the immunity and secrecy under which ‘security force’ members operated. It noted that detainees were often kept in makeshift corrugated-iron detention cells in the blazing sun and that, despite the many reported cases of torture, very few incidents were ever brought to court.

d In 1989 a Dutch group, acting on behalf of the inter-denominational faith group Kairos, undertook a mission to South West Africa to investigate torture allegations specifically relating to the activities of Koevoet.

96 The systematic pattern of torture, which was institutionalised as an operational military and policing technique, resulted in few prosecutions or official efforts to eradicate the practice. Indeed, where military and police officials were found guilty, they were often given derisory sentences. In a 1984 case, two SWATF members were each fined R50 after being found guilty of assaulting sixty-three-year-old Mr Ndara Kapitango, whom they roasted over an open fire, causing extensive injuries. In another case in 1983, two Koevoet members were given similarly small fines after the death of a detainee, Mr Kadmimu Katanga, whom they had beaten with an ox yoke.

97 In general, police and soldiers could escape prosecution under section 103 of the Defence Act, which granted immunity to members of the security forces for any acts carried out under operational conditions, providing they were done "in good faith". South African State President PW Botha invoked this clause twice; first in 1986 to stop the trial of four soldiers accused of beating a detainee, Mr Frans Uapota, to death; and again in 1988 to stop the trial of six South African soldiers charged with the murder of SWAPO leader Mr Immanuel Shifidi, who was assassinated at a public meeting.

Extra-judicial executions and killings

98 The powers granted to security force personnel, and the secrecy in which they operated, created conditions for summary executions and killings for which they did not have to account. Usually, inquests into deaths were not held in operational areas. When they were, they were usually brief and inadequate, and responsibility was commonly attributed to "persons unknown". It was common practice for the security forces to leave bodies where they lay or to bury them in shallow graves at the place of death.

99 Koevoet in particular kept no proper or official records of the identities, numbers or whereabouts of people it killed. It seems that the unit was only really interested in keeping scorecards of those it killed for bounty. These practices were confirmed by journalists who were allowed to travel with security force units, as well as by court testimonies by security force members. At the height of the war, in the early to mid-1980s, Koevoet alone claimed a kill rate of around 300 to 500 people a year, for which its members were paid a bounty per corpse. Rough ‘body counts’ were periodically issued by military headquarters, but there was never any independent confirmation as to whether these figures were accurate or whether the victims were civilians or SWAPO fighters.

100 The South African authorities refused to accord prisoner-of-war status to captured SWAPO combatants, despite the 1977 Protocol 1 to the Geneva Conventions, which extended the provisions of the Conventions to anti-colonial struggles and wars of national liberation and self-determination.

101 While combatants were initially put on trial and imprisoned (see above), there is considerable evidence that, as the war progressed, South African security forces, especially Koevoet, resorted increasingly to summary executions of captured combatants. The payment of bounty served as an incentive for the extra-judicial murder of captives. The representative of the International Committee of the Red Cross in South West Africa said in 1981 that "it simply does not happen in any conflict or battle that you have a clash with 200 people and forty-five killed and no prisoners or wounded are taken"3.

102 In other cases, captured combatants were kept in makeshift detention centres, such as the camp at Osiri, 160 kilometres north of Windhoek, where they were interrogated and often tortured. Reporters found that Osiri detainees were held in tiny corrugated-iron cages and were always blindfolded in the presence of their interrogators. Though periods of detention were limited to thirty days, detention orders could be renewed indefinitely, running into months and even years.

103 After the formation of Koevoet, it became standard practice to ‘persuade’ captured guerrillas to ‘turn’ and become askaris assisting Koevoet in the conflict against their former comrades. This was a practice pioneered by the Rhodesian Selous Scouts, the archetypal model for Koevoet and the unit within which most of Koevoet’s founding members had learnt their counter-insurgency skills. There is considerable evidence that the process of ‘turning’ was accompanied by torture and that the price of non-compliance was summary execution. Once ‘turned’, these askaris and other Koevoet members are said to have carried out atrocities while disguised as SWAPO fighters in order to discredit the liberation movement, as the Selous Scouts had done during the Rhodesian war.

104 One such incident occurred at the village of Oshipanda near Oshikuku in June 1983. A group of armed black men in camouflage uniforms raided the kraal of Mr Hubertus Mateus Neporo, a modestly prosperous shop-owner who was suspected of giving financial backing to SWAPO. They ransacked the kraal, stealing cash, clothes and a radio, and smashing the family vehicle. Neporo was away that night, but the rest of his family and other occupants of the kraal were lined up against a wall and shot. Neporo lost his wife, children, mother, brother and friends – in all, eight civilians were murdered. The official line from the police, media and magistracy was that they were killed by ‘terrorists’. However, one kraal resident survived and identified the attacker as a certain Nakale from the Koevoet base at Okalongo. Within months of this incident, Koevoet members were reported to be boasting of this successful false-flag operation.

105 In an amnesty application, SADF conscript Mr Kevin Hall [AM1383/96] provides an insight into the brutal nature of the conflict with SWAPO as well as the routine use of torture. In May/June 1975, he was stationed at the Mapungeerela base in northern South West Africa. He recounts an incident where he was sent out on a seven-day patrol with instructions "to eliminate or arrest any terrorists". On the first night of the patrol, the group was "attacked by unknown forces and came under heavy gunfire". A lengthy gunfight ensued. In the morning, several bodies were discovered, as well as three badly wounded combatants. According to Hall:

I realised that none of them could survive and to save them any further suffering I shot and killed the three of them. When I shot them, I turned my head away as I could not bear their suffering any more.

106 Hall’s action in killing the three was a gross violation of human rights for which he applied for amnesty. His commanding officer, whom he names but whom the Commission cannot identify as he has not applied for amnesty, is likewise accountable. So too is the command structure of the SADF at the time.

107 A day or two later, while on the same patrol, the group came across four unarmed "terrorists" and arrested them. On their return to base, the four were placed in a hole in the centre of the base "approximately eight foot square by about seven foot deep" which served "as a place of safekeeping of all arrested terrorists". Hall continued:

They were the only ones in the hole. Whilst I was guarding them some of the troops poured boiling water over their heads; another troop of whom I cannot remember the name jumped into the hole and cut off the left ear and centre finger of the right hand of one of the [still living] prisoners.

108 Amnesty applicant Captain Eugene Fourie [AM3767/96] was a member of the Security Branch of the SAP stationed in Oshakati between June 1980 and January 1982. His application covers the death of a "SWAPO terrorist" whom he was interrogating. Fourie describes how, after administering electric shocks, the victim "het op ‘n stadium inmekaar gesak en sy bewyssun verloor en hy is toe agterna oorlede" (collapsed at one stage and lost consciousness and later died).

109 Former security policeman Warrant Officer Paul Francis Erasmus [AM3690/96] served in Owamboland in 1981. He applied for amnesty for the murder of a suspected SWAPO medical officer and for "suspects who were tortured on a regular basis by myself and other SB [Security Branch] members with the full knowledge and consent of commanding officers".

110 Another former member of the Security Branch based at Oshakati, Warrant Officer John Deegan, gave a lengthy statement to the Commission on his experiences with the SAP and Koevoet in South West Africa and Angola. He describes visiting the Security Branch offices in Oshakati in January 1981 where a round-the-clock interrogation session had been underway for about a week.

We would work in shifts and the prisoners were kept awake, beaten, shouted at, deprived of food and water and toilet facilities and electric shocks were applied …

111 He describes how, during this particular session, the base came under rocket attack. The interrogators reacted by severely assaulting the teacher they were questioning at the time. Deegan joined the assault. Later that night the victim died.

When I was told about his death I was scared and realised that I was a murderer now, but the official lack of response to the incident made me realise that this had happened in reality before; no charges were brought against us and no official inquiry was ever held. Since then even to this day I have vivid nightmares about this man.

Killings, intimidation and harassment of civilians

112 Civilians were routinely harassed, intimidated and beaten by security forces in the operational areas, especially by Koevoet members in pursuit of SWAPO guerrillas. Many were killed during such operations, either by accident (caught in crossfire) or deliberately. Sometimes the human rights abuses involved detention under emergency proclamations, although it was often difficult to determine when the emergency regulations had been invoked as, under their provisions, any member of the security forces could summarily detain any South West African. Often intimidation happened as a result of a belief that the local population were assisting guerrillas and knew their whereabouts, although retribution against suspected SWAPO supporters was also a factor. The Bishops’ Conference reported in 1982:

The Security Forces stop at nothing to force information out of people. They break into homes, beat up residents, shoot people, steal and kill cattle and often pillage stores and tea rooms. When the tracks of SWAPO guerrillas are discovered by the Security Forces, the local people are in danger. Harsh measures are intensified. People are blindfolded, taken from their homes and left beaten up and even dead by the roadside. Women are often raped …There is no redress because reporting irregularities or atrocities to commanders is considered a dangerous or fruitless exercise4.

113 In 1986, in a ‘false-flag operation’ that went awry, a group of recces placed a bomb near a bank in Oshakati and detonated it. The intention was to make it look like a SWAPO operation in order to justify harsh measures planned against the organisation. The bomb killed one bank employee who turned out to be the wife of a member of the same Recce detachment. No charges were ever brought.

114 During 1973, following mass detentions in Owamboland, the SAP began to hand over alleged SWAPO supporters to the bantustan authorities. After cursory hearings, the victims were publicly flogged with epokolos, the central ribs of makalani palms. Both women and men were subjected to these ‘traditional’ punishments, which resulted in extensive cuts and bruising, as well as public humiliation.

115 Dusk-to-dawn curfews were imposed on much of northern South West Africa for most of the duration of the war, although the application varied from time to time and from place to place. This was a major grievance of the local population, as the curfews caused considerable disruption of day-to-day life, and also gave rise to many killings and assaults as troops and police tried to enforce restrictions on movement. In some areas, security force members were under orders to shoot on sight during curfew hours, and there are many reported incidents of civilians being shot while going to the toilet, seeking medical attention or looking for livestock after dark.

Special Operations K Unit of the South African security police (Koevoet)

The crowbar which prises terrorists out of the bushveld like nails from rotten wood. (Minister of Law and Order, Louis le Grange.)

We were basically automatons. We would just kill. That’s how we got our kicks. We were adrenaline junkies. (John Deegan, 1997.)

116 The police unit Koevoet, as noted above, was responsible for many human rights abuses in South West Africa. The unit was set up by Brigadier Hans Dreyer of the SAP Security Branch in June 1979. While its officers were mainly white South African policemen, the unit recruited mostly from the local black South West African population and eventually numbered about 1 000. Cast in the mould of the Portuguese Flechas and Rhodesian Selous Scouts, Koevoet was established as a mobile unit, using specially designed Casspirs (armoured personnel carriers) to gather intelligence, track guerrillas and then kill them.

117 Koevoet was established as a consequence of a failed attempt to create a South West African surrogate force along the lines of RENAMO (the Mozambique National Resistance). This project, known as Operation Vanguard, involved the training of local Owambos at Fort Doppies in the Caprivi but failed when it could make no inroads in Owamboland because of overwhelming local support for SWAPO. Once Vanguard was abandoned, a small group of locals, many ‘turned’ ex-SWAPO fighters (known locally as makakunyanas which means literally blood suckers) and former members of the Front for the National Liberation of Angola (FNLA), were selected and trained as the nucleus of Koevoet. Over time, other groups were established, each made up of ten to fifteen makakunyanas under a white officer with considerable counter-insurgency experience. At its largest, Koevoet comprised approximately 250 white officers and 750–800 Owambos.

118 Speaking in Parliament in Cape Town in 1984, in the only public debate ever permitted on the activities of this unit, Minister le Grange explained the reasons for its formation:

As ... the ordinary conventional methods of warfare appeared to be ineffective in combating terrorism in Owambo and the rest of South West Africa, it was decided after consultation between the SADF ... and the SAP to form a special unit to gather information and make it possible for the security police to track down and wipe out terrorist gangs ... [W]ith the passage of time it became clear that the initial basis on which the unit had come into existence, and according to which it would transmit all information it obtained about terrorist movements to the combat units of the security forces while the latter would carry out the pursuit operations, gave rise to problems in practice ... the unit in due course began to operate as a combat unit5.

119 Koevoet was in many respects an archetypal counter-revolutionary unit, a means of fighting fire with fire. Its top echelon comprised battle-hardened veterans of the Rhodesian war. Amongst these were Dreyer, Colonels Eugene de Kock and Eric Winter, Captains Sakkie van Zyl and ‘Beachball’ Vorster, Lieutenant Frans Conradie and Warrant Officer ‘Snakes’ Greyling. Koevoet soon gained a reputation for brutality, largely because of its methods of interrogating local people, which invariably involved torture, and for the way its members careered around the operational areas in Casspirs, laying down heavy fire, flattening fences, driving straight through fields of crops, and even people’s homes, whenever they suspected a guerrilla contact.

120 In his amnesty application, Lance Corporal Sean Callaghan [AM4026/96] described his experiences on attachment to Koevoet:

A Koevoet team spent a week in the bush and a week back in camp. I think I was in a contact every week. There was a scoreboard and a map in the operations room in the Koevoet base and on the weeks that we were not in the bush, we were checking the scores of the teams that were in the bush. Koevoet was much more effective than SADF units because of its bounty policy.

121 Koevoet’s operational mode involved monetary rewards for killings, captures and the discovery of arms on a graduated scale which rated and rewarded killings most highly. Corpses were also used for purposes of spreading terror and intimidating villagers. Callaghan described another incident:

I can remember … loading bodies onto and off Casspirs. After a contact bodies were tied onto spare tyres, bumpers, mudguards and were left there until we got back to the base camp, until they could be unloaded. This could be days of driving through thick bush, and the skin could be worn right off the bodies.

122 Space does not permit a detailed description of the violence and torture used by Koevoet. However, the Kairos report contains extensive documentation on physical beatings, the destruction of property, sexual assault and various forms of torture (such as solitary confinement, hooding, electric shock, submersion in water, mock burials, mock executions, roasting over fire, and sleep, food and water deprivation) as a means of coercion, intimidation and the extraction of information. Rape was common, and women and girls of all ages were victims. In the Kairos documentation is an account of the rape of an eighty-year-old woman by a Koevoet member, and one referring to the rape of a four-year-old girl.

The case of Jona Hamukwaya

Mr Jona Hamukwaya was a thirty-year-old teacher in western Kavango province when he was detained by Koevoet on 18 November 1982. He died the same day.

Hamukwaya was arrested by a Koevoet team headed by Sergeant Norman Abrahams. Abrahams claimed that Hamukwaya was initially taken to a river close to the school from which he had been taken, and briefly interrogated.

He was then taken to a police station at Nkurenkuru where, according to Koevoet members, he slipped on the top step of a seven-step stairway and hit the concrete floor with his head. Though he appeared initially to be unhurt, some twenty minutes later, according to Abrahams, he started "making gurgling sounds … I tried to give him chest massage. I was under the impression that he was having a heart attack … then I found that he had already expired."6

Hamukwaya’s wife and other villagers gave a different version. They claimed that, while washing clothes in the river, they heard sounds of beating and screams which Mrs Hamukwaya recognised as coming from her husband. When she and others tried to go to his assistance, they were prevented from doing so by three Koevoet members.

Hamukwaya was a prominent community member and his case received considerable prominence and was taken up by a number of groups. The services of pathologists from Groote Schuur hospital in Cape Town were acquired and the Catholic Archbishop of Durban, Denis Hurley, criticised Koevoet for atrocities committed against the civilian population of northern South West Africa, specifically citing this case. For this he was charged with the offence of "falsely accusing the police".

At the inquest, the pathologists testified that Hamukwaya’s injuries "were incompatible with the story of the fall"7 and that he had been "subjected to massive trauma on his back, probably inflicted by a blunt instrument"8. This evidence was accepted by the magistrate who ruled that the death "was caused by "an act or omission that must be seen as a crime on the part of members of the unit known as Koevoet".

Despite this ruling, no charges were ever brought. Ms Hamukwaya sued the security forces for compensation and was paid R58 000 in an out-of-court settlement. The charges against Archbishop Hurley were dropped.

123 In operational terms, Koevoet was a highly effective unit. It is said to have achieved a killing ratio of some one to twenty-five. According to an article by Mr Helmoed Heitman in Armed Forces (December/January 1984), in its first year of operations Koevoet lost twenty-three members and killed 511 "insurgents" – a killing ratio of one to forty-two.

124 A document supplied to the Commission by a one-time Koevoet member gives details of 1 666 "contacts" over a ten-year period by some 250 white former officers and is positive proof that the bounty system encouraged the killing of opponents and discouraged the taking of prisoners. Of these Koevoet members, fourteen were involved in more than one hundred contacts. One member, Warrant Officer L Kilino, notched up 221 contacts in which 346 people were killed and only twenty-three captured. For an unknown reason, this document did not include Eugene de Kock, who put his number of contacts at about 400.9 He gives no details of his killing rate but popular legend has it that it was the highest of all Koevoet operatives.

125 In toto, these fourteen officers were involved in 1 754 contacts in which 3 323 individuals were killed (an average of nearly two per contact) and only 104 prisoners were taken. The ratio of prisoners to fatalities was thus in the region of 1:32. Heitman describes as Koevoet’s "most successful single contact" an encounter in which "34 out of 34 insurgents"10 were killed.

THE COMMISSION FINDS THAT THE KOEVOET UNIT WAS RESPONSIBLE FOR THE PERPETRATION OF GROSS HUMAN RIGHTS VIOLATIONS IN SOUTH WEST AFRICA AND ANGOLA. THESE VIOLATIONS AMOUNTED TO A SYSTEMATIC PATTERN OF ABUSE WHICH ENTAILED DELIBERATE PLANNING BY THE LEADERSHIP OF THE SAP. THE COMMISSION FINDS THE SOUTH AFRICAN GOVERNMENT, THE SAP AND THE MINISTER OF LAW AND ORDER ACCOUNTABLE. THE COMMISSION FINDS FURTHER THAT THE BOUNTY POLICY OF THE SAP, BY WHICH MEMBERS OF KOEVOET WERE MONETARILY REWARDED FOR CERTAIN OF THEIR ACTIONS, SERVED AS A POSITIVE INDUCEMENT FOR THE COMMISSION OF GROSS VIOLATIONS OF HUMAN RIGHTS, INCLUDING KILLING.

Pre-election events in 1989 and the assassination of Anton Lubowski

126 The New York Accords signed on 22 December 1988 set in motion the implementation of UN Resolution 435 adopted ten years earlier. Actual implementation by a United Nations Transitional Assistance Group (UNTAG) was set to commence on 1 April 1989, culminating in an election seven months later. The agreement also provided for a cease-fire, a phased withdrawal of Cuban and South African troops from Angola and South West Africa/Namibia respectively (to be completed only after the election), the dissolution of Koevoet and the deployment of both SWAPO guerrillas and South African forces to designated assembly points or bases.

127 With its acceptance of 435 and the end of the armed phase of the conflict, the South African government reverted to other methods – developing a multi-faceted effort to weaken and damage SWAPO and disrupt its electoral campaign. This took a number of forms.

Breach of the cease-fire agreement

128 The first of these was the breach of the cease-fire agreement. On 1 April, UN forces moved into South West Africa/Namibia to oversee the transition process. From that day on, both SWAPO and the South African forces were expected to abide by the cease-fire and be confined to base. Instead, the launch of this new era was marred by a series of military battles across a 320-kilometre front in Owamboland. The fighting was prompted by what the South African government referred to as a large-scale SWAPO incursion from Angola. The new UN administration conceded to pressure from British Prime Minister Margaret Thatcher and South African Foreign Minister Pik Botha and permitted the redeployment of Koevoet and other military units into northern South West Africa/Namibia.

129 SWAPO denied the allegations of an incursion and claimed its guerrillas had crossed the border to link up with UNTAG elements who, in turn, were to deploy them to the bases in terms of the New York agreement.

130 Whatever the cause of the fighting, the consequence was the death of more than 300 South West Africans. While the South African authorities argued these were all SWAPO combatants, local residents claimed that some of the dead were civilians. There is prima facie evidence that some of the dead may have been summarily executed as many of the victims had single bullet holes to the back of the head. It is possible that some or many of these were SWAPO prisoners shot on the day they were scheduled for release. This suspicion is expressed in the statement of Sean Callaghan:

One of the real questions I have in my mind is what happened to these people when Koevoet pulled out. My suspicion is that they were assassinated … I further suspect that when Koevoet went on their last killing rampage, breaking the UN cease-fire in April 1989, that these prisoners were killed at the same time. The reason I think this is because it was their last opportunity to get rid of them. I can’t believe they let them all go free.

131 The Commission has not been able to verify this suspicion.

Operations Heyday and Victor

132 A second dimension to the South African government’s manipulation of the election took the form of covert disinformation campaigns. Secret funds amounting to R185,5 million were made available and used both to promote the electoral chances of the Democratic Turnhalle Alliance (DTA) and to damage those of SWAPO. This was a multi-departmental effort involving the SADF, security police, National Intelligence Agency, Department of Foreign Affairs and Department of Information.

133 Operation Heyday was the SADF’s contribution to the campaign, and was allocated R125 million – 70 per cent of the total. Run by Brigadier Ferdi van Wyk, it involved gathering intelligence on SWAPO members and supporters for use in disinformation campaigns, intimidating SWAPO’s supporters, disrupting SWAPO’s election meetings and so on.

134 Another component of the SADF’s contribution was a sophisticated media centre run by Major Nico Basson of the South African Army Troop Information Unit. Set up on the authority of the then chiefs of the SADF and army, Generals Geldenhuys and Liebenberg, it operated in civilian guise from an up-market Windhoek hotel. Basson’s Africa Communications Project became the first point of reference for the foreign media and UNTAG officials.

135 Operation Victor was the security police’s contribution, to which R36,5 million was allocated. Run by Brigadier Hein Oliver, it involved the setting up of two front companies through which vehicles and vast amounts of office equipment were purchased. In order to make it appear as if one of these companies, the Namib Foundation, was supported by public contributions, a member of the security police was sent from Pretoria to Durban depositing amounts ranging from R1 to R10 000 in every bank en route.

136 A large proportion of the 450 vehicles purchased for the campaign were used by ex-Koevoet members to transport residents of Owamboland to the election meetings of parties opposed to SWAPO.

Aksie Kontra 435

137 There is also evidence that these funds were used to mobilise the white right wing, which formed an organisation called Aksie Kontra 435. One of its members, Mr Horst Klenz, applied for amnesty [AM0316/96] for his involvement, with two others, (Mr Leonard Veneendal and Mr Darryl Stopforth) in a grenade attack on an UNTAG regional office at Outjo in northern South West Africa during which a security guard was killed. Arrested soon afterwards, the three escaped back to South Africa when the police van in which they were travelling was ambushed by two white men (known to the Commission only by the pseudonyms of ‘Archer’ and ‘Barker’) who killed one of the police escorts11.

138 All five were arrested in South Africa and held incommunicado under section 29 of the Terrorism Act. No charges were ever laid and they were all eventually released. During their detention the five were never interviewed by South West African authorities.

CCB covert operations

139 Another dimension involved the deployment of CCB members to South West Africa. Testimony in regard to the CCB operation was given to the Commission by the CCB’s intelligence head, Colonel Christoffel Nel. He told the Commission:

Prior to the election in Namibia all the regions [of the CCB] were told to do something there. No matter where you worked. And this was a recipe for disaster. Because people who used to work in Europe now had all of a sudden to do covert work in Namibia. Where it normally takes about five years to get a covert system set up, these guys had to do this overnight. And it was not surprising when a kitbag full of limpet mines was found in northern Namibia with a Special Forces’ golf membership card in it. It was not surprising to see [people] roaming the blocks around Anton Lubowski’s house and still the confusion today about who shot him. Because it could have been anybody from any of these other regions. I was in region one [Botswana] and region four [Angola, Zambia and Tanzania] primarily, we were doing a job in northern Namibia. We had no interest, we had no knowledge of the area but we had to do it. Because we were told double up your production and you will get a production bonus.’ (Section 29 hearing, 18 May 1998.)

140 There is evidence in the Commission’s possession that the task of attacking the UNTAG facility was assigned to a senior CCB member. Captain Pieter Botes. whose area of CCB responsibility was Swaziland and Mozambique. He was assigned five operatives for the mission. As noted above, five detainees were held in South Africa in connection with the UNTAG case.

The killing of Anton Lubowski

141 On 12 September 1989, Advocate Anton Lubowski was shot dead outside his home in Windhoek. At the time, he was the secretary general of SWAPO and the highest-ranking white person in the organisation. One human rights violation submission and two amnesty applications were made to the Commission on this case. The human rights violation submission was made by Ms Molly Lubowski, the deceased’s mother. She appealed to the Commission to identify her son’s killers and to clear him of allegations that he was a South African MI agent.

142 Considerable attention was given to this case, including a trip to Namibia and meetings with the judicial authorities there. A vast amount of documentation was supplied to the Commission by various parties.

143 Neither of the amnesty applications – by Mr Derrick Nielsen [AM 4792/97] and Mr Horst Klenz [AM0316/96] – provided any evidence of substance. Nielsen originally applied for amnesty for the murders of both Lubowski and David Webster but supplied no details. Later he sent letters to the Commission. In one of these, dated 4 December 1996, he alleges that he supplied an AK-47 to Mr Ferdie Barnard for "a hit" and that three days later Barnard "bragged that they had got rid of a kafferboetie". He said that the name Lubowski was mentioned. The Commission paid several visits to Pollsmoor prison where Nielsen was serving a sentence for a traffic offence in 1997, but he refused to discuss his application and divulged no further details. His main interest seemed to be to bargain information for a speedy release. Given these facts and the developments pertaining to the Webster murder, the Commission is of the view that little credence can be given to this application.

144 Horst Klenz’s application contained only some hearsay information to the effect that SWAPO had killed Lubowski but contained no corroborating evidence.

145 In an amnesty application [AM1909/96] not directly related to this murder, Mr Kevin Trytsman, an associate of Ferdi Barnard, claimed that Barnard had told him that the CCB had committed the murder. This is also the view of Christoffel Nel as expressed in the quote cited earlier. Elsewhere in his hearing, Nel described the Lubowski murder, along with the killings of Ms Dulcie September and Mr David Webster, as one of the CCB’s "successes".

146 This was also the conclusion of Judge J Levy of the Namibian Supreme Court, who conducted a lengthy inquest into the case. In a 144-page judgement, Levy named Irish mercenary Donald Acheson as the assassin and, as accomplices, CCB members Joe Verster, Staal Burger, Abraham ‘Slang’ van Zyl, Calla Botha, Leon ‘Chappies’ Maree, Johan Niemoller jr, Captain Wouter Basson (aka Christo Britz), Ferdi Barnard, and Charles Wildschudt (formerly Neelse).

THE COMMISSION BELIEVES THERE ARE NO GROUNDS TO CONTRADICT JUDGE LEVY’S GENERAL FINDING PERTAINING TO THE INVOLVEMENT OF THE CCB AND ITS RESPONSIBILITY FOR THE CONSPIRACY THAT LED TO THIS KILLING.

147 The evidence of the CCB’s intelligence head, Christoffel Nel, is regarded by the Commission as important corroborating evidence of the CCB’s role.

148 It is well established that the CCB was set up as a covert grouping with the purpose, among others, of killing political opponents. So, too, is the fact that prior to Lubowski’s killing, such opponents were subject to extensive ‘target identification’ or intelligence-gathering operations. It has been established that Lubowski was the subject of such an exercise, an operation conducted by the West Front (South West Africa and Angola) section of the Directorate of Covert Collection (DCC) then headed by Brigadier Koos Louw. His deputy, Major Geoffrey Burton Price (aka Arthur Wilshire), was in charge of the South West African arena.

149 The actual eavesdropping operation was undertaken by Lieutenant Johan Frederich Verster, who provided this information to the Commission12. Verster was still running the surveillance/eavesdropping exercise at the time Lubowski was shot. He told the Commission that he was shocked by the killing "because we were busy recruiting him and it was the wrong person ever to have shot". He was instructed to return to South Africa immediately after the shooting.

When I got into Pretoria I was summonsed to go and see Koos Louw … and also Tolletjie Botha [head of DCC]. I explained to them what had happened … They went to an office with me … he [Botha] picked up the telephone and he phoned South West Africa, the head of the Prison Department and also the Police and he said we’ll have to use the old boys network … they phoned up the Brigadier in South West Africa and asked what happened, have they got Acheson? They said yes, they have …

150 This would seem to lend credence to the view expressed by Judge Levy that Acheson was the assassin. The Commission had some reservations about this view. Its investigation suggested that, although Acheson may have been the intended killer and was certainly in the vehicle used for the operation (probably the driver as he had hired the vehicle, a red Toyota Conquest), the fatal shots may have been fired by a passenger, who was probably his CCB handler for this operation. The Commission has been given the handler’s name but lacks the corroborative evidence to name him as the killer.

151 In a statement to the police which is in the possession of the Commission, Acheson denied that he killed Lubowski. He confirmed that he had been recruited by Chappies Maree and Ferdi Barnard into an organisation which he later learnt was the CCB, and deployed to South West Africa, where he said that one of his tasks was to kill Ms Gwen Lister, the editor of The Namibian newspaper, regarded by South African security as a pro-SWAPO organ. He was to use "slow acting poison which could be injected into her toothpaste, placed on her Tampax or put into anything she would eat or drink … she would keel over in forty-eight hours". Although he tried to fulfil his assignment, he said he was unable to access her household. After his release and return to South Africa, Acheson said he was paid R20 000 and US$ 4 000 and sent to a hideout in the Greek islands where he remained for several months.

152 The allegation that Lubowski was a paid informer for MI was first voiced by General Magnus Malan in 1990 – in Parliament, where he enjoyed the protection of parliamentary privilege. Appearing before the Commission in 1997, he repeated the statement. One of Malan’s sources for this allegation may have been Brigadier Koos Louw, the head of the West Front of DCC. When Louw was interviewed by investigators from the Commission, he claimed to be Lubowski’s handler. He refused, however, to divulge the names of any other handlers and stated that all the documents pertaining to Lubowski had been destroyed.

153 Louw also stated that he paid a sum of R100 000 into Lubowski’s account in June 1989 through Global Investments, an MI front company. Despite the alleged destruction of the Lubowski documents, Louw was able to produce the original microfiche of the bank clearing house that processed the payment. Further, Lubowski never used this money. This would seem strange given the claim put out by MI that he was in financial difficulties, a factor which they claim was used to recruit him.

154 The allegation is not universally accepted, even by some MI members themselves. Verster, who worked in Louw’s DCC section, claims that the documentation was forged:

Signatures were made by Anton and that was carefully changed by artists inside Military Intelligence to make it look like Anton was an agent of ours. But the reports that I had given in, also the tapes and the conversations that I had listened to, was never (sic) that Anton was an agent. Because why, if Anton was an agent, why would Bleny (nickname for Army Intelligence, GS2), the heart of it, want all the tapes and information about him, and the inside stuff.

155 Verster’s doubts were echoed by Major Nico Basson, who was running a disinformation campaign at the time of the killing. Basson’s view is based on the fact that Lubowski had rejected his offer of R250 000 to leave SWAPO and join the DTA. Operative Clive Brink, who operated in South West Africa in the pre-election period, also expressed scepticism and stated that the paying of money into bank accounts to compromise recipients at a later date was a not unusual intelligence practice.

156 This latter view is shared by Mr Julian Knight, the Lubowski family lawyer, who argued that the payment was made either as an anticipatory cover-up, a pre-arranged alibi for the planned assassination, or fraudulently as part of a post-killing cover-up.

157 Finally, the Commission took note of two factors. The first is the lack of consensus on this issue among those connected to MI structures in South West Africa. The second is a question as to why the agency would have paid a considerable sum of money to someone they were on the point of killing.

THE COMMISSION FINDS THAT THE ALLEGATION THAT MR ANTON LUBOWSKI WAS A PAID INFORMER OF SOUTH AFRICAN MILITARY INTELLIGENCE IS UNPROVEN AND THAT HE IS CLEARED OF THE ALLEGATION.

THE COMMISSION FURTHER FINDS THAT THE ACTIVITIES OF THE SADF AND THE SAP IN THE FORMER SOUTH WEST AFRICA BETWEEN 1966 AND 1989 LED TO GROSS VIOLATIONS OF HUMAN RIGHTS ON A VAST SCALE. THE COMMISSION FINDS THAT SUCH ACTIVITIES CONSTITUTED A SYSTEMATIC PATTERN OF ABUSE WHICH ENTAILED DELIBERATE PLANNING ON THE PART OF THE FORMER CABINET, THE SSC AND THE LEADERSHIP OF THE SAP AND SADF. THE COMMISSION FINDS THAT THESE INSTITUTIONS AND THEIR MEMBERS ARE ACCOUNTABLE FOR THE AFORESAID GROSS VIOLATIONS OF HUMAN RIGHTS.

· POLICE AND MILITARY COUNTER-INSURGENCY OPERATIONS IN ZAMBIA

158 Operation Plathond involved training by the SADF of a surrogate force for operations in Zambia in the early 1970s. At the Commonwealth heads of state meeting in 1981, the Zambian Foreign Minister claimed that between 500 and 600 Zambians were being trained in the Caprivi for operations inside Zambia. There is no evidence to support the claim. While there certainly were foreign nationals being trained in the Caprivi for military operations in the region, they are more likely to have been Mozambican members of RENAMO than large numbers of Zambian dissidents.

159 Zambia was a target of South African aggression. Between 1978 and 1980, the SADF undertook several conventional military operations inside Zambia, aimed largely at SWAPO installations in the Western Province. The first occurred after a SWAPO rocket and mortar attack on the SADF headquarters base at Katimo Mulilo in the eastern Caprivi in which ten servicemen were killed. The SADF responded by sending combat units 250 kilometres inside Zambia to attack SWAPO camps.

160 During the SADF’s Operations Saffran and Rekstok in 1979, SWAPO bases in Western Zambia were again attacked. The recurring nature of these attacks in 1979/80, and the civilian casualties they caused, as well as the disruption to rural life (burning of crops, poisoning of local water supplies, killing of cattle, mining of roads) led SWAPO to abandon its camps adjacent to the Caprivi and move further north. However, with Western Zambia under virtual occupation by two SADF battalions for nine months after Zimbabwe’s independence in April 1980, and with malnutrition and starvation rife, the Zambian government eventually banned SWAPO from operating any military bases in the country – an early success for the strategy of counter-revolutionary warfare. What was also significant, and a lesson the SADF noted, was that this success owed more to the disruption to civilian life than to the damage inflicted on the military capacity of the ‘enemy’.

161 With SWAPO now concentrating its military facilities in Angola, large-scale SADF operations inside Zambia largely ceased. One exception was the SAAF raid near Lusaka in May 1986.

162 Further information supplied to the Commission by the NIA revealed that, as of June 1985, four people were being held in Zambian prisons for spying for South Africa. Two were current members of the SADF, one a former SADF member and the fourth a Caprivian working for South African MI. One of the SADF members was Sergeant Isaiah Moyo, a former member of the Rhodesian African Rifles, who had joined the SADF after 1980 and who was placed in Zambia by MI in about 1984/5. He was sentenced to twenty-five years for spying but released in 1991.

 

n POLICE AND MILITARY COUNTER-INSURGENCY OPERATIONS IN PRE-INDEPENDENCE RHODESIA AND IN ZIMBABWE

163 From the time of the unilateral declaration of independence in Rhodesia in November 1965, the security situation in that country was a major concern of the South African government. With the launch of joint ANC/ZAPU13 military operations in areas of north-west Rhodesia in August 1967, South African police units were deployed inside Rhodesia where they stayed for the next eight years. As a gesture of support for the 1975 Kissinger diplomatic initiative over Rhodesia, South Africa withdrew its police units but left behind all its equipment, which included helicopters, Dakotas, small arms and ammunition. In addition, the South African government met the costs of 50 per cent of the Rhodesian defence budget for 1975–76. This was followed by Operation Polo, a secret agreement in terms of which the SADF assisted in the construction of five new military airfields in Rhodesia.

164 By 1978 the SADF was supplying sophisticated Mirage III fighters and Impala strike planes, as well as Alouette and ‘Huey’ helicopters. It was also secretly deploying troops into southern Rhodesia from bases inside South Africa and sending conscripts to Rhodesia to fight in local uniforms as ‘members’ of Rhodesian army units.

165 Colonel Craig Williamson told the Commission (3 February 1998) that the South African Security Branch also funded out of its secret account the police counter-insurgency unit, the Selous Scouts, in which numerous SAP members also served.

166 At its meeting on 26 March 1979, the SSC approved both the setting-up of a Rhodesian Joint Management Centre (JMC) to operate from the South African diplomatic mission in Salisbury, as well as a short-term strategy for Rhodesia. This recommended, inter alia, clandestine support (logistic as well as special forces) for the Rhodesian security forces. In July 1979, the SSC approved a stepping-up of military assistance, including covert air support for offensive measures against ‘terrorist’ and other targets in their host states ("gasheerlande"); unspecified military support with electronic warfare; aerial reconnaissance and support of special operations undertaken by the Rhodesian forces.

167 Six weeks later, at an SSC meeting on 27 August 1979, General Malan reported that the situation in Rhodesia had reached a watershed and that it needed further military help. As a result, the SSC authorised special clandestine actions, ordering that these be mounted within the context of a co-ordinated strategy. To this end, it established a Mozambican JMC comprised of representatives of the SADF, the Department of Foreign Affairs (DFA), South African Railways (SAR), the National Intelligence Service (NIS), the South African Broadcasting Corporation (SABC), the SAP and the Departments of Finance, Trade and Industry and the Information Service. The presence of the SABC should be noted.

168 Special Forces operative Johan Verster told the Commission that, in 1979, he participated in parabat attacks on guerrillas moving into the cease-fire assembly points in the Tshipise Tribal Trust area. His group operated from a camp "on the side of the river at Gumbu Mine" in Botswana. The attacks, Verster claimed, were ordered by "military headquarters" using intelligence provided by the Selous Scouts.

169 These attacks may have been prompted by the fact that ANC/MK guerrillas were infiltrating Zimbabwe along with returning ZIPRA14 fighters. A list of ANC members who died in exile was supplied to the Commission and includes the names of fourteen "comrades killed in Rhodesia in 1979". It is possible they were victims of these attacks.

170 Most of the ANC infiltrators were eventually returned to Zambia by the new government, but it was largely in response to this MK inflow that the SADF moved a unit of its troops through the Beit Bridge border post towards the end of 1979. According to Mr Pik Botha’s statement to the Commission, the movement of troops across the bridge was done with the concurrence of the Muzorewa Government.

171 In the run-up to the March 1980 pre-independence election, Rhodesia remained at the top of the SSC agenda. Excerpts from the minutes of the SSC meeting of 28 January 1980 provide an insight into the state’s strategic thinking at the time. General Malan asked what would be done if Rhodesia "verkeerd gaan" (goes wrong) and argued that Rhodesia and South West Africa were key to South Africa’s defence. Arguing for a proactive defence strategy, he asserted that the country’s first line of defence had to be beyond the Republic – "ons moet die tyd en die plek kan kies" (we must be able to choose the time and place). Mr PW Botha assured the General that the meeting shared his views, arguing that "as ons op die Limpopo en die Oranje veg, kan die vyand ons hartland aanval" (if we fight on the Limpopo and the Orange rivers, the enemy can attack our country).

172 Early in February 1980, the SSC dispatched a special task team to review the situation in Rhodesia. The most significant of its ten recommendations read: "Die implikasie van eliminasie van politieke figure in Rhodesie moet voortdurend onder oë gehou word" (The implication of the elimination of political figures in Rhodesia must be constantly kept in mind.) There is an ambiguity about this statement in that the reference could be to other countries’ attempts to assassinate Mr Robert Mugabe. This was the interpretation that both Mr Pik Botha and Dr Niel Barnard gave in their appearances before the Commission and there is no evidence to suggest that South African security forces ever attempted to assassinate Mugabe in the period prior to the election.

173 The South African government raised in excess of R12 million in support of Bishop Muzorewa’s United African National Council (UANC) in the March 1980 election, approximately half of which came from state coffers, while the rest was raised from the private sector by Foreign Minister Pik Botha. At independence in April 1980, the government of Zimbabwe inherited a total debt over R4 000 million which South Africa was to insist be repaid. Moreover, the long tradition of direct South African involvement in the country’s security affairs did not end at independence; it merely changed its form.

Surrogate-force insurgency operations

174 The SADF’s surrogate-force operations in the 1980s fell under the Directorate of Special Tasks (DST) in the office of the chief of staff intelligence (CSI). (These bodies are discussed in more detail elsewhere in this volume).

175 The DST had its origins in the 1976 decision to channel assistance to UNITA (Operation Silwer) and a special office was set up in Rundu headed by Colonel (later Major General) Marius Oelschig. In the early 1980s, DST was set up and located as a secret project in Pretoria. Its first head was Colonel (later Brigadier) ‘Cor’ van Niekerk, who was also responsible for managing the RENAMO project in the early 1980s. By the mid-1980s, DST had been incorporated into the Operational Intelligence Directorate headed by General Niels van Tonder. In the mid-1980s, DST developed an internal dimension in the form of Operations Marion (assistance to Inkatha) and Katzen (targeted at the Transkei and Ciskei).

Zimbabwe: Operation Drama

176 The outcome of the independence election was not quite the worst-case scenario feared by South Africa. That would have been a ZAPU victory. Nonetheless, the failure of Muzorewa’s UANC to secure a place in the ZANU/ZAPU coalition was a setback. Its initial public response was diplomatically correct; its covert response was counter-revolutionary. At its first post-election meeting on 10 March 1980, the SSC declared Messina an "SADF operational area". This was in order to give the SADF "meer beweergruimte" (more room to manoeuvre) to facilitate the clandestine transfer of RENAMO to South Africa which, according to the SANDF’s second submission to the Commission, began in March 1980.

177 The deployment of RENAMO was part of a much larger exercise involving the transfer to South Africa of various parts of Rhodesia’s pre-independence security apparatus. This included several hundred black members of Bishop Muzorewa’s Security Force Auxiliaries who were deployed to a farm near Pretoria. Simultaneously, the SADF launched Operation Winter to recruit mainly white members of Rhodesia’s various counter-insurgency units. The operation was directed by Major General FW Loots, then general officer commanding of Special Forces, who personally travelled to Rhodesia in the last days of the Smith regime to screen potential recruits.

178 In all, it is estimated that about 5 000 Rhodesian military personnel were recruited into the SADF in this period. Apart from skilled counter-insurgency specialists, other security personnel who joined this southern exodus at independence or soon afterwards included some Special Branch police officers and intelligence personnel from the Central Intelligence Organisation (CIO). Amongst these was Mr Gray Branfield, who was assigned to Daisy farm adjacent to Vlakplaas, to run a Zimbabwe Special Operations Unit. Branfield ran a string of agents inside Zimbabwe, the most important of whom were Mr Christopher ‘Kit’ Bawden, his cousin Mr Barry Bawden, and Mr Michael Smith.

179 Other security and intelligence personnel who moved south and were integrated into MI were Mr Pat Keyser, Mr Eric May, Mr Bob Wishart, Mr Peter Stanton and former Selous Scout Peter Grant. They were integrated either into Special Forces, DST, which ran the surrogate forces, or the DCC. Stanton later became a member of the CCB.

180 Their departure notwithstanding, a fifth column of South African agents remained intact inside Zimbabwe, strategically located within the military, the police and the CIO. Possibly the most sensitive of these was CIO operative Geoffrey Burton Price, retained by President Mugabe as his head of close security after independence. Others who have been named as agents working from inside in the immediate post-independence period are CIO members Colin Evans and Philip Hartlebury, and security police officials Alan Trowsdale, Alec West and the CIO head in Bulawayo, Matthew ‘Matt’ Calloway. Another CIO member who admits to having assisted some of these operatives was Mr Kevin Woods.

181 With the above infrastructure in place and large numbers of ex-Rhodesian soldiers in camps in the northern Transvaal, the SADF was well placed to launch Operation Drama – a militarily-driven project aimed at destabilising the new independent government of Zimbabwe. Its objective was, inter alia, to ensure that the government did not provide concrete support to the ANC and PAC in their armed struggles. To this end, it recruited and trained Zimbabweans, primarily for sabotage operations designed to destroy infrastructure, damage the economy and undermine the military capacity of Zimbabwe’s armed forces.

182 In a statement to the Commission, Lieutenant Kenneth Gwenzi, who j