Simpson, G. (1994). Business and Endemic Violence in South Africa: Surviving the disaster or managing the transition? Paper presented at The 6th South African Disaster Recovery Conference, Midrand, 16 - 17 March.


Graeme Simpson

Paper presented at the 6th South African Disaster Recovery Conference, Volkswagen Conference Centre, Midrand, 16-17 March 1994.

Graeme Simpson is a founder and former Executive Director of the Centre for the Study of Violence and Reconciliation.


It has been argued that the legacy of apartheid has bequeathed to South Africa a 'culture of violence'. This has been rooted in the notion that violence in South Africa has become normative rather than deviant and it has come to be regarded as an appropriate means of resolving social, political and even domestic conflict. This is quite easily visible across the entire political spectrum, where violence has been sanctioned as a means both of maintaining political power, as well as an accepted means of attaining change or resolving conflict.

The consequence has been that, despite the prospects of peace heralded by the process of national political negotiations, the past four years have been amongst the most brutal in this country's history.

Part of the explanation for this lies in the fact that the era of negotiations established the terrain for an intensified political context. For the first time in this country's history, the key political interest groups had to establish their credentials in terms of their national representivity if they were to occupy a seat at the negotiation table with any degree of clout. The resultant power struggle, involving all the key political interest groups has, almost inevitably, played itself out through the style of violent confrontation so firmly established in the preceding decades.

The racially-based, hostile stereotypes generated by apartheid, coupled with the resultant political intolerance, have continued to articulate closely with the experiences of economic impoverishment and encroaching poverty for the majority of South Africans. In the absence of an effective social welfare net and in the context of dramatic levels of unemployment, conditions are created which offer a solid foundation for the social, political and criminal violence which pervades South Africa.

Yet, considering the extent to which bread and butter issues such as housing, jobs, rents, wages and education, have become politicised in South Africa, it should come as no surprise that people have developed very high expectations of the process of political change. The constant stalling of negotiations has created a climate of extreme impatience and frustration which has enhanced the prospect of violence. Once the collective inhibitions of the resort to violence as a means of resolving conflict have been overcome within this broad political culture, it is inevitable that the resultant violence will begin to spill over into the social and domestic arenas of society - the workplace, the home and in the communities. The result is that violence begins to intrude into all these dimensions of social life, often manifesting itself through conflict over the most basic resources within impoverished communities.

Individuals, feeling powerless or helpless in the face of dramatic social and economic upheaval, frequently symbolically reassert their power through violence in those dimensions of their lives which they still feel they hold sway. This results in much aggression which, although social or political at root, is expressed through displaced violence within the family and in the home.

It is not possible in a paper as brief as this, to document, analyse, or even describe the violence in South Africa in all its complexity. What this brief outline should demonstrate is that the search for mono-causal explanation is fruitless. The convenient terms in which the violence has been labelled, by politicians and the commercial media, often does more to disguise complex causation than it does to explain it.

The violence has been variously labelled as 'black on black' violence, ethnic conflict, conflict between hostel dwellers and squatters or township residents, conflict between ANC and IFP supporters, or between the police and township residents. It has been referred to as violence between the poor and the very poor, conflict generated by government or by a 'third force', or it has been described simply as violent crime. None of these descriptions is simply wrong. Yet none, on its own, will properly explain this complex situation. It is only when South Africans accept that we are dealing with a host of overlapping causal factors that we can begin to address the problems constructively.

Even statistical information is an inadequate yardstick of the problem. The notorious lack of reliability and the politically contested nature of most of the statistics which are generated about political and criminal violence, make it difficult to rely on this sort of information. There has been a prolonged 'battle of the statisticians' on this issue, relating not only to political violence, but to crime statistics as well. It is argued here that, at very best, the sort of 'body-count' statistics generated, if taken together, only give us a limited indication of what is actually going on. They are generally plagued by the problems of under-reporting or by perceptions of bias by those gathering the data. At worst, it could be argued that the picture which is painted in statistical terms, has the ironic effect of numbing us to the human suffering which the statistics ostensibly convey - instead of guiding us in the development of the remedial and preventative steps which need to be taken.

Some Recent Trends in the Violence

Despite what has been written above, some trends over the past years remain discernable and are instructive. Amidst the competing claims over who is most responsible and over whose political interests are best served by the violence, certain trends remain clear:

  1. The transitionary phase in South Africa's history, heralded by the move towards a negotiated settlement, has had the effect of discrediting the traditional apartheid-based mechanisms of authority regulating the society. In particular, this has meant that old forms of repressive control exercised through the deployment of security forces, have become untenable during this era. However, the failure - or at least the haphazard nature - of the negotiation process, has meant that alternative forms of consensus-based social regulation have not been effectively forged. The result is a 'window period' in our history in which the society is under-regulated by any legitimate source of authority. This has generated a climate of lawlessness and has facilitated the resort to 'legal self-help', thereby contributing to the spiral of revenge, retribution and increased violence.

  2. The contention that much of the violence can be explained away in terms of ethnic conflict requires some discussion here. It would indeed be somewhat surprising if the ethnic orientation of 40 years of Apartheid did not, in some way, shape the identities of people living under the system. However, it is really where racial and ethnic identities correlate with conflicts over material resources or political allegiances that their potential as vehicles of violence becomes explicable. The short-hand description of township conflict as 'black on black' violence, so often resorted to by the commercial media, therefore serves to disguise the underlying causes of violence rather than to elucidate them. It reinforces the unhelpful notions of 'black barbarism' and feeds the fear which for decades has been the foundation of white politics under apartheid. The real danger, however, resides in strategies of political organisations such as the IFP, the National Party, far right political organisations and to some extent the Pan African Congress, which deliberately mobilise political support in ethnic or cultural terms. Once mobilised in violent situations, such ethnic-based political identities, like religious identities, become particularly volatile, defensive and extremely difficult to control or to demobilise.

  3. Third force destabilisation. Many researchers have argued that the close correlation of upsurges in the violence with key points in the negotiation process, indicates the clear involvement of a 'third force' which has a political interest in destabilising and, if possible, derailing the negotiation process. It is argued and seems clear that both within the state security establishment, as well as within right wing groupings and even within the government, there are those who have such an interest in undermining the process of change. However, there is a danger of an over-simplified conspiracy theory which does not adequately explain the nature of such destabilisation. For example, it is clear that it is not simply these white political activist groups who are 'pulling the triggers' - that is, who are directly involved in perpetrating such acts of violence. It is argued here that a properly sophisticated understanding of 'third force' involvement in the violence, must entail a distinction between such politically motivated interests in destabilisation on one hand, and the 'war-based', materially-rooted interests on the other. Those who seek to disrupt the process for political purposes, are clearly reliant on the assistance of surrogates who have developed a more materialist interest in ongoing violence. In poverty-markets, violent conflict quickly generates its own sub-economy - based on the trade in arms, assassinations and protection. The insecure climate, coupled with the quest for defence and the resultant proliferation of private armies, generates ready-made foot-soldiers in the conflicts described. Candidates are readily found amongst disillusioned youth, criminal gangs and within the defensive and insecure hostel populations.

  4. The dramatic growth in right wing violence. The essential context to violence perpetrated by ultra-right groupings is the proliferation of both mainstream and fringe groupings in the wake of the un-banning of the liberation movements in 1990. The choice to remain outside of the negotiation process has also locked these political groups into a highly militarised and conflict-orientated political discourse which has been further serviced by the popular ethnic-based, racist explanations of township conflict. The overwhelming defeat of parliamentary right-wing groupings by the Nationalist government in the all-white referendum of March 17 1992, probably served to consolidate this position. The Human Rights Commission has argued that: '… [r]ight wing actions [are] a mixture of organised premeditated attacks, and of impulsive and irrational outbursts against innocent victims.' Perhaps the most significant recent attacks have been the invasion of the multi-party talks at Kempton Park, the involvement of high-profiled right wingers in the assassination of South African Communist Party leader, Chris Hani and the spate of acts of sabotage which seem likely to continue into the election period.

  5. Despite competing claims that particular political interest groups have been responsible for 'switching on the violence' in order to control the negotiation process, it is clear that the violence has developed a momentum of its own. This has rendered it much less possible for any of these interest groups to simply 'switch the violence off'. In the final analysis, it is clear that the levels of violence have become central to further stalling the national negotiation process.

  6. The independent dynamic which the violence has developed is reflected in the increasingly blurred dividing line between political and criminal violence. This contributes to explaining both the dramatic increase in criminal violence and the extent to which criminal violence has become politicised. There have been reports of a dramatic increase in the levels of violent crime over the past two years. It was reported by the Minister of Law and Order that in 1992 there were 20 135 murders in South Africa (up from 14 693 in the previous year), more than 24 700 rapes and 79 927 robberies. In total this means that there were 77 South Africans murdered, 68 raped and 775 assaulted, on average, every day during 1992. There were 219 robberies, 201 car thefts and 709 household burglaries each day. The South African murder rate of approximately 50 people per 100 000 in 1992, was 5.5 times higher than the US average for the same year.

  7. The above trend is facilitated by the poverty, unemployment and consequent conflict over scarce resources which dominates township communities and which often (although not exclusively) lies at root of the involvement of the most economically impoverished communities such as squatters or hostel dwellers. An illustrative example of this has been the increase in 'taxi wars' in the PWV and the Western Cape. At origin a conflict over relatively lucrative transport routes in the deregulated transport sector, the taxi war in the Western Cape rapidly translated into a conflict of political affiliations. Ironically the effect of the structures of the National Peace Accord has been to occasionally compound this problem through framing such conflicts (and the resolution of them) in terms of politically defined interest groups and parties. The result is that violence emanating from a range of ostensibly non-political sources, rapidly becomes politicised.

  8. A related trend has been the failure of the state's law enforcement agencies to establish their good faith and thus to develop any trust-based relationship with township communities through meaningful reform of the SAP. This has resulted in an escalation of violence through the resort to informal justice with a consequential spiral of revenge and retribution. In the process, the dividing line between political and criminal violence has once again become increasingly blurred, resulting, whether by accident or by design, in the cumulative effect of further political destabilisation. This also lies at root of the ongoing attacks on members of the security forces, who themselves have been regularly implicated in alleged partisan behaviour in the violence.

  9. Perhaps the most significant trend has been the increasingly arbitrary and random nature of the violence as victims have been less and less politically selected. One example of this has been the frequent random attacks on mourners at funerals and vigils, and massacres at beer-halls in the PWV region. An even more striking example has been the violence on the trains. The recent upsurge in arbitrary attacks, the renewal of massacres on the trains and the increase in allegedly racially motivated attacks on white South Africans such as the Kenilworth Church and Heidelberg Tavern massacres, are further indicators of this trend. In particular, the massive upsurge of violence on the East Rand reflects the continuation of many of the processes outlined above, rather than an entirely new phenomenon. The immediate result of the arbitrary and random nature of violence of this sort, is that it further entrenches widespread feelings of insecurity and fear, which in turn often lead to forms of violence which are rationalised as being defensive in nature.

  10. It has already been noted that much of the social tension, fear and insecurity, manifests in the form of displaced aggression within the domestic or family arena. In this manner, we are seeing the effective domestication of much of the conflict in South Africa. Here, out of the view of the public eye, it is those structurally weaker members of our society - women, children, the elderly - who bear the brunt of a violent victimisation which the crime statistics cannot even begin to penetrate. Child abuse, marital rape, wife battery, etc. are all encountered more and more frequently. Such domestic violence is at its most brutal when it coincides with racial tensions such as in the ever-increasing attacks by white homeowners on their black domestic workers, on farm workers, or the regularly reported slaying of elderly white farmers.

  11. Another arena into which violence threatens to spread is the working environment. Despite the achievements of the system of collective bargaining in institutionalising conflict in the industrial context, the relative peace of the workplace cannot be taken for granted. The relationship between the community and the workplace is a reciprocal one and there is the real danger that, unless pro-active steps are taken, community violence will pollute the relative peace of the workplace. The manifestations are often less obvious than may be expected - ordinary workplace relations may become the vehicle for expressing frustration, low levels of concentration may result in increased industrial accidents, alcohol and substance abuse may become more prevalent, absenteeism may increase or productivity may drop considerably or, at worst, the workplace could become the scene of actual violence. In any event, the costs of high levels of violence in financial terms is astronomical. In one article, it was conservatively estimated that the costs of political unrest alone, prior even to the outbreak of the so-called Transvaal War in August 1990, amounted to as much as R3 billion. The cost in human terms is clearly much greater.

Transition Management - Spreading the peace1

In the past five or six year, employers, trade unionists, researchers and South Africa's criminal and industrial courts, have all confronted the escalating magnitude and brutality of violence in the course of industrial conflict.2 However, these are merely some of the more extreme incidences of the intrusion of violence within industrial relations in South Africa and this paper will not specifically address this issue of strike-related violence.

However, in examining some of these conflicts, authors such as Webster and Simpson have argued that, considering the extent to which violence has dominated the political culture of South African society - and considering the historical politicisation of the workplace - the intrusion of violence in industrial relations is an inevitable consequence of the reciprocal relationship between community and workplace, rather than merely the consequence of the psychological profiles of the individual perpetrators. In this vein, Webster and Simpson argued that attempts to refine collective bargaining so as to institutionalise conflict at the workplace could, at best, only have a limited effect in preventing violence at work - if this was not accompanied by the broader empowerment of black working class communities as a whole.3

However, in the context of dramatic political change since 1990 - which has not surprisingly been associated with increased rather than decreased levels of political and criminal violence - the vision of most employers has remained rather conservatively limited to futile attempts to 'insulate' or 'protect' the workplace from encroaching violence, rather than engaging in any way with the origins of the problem beyond the factory gates. Furthermore, where they have been evident, employer concerns over community violence have frequently manifested themselves as limited concerns with 'productivity' and have more often than not been 'relegated' to the arena of employee care programmes aimed at treating individual victims within the workforce. The result is that in-house programmes have tended to address the symptoms rather than the causes of violence and, in so doing, have been inclined to individualise the problem instead of engaging with its collective implications and its broader influence on the industrial relations context. Another consequence of this conservative approach has been to effectively draw a clear line of distinction between company 'insiders' and those members of the wider community who have no such direct access to corporate care programmes. Perhaps most important is the fact that this conservative view is premised on the assumption that violence only affects workplace relations though direct victimisation or when industrial conflict is most extreme, such as in strike situations.

It is argued here that this traditional approach ignores the influence of both direct and indirect experiences of violence on workplace relationships and ultimately constitutes the 'non-management' of the transition process in South Africa on the part of employers and managers - despite frequent demands and pleas from worker representatives and victimised communities for business organisations to get involved. The result is that the potential of the workplace as an agency of social change is severely under-utilised, and the narrowly conceived strategies to insulate industrial relations thus actively undermines the potential of harnessing the 'relative peace' of the workplace (that is, relative to the pervasive violence which dominates community life) as a vehicle for generating a broader peace process.

This fact appeared to be partially recognised in the 1993 draft bill on Occupational Health and Safety4 which, under Section 13(1), proposed to impose extensive duties on employers to take pro-active measures to protect employees and to prevent violence at the workplace. These proposed duties included:

… an action plan which provides for the measures necessary to protect such employees including, but not limited to, training, the provision of security measures, the adaptation of procedures, aid to victims and, generally, all other measures necessary under the circumstances to give effect to the employer's duty contemplated in this section.5

Despite the limited ambit of the bill in addressing violence at work, the final Occupational Health and Safety Act of 1993 excluded these proposed measures altogether, merely providing for a general duty of care on the part of employers, including a duty to ensure a safe working environment.6

Yet recent research conducted by the Centre for the Study of Violence and Reconciliation has shown that community-based violence (whether political or criminal) and the trauma associated with both direct victimisation and with stress resulting from potential or indirect victimisation, has had the effect of broadly 'polluting' workplace relationships. These stress factors, affecting shop-floor employees, middle managers and senior managers alike, serve to undermine channels of communication in the workplace, detrimentally intensify ordinary workplace conflicts and dramatically intrude in the capacity to generate non-adversarial collective bargaining approaches. In all these respects, the situation cries out for effective pro-active strategies which take cognisance of the extensive and collective impact of community violence on industrial relations and which are rooted in the informed monitoring of the frequently disguised effects of violence on workplace relationships and performance.

In needs assessment workshops conducted within a range of business organisations in several regions in the country and involving workers and worker representatives as well as supervisors and senior managers, a number of 'disguised' manifestations of violence-related stress became evident in the workplace and were identified as negatively affecting the industrial relations context. These included such issues as increased racial tensions, conflicts between workers (including between union and no-union members or between members of competing trade unions), loss of concentration resulting in increased industrial accidents, sleeping on the job, increased alcohol-related disciplinary infringements and a range of other issues. It was strikingly significant that incidents of direct political violence at work were relatively isolated. Even more significant was the fact that despite differing lifestyles and social experiences, the fears and concerns of senior managers and shop-floor workers were strikingly similar in respect of both criminal and political violence beyond the factory gates.

The ironic indicator is that the very violence which is seen as a most divisive feature of South African social life, actually operates as a shared problem in those workplaces where there is some willingness to engage with the issue in terms which acknowledge its impact both within industrial relations in general and beyond the shop-floor. This has the potential to facilitate joint problem solving strategies which service greater worker participation in managing the stress associated with social transition. Equally important is the extent to which open workplace discussion forums on violence, also facilitate non-adversarial engagement with issues of housing, transportation, insurance schemes, medical and legal aid schemes (amongst others) as partial solutions to the problems of violence-related trauma - where these very issues have been stumbling blocks in collective bargaining processes.

There is a vital need - and equally vital advantages - for business and trade union leaders to jointly engage in interactive planning within the industrial relations context in order to harness the potential of the working environment as a proactive arena of peaceful social change, whilst simultaneously addressing the concrete needs of the most victimised township communities.7 This entails some commitment to victim-driven programmes and, therefore, to increased worker participation in generating solutions within the workplace.

To this end the Centre for the Study of Violence and Reconciliation has evolved a five-pronged approach for dealing with the impact of violence on industrial relations:8

  • Monitoring and Budgeting

    It is critical that business organisations effectively monitor the disguised impact of extra-workplace violence on their operations and on workplace relationships. This must be done systematically and may entail additional in-house training to ensure that workplace personnel are adequately equipped to identify and measure the impact of the hidden symptoms discussed above. Perhaps most important is the need to scientifically cost the effects of violence on the organisation and to consciously build such hidden cost factors into the organisation's budgets and financial planning processes.

  • Victim Aid

    Companies should recognise the workplace as a vital point of contact for initiating treatment of violence-related trauma. Whilst such treatment may appear to be reactive, if appropriately developed and administered, it is a pro-active mechanism of empowerment which plays a crucial role in intervening in the cyclical nature of violence in the society and in the workplace.

  • Training

    Employees must be trained in order to support and counsel traumatised co-workers. More importantly, training is necessary in order to develop the skills in monitoring and identifying the 'disguised' impact of violence related-stress on workplace relationships, so as to facilitate directed joint problem-solving strategies to deal with problems as they arise. In addition to the training of in-house 'care workers', there is a vital role for the training of managers as change agents - equipping managers to effectively manage the process of transition.

  • Education and Communication

    Cross-sectional communication forums are essential in generating information and sensitively to the shared problems of violence and their influence on the industrial relations context. It is essential that such forums provide victims of violence with a voice within the workplace and that they operate as a vehicle for management education in relation to violence. It is these forums which can operate as the foundation stone for generating creative in-house and broader strategies and for building reciprocal understanding in the workplace which will become functional to a more peaceful industrial relations context. Most importantly, such forums can operate as the foundation stone of greater workplace democracy through facilitating broader participation by the workforce in the search for practical solutions.

  • Corporate Community Involvement

    Consistent with any attempt to address the impact of violence in the workplace, is a programme of community engagement. This may take several forms including: employer and trade union involvement in national and/or local peace structures, involvement in conflict resolution processes beyond the shop-floor, violence monitoring and community development initiatives. In respect of the latter, it is equally important that worker or community representatives become party to the strategic decision-making processes in respect of the allocation of those resources earmarked for corporate social upliftment programmes.

In conclusion, the potential of violence in industrial relations and the experiences of trauma resulting from high levels of community violence are fundamentally inseparable. The challenge in dealing with the problem touches directly on the role of the industrial community beyond the confines of the workplace. Ultimately, if business is to play a role to its full potential, it needs to do so as an active agency for peaceful social change.


1 This section of the paper draws extensively upon another paper written by the author. See: Simpson, G. "Violence and Industrial Relations", in Employment Law, Vol.10, No.3 (January 1994), pp.55-58.

2 See: Anstey, M. "Process Analysis of the SATS Strike in Terms of Conflict Escalation", Report prepared in State v Sibisi and 17 Others, Unreported; Bendix, W., Bibb, P., and Ibbotsen, J. "The 1987 SATS Strike: An IR Institutional Environmental Perspective", South African Journal of Industrial Relations, First Quarter (1988); Segal, L. And Simpson, G. "'Off the Rails': Violence in the Railway Strike of 1987", ASSA Conference Paper (July 1990); Von Holdt, K. "Violence in the Workplace", South African Labour Bulletin, Vol.14, No. 3 (1989); Vogelman, L. "Some Psychological Factors to Consider in Strikes, Collective Violence and the Killing of Non-Strikers", Violence in Contemporary South Africa, South African Breweries (SAB) conference proceedings, Johannesburg, September; Vol. 14, No. 3 (1989); and Webster, E. and Simpson, G. "Crossing the Picket Line: Violence in Industrial Conflict", South African Journal of Industrial Relations, Vol. 11, No.4 (1991).

3 Opcit., pp.29-31.

4 Government Gazette No. 13978, 11 May 1992.

5 S 13 (2) (b).

6 Act No. 85, Section 8

7 On this issue see: Dor, G. "East Rand Violence: Building Unity at Work", South African Labour Bulletin, Vol.17, No.5 (September/October 1993), especially at pp. 40-41.

8 For more detailed information on this programmatic approach, see: Simpson, G. And Vogelman, L. "Overcoming Violence: The Role of Business", Innes Labour Brief, Special Brief (October 1992); and Simpson, G. "Violence and Social Change: Some Effects on the Workplace and Some Possible Solutions", Paper presented to the national EAP Committee of the Institute for Personnel Management Conference on "The Role of Employee Assistance Programmes in Managing Trauma", Johannesburg (August 1991).

© Centre for the Study of Violence and Reconciliation

Copyright © 2021 CSVR. All Rights Reserved.