About this site

This resource is hosted by the Nelson Mandela Centre of Memory, but was compiled and authored by Padraig O’Malley. It is the product of almost two decades of research and includes analyses, chronologies, historical documents, and interviews from the apartheid and post-apartheid eras.

Chapter 7: Causes, Motives and Perspectives of Perpetrators1

INTRODUCTION

1. The Act governing the work of the Commission requires not only that it establish as complete a picture as possible of activities during the years falling under its mandate and that it identify perpetrators of gross human rights violations, but also that it establish the "antecedent circumstances, factors and context of [gross human rights] violations as well as the causes, motives and perspectives of the persons responsible".2

2. Who were the perpetrators and what 'made' them, moved them or contributed to their behaviour? It is essential to examine perpetrators as multi-dimensional and rounded individuals rather than simply characterising them as purveyors of horrendous acts. Building on the factual history presented in earlier chapters, and utilising research developed in relation to the Holocaust and other situations of sustained oppression that gave rise to systematic acts of gross human rights violations, this chapter attempts to explain why and how these violations transpired, as a basis for informing the debate concerning reconciliation and recommendations to prevent violations in future. A diagnosis of persons and institutions responsible for gross human rights violations is of paramount importance in seeking to prevent future gross human rights violations.

3. It could be argued that prevention is only effective in the early stages of the development of a culture of gross human rights violations, and that signs and symptoms of the 'syndrome' should be made known widely. Itzhak Fried3 has suggested that:

Individuals in most societies know that a constellation of high fever and coughing may indicate pneumonia. In the same sense, people might become aware that symptoms of an emerging obsessive ideology, hyper-arousal, diminished affective reactivity, and group dependent aggression, directed at members of other sub-groups may signify a situation which needs immediate political, social, and social scientific attention.

4. The tale to be told in this chapter is incomplete and in some respects premature for, although the Commission heard evidence over two and a half years, the bulk of this material came from victims. The work of the Amnesty Committee, which involved hearing testimony from alleged perpetrators, was incomplete at the time of reporting and was scheduled to continue for approximately another twelve months until July 1999. Hence, a complete picture would emerge only when that task was done and the testimonies thoroughly studied to reveal the full pattern of motives and perspectives of all perpetrators. Even then the tale will be incomplete, since not all categories of perpetrators, for instance 'necklace' murder cases, are likely to be fully represented.

5. Further limitations of this chapter need to be clearly stated. The Commission is aware that the quotations in this chapter are not necessarily even-handed extracts from all parties concerned. Quotations are used to illustrate the processes described and the perspectives of various groups. This could result in allegations of bias. However, in some instances, this was unavoidable. Not all groups or parties co-operated equally with the Commission. As a result, comprehensive searches through documents revealed few instances of statements from Inkatha perpetrators, from township 'vigilante groups', from torturers or from African National Congress (ANC)-aligned self-defence units (SDUs). Documentary evidence is thus only partly available. Furthermore, the precise question of motives of perpetrators was often not fully canvassed by amnesty panels, nor by special hearings of the Commission. These shortcomings should be attributed to partial failings of the Commission itself, rather than to systematic bias.

6. As a premature effort, the chapter draws on a wide-ranging literature, from local as well as international sources, in an attempt to understand the position of perpetrators. It would be helpful to understand this chapter as being in the form of an agenda for future research and verification rather than a closed book. Nonetheless, partial understandings may be better than none.

7. The chapter begins with a general perspective on patterns of violence, with examples drawn from the information given to the Commission. This is followed by a discussion on the problem of perspectives on the conflict and its participants. It then explores various possible explanations of causes and motives of perpetrators, giving attention both to the political context of the violations and to individual psychological explanations of perpetrators' actions and the situations in which violations occurred. Consideration is then given to the role of language, the existence of secret societies and silences in the conflict. The information set out above will then be applied to two case examples, exploring the causes, motives and perspectives of torturers and of participants in crowd violence. The chapter concludes with some ideas for the future prevention of atrocities.

A general perspective on patterns of violence

8. A number of general patterns are discernible from the huge body of materials collected by the Commission. A description of these general patterns is essential to an understanding of the particular that is, the acts of individual perpetrators. Acts of violence are in many ways quite different from each other: they range from careful calculated intentional actions to unintentional, unplanned acts that occurred because things 'went wrong'.

Intentional military actions: "We were at war."

9. As is apparent from the testimony of the former head of ANC special operations, Mr Aboobaker Ismail, in the amnesty hearing on the Church Street bombing of the South African Air Force headquarters (in which nineteen people were killed and over 200 injured), many acts were carefully calculated actions of war:

Special operations were set up in 1979 to undertake high-profile acts of sabotage on key economic installations. This structure reported directly to [then ANC president] OR Tambo.

10. Later, in terms of a shift in ANC policy which resulted in the inclusion of military personnel as justifiable targets, the Church Street operation in Pretoria was carried out (with the stated approval of Mr Tambo) on 20 May 1983. The operation was conducted

in the wake of the SADF cross-border raid into Lesotho, killing forty-two ANC supporters and Lesotho civilians, and also in the wake of the assassination of Ms Ruth First in Maputo by the security forces.

In terms of stated ANC policy, military targets including personnel [were] justifiable, even if these entailed limited loss of civilian life.

11. Acts such as these were quite clearly rational, intentional and thoroughly planned (although mistakes did occur; for example, the two ANC operatives in the Church Street bombing were themselves killed in the attack). Often, as indicated above, they occurred in retaliation against state security violence. Following the Church Street bombing, the South African Defence Force (SADF) conducted various attacks, including air raids on Maputo. The sequential nature of such calculated attacks constituted something of a 'dialogue' or a 'spiral of violence'.

12. As a further illustration of calculated attacks, Mr Ismail described the Dolphin Unit, established in 1982, which "had been established inside the country to carry out operations within the broad mandate of the ANC and MK [the ANC's armed wing, Umkhonto weSizwe]". In an amnesty hearing in May 1998, Mr Mohammed Shaik, head of the Dolphin Unit, described thirty-two carefully planned operations against police, embassy, magistrate's court and state department targets within South Africa. Mr Shaik stated in conclusion that:

At all times I acted within the policy and guidelines laid down by the ANC; I was comprehensively briefed on the modus operandi of special operations in MK. I accordingly attempted to avoid or minimise civilian casualties whenever I conducted operations. To this end, whenever circumstances permitted, I timed my operations after hours, when targeted buildings had been vacated by civilians. I accept that, in the end, there was always a possibility of civilian casualties.

Where there were civilian casualties these were never at any stage intended to be targets, but were rather caught in the crossfire. To the extent that there were civilian casualties, I express my deep regret to those who experienced pain and suffering. The apartheid state left us no choice but to take up arms.

13. These rational and calculated acts of violence were justified on each side by the statement that "we were at war". General Andrew Masondo was national political commissar of the ANC between 1977 and 1985, and earlier a Robben Island prisoner. In a section 29 hearing, when responding to enquiries about atrocities, including executions in the Angolan Quatro camp, he repeatedly stated that "we were at war":

You remember I said we were at war There might be times that I will use third degree, in spite of the fact that it is not policy.

People who it was found that they were enemy agents, we executed them, and I wouldn't make an apology. We were at war.

14. General Constand Viljoen, former chief of the SADF, expressed it this way in a public Commission debate on the notion of a 'just war':

The liberation struggle used revolutionary methods to coerce. This was a new kind of total war, not total in its destructiveness but total in its means of applying different ways of coercion: political, psychological, economic, propaganda. It was a new kind of war.

This war, if it could be called a war, is so unique that the traditional 'just war' theory cannot be easily applied.

15. Even the conflict between Inkatha and the ANC was repeatedly described as a war situation. Inkatha Freedom Party (IFP) amnesty applicant Mr Victor Mthembu stated:

If it had not been a war situation between the IFP and the ANC, I would not have participated.

16. In a situation regarded as war, violent actions were undertaken with pride rather than with distress and embarrassment. In this regard former senior security force member Major Craig Williamson said:

The psychological effect of fighting such a counter-revolutionary war should not be underestimated, especially when this entailed long periods of covert operations. The members of the security forces, especially in covert units saw themselves as the elite frontline troops in a critically important theatre of the overall war. Security force successes produced praise, pride and relief from pressure.

17. Even a self-confessed torturer such as Captain Jeffrey Benzien admitted to pride in his work when cross-examined by Mr Tony Yengeni (ANC) whom he had tortured. After saying, in respect of a particular torture method, that "I applied it well and with caution", Captain Benzien went on to make this extraordinary statement:

Mr Yengeni, with my absolutely unorthodox methods and by removing your weaponry from you, I am wholly convinced that I prevented you and your colleagues I may have prevented you from being branded a murderer nowadays.

Denial: the gap between authorities and followers

18. From a range of different quarters, there was denial from senior persons in authority that they knew what was happening, or denial that they gave specific orders, even while their supporters or followers were claiming to have acted under instructions. In other words, there was a gap between the perceptions of leaders and followers. In the second National Party (NP) submission, Mr FW de Klerk said:

but things happened which were not authorised, not intended, or of which we were not aware I have never condoned gross violations of human rights and reject any insinuation that it was ever the policy of my party or government.

19. In contrast, there is the perception of convicted Vlakplaas killer Colonel Eugene de Kock in the closing pages of his book:4

Yet the person who sticks most of all in my throat is former State President FW de Klerk. Not because I can prove, without a shadow of doubt, that he ordered the death of X or cross-border raid Y. Not even because of the holier than thou attitude that is discernible in the evidence he gave before the [Commission] on behalf of the National Party.

It is because, in that evidence, he simply did not have the courage to declare: "yes we at the top levels condoned what was done on our behalf by the security forces. What's more, we instructed that it should be implemented. Or if we did not actually give instructions we turned a blind eye. We didn't move heaven and earth to stop the ghastliness. Therefore let the foot soldiers be excused".

20. From another side of the conflict comes the position of the president of the IFP, Chief Mangosuthu Buthelezi.

On no occasion has the IFP's leadership ever made any decision anywhere at any time to use violence for political purposes. I have always abhorred violence now and will die abhorring violence. I personally have never made any decision to employ violence anywhere for any purpose whatsoever.

21. By contrast, here are extracts from statements by Inkatha members applying for amnesty on grounds of numerous murders:

Mr Wills: Now what was the purpose of this training?

Mr Hlongwane: It was to protect IFP or areas controlled by Chiefs, as well as to kill the ANC.

22. IFP member, Mr Dlamini, said:

I will say that it is painful to me that after all these activities that we committed that people should deny our existence and call us criminals. When I went for training at Caprivi, nobody called me a criminal. When I killed people here, I was not called a criminal. Today they do call us criminals and deny knowledge of our activities and ourselves. No IFP leader is prepared to stand before this Commission and admit to these activities. We decided among ourselves to expose these activities. We in fact were not mad persons who just took weapons and started shooting people at random. Therefore it hurts me very much for the IFP to desert us and say that they do not know anything about us when they know that they were in fact responsible for all these things.

23. In yet another context, in the section 29 hearing into the violent activities of the Mandela United Football Club, Ms Winnie Madikizela-Mandela repeatedly denied, in the face of allegations of her awareness, that she had knowledge of events:

I knew nothing about these activities.

I did not monitor them when they were in and out of my house.

I did not know who recruited who into the Mandela United.

I knew nothing about who took what decision. I had nothing to do with the activities of the boys.

24. The gap between leadership and foot soldiers, particularly amongst the youth, was also described in the United Democratic Front (UDF) submission to the Commission. A gap of this sort means there were possibilities of misinterpretation that led to atrocities on the part of youthful activists.

In this context, many activists interpreted statements by the UDF and its allies making reference to the breaking down of apartheid to mean that this should be done by means of violence.

25. Regarding questions about the brutal enforcement of labour, consumer and student boycotts that involved gross violations of human rights, the UDF submission argued that such acts should be seen against

the background of emergency when most of the UDF leadership was in detention or on the run. The acts were committed by youths acting on their own, even though some may have believed that this was being done in the interests of the struggle.

26. Similarly, within the ANC and MK, although for different reasons (not least of which was the physical distance between leadership in exile and operatives within South Africa), there was recognition of a gap, a distance, between top and bottom. In this case, it is not expressed as a denial but as a concession of problems caused by this gap. An MK leader testified at an amnesty hearing as follows:

There were long and insecure lines of communication, command and control. Many of the established MK units had been allowed a degree of initiative in executing operations, as long as these remained within policy guidelines. In contrast with the conventional military force in which planning takes place at headquarters levels, in guerrilla warfare most of the detailed planning takes place at the lowest level There is no so-called hot-line to higher structures to ask for guidance. Communication could and at times did result in deaths, given the degree to which communication lines were monitored.

27. Overall, across different parties in the conflict, the above quotations indicate that, although particular contexts varied, a common problem existed in terms of a distance between top authorities and field soldiers, supporters or followers. Atrocities, it is suggested here, emerged precisely because of this gap, opening up possibilities of miscommunication, misinterpretation and possibly, as Mr FW de Klerk suggested, male fides.

What went wrong? "We made mistakes."

28. Different parties to the conflict admitted that there were errors, mistakes and unintended consequences. Several parties contended that violence occurred precisely because of the grey areas that developed.

29. At one of the hearings on children and youth, Mr George Ndlozi, reporting on the activities of SDUs, said things "went wrong" because they

had to depend on criminals and people took advantage of the situation. They ended up operating out of personal gain.

30. Mr Niel Barnard, former head of National Intelligence, said at the hearings on the State Security Council:

It is true that instructions and mandates were sometimes vague and were communicated poorly [and] in large bureaucratic institutions such as the public sector there is a danger that decisions and instructions are not      formulated, conveyed and interpreted in a correct way.

31. Mr Johan van der Merwe, former commissioner of police, said at the State Security Council hearings that "we had to move outside the boundaries of our law", leading to all sorts of blurred distinctions between acceptable and unacceptable methods. This point was also conceded by Mr FW de Klerk in the NP submission.

32. General Andrew Masondo, former political commissar of the ANC, admitted that they "could have made mistakes" as a result of disinformation or when they had to rely on young, inexperienced people in authority in the Quatro camp.

33. Mr FW de Klerk, answering questions on widespread torture during the 1980s in the second NP submission, said:

I'm not saying we were perfect I'm not saying we didn't make mistakes.

Detailed operationalisation (of security policy) takes place at a much lower level that is where, either through over-zealousness or a male fide approach, where things get out of hand.

History has subsequently shown that, as far as the policy of apartheid was concerned, they were deeply mistaken.

None of these unconventional projects was intended to lead to any gross violations of human rights but they did create an atmosphere conducive to abuses.

34. Former MK leader Mr Ronnie Kasrils, speaking during the Commission's public debate on the notion of 'just war', said: "I'm not saying that there weren't certain departures, certain aberrations". Similarly, the ANC submission to the Commission reported concern in late 1987 regarding an increase in "attacks which did not accord with ANC policy", conceding that "some incidents not entirely consistent with ANC policy did take place". In its second submission, the ANC repeated that "mistakes were made". In similar vein the UDF stated that the

activities of the UDF and its allies, while making invaluable contributions to the democratisation of South African society, had many regrettable consequences.

35. Even in the details of operations of bomb attacks, things could go wrong, mistakes were made. Regarding the explosion at the Krugersdorp magistrate's court adjacent to a "notorious security police branch", Mr Mohammed Shaik told the amnesty hearing:

I prepared two charges; one being a decoy which I placed in the toilet used by police officers in the court complex, the other being a car bomb. The decoy was to have exploded first, drawing out police officers, who in a few minutes would have cordoned, cleared and secured the area. Their presence would have been very near to the car bomb which was to explode minutes later. Unfortunately the decoy failed to explode due to some malfunction. The car bomb detonated as planned. The intended aim of a large number of enemy personnel being killed, injured was not achieved. A civilian and two security branch members were killed.

36. In the Freedom Front submission, General Constand Viljoen also reported on mistakes of the former government. Referring to the NP government, in which he was chief of the defence force, he testified:

Forty years of governmental control made them power-drunk. Expediency, manipulation, propaganda and in the end the ruthless tactics of an unconventional sort to retain power all these things are not necessarily part and parcel of the original concept of differentiation that prevailed within Afrikaner political thinking. The original motivation of the Afrikaner was not to rule others.

37. He argued further that errors were made due to the arbitrary powers given to ministers and "even officials in the security forces" during the states of emergency.

because of the absence of normal checks and balances that would avoid misuse of these powers most cases of gross violations of human rights resulted from these practices and they had the serious additional effect of keeping the public in the dark on these activities and creating a sense of fear and bondage in general that was not conducive to free and responsible citizenship.

38. The Pan Africanist Congress (PAC), in its submission to the Commission, also admitted to mistakes. Reporting on "a new pattern in the 1990s where civilians within the white community were attacked", the PAC submission stated:

In the nature of guerrilla war, which is unlike conventional warfare, detailed plans could not be made from Dar-es-Salaam. The actual targets were decided by local commanders In the militarised environment in our country in the 1990s internally based operatives often made errors that APLA [the Azanian People's Liberation Army] had earlier avoided. These are the causes of the departures in the 1990s.

39. Specifically on the murder of Ms Amy Biehl, the PAC submission regretted its error, stating that:

PASO was not a part of APLA. They are a component part of the PAC not involved in armed struggle. They wrongly targeted and killed Amy Biehl. We expressed our regret and condolences.

Lack of discipline/restraint: "Us and them"

40. A lack of discipline exercised by the state over its security officials, and by other parties over their activists, could clearly have contributed to the escalating spiral of violence. The Commission specifically questioned the leading parties in the conflict regarding their tolerance for violent acts among their own members, and efforts they made to discipline transgressors. Psychological factors appear to have played a role in the general pattern, on all sides, of condoning lack of restraint in their own members, and the relative absence of tough discipline regarding violent offenders. The UDF submission, in an honest attempt to get to grips with what they admitted was a "dilemma", put the case most clearly:

The way we approached this question is like a father, like parents would approach, let's say, an aberrant child: that child is part of your family, these were people who were oppressed people, part of this history.

Now if a child misbehaves and hopefully [the parent] disciplines that child and shouts at the child and does what is possible within the limits of the family alright, but they don't disown these people. For us to disown these people would mean that we don't understand the history of these people [who] tended to do these sort of things.

So they were undisciplined in some instances. When they did that they were not acting within the UDF policy but we own them, they are part of us, and they are part of our history and we accept them as part of our family.

41. The dilemma posed for the UDF, and similarly for other organisations, is highlighted in the response to a question about the 'young lions'. The UDF expressed pride in the efforts of this generation:

In general, the generations of young people from 1976 onwards have been should be amongst the most admired citizens of this country. They laid down life and limb to wage the struggle Generally the young lions were doing a wonderful piece of work.

42. Although various efforts were made to impose restraint and discipline in all parties concerned (for example, the ANC and MK developed codes of conduct, and attempted to discipline conduct through political education), evidence before the Commission showed that all parties fell short, in some respects, in imposing restraints and discipline on their own members, followers and supporters. The dilemma, as clearly indicated in the UDF quote above, is that harsh discipline imposed on "our own" (even where it was feasible to track down transgressors) would have risked alienating their own supporters.

43. This phenomenon is exacerbated in a highly conflictual war-like situation. Militaristic authorities were clearly at times covertly proud of the violent actions of 'their own'.

44. Sometimes, this pride was not even covert. Colonel Eugene de Kock was awarded numerous medals for his actions. It is claimed that the former Minister of Law and Order, Mr Adriaan Vlok, attended a party with Vlakplaas operatives after they had blown up Khotso House.

The problem of perspectives

45. It is important to state from the outset that emotions may run high when considering the actions of perpetrators, and that perspectives may differ sharply, leading to difficulties with reconciliation. Some will tend to blame, condemn and feel bitterness towards perpetrators while others are able to demonstrate empathy, understanding, sympathy or even praise for those who did some of these deeds. Given the divisions of the past, such varying perspectives towards perpetrators from the varying sides of the struggle are not surprising. It is neither simple nor easy to take a neutral or wholly objective stance towards perpetrators of evil deeds. Nevertheless, this part of the report needs to provide an understanding of dreadful deeds, without condemnation. At the same time, as Browning, a leading Holocaust scholar, puts it: "Explaining is not excusing, understanding is not forgiving"5. The Commission, in this chapter, is seeking to fulfil its objective to

promote national unity and reconciliation in a spirit of understanding which transcends the conflicts and divisions of the past.6

46. In an effort to grasp and understand, rather than to condemn or excuse the actions of perpetrators, it is important to be aware of difficulties of perspective.

The problem of perspectives

47. The Act makes a clear distinction between "the perspectives of victims and the motives and perspectives of the persons responsible for the commission of the violations"7. Baumeister, in a recent major study8, describes this as the "magnitude gap": the discrepancy between the "importance of the act to the perpetrator and to the victim". This magnitude gap has a number of features:

9, where repeated acts of violence are described in a matter-of-fact manner:

why? Victims' accounts show two versions, one which emphasises sheer incomprehensibility the perpetrator had no reason at all and the other which presents the perpetrator's action as deliberately malicious, as sadistic, as an end in itself. By contrast, the vast majority of perpetrators, even if they admit wrongfulness, provide comprehensible reasons for their actions, and almost never admit to being motivated by sheer maliciousness or the wish to inflict harm as an end in itself.

48. This perspective gap may be illustrated by the case of Mr John Deegan, a former member of the Security Branch and a Koevoet operative responsible for various atrocities. In a testimony dated 30 June 1996, he reports as follows on the recent death of his father:

He was cold-bloodedly shot dead and his murderers escaped. I cannot come to terms with his death in that it was a senseless act of violence in the pursuit of greed. This is the first time that my family and I have come so close to experiencing the horror of violence so directly in this country.

49. Here is the perspective of the victim. But this is the same man who, in a report dated 23 August 1993, appeared as perpetrator, a Koevoet member in then South West Africa, dealing with a wounded SWAPO10 operative:

Even at that stage he was denying everything and I just started to go into this uncontrollable rage and he started going floppy and I remember thinking "how dare you" and then this is what I was told afterwards I started ripping. I ripped all the bandages, the drip which Sean had put into this guy pulled out my 9mm put the barrel between his eyes and fucking boom I executed him. I got on the radio and said to Colonel X "We floored one we are all tired and I want to come in."

50. From the point of view of the victim, violence is a "senseless act" and he experiences it with horror (the first time he claims to have had such an experience), yet as the perpetrator he has reasons and strong emotions, even expressing outrage (a moral stance) at the apparent defiance of his captive. The magnitude gap is a discrepancy between two quite different and irreconcilable positions.

Third parties: the perspective of the Commission

51. There is a third perspective: that of the onlooker, the outsider, the observer, the recorder, the evaluator, the scientist. That is the position of the Commission. Neither victim nor perpetrator, but charged with the task of understanding such acts of evil and helping to prevent them in future, it is a difficult stance. From the observer's difficult position, it is both helpful and instructive to grasp the notion that the perspectives of victims and perpetrators may differ sharply.

52. While its overall aim is to be even-handed and as objective as possible, to view the Commission as homogeneous, as all of one piece, is a rather oversimplified approach. The Commission is made up of many people with different perspectives. Members have had sympathy with the victims not least because of the harrowing process of hearing month after month of testimony from victims. In addition, some Commission members have shared the perspectives of victims in their own past experience. Depending on the context, members may also have had some empathy with perpetrators, perceiving them to some extent also as victims. This is not to decry the efforts of the Commission to be objective. It is an honest admission that the perspective of the Commission and its members is a complex one.

Perpetrators as victims?

53. A further problem of perspective is the thorny question of whether perpetrators may also be viewed as victims. Although one may wish to have a clear-cut position on perpetrators, its is possible that there are grey areas. Perpetrators may be seen as acting under orders, as subjects of indoctrination, as subjected to threats, as outcomes of earlier doctrinaire education. In the most pernicious situation, askaris (former ANC cadres who were 'turned', frequently through torture, threats and brutality, into state agents) are themselves transformed into killers and torturers. Military conscripts could view themselves in part as victims of a state system. Kitskonstabels (special constables) could see themselves as victims of poverty, in need of a job.

54. To understand these potential grey areas involves being drawn into a position of some sympathy with the perpetrator. The dangers of this are twofold: first to forget and ignore the suffering of victims of abuse, and second, to exonerate the doer of violent deeds. From the third perspective of the Commission, difficulties are once again manifest. Two statements may be fruitful. First, it is important to recognise that perpetrators may in part be victims. Second, recognition of the grey areas should not be regarded as absolving perpetrators of responsibility for their deeds.

55. The position of the Commission regarding accountability and responsibility is quite clear and was repeatedly stated by the chairperson of the Commission. While acts of gross violations of human rights may be regarded as demonic, it is counter-productive to regard persons who perpetrated those acts as necessarily demonic. The work of the Commission towards reconciliation would be useless if such a stance were to be upheld.

Violence of the powerful and the powerless

56. As Frantz Fanon recognised some forty years ago, violence of the powerful and the powerless is not equivalent. An unhappy characteristic of oppression is that violence is often committed by the powerless against other oppressed groups. Bishop Peter Storey expressed this succinctly in a Commission hearing into the activities of the Mandela United Football Club:

The primary cancer will always be the apartheid oppression, but the secondary infection has touched many of apartheid's opponents and eroded their knowledge of good and evil.

57. The phenomenon whereby the oppressed turn their violence against each other was expressed in many forms in South Africa: between the Azanian People's Organisation (AZAPO) and the UDF, between township vigilante groups and more youthful 'comrades', between township groupings in the enforcement of boycotts, in the case of the gruesome 'necklace' murders which numbered many hundreds from the mid-1980s onwards, in the case of violence by the Mandela United Football Club, in the case of askaris, and most prominently in the murderous violence between Inkatha on the one hand and Charterist groupings ANC, UDF, SDUs on the other, all of which often took the form of cycles of revenge. An IFP amnesty applicant in respect of the Boipatong massacre, Mr Victor Mthembu, expressed it as follows:

We would not have done these things if the people of Boipatong did not terrorise IFP members, if the comrades had not killed IFP members. If it had not been a war situation between the IFP and the ANC I would not have participated.

58. Violence among the oppressed has often been described as 'black on black' violence. This is an unfortunate and potentially racist depiction since it camouflages the role of the state in orchestrating or steering such divisions. It is common knowledge that the state provided covert support for homeland leaders and for Inkatha. The security police gave support for conservative, anti-UDF 'vigilante' groupings. In its submission to the Commission, the UDF said:

The State repeatedly distributed leaflets all around the country in the names of various organisations with the aim of causing confusion and fermenting violence between the UDF and AZAPO. Unfortunately we say that this sometimes succeeded in doing precisely that.

attitudes of intolerance had to be seen against the background of a climate of suspicion and intolerance that was created by the regime People who are constantly fearful of attack or arrest not surprisingly develop attitudes that are not only intolerant but also undemocratic in such a climate.

Even-handedness

59. There is a final major area regarding the problem of perspective: the question of even-handedness in understanding perpetrators from the multiple and varied sides of the struggle. Perpetrators of gross violations of human rights came from all sides: the security forces, military conscripts, the liberation movements and their armed wings, Inkatha and the UDF, from askaris and kitskonstabels, from township vigilante groups, youth organisations, from torturers and assassination squads, from the far-right, and from township crowds responsible for 'necklace' killings. It is probably not possible to provide a neat, tidy or exhaustive classification of perpetrators.

60. In this respect, the Commission wishes to state that:

11. To be even-handed in understanding the motives of perpetrators also requires full recognition that violence of the powerful, the South African state, was not necessarily equal with violence of the powerless, the disenfranchised, oppressed and relatively voiceless black majority. While each side may put forward reasonable and quite understandable explanations or justifications for such actions, the task of the third perspective, that of the Commission, is to recognise that these accounts are not necessarily equivalent. This non-equivalence means that protagonists in the thirty-year conflict were motivated by quite different political perspectives.

Explanations of Motives and Causes

A political understanding of causes

61. To understand perpetrators' actions, it is essential to locate them within a particular pattern or sequence of events. One aspect stands out above all others when one looks back at the patterns of numerous atrocities over the twentieth century. They begin under certain political configurations, increase rapidly in scope, diversity and spread, peak, then decline quite rapidly when political circumstances change. In the South African case, this pattern spanned the historical period 1960 to 1994, although the majority of violent acts occurred mainly towards the tail-end of this period, between 1983 and 1994. This means that the motives of perpetrators need to be understood primarily in historico-political terms; that is, psychological explanations and motives, while not irrelevant, are secondary.

62. None of this means, however, that there were no atrocities beforehand. Historical circumstances build over a long period and, in South Africa, conflict goes back to the initial appearance of invader-settlers. Nor does it mean that atrocities decline entirely following a change in political dispensation. There have, indeed, been isolated incidents of violence from far-right-wing groups and ominous recent attacks against farmers. Violence continues in KwaZulu-Natal and, of course, criminal violence and violence against women have not abated.

63. Yet the pattern is different. Atrocities are widespread and rampant at particular times, then decline and dribble away to sporadic cases. Types of violence change; hit squads, torture, abductions, cross-border raids, assassinations, guerrilla bombings decline and disappear. Criminal activities, and violence against women, have different motives.

The primacy of the political motive

64. The primacy of the political context as an explanation for violence was persuasively put by General Constand Viljoen in the submission of the Freedom Front to the Commission. Regarding accountability, General Viljoen said:

I still maintain it is unfair that the operators be exposed as the chief perpetrators of atrocities and violence in general when the politicians and strategy managers hide behind their status and positions. The iniquity of our past was of a political nature first, and mainly in that way a moral problem on an individual level.

65. Mr FW de Klerk also confirmed the primacy of the political in creating the overall climate for subsequent violence. In the NP's submission statement to the Commission in May 1997, he said:

Let me state clearly that the National Party and I accept full responsibility for all our policies, decisions and actions. We stand by our security forces who implemented such policies. We accept that our security legislation and the state of emergency created circumstances which were conducive to many of the abuses and transgressions against human rights We acknowledge that our implementation of unconventional projects and strategies likewise created such an atmosphere.

Contexts of political motives

66. What were the political motives? While apartheid, rooted in colonialism, may be the primary context for the struggle, two other, wider, contexts combined to produce the particularly volatile mix in South Africa.

The cold war context

67. The first was the international climate of cold war, in particular the virulent form of anti-Communism and anti-Marxism that took root after the 1948 election victory of the NP. According to former Minister of Law and Order Adriaan Vlok at the Commission hearing of 14 October 1997:

The mother organisations of the liberation movements, the ANC-PAC, were seen with justification as fronts and tools of the Marxist-Communist threat against the country I believed and still believe that if the forces of Communism and Marxism since the 1950s were allowed to take over South Africa, our country would today be destroyed, impoverished and a backward country with an atheist communist ideology as the government policy I saw it as part of my duty to fight against such thoughts, programmes or initiatives and to ensure that these objectives were not successful.

68. Not only leaders but countless foot-soldiers were fed on a diet of this sort of propaganda over a long period. In the same testimony, Mr Vlok says clearly:

We actually still referred to them as the enemy in those days; the enemy was doing this that and the other.

The anti-colonial context

69. The second wider context was the anti-colonial resistance movement in Africa, particularly in the neighbouring territories of Zimbabwe, Namibia, Angola and Mozambique. This occurred over the same period and became deeply entangled with the South African struggle. Although the liberation movement was dominated by the non-racialism of the ANC and anti-racism of other movements such as Black Consciousness, some organisations interpreted it as a struggle against whites. Mr Ntobeki Peni, a member of the Pan Africanist Students' Organisation (PASO) who was involved in the murder of Ms Amy Biehl in Gugulethu in August 1993, expressed it thus:

These speeches were closed with the slogan "one settler, one bullet". I understood this slogan to apply to every white person who came into the line of fire during an APLA operation, or an operation to support APLA, or where we, as PASO members were to assist in making the country ungovernable.

70. While both the ANC and the PAC made it clear in their political submissions that their primary motives were in response to the apartheid regime, it is also clear in their joint campaign and their alliance with others in the front-line states, together with their involvement in organisations such as the Non-Aligned Movement and the Organisation for African Unity, that the local struggle was part of the wider anti-colonial movement in Africa. The PAC submission stated:

When the Organisation of African Unity was formed in May 1963, it gave support to armed struggle through its Liberation Committee based in Dar-es-Salaam. Many countries in Europe and Asia channelled their aid to our people through the OAU Liberation Committee.

71. The ANC submission quite succinctly identifies anti-colonialism as the central political motive:

At the root of South Africa's conflict was the system of colonial subjugation. Like other colonial countries, South Africa was victim to the rapacious licence of an era that defined might as right, an epoch of international morality that justified dispossession and turned owner into thief, victim into aggressor, and humble host into ungodly infidel.

72. Further on, the ANC submission states:

Thus ranged against one another, in intensifying conflict, were the oppressor and the oppressed, the owners of wealth of the country and the dispossessed, the rightless and the privileged. The ANC was a product of this history and this conflict, not their creator.

73. The issues surrounding resistance to colonial domination in South Africa were further complicated by the perception on the part of many of the dominant political forces that the Afrikaner population, too, had been injured by attempts at colonial subjugation by the British. This perspective was carefully laid out in the political party submission of the Freedom Front. Describing the effects of the discovery of South Africa's mineral wealth, General Constand Viljoen states:

It invited the greedy attentions of colonialist powers, particularly Great Britain who in its imperialistic drive soon had second thoughts on its endorsement of the independence of existing Boer republics The result was war and trauma of a sort that have in a way not yet been resolved.

For it conditioned the white tribe of Africa the Afrikaners to consolidate in a nation around the dangerous sentiments of a collective sense of injustice, discrimination and deprivation.

74. In the first Freedom Front submission, General Viljoen suggests:

We may have redirected our quarrel with the British to our compatriots in South Africa.

The apartheid context

75. The third and most direct political context was the NP policy of apartheid, long rooted in colonialism and segregation, but increasingly from 1948 and particularly after the banning of the PAC and the ANC in 1960 involving a direct struggle between oppressed and oppressor: an armed conflict which gradually intensified over the subsequent years. Here of course the political perspectives differed widely. For the PAC the conflict was:

A national liberation struggle against settler colonialism for the restoration of our land to its rightful owners the African majority.

76. For the ANC, apartheid was, quite starkly: "a crime against humanity".

77. By contrast, for supporters of the NP, 'separate development' was a defensible policy fashioned in order to solve local problems. The NP submission to the Commission on 21 August 1996 states that:

We thought we could solve the complex problems that confronted us by giving each of the ten distinguishable Black South African nations self-government and independence in the core areas that they had traditionally occupied The underlying principle of territorial partition to assure self-determination for different peoples living in a common area was widely accepted.

78. Further on, the same NP submission says:

The great majority of those who served in the security forces during the conflict were honourable, professional and dedicated men and women.      They were convinced that their cause was just, necessary and legitimate.

The perception of those on the side of the Government was accordingly that the installation of an ANC Government would lead to Communist domination. They believed that in conducting their struggle against the ANC, they were playing an important role in the West's global resistance to the expansion of Soviet Communism.

79. How did the purported idealism of the apparently righteous struggle of the Afrikaners for self-determination go wrong? Here again, not surprisingly, there are differing political perspectives. For Mr FW de Klerk, who repeatedly stated that he had no knowledge of NP or cabinet authorisation of gross human rights violations, things went wrong because:

You cannot fight that type of thing in the normal way.

80. The result, according to the NP submission, was:

more authority to the security people to fight a very specific kind of war, and across the world where this type of war occurred there have been these aberrations.

81. The version of the Freedom Front submission was that Afrikaners, rooted originally in the ideals of 'freedom from bondage', gradually lost their way and, during this process, the NP denied "on a racial basis, democratic rights to others" and found themselves "far removed from their traditional value systems". According to General Constand Viljoen, the NP

started slipping and they had to resort to unconventional devices, propaganda and group force in order to keep political control.

82. The ANC submission puts a different argument:

Apartheid oppression and repression was therefore not an aberration of a well-intentioned undertaking that went horribly wrong. Neither was it, as we were told later, an attempt to stave off the 'evil of communism'. Its ideological underpinnings and the programmes set in motion constituted a deliberate and systematic mission of a ruling clique that saw itself the champion of a 'super-race'. In order to maintain and reproduce a political and social order which is premised upon large-scale denial of human rights, far reaching and vicious criminal, security and penal codes were necessary the system increasingly relied upon intimidation, coercion and violence to curb and eliminate the opposition that apartheid inevitably engendered.

83. Racism, as a central ideological ingredient at the core of the political struggle, was unfortunately underplayed in the NP submission. Racism as an ideology, a means of domination and oppression, provided the central grounds for the systematic exclusion, segregation and denigration of the black majority. Racism is a systematic ideological doctrine which creates the 'other' as essentially different. In South Africa this was the rhetorical basis for apartheid and 'separate development': blacks required development, but at their own, slower and different pace, since (as the argument went) they were essentially different from the more civilised, developed people of European origins. Not only politicians but also leading academics, scientists, theologians and churches such as the Dutch Reformed Church (DRC) provided constant fuel to bolster such ideological positions. Racism, it hardly bears reminding, also served to distance and to dehumanise the black 'other', a process which opened the way for violence. In the practice of torture, for instance, black people were more severely brutalised in the main than white people.

84. These three political frameworks, the cold war, anti-colonialism and the racist and oppressive apartheid regime, ideologically fuelled by Christian-Nationalism and increasing militarism, provided the arguments and justifications, the passions and the furies for the eventual commission of dreadful deeds. If political frameworks provide the fuel for atrocities, they must also form the focus of primary attention for future prevention. Political contexts do not, however, provide the full set of explanations.

Other explanations of motives

Individualistic psychological explanations

85. In a comprehensive analysis of many and varied forms of evil deeds, Baumeister distinguishes between four main clusters of motives underlying the actions of perpetrators12. In this section, a slightly different but overlapping scheme is followed, also listing four main forms of explanation. It ends with an enquiry as to which forms of explanation best fit the South African situation and which apply best to particular forms of violent action for example, torture and 'mob' killings. Both the popular and more scientific understandings of perpetrators are interrogated. In addressing motives, it is important to be mindful that reasons are likely to be pluralistic, overlapping, multi-layered and contingent on particular and local circumstances.

86. Popular accounts sooner or later raise the suggestion that violence is due either to deep, ingrained aspects of human nature ("it is in our nature to be violent" or "he is inherently evil") or to various forms of psychological malady ("these actions are crazy, mad or mindless" or "these torturers must be sadists"). Since these everyday examples are so widespread and commonplace, they warrant asking questions about the substance of such claims. Much of the social psychological evidence points against these hypotheses, however.

Human nature

87. Let us examine the first claim, that violence is in our human nature. The notion is usually couched in some form of evolutionary or biological explanation that we have descended from animals, that current violence is a throwback to more primordial, primitive or regressed forms, or that there are particular biological mechanisms (genes, primitive brain stems, hormones) that predispose us to repeated episodes of atrocities. Against this, one should consider the following. Why is it that mass atrocities appear only at certain historical times and not others? Why is it that some societies or cultures are in the main peaceable? Why is it that half of humanity, women, are not particularly violent and are seldom involved in large-scale atrocities? Even if one could identify atavistic predisposing factors, this would not explain the plain fact that large-scale killings (as in Rwanda, Bosnia, the Holocaust, the cultural revolution in China and Cambodia) occur over relatively brief historical periods and then cease. Atavistic accounts are simply not adequate.

Psychological abnormalities

88. The claim that violence is due to psychological dysfunction appears to warrant more attention. Nevertheless, the bulk of the international literature on atrocities and perpetrators13 reports little evidence of severe abnormality. Even in regard to sadism (Baumeister's final potential motive of perpetrators: pleasure in hurting others), the general consensus is that, while it cannot be entirely dismissed, only about 5 per cent of all types of perpetrators (for example, serial killers, torturers, rapists) may be classed as sadists, and furthermore that this motive is gradually and slowly acquired over time. It is not inherent, but a consequence rather of earlier perpetration of violence. Although there is scant evidence that perpetrators suffer from major or severe psychological pathologies, other studies suggest that perpetrators may experience severe stress and anxiety along with denial, disassociation, 'doubling' and other defence mechanisms14.

89. In the South African case, some submissions to the Commission have made claims of post-traumatic stress disorder among some perpetrators. However, the diagnostic manual on post-traumatic stress disorder attributes this state to victims, not to perpetrators. Furthermore, post-traumatic stress disorder, even if diagnosed among perpetrators, is far more likely to be a consequence of appalling actions, not primarily a causal factor. There is sizeable evidence that perpetrators experience severe stress reactions, and take to heavy bouts of drinking and drug-taking: but these are consequences.

90. An extract from the written statement of Koevoet member Mr John Deegan states:

I really had bad dreams I have dreams of bodies, or parts of bodies like an arm this is a recurring dream I still have now an arm sticking out of the ground and I'm trying to cover it up and there were people around and I know that I killed them, whatever is down there and its been down there for weeks and it is this intense feeling of guilt and horror that this thing has come out of the ground again and I had a dream that I actually met a guy that I shot.

91. While it is premature for the Commission to draw any final conclusions on this matter, the considerable bulk of international literature, and also the testimony submitted to the Commission, suggests that severe psychological dysfunction is not a primary cause of atrocities. Instead, most commentators have emphasised the ordinary, rather unexceptional character of perpetrators, typified in Hannah Arendt's celebrated phrase, the "banality of evil", or in Browning's term "ordinary men"15. In this regard, it is instructive to quote Colonel Eugene de Kock, the Vlakplaas multiple killer who distances himself from psychological accounts that put the blame on childhood experiences, another form of explanation which seeks dysfunctions.

I know it has become fashionable to blame a person's adulthood on his childhood But such an approach makes me uncomfortable. I do not believe my childhood was especially bizarre. To be sure, my father was the proverbial hard man and he drank too much. So what? Many sons had hard men and drinkers for fathers I find it unacceptable to blame my father and my home life for me.16

92. Along similar lines, in explanation of ANC-inspired SDUs in Gauteng townships, the Commission heard testimony that such persons were not dysfunctional but quite dutiful citizens. At the special hearing on children and youth, the Commission heard that

far from being a bunch of undisciplined comrades or the last generation, SDUs were in many ways the backbone of defence in certain townships.

93. Rather it was the social system and wider context that changed people. Mr Jimmy Nkondo, who joined an SDU at age thirteen

changed from a carefree young man who enjoyed school and sport to a person with no mercy. Instead of being nurtured in the family home he became a killing machine. There was no choice, it was kill or be killed.

Authoritarianism

94. Before leaving individualistic psychological explanations, it is worth asking whether perpetrators are not self-selected, that is, people with particular kinds of personality predispositions who are drawn to certain careers and circumstances to emerge as killers and torturers. The theory of the authoritarian personality presents such a view. Rooted in earlier thinking from the 1920s, the authoritarian thesis was made famous in a major book published in 195017. A particular kind of person, the 'authoritarian personality', it was claimed, emerged from rigid and punitive family structures. Drawing on psychoanalytic theory, it was argued that children of such families split off and repressed feelings of hostility and resentment towards their parents and projected these hostile feelings towards members of weaker and stigmatised groups. As Adorno once put it, authoritarians metaphorically resemble cyclists: "above they bow, below they kick". Authoritarians as a type also manifest a particular pattern of social attitudes: ethnocentrism (or favouritism towards their own group), prejudice towards out-group members, anti-democratic views and a propensity towards conservatism and fascist ideology.

95. There is solid and reliable recent evidence that authoritarianism was manifest among white South Africans, that it was correlated with anti-black prejudice and anti-democratic tendencies, and was more prevalent among Afrikaans rather than English-speakers, and among those who regarded themselves as conservative.

96. In recent years, the theoretical base of authoritarianism has been revised. It is no longer seen as rooted primarily in either intrapsychic conflict or in strict, hierarchical and rigid family socialisation. Rather, it is a set of beliefs expressing strong and loyal identification with one's own social or cultural group ethnic, national, racial, religious with strong demands on group togetherness (cohesion). Respect and unconditional obedience is given to own-group leaders and authorities (an attitude of bowing to the top) while authoritarian aggression reflects negative prejudice, intolerance and punitiveness towards out-group members (the kicks below). Authoritarianism in this view is a form of social identity rather than a particular personality type. It is nevertheless a modality of identity with tendencies towards hostility towards 'others' expressed in South Africa as the intolerance and hostility of whites towards blacks and those on the left18. It is certainly not far-fetched to argue that such people may be drawn towards lives in the police and the army, themselves rather rigid and hierarchical institutions, which then reinforce authoritarian tendencies that are already present. Hence, self-selection on psychological grounds is quite feasible.

97. But does this offer an explanation for a predisposition to commit atrocities? Evidence is really rather thin. There is no direct evidence that shows that authoritarian identities are actually more violent in terms of behaviour. Research of this sort has shown evidence of expressed intolerance, prejudice, racism and anti-democratic tendencies but not direct evidence of murderous actions. It is dubious in the additional sense that there may be many authoritarians in a given social order, but far fewer perpetrators of violent deeds. Therefore the links to violence are neither direct nor proven.

98. However, authoritarianism may form a contributing factor in the propensity towards violence in that both central elements of the authoritarian personality obedience and loyalties to in-group authorities, and hostile distancing from others (a tendency towards dehumanisation of the 'enemy') have been directly implicated in aggression. It is a feasible link in the chain; it is not the whole story.

99. In evidence before the Commission it would seem that many perpetrators,      particularly from the security forces and right-wing organisations, would fit rather closely the description of authoritarian identity. Certainly there is evidence that security force training, perhaps particularly of the sort found in South Africa prior to 1994, may facilitate such tendencies. Here again are extracts from Mr John Deegan's testimony, describing police training:

During police training at the college in Pretoria, the 'code of silence' was soon inculcated into new recruits through various methods of indoctrination and brainwashing and reprisals for not being one of the group. If one person stepped out of line, no matter how trivial the offence, the whole platoon or company would be punished. Individualistic behaviour was punishable not only by the system of instructors, trainers and officers, but by your peers as well fellow trainees eventually through fear of punishment would punish fellow students before infringements came to the notice of superiors I learnt early on in my training that individualism was out.

100. It should be apparent from the above that violence is not a matter of individual psychology alone. It is the combination of personal biographies interwoven with institutional forms (organisations, military structures, hierarchical arrangements of power) and an escalation of events in historical terms that provides the assemblages or configurations that produce awful deeds. It is not a mechanistic formula. Since authoritarianism in this view is a particular form of identification, social identity frameworks as potential motives for violence will now be examined.

Social identities: Preconditions for violations

101. Put most simply, people do not act only due to personal or individual attributes. We also act in terms of the norms, values, standards of groups that provide us with social identities (racial, national, ethnic, gendered). When groups are in hostile, suspicious relations with each other, we are capable of acting towards others in a dehumanised fashion, treating them as the enemy, as described in Mr Adriaan Vlok's statement:

We also have to remember that we were engaged in war and that makes it even more difficult to really do what you ought to do. And we made a mistake, we should have listened to people but we were engaged in a war, we had all been indoctrinated not to listen to each other

102. In such situations, people act primarily in terms of their social identities rather than personal attributes. What implications does this have for explaining violence? The critical implication is that the psychological dynamics of inter-group relations are of a different order from those of interpersonal relations. Perpetrators' actions are instances of inter-group rather than interpersonal relations, and require a different order of psychological explanation. A range of possible options may make the contrast clear.

103. One class of account places the emphasis on a loss of personal identity, or a loss of moral restraints. This understanding is common enough in everyday descriptions such as "killing frenzy", "mob madness" "war brutalisation" or "losing control". This "loss", or dysfunctional, class of understanding is also prevalent in formal psychological theorising such as de-individuation, extreme stress, frustration, aggression and the like. These versions imply a move away (disintegration) from a normal personal rationality into a mode of irrationality, or a regressed, more 'primitive' state.

104. A second class of understanding motives explains violence as a product of personal or interpersonal psychology. Violence is due either to an intrinsic personality trait or type such as a sadist, psychopath, Machiavellian or authoritarian type, or conversely an inner psychological state or mood (rage, jealousy, frustration, revenge, provocation induced via interpersonal processes and interactions). Such explanations do not account adequately for violence against categories of people with whom we have little or no interaction, as in the case of warfare. The two domains, interpersonal and inter-group, are controlled by different processes.

105. A third class of explanations is located at an inter-group level. We act towards or against others because they are construed as members of other groups/categories: the 'enemy' or the out-group. Processes which contrast group differences, stereotype the other and promote ethnocentrism all serve to differentiate, distance from and ultimately dehumanise members of other groups. This goes hand in hand with strong feelings. This is put very succinctly by Mr Adriaan Vlok in his testimony to the Commission:

I believe that most policemen who found themselves in such a situation, where he found himself obliged to act in an illegal way, probably did this by virtue of his position as a policeman and not from personal considerations.

106. Numerous and multiply overlapping influences reinforce and manufacture these particular social identities. Here again is former Minister Vlok:

There was a plethora of various influences on a typical Afrikaans-speaking conservative, Christian person, for instance teachers at schools, parents and the way they brought up their children, professors and teachers at university, eminent people in society by means of statements and documents, the press, politicians in their statements and policies and the ministers in their churches.

107. The generalised categorisation and dehumanisation of the other was chillingly expressed by amnesty applicant Victor Mthembu when asked why children as young as nine months of age were attacked. He replied: "a snake gives birth to another snake". While the expressed aim was given as an attack against the ANC and particularly the SDUs organised by the ANC, Mr Mthembu said: " we did not discriminate, it was not discriminated against as to who was attacked".

108. It may be noted that social identity theory does not explain violence itself, but the preconditions of violence. It is employed here in order to emphasise the necessity of locating explanations of mass violence at the inter-group rather than the interpersonal or intrapsychic levels alone. It is obvious enough that racialised identities loomed very large on the South African landscape. There is plenty of evidence of racialised prejudice, stereotyping and distancing. Here are three examples from reports to the Commission:

I'd say that Apartheid turned me from being a human being into a white man, and so for me the motivation of joining the struggle against Apartheid was to seek to recover my own humanity I'd been robbed of by Apartheid under Apartheid I found it impossible to be a neighbour to a person of a different colour I was locked into the relationship of oppressors and oppressed. (Father Michael Lapsley, victim of a parcel bomb.)

At the time of the killing we were in very high spirits and the white people were oppressive, we had no mercy on the white people. A white person was a white person to our eyes. (Mr Ntombeki Peni, granted amnesty for the murder of Amy Biehl.)

the Lord wished separate peoples to maintain their separateness (Apartheid) respect for the principles of Apartheid had God's blessing. (DRC's submission to the Commission.)

109. While racial and ethnic identities (which also promoted division among black people) were made particularly salient as a systematic part of apartheid and Christian-National ideology, these were not the only pertinent identities. Religious identities, for instance, became intertwined with the military defence of apartheid:

Through the idea of the total onslaught, the Church immediately became an ally in the war. The total onslaught concept assumed that only twenty per cent of the onslaught was of military nature, eighty per cent was directed against the economical and spiritual welfare of the people the Church was now totally convinced that we were fighting a just war. Almost every synod of the DRC during this time supported the military effort in their prayers. (Reverend Neels du Plooy, during the hearings on compulsory military service.)

110. An additional form of salient identity, often ignored in explanation of mass atrocities, is that of masculinity. It is most surprising to find, in masses of literature on atrocities of many kinds, the sheer neglect of a simple fact: most of these acts are committed almost exclusively by men. Few women were found among perpetrators in the South African case. Although this is a matter which requires further investigation, this pattern of overwhelming male predominance among perpetrators appears to be confirmed in the preliminary analysis of the period under review by the Commission.

111. What is the relation between masculinity and violence? There have been many and varied efforts to explain male aggression: genes, hormones, socialisation, roles, essence, archetypes, peer pressure, status, careers, warrior mythology, the Oedipal complex and more. Patriarchy, the ideology of male domination, portrays men as protectors and defenders of women, property, territory and nation. Patriarchy is a significant explanation of the male's apparent propensity towards violence, but patriarchy as ideology itself requires explanation. It is beyond the scope of this report to explore the issue fully, but it remains an important part of the understanding of violence. In South Africa, it is clear that patriarchy and the cult of masculinity has been embedded deeply in each of the various cultural streams: black, Boer, British. Its significance as a contributing factor should not be undermined.

112. Masculinity intertwines with other identities, for instance those forged in military establishments. Baumeister lists egotism (self-pride or group-pride, bordering on self-aggrandisement) as one of the key motives of perpetrators. Masculinity intertwined with militarism jointly act as constituents of potentially lethal forms of egotism. Here again is Mr John Deegan, later a Koevoet operative, talking about police training. Apart from fear, discipline and propaganda there was also pride.

Pride was also worked into the equation and in the closed smallness of our lives under training, pride became of paramount importance. Pride in ourselves and our platoons. Pride in the company of platoons. Pride in the college. Pride in the police force, the nation, the country, the flag. Pride and patriotism. By the end of our training we were fully indoctrinated in the functions of the established system they strip away your individuality and they make you a man, kind of thing.

113. Here is a more stark statement given at the special hearing on compulsory military service:

Action, especially for young national servicemen, is often a thrill, an ego-trip. There is a tremendous sense of power in beating someone up even if you are the most put-upon dumb sonofabitch, you are still better than a kaffir and can beat him up to prove it.

114. This last quote is a stark example of the intertwining of multiple identity forms to produce violence. A threatened sense of masculinity is interwoven with a racialised identity and militarism to effect a volatile mixture. It bears repeating that it is not merely a single identity form that leads to violence. Multiple social identities such as masculinity and racial, militaristic and national patriotism combine with religious, ethnic and political identities to render people quite willingly capable of murderous deeds in the play of egotism and pride. If the construction of particular identities provides the preconditions for violence, it is the contradictory pushes and pulls, sequences and spirals of situations that provide the triggers.

Situations: triggers of violence

115. If there is a single dominant message emerging from psychological research over the past fifty years, it is a tale that emphasises the persuasive power of the immediate situation. While it is dangerous to regard situational forces as inevitable since there are always possibilities of resistance, it would be as much of an error to see resistance to situational forces as merely freedom of choice, strength of character or individual moral maturity. We are social creatures, and resistance to situational powers is also a matter of positioning in relation to others. For instance, resistance to the powers of group pressure is easier if you are part of a small group standing together, than on your own.

116. The literature in this area is quite technical and complex and a more detailed account is given in Foster's paper to the Commission19. Centrally at issue here is a question of the motives involved in 'binding people' into groups and to authority. There are three main motivational processes: compliance, identification and internalisation.

Binding forces

Compliance is a process of going along with a group because we wish to avoid censure (avoid sticking out like a sore thumb) or gain approval, and because groups provide us with information, they shape reality.

Identification is a second process of binding a person to authority, in which one 'goes along' because one feels the same identity (group, culture, racial, national) as the authority. This is the version of social identity theory, given above.

Internalisation is a process in which one goes along, complies with a particular institutional authority because it is consonant, in agreement with one's values.

117. While these three processes begin to explain why we become bound into groups, institutions and authorities, they do not yet suggest violence. Stanley Milgram's experiments, in which ordinary people gave high levels of electric shocks to innocent people in a laboratory, point out further processes in the steps towards violent actions.20

118. Two intertwined sets of processes are discernible from Milgram's work. On the one hand, there are those forces that bind the person into the situation. On the other hand, there are processes that distance us from the victim. These two operate in tandem. The 'binding in' processes turn on the hierarchy, surveillance and legitimacy of authority. Obedience to authority even to the point of acting violently is more likely when authorities are powerful, act as a group (consensually), are regarded as legitimate and have increased surveillance. On the other side, the greater the degree of psychological distance from the victim, and the more the perpetrator is reduced to a link in the chain of orders (the bureaucratic process), the more substantial is the degree of violence. Obedience to authority is not inevitable however. When circumstances were so arranged that two peers rebelled together, obedience dropped dramatically. Milgram commented as follows:

Revolt against malevolent authority is most effectively brought about by collective rather than individual action. (1974, p. 116.)

119. Two further aspects are pertinent to an understanding of the binding to authority. Both refer to subtle, almost imperceptible, but powerful 'rules' that operate in everyday life. The first refers to the sheer embarrassment involved in refusal. It requires making a scene, disruption, argument, making a fuss. The second process involves sequencing: a step-by-step interactive spiral that draws the person in by volunteering in the first instance, by accepting the early steps, by being drawn into the experimenter's definition of the situation and by 'tuning in' to the authority rather than the victim.

120. These two processes, working together, operate in subtle, quite normal ways to suck a person into a positioning of obedience, rendering refusals and resistance difficult. Regarding this process of sequencing, here we extract from John Deegan's story.

Slowly the artistic side started slipping away and I started getting into the kind of conventional, macho world of things.

I really didn't understand the function of the Special Branch until I was in it I thought I could still hold onto the real person in me, the artist, the sensitive idealist I thought I could actually do good within the police force but the system changed me, and it was a long process of erosion and mixing with these people and becoming part of the culture.

121. As Bauman stated in his study of the Holocaust: "Cruelty is social in its origins much more than it is characterological"21. Other studies have shown that it is particular roles and positions that people are placed in, rather than their personal characters, that lead to abusive actions. Perpetrators themselves may be in part victims of their circumstances; lines may blur and grey areas appear.

122. Crimes of obedience occur due to three main reasons, reiterating themes already discussed above.

Authorisation is the process in which authorities order, implicitly encourage, or tacitly approve of violence. The impulse to obey orders, to follow rules even with pride (to do one's duty) propels perpetrators forward.

Routinisation is understood as a sequence in which events are organised as routine, repetitive, programmed: little in the way of serious thinking or decision is required.

Dehumanisation is a process in which victims are transformed into creatures to whom normal morality does not apply.

123. Although the top political structures of the apartheid government repeatedly denied giving orders, as in the words of former Minister Vlok, "I never gained the impression anybody proposed an instruction or issued and instruction with such a sinister objective", it is also quite clear that in the minds of operatives such as Colonel Eugene de Kock, they were acting under orders. There seems to be ample evidence of such orders in De Kock's recent book. It is also clear from De Kock's account that all these factors implicated in crimes of obedience that is, authorisation, routinisation, and dehumanisation systematically became part of the security force culture.

Language and ideology

124. It is commonplace to treat language as mere words, not deeds, therefore language is taken to play a minimal role in understanding violence. The Commission wishes to take a different view here. Language, discourse and rhetoric does things: it constructs social categories, it gives orders, it persuades us, it justifies, explains, gives reasons, excuses. It constructs reality. It moves certain people against other people.

125. Apartheid discourse constructed socialised categories, enshrined in the language of laws, which forged differences and distance between groups. As the spiral of conflict escalated and the ANC and PAC turned to armed struggle in the 1960s, so the language of the apartheid security apparatus broadened. From the late 1970s onwards, the language of 'total onslaught and total strategy' enmeshed people increasingly in a discourse of militarism, side-taking and construction of 'enemies'. From the side of the liberation movements, the apartheid regime was similarly constructed as the 'enemy'. A spiral of discourses increasingly dehumanised the 'other', creating the conditions for violence.

126. Language calls people up, motivates people for action. Mr Clive Derby-Lewis testified in his amnesty application for the murder of ANC leader Mr Chris Hani:

Dr Treurnicht had called us up for the third freedom struggle, Mr Chairman, which in Afrikaner history means only one thing.

127. Language instructs and advises people. Here again is Mr Clive Derby-Lewis:

In terms of the Bible teachings we as Christians are told that it is our duty to fight the anti-Christ in whichever way we can the impression I got from Dr Treurnicht was that under certain circumstances it would be permissible to even kill in the battle against the anti-Christ.

128. Former Minister of Law and Order Adriaan Vlok noted with some surprise in hindsight that language could potentially construct a climate of violence, but he conceded eventually that this could be so.

It is a fact that our country, especially during the conflict of the past, was plunged into a war psychosis where words and expressions which were derived from the military became part of the vernacular, just as other expressions with the same import became part of the revolutionary language. At that stage there was nothing unnatural or unusual in the use of these expressions. It is however so, as already said, that with the benefit of hindsight, it is an indisputable fact that there wasn't necessary consideration of the perspectives in interpretations of other people who did not attend those meetings.

I realise with shock now, with shock and dismay that this language usage obviously and apparently gave rise to illegal actions by policemen whereby not only victims were prejudiced but from which also certain negative results came for policemen and their families.

I don't know how the man on the ground saw the position. Perhaps because of the greater pressure we exerted on them, they experienced greater pressure to act illegally and perhaps then we pressurised them to such an extent that it led to people being killed and that policemen landed up in problem situations. Once again it was a case of perceptions which we perhaps had a hand in creating because I said to the policemen and the men on the ground, you have to achieve and perform, you have to solve this problem and this matter. So perhaps, if that led to that kind of pressure, I'm sorry.

129. If Minister Vlok is surprised at the manner in which language could create impressions, Eugene de Kock is quite adamant about the meaning of certain expressions. In testimony, De Kock indicated that orders were usually given in the form of euphemisms such as "go for a drive" ( a person wouldn't return), "had to be removed", "neutralise" "make a plan with these people". De Kock laconically commented that the phrase "take them out" did not mean that "you had to take the person out and entertain them". Referring to the orders to bomb Khotso House, De Kock expressed surprise that in this case the orders were quite clearly to "blow up". Usually, he said, the instruction would be "to shake up a little" or "to put a couple of cracks in the wall". Although the link between language and violence in the South African case has not yet been studied sufficiently and must form part of a future research agenda, the above examples point to the importance of the topic.

130. In the UDF submission to the Commission, the question of language and violence is discussed as follows:

The usage of militant language within the Front took place against a background of increasing struggle and general escalation of violence. We were concerned about this development and discouraged the use of militant rhetoric. But, having looked at this question hard and for a long time among us, we concede that the language used by some of us from time to time could have provided the reasonable basis for some of our members to infer that violence and even killing was acceptable.

131. Ideology is a form of power in which meaning (signification) serves to sustain and reproduce relations of domination. Language, in its many and varied forms, is the central element in ideology as power. As language, ideology 'does things'. In the South African context, it is important to understand how multiple discourses combined, intersected and intertwined to create climates of violence. In this respect, the ideologies of racism, patriarchy, religions, capitalism, apartheid and militarism all intertwined to 'manufacture' people capable of violence. Ideologies in these sorts of combinations provide the means and grounds for people to act violently and yet, ironically, believe they are acting in terms of worthy, noble and morally righteous principles. Thus some Afrikaner nationalists could claim a 'just war' not against black people, but against Communism. There are examples of such rhetoric above. On the other side, with greater legitimacy, the liberation movements could justify violence as a means to a greater end, 'freedom and democracy'. Although that has indeed been the result, the language and slogans deployed could nonetheless justify atrocities of various forms.

132. In this sense, the language of violence takes a form akin to a dialogue, an arrangement of sequences and spirals that enmesh each side and in turn increase the likelihood of violent acts. These steps and sequences have been described as a process of "ideological acceleration". People in political movements take a series of steps which increasingly commit them to their ideological arguments and lead them to distance themselves from outsiders.

The sequence consists of acts of increasing violent contempt for outsiders. It may start with words and uniforms and end in killing.22

133. It is sufficiently plain that language, discourses and ideological processes are important factors in the understanding of the motives of perpetrators. Human beings act in terms of the meanings of particular situations.

Two neglected factors

134. It is frequently unremarked that violence is perpetrated mainly by men. While it needs more research, this chapter has dealt with this neglected area above. Two further factors are also often neglected. The first is the place of special organisations, the second the role of secrecy and silence. Taken together, attention to these matters may enhance understanding of particular contexts of atrocities as well as pointing towards possible remedial actions.

Special organisations

135. Surely it is only some people, not others, and then only a relatively small number, who actually committed atrocities in South Africa. One may be tempted back to characterological explanations, but these, as we have seen, generally run into infertile ground. More fertile soil presents itself in the form of special organisations. People join up or are recruited, and are then selectively drawn deeper into the organisational culture in sequential steps of training, specialised allocation and 'ideological acceleration'. South African history is littered with special groupings of a semi-secretive nature, designed to do either ideological work (the Broederbond) or repressive work, or sometimes both.

136. The military and the police are habitually semi-closed establishments, but within them, given the specialised divisions of labour of modernity, some sections are given even more clandestine tasks: the Bureau of State Security (BOSS), the State Security Council (SSC), the National Security Management System (NSMS), the National Intelligence Service (NIS), Joint Management Centres (JMCs), the Civil Co-operation Bureau (CCB) a euphemism if ever there was one Koevoet, Vlakplaas, the Roodeplaat Research Laboratory, the Afrikaner Weerstandsbeweging (AWB), C10, Stratcom and others that may yet be unearthed. On the other side of the struggle, for somewhat different reasons, there were also specialised organisations. Not least of these were the armed wings, such as MK and APLA, as well as SDUs, which also operated in clandestine ways. Special organisations within Inkatha, such as the Caprivi-trained group, as well as numerous township vigilante groups (such as witdoeke), constitute further examples. Within many of these organisations, yet smaller groups were given the task of special operations. According to John Deegan:

becoming part of the culture you have the police force culture and the plainclothes culture and then you have the Special Branch and within the Special Branch you have still smaller and smaller cliques and inner circles and really there is such clandestine stuff.

137. There is appreciable evidence of the involvement of these organisations and special operations groups in atrocities. Further investigation is needed to explore the modus operandi of such special groups: methods of training, recruitment, hierarchical responsibilities, psychological profiles and the like. Until the work of the Amnesty Committee is complete, final conclusions would be premature. The role and place of specialised groupings in murderous deeds remains an important avenue for future research.

Secrecy and silence

138. Secrecy was particularly characteristic of apartheid rule. The massive curtailment of press freedom, restrictions on academic freedom, a considerable increase in censorship, the banning of organisations all these went hand in hand with secrecy of the security apparatus and even of cabinet and parliamentary procedures. Along with secrecy went silence, and much of the country's populace was silent, through fear, apathy, indifference or genuine lack of information. Finally, some of the victims were silenced through death, being killed because they knew too much, or through imprisonment, detention, threats and torture. Collaborators, spies and double agents did their work in secrecy and silence. The silence of the media, the state and collaborators, along with the secrets held inside sequestered special organisations, all helped to jog the terrible process onwards. Much has emerged through the Commission; the amnesty process is still in progress. Yet many secrets and silences remain in various closets.

139. The antidotes are simple and clear. Open, transparent, accountable government should remain a central priority. Academic freedom and freedom of the press should be inviolable principles. Security forces should be prised open; their operations, budgets and methods of training opened to public scrutiny. Non-accountable vigilante groupings should be regarded with suspicion and concern. If atrocities thrive in the soil and climate of silence and secrecy, one must remove the conditions in which they flower. Much has already been effected through the new Constitution. More remains to be done to cultivate a climate favourable to human rights, in all social institutions.

Two case examples: torturers and 'Mobs'

140. Having considered the array of motives, perspectives and explanations outlined above, let us speculatively apply them in the case of two rather different forms of human rights abuses. Perpetrators' actions cannot necessarily be understood in terms of the same set of factors.

Torturers

141. Although some torture took place at the hands of liberation movements, the bulk of torturers were members of the security police paid state officials using state powers, including draconian laws providing for protracted periods of detention for purposes of interrogation23. In such cases, victims were almost entirely at the mercy of their captors. It was a closed system of state powers, governed by the legal apparatus. As such, it involved forms and paperwork, working shifts, possibilities of job promotion, and lines of hierarchical authority going up, in principle, to ministerial level. Despite the official lines of authority, it was, due to its secrecy, also open to falsification, fabrication of documents, lies and distortions, as evidenced by the operations of Vlakplaas. Senior officials, while ultimately responsible, did not always want to know the details. As Eugene de Kock put it in testimony to the Commission:

They [senior officials] would have a document that, should anything happen, that "this was only a suggestion, we never did it". In other words by means of [euphemistic] language, they removed themselves from the death, from the attack itself. And I'm not trying to place a burden on people, I'm just telling you how it worked in those days.

142. How to describe the motives of torturers? From testimony to the Commission it is clear that most, deeply informed by the ideologies and discourses of apartheid, total onslaught rhetoric, and the masculine and militarised culture of the Security Branch, believed that they were doing their duty. Clearly they perceived themselves as authorised from above. Such people were praised, promoted and received awards for such activities (Eugene de Kock was, for instance, repeatedly decorated). Compliance with the norms of police culture constituted additional binding practices. Egotism and pride in doing the task added positive emotions. Only a minority would have been 'true believers' and only a minority would have learned to become sadists experiencing joy out of hurting; more would have enjoyed the sense of power in such situations. It was not a job done unwillingly.

143. The perspective of torturers and victims would have been grossly discrepant. For the latter, the situation would be engulfed in fear, helplessness and terror. For the torturer, the situation would have been a means to an end (breaking a 'bolshie' victim, extracting information, exerting power, doing the job) and somewhat routinised and banal, done in shifts. A combination of such factors, differing to various extents among individuals, would have been sufficient to lead to repeated atrocities. There was little evidence before the Commission that any such perpetrators were suffering from severe psychological abnormalities. Stress, certainly quite commonly expressed, would have been a consequence rather than a cause. Many may have felt shame, remorse and regret. Under entirely changed circumstances, there would be little likelihood of the recurrence of such actions.

Crowd violence

144. This constituted a very different situation. In the majority of cases of 'necklace' murders, for instance, victims were members of the same community. Events were fast-paced, apparently emotionally charged and relatively spontaneous. No legal apparatus, bureaucracy and hierarchical authorisation was involved. Perpetrators were, in the main, young men, aligned to liberation movements and linked to youth organisations, bearing the social identities of 'comrades'. Targets of attack were repeatedly people seen as linked to the apartheid system (councillors or their families, police, sell-outs) and invariably rumoured to be, or identified whether justifiably or not as impimpis (informers).

145. The dominant explanation of such atrocities is that of the 'crazed mob'; of people who are out of control, irrational, over-emotional; in the formal nomenclature of psychology, in a state of deindividuation. Violence is apparently due to a loss, a lack, a reduction or regression to more 'primitive' forms of behaviour. There is however an alternative picture of crowds: the sequence of action was far more patterned, directed and limited than usually depicted. Crowd violence was invariably preceded by a series of violent incidents, mostly at the hands of security forces and often leading to the deaths of community members. Crowd violence was directed only at particular targets: people believed to be impimpis, or places symbolic of apartheid oppression beer halls, local council buildings, police vehicles. It was not simply random violence.

146. A better explanation comes from social identity theory. While there certainly is a switch that makes people see, think and act in a manner quite different from that of an isolated individual, it is a shift from personal to social identity rather than from individual rationality to a loss of identity or control. Crowd violence is an instance of inter-group action in which particular, local identities (for example, 'comrades' versus 'sell-outs') become salient. People act violently not because they are out of their minds, but because they are acting in terms of a social frame of reference. Emotions ran high because the struggle against apartheid was seen in strongly emotional terms of taking sides against the 'enemy' or against the 'system' of oppression. Lives, quite literally, were on the line. Within such situations, perpetrators became bound and 'sucked in' by the sequences and meanings of the particular events, but it is the salience of local identities, on different sides, that structure the situation. Again it is not psychological dysfunctions that account for the actions. Social explanations are both more plausible and more coherent. Implications are that with changed circumstances, perpetrators are not likely to commit such offences again.

Prevention of atrocities

147. If the above descriptions of motives and explanations have merit, then steps towards prevention of future atrocities are quite clear. If political circumstances literally power arrangements in a social order constitute the primary explanation, such circumstances must be changed. In South Africa this has already been effected. The dramatic changes which have ushered in the new principles of democracy, non-racialism, non-sexism, and equal opportunity citizenship in a unified state are major steps in the right direction. However, until real economic inequalities are eliminated, until equal opportunities become feasible realities, such noble ideas and principles remain under partial threat.

148. As an important first step towards the prevention of future possibilities of crimes of obedience, the South African Constitution states in Chapter 11, section 199, that:

The security services must act, and must teach and require their members to act, in accordance with the Constitution and the law.

No member of any security service may obey a manifestly illegal order.

149. If secrecy and silence and clandestine organisations provide fertile ground for evil deeds, then solutions lie in open, transparent and accountable social institutions. Since security forces, private armies and vigilante groups constitute particular sites of recruitment, training, propaganda and promotion of violence, these sites demand special scrutiny. Open scrutiny by the public seems the most powerful rehabilitation device. Freedom of the mass media, academic freedom, and the role of civil society as watchdogs are all vital.

150. Since ideologies, discourses and language codes are the constituent grounds for social identities of difference, disparagement and disgust and for inter-group cleavages based on hostility, resentment, suspicion and revenge, these factors demand sharp vigilance and radical change. The various Commissions recently established provide a good start. The vigorous promotion of a culture of human rights, of equality and mutual respect in every sector, is of paramount importance. Particular attention needs to be given to language codes that promote, quite subtly, images of hatred, distance and disparagement between groups.

151. Obedience to authorities, compliance with group norms and the power of the immediate situation were all identified as dangers. Encouragement of dissent, the power of minor influence and the promotion of dialogue, negotiation and multi-vocality all constitute steps toward prevention.

152. If crowds are a potential seedbed for violence, they require adequate channels for expressing voice and opening dialogue. The new Regulation of Gatherings Act is a vast improvement. This Act will require further education and promotion to establish freedom of association, the right of protest and effective channels for dialogue as part of the daily bread of the fledgling democratic order.

153. These few ideas, neither too lengthy nor too cumbersome, would seem to be but a limited burden to effect the future prevention of atrocities.

This resource is hosted by the Nelson Mandela Centre of Memory, but was compiled and authored by Padraig O’Malley. Return to the Nelson Mandela Centre of Memory site.