Crossing the Picket Line: Violence in industrial conflict -the case of the Afcol strike

Crossing the Picket Line: Violence in industrial conflict -the case of the Afcol strike

Webster, E. & Simpson, G. (1990). Crossing the Picket Line: Violence in industrial conflict -the case of the Afcol strike. In Industrial Relations Journal of South Africa, Vol. 11, No. 4.


Eddie Webster & Graeme Simpson

In Industrial Relations Journal of South Africa, Vol. 11, No. 4, pp. 15-32, 1990.

Eddie Webster is Head of the Sociology Department at the University of the Witwatersrand.

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


The mid-1980s has witnessed increasing violence in the course of industrial disputes. This violence has not only increased quantitatively, but has also become increasingly brutal. This paper is concerned with physical violence directed against people rather than property, as there is an identifiable trend towards more attacks against people over the last five or six years in South Africa. The result has been an escalation in criminal charges emanating from labour disputes. This is turn has resulted in a greater focus on sociological and psychological explanations for violence in industrial disputes which have come before the courts in the form of arguments in mitigation or extenuation.

The Afcol case (State v Elias Phasha and Four Others) is just one recent example where both sociological and psychological evidence was led as extenuation before sentencing in a criminal case emanating from a labour dispute. This case revolved around the murder of five 'scabs' by striking workers during the period 10 to 17 June 1988. Everyone of the five were killed quite brutally, receiving multiple stab wounds before being thrown off moving trains.

This paper will deal only with the sociological evidence brought before the court, although psychological evidence was also led in extenuation. In extenuation, the accused were sentenced to between four and eleven years in prison, in the latter case on three counts of murder.

The fact that these social scientific arguments are subjected to the scrutiny of the courts, poses a number of problems for sociologists and psychologists. Defence attorneys are inclined to be more concerned with making a reasonable case than with the nuance usually required in fully explaining the relationship between collective experience and individual actions. the result may be an inclination on the part of the legal team to use such evidence as serves their purposes, and sociologists and psychologists have to guard against legal shortcuts being substituted for scientific explanation outside of the courtroom environment. This study was largely based on evidence before the court, including an 'agreed statement of facts'. The result may be a limitation in as much as a range of participants in the strike and the ensuing events remain outside the scope of the paper for the simple reason that the court did not see it as relevant to assess their guilt or innocence.

A perfect example of the dangers implicit in such legal shortcuts resided in the potential contradictions which may exist between sociological and psychological explanations of why the perpetrators of violence in industrial disputes are who they are.

Before outlining the sociological argument used in the Afcol case, some introductory comments on the relationship between sociological and psychological explanations of violence in industrial conflict should be made.

There are several recent cases in which both sociological and psychological evidence have been used side by side in extenuation cases. The Afcol case is one such example. Another is the SATS trial which resulted from the murder of 4 'scabs' in the course of the railways strike of 1987. In many such cases, psychological evidence has been used to explain why certain individuals resorted to violence through the use of the notion of 'deindividuation' (Diener, 1979). This notion has usually been invoked to explain violence in a crowd situation characterised by heightened levels of arousal, anonymity, a consequent loss of self-awareness and breakdown of self-control. The result, it is argued, is disinhibition and the resort to uncharacteristic or 'irrational' forms of behaviour, such as the perpetration of violence.

By contrast, sociological explanations of violence in industrial conflict has often focussed on the high levels of frustration experiences, the lack of channels for the expression of these frustrations and failure to institutionalise conflict resulting in its escalation and the increased likelihood of violence. In contrast to psychological explanations, sociologists are inclined to explain violence in the identical situation as a 'rational' and possibly even inevitable consequence of given circumstances.

This apparent contradiction is often lost on the court, which might be quite arbitrary in finding extenuation to exist in some cases and not in others. However, for sociologists and psychologists, the dilemma has to be confronted and handled rather more sensitively.

It is appropriate to used deindividuation as a means of explanation in some cases and not in others. However, where it is used to explain violence, the deindividuation argument should not be separated from sociological explanation, but should rather be a part of it. What is irrational about violence in industrial disputes is often not the commission of violence per se, as much as the brutal or extreme nature of the violence committed. In fact, it is arguable that even the most brutal violence can, on occasion, be explained as entirely rational. If sociological and psychological explanations are not integrated in this manner, then they are likely to beg more questions about the relationship between individual action and collective experience than they are able to answer.

This paper is limited to the examination of just one particular strike and the violence which occurred within it. Furthermore, the inability to examine the individual profiles of the accused, as well as those of the other actors makes it difficult to offer a truly integrated psycho-sociological analysis. However, it is attempted by means of this case study to explain the dynamic interplay between social influences and individual experience in the strike situation more generally. These arguments are in no sense an attempt to condone violence in the workplace, but merely to attempt to explain the social context within which the perpetrators of violence act. In so doing, it is believed that a vital gap in the current system of dispute resolution and of public order maintenance in South African industrial conflict would have been identified.

The paper roughly follows the structure of the report submitted to the Supreme Court (Webster, 1989) and is divided into four sections. In the first, it is shown how the right to conduct an effective picket is the central mechanism for containing frustration and possible violence in a strike situation. In section 2, it is attempted to show how frustration increases and violence escalates when an effective picket is not possible. In section 3, the model developed in section 2 is the case of the Afcol strike. Finally, it is attempted to draw some conclusions and to outline some areas for further research which look beyond the Afcol case.

The Argument

Industrial relations involves the strategic use of power, especially in strike situations. Management will generally attempt to divide the workforce whilst the union will attempt to establish and maintain collective solidarity amongst its members.

In practice this means that the union requires some sort of sanction which can be used to unsure that the majority decision is carried out. When a minority of workers rejects a democratic decision and attempts to 'scab', or where 'black-legs' are employed to replace striking workers, thereby threatening to break the strike, this usually results in great bitterness. It is important that institutionalised ways should exist for dealing with disputes of this kind, otherwise they are likely to be resolved by violence. Violence against non-strikers is a universal feature of industrial history. One of the most important institutions for containing this violence is the picket.

The picket is a technique for bringing moral pressure to bear on would-be non-strikers. From the point of view of the strikers, the individual worker should be morally bound by the majority decision to strike, and a peaceful picket of workers at the entrance to the place of work is designed to make this moral duty clearly visible. At no point, however, does this collective moral appeal override the right of an individual to work (Webster, 1979).

Moral pressure can be brought to bear by picketers standing in line at a factory gate and displaying placards informing the public of the dispute, or by marching in a procession while displaying placards. Picketers might also use verbal please, sing songs, chant or they might distribute leaflets to the public in general or to a specific group of people in order to try to persuade them of the merits of their cause (Napier & McBride, 1986). The result of an effective picket is, therefore, to give the participants in the strike a feeling of controlling their involvement in the strike and a commitment to the negotiating process.

Industrial picketing has for several decades been considered a legitimate device in the industrial environments of the United States, the United Kingdom and other Western countries to reinforce the strike weapon. The legality of picketing was established over a fairly long period in the United States, beginning with the passing of the Labour Management Relations (Taft-Harley) Act of 1947. Picketing in Britain is regulated and protected by Section 15 of the Trade and Labour Relations Act of 1947, as amended by the Employments Act of 1980 and 1982. In the United Kingdom, the right to picket is really established within common law rather than within statutory enactments.

The right to picket, along with political enfranchisement, played a crucial role in the decline of violence in industrial disputes in the United Kingdom. Indeed, in the last decade of the 19th century, on of the key features which gave rise to the formation of the British Labour Party was the demand for the right to picket (Pelling, 1976). Geary (1985) argues that the "… constitutionalisation of the working class …" through the development of a strong trade union movement associated in large part with the Labour Party has, on the whole, served to restrict industrial violence.

Since picketing became an established practice in the 1890s in the United Kingdom, only one non-striker has been killed. It would be an exaggeration to attribute the peaceful resolution of conflicts solely to the fact that the right to picket exists. The close association of the trade union movement with a political party committed to democratic reform rather than revolutionary change, has had a restraining effect on the behaviour of strikers. Union officials and strike leaders have been anxious to avoid violence during industrial disputes on the probably correct assumption that such action would have a detrimental effect on the Labour Party's fortunes.

This contrasts substantially with the current situation in South Africa where no such restraining effect exists. Indeed, the lack of political rights among black workers has led to the opposite effect to that described by Geary (1985) in the British example. Instead of a political party acting as a restraining force, the frustrations and lessons of the wider community have played a key role in determining the path of industrial relations disputes, at least since the mid-1980s. This point can best be understood by a brief periodisation of the recent history of black trade unionism.

It is possible to divide the recent history of black South African trade unionism into three broad periods:

  • From 1973 to 1979, these unions were involved in a struggle for recognition. Although a number of strikes took place, these were of short duration as the unions were small and weak.

  • In 1979, there emerging unions won statutory recognition with the amendment to the Labour Relations Act. These amendments started a different phase in the development of these unions as they began to consolidate their organisational presence in South African workplaces. In retrospect, this period was something of a 'honeymoon' period for the South African industrial relations system as the workplace remained relatively isolated from the wider societal tensions building up during the early 1980s.

  • A third period began in 1984 when the trade union movement was drawn into the wider political struggles that were to engulf black society. Struggles in three separate spheres – in the townships over the Community Councils, in the schools over Bantu Education, and in the workplaces over wages and working conditions – culminated in a large scale stay-away in the Transvaal in November 1984 (Labour Monitoring Group, 1985). The politicisation of the factory floor (Lambert & Webster, 1988) and of South African trade unionism since 1984 has brought into the industrial relations environment the frustration, aggression and violence of the wider society. So, too, does it witness the permeation and influence from the wider society of the idea that social problems can be solved by resorting to violence (Cock & Nathan, 1989).

Ralf Dahrendorf (1959) concludes his classic study on industrial conflict by arguing that "… industrial conflict has become less violent because its existence has been accepted and its manifestations have been socially regulated". In the United States, where large-scale stoppages often accompany the negotiation of company-wide agreements, these have been ritualistic engagements: with unions co-operating in an orderly shutdown of production and companies often providing shelter and refreshments for pickets. Even the British General Strike of 1926 – symbolically the apogee of union militancy – is today, at times, remembered primarily as the occasion for football matches between police and strikers.

Under the provisions of the 1980 Employment Act, the British government published a 'Code of Practice for Picketing' to supplement the provisions of the Act itself. The Code outlines the relevant law and, in addition, gives practical guidance on picketing procedures. The Code sets out the role of the police, provides guidelines on limiting the number of pickets and provides for the organisation of picketing. The result is that a friendly relationship often develops between the pickets and the police. For example, at one colliery during the 1984-85 miners' strike, the village policeman even gave the strikers lifts in his police car to and from the picket line (Geary, 1985).

If the right to strike is accepted as integral to collective bargaining, then the right to picket must be accepted as an ancillary to the right to strike. Indeed, Jacobus (1989) argues that the right to picket is part and parcel of our common law in South Africa, but that this right has bee severely restricted by other legislation such as the Riotous Assemblies Act of 1956, the Internal Security Act 74 of 1982, the Trespass Act of 1959, the Public Safety Act (emergency regulations) and in terms of the Intimidation Act 72 of 1982. These statutes do not just interfere in the collective bargaining relationship, they drastically disturb the balance in bargaining power in favour of employers (Jacobus, 1989).

The relationship of picketing to violence remains a particularly difficult issue. In many cases it is argued that rights to picket will inevitably lead to violence. However, the fact that picketing, by definition, involves some form of gathering does not mean that it will inevitably result in violence. On the contrary, it can be argued that the right to picket may well facilitate much more efficient policing of strike situations through collectivising and institutionalising the points of conflict during a strike.

Indeed, the very absence of the right to picket may itself lead to outbreaks of violence which may otherwise have been avoided or controlled. This fact is even acknowledged by the National Manpower Commission (Jacobus, 1989: 56) which, in its report titled 'Different Levels of Collective Bargaining, Dispute Settlement Machinery and Related Matters', submitted to the Minister in July 1986, commented on picketing as follows:

While accepting that picketing, especially if it is uncontrolled, could lead to the protraction of tension and to the undermining of public order, the National Manpower Commission is of the opinion that the present restrictions on picketing should be lifted to at least some degree. This is based on the consideration that the limitation of the right or freedom to picket can have other prejudicial consequences, i.e. 'sit strikes' or 'occupations' on factories or premises and intimidation at places other than the workplace. It should also be pointed out that picketing (subject to certain limitations) is permitted in many countries and is accepted by the ILO.

There is another fundamental ancillary to the right to strike which must be guaranteed if violence is to be avoided in strike situations. This is the question of union access to its striking members. Once strike action has been decided on, the need of the union to meet with its membership becomes crucial. The union must be able to convene meetings of the striking workers on the premises in order to report back on the progress in negotiations with management. This is really the only manner in which the union can effectively exercise its discipline and leadership, thereby best guaranteeing that it is the negotiation process rather than violent solutions which hold sway.

It is patently obvious that the laws of South Africa impose a major limitation on industrial picketing. Workers and the general public know how harsh the enforcement of legislative measures can be. Intending picketers, Napier and McBride (1986) argue, also perceive the South African government and law enforcement officials to associate picketing with unruly and riotous behaviour. This perception has been reinforced by the actions of state officials who have instilled fear into trade unionists by having a number of them arrested and detained over the years under the country's security laws.

A further reason why picketing might be so infrequently encountered in South Africa, is that the industrial relations culture in the country is relatively undeveloped and has never really been exposed to the idea of a picket. Since the passing of the Labour Relations Act of 1979, the number of black unions in existence and occurrences of strike activity have grown enormously. Workers are still feeling their way and becoming used to the idea of union membership and the potential power of unions. For example, in the case of Afcol, the union had only started recruiting in early 1987, twelve months before the dispute began.

However, in spite of the regulations against picketing in South Africa, there are a few recorded incidents of picketing in recent times. Although these cases, such as the Armourplate Safety Glass strike of 1976, and outside Edgars, Jet and Sales House stores in 1985, the picketers were arrested, there were no assaults against the non-strikers and violence was avoided (South African Labour Bulletin, 1977 & 1985). Possibly the best recent example is that of the Hextex strike in 1989 which lasted 59 days, and which was preceded by an agreement between management and the South African Clothing and Textile Workers' Union (SACTWU) allowing for on-site picket facilities for strikers. There was no violence in the course of the strike (Patel, 1990).

Owing to the intense fear of the law, workers in industrial disputes have resorted to other tactics of a more covert nature in order to reinforce the strike weapon. These tactics may include industrial sabotage, consumer boycotts and intimidation. By resorting to these tactics, the participants cannot very easily be pinned down or traced and charged under one of the country's laws.

The objective of intimidating non-strikers is two-fold. Clearly, the first objective is to maintain the strength and unity of the striking workers. However, violence is also a warning against neutrality and a clear signal of the cost of siding with management in a particularly emotional or vital dispute. This is best captured by quoting from an interview with a striker published in the South African Labour Bulletin (1989: 30):

Workers sometimes think that violence can make a strike stronger. They want to make non-strikers understand that a strike is a battle between management and workers. No worker can be neutral. Strikes are an educational process and create unity among workers.

There are cases where the threat of violence has forced non-strikers to come to a workers' meeting and explain their position on the strike. What has come out of such workers is their fear to lose jobs, their understanding of management as a powerful force which cannot be shaken by workers. However, after discussions on the experience of other workers, some workers begin to understand that their fears are also shared by strikers, and it is a question of battle.

Strikes are not personal problems. But it is widely known that management always tells workers they are individuals and must take their own decision. It is sometimes difficult to get rid of this perception. Some workers will resort to violence to fight individualism.

It is unfortunate that unity is sometimes forged by violent means. But you have to understand the factory situation. We are always exposed to violence, in the form of the authoritarian hierarchy in the factory system, which gives supervisors all the power to do as they like, and other workers get the task of spying on other workers.

There remains, of course, the danger that violence may have exactly the opposite effect to that which is desired. First, it may alienate strikers who shy away from violence. Secondly, it may well pave the way for extreme repressive measures on the part of management and may offer justification for calling in the police. This is another reason why unions are opposed to the use of violence and why the Paper, Printing, Wood and Allied Workers' Union (PPWAWU) which was a union involved in the Afcol strike was additionally concerned to publicly state its opposition to the use of violence during the dispute (The Star, 1988).

As a result of the negative effects of strike violence, unions are beginning to confront the problem of intimidation and have begun to experiment with strike rules where intimidation or violence is expressly forbidden. For example, the National Union of Mineworkers (NUM) is in the process of negotiating "a code of conduct" with the Anglo American Corporation to promote acceptable norms of behaviour during an industrial dispute. The most far reaching of these agreements are those that have been entered into by a number of German companies. These agreements provide both for the right of union access to company premises during the strike and for the right to peaceful picketing on company premises (IG Metall, 1988).

While strike rules are an important development which provide a guide to worker leadership on the ground, they cannot on their own solve the problem of violence. This is because they cannot address the full range of factors in the strike situation that contribute to violence. There are several factors operative outside the workplace which undoubtedly also contribute to the use of violence in relation to industrial disputes. The high cost of living, rampant inflation and the failure of wages to keep pace with the rising costs faced by black township residents, substantially heightens frustrations around any wage dispute, especially when a strike ensues which completely eliminates access to even the most meagre of resources. When the strike continues for a long period of time, especially if management appears to workers to be delaying the negotiating process, then these pressures further heighten frustration levels. These frustrations are compounded where the police act in what workers perceive to be a one-sided way by arresting striking workers, or detaining worker leadership during a strike.

Any attempt to grapple with the relationship between individual action and collective experience in a strike situation must take full account of the influence of those stressors which are extraneous to the factory floor, and which may affect different workers differently. Probably one of the most prevalent stressors amongst workers in a strike situation is the stress derived from fear of losing their jobs. Yet even this stress inducing experience may affect different workers differently depending on their individual circumstances outside the workplace.

In practice, strikes will be particularly stressful for those with histories of familial responsibility, in particular a sole breadwinner. Because of limited strike funds, strikers have little access to money during such disputes. Strikers in this situation feel guilty and fearful which results in high levels of stress and anxiety. In the case of migrant workers whose families are usually far away and even more reliant on the repatriation of part of the migrant's earnings, these feelings may be even more intensified. Also in the case of male workers who experience certain gender expectations of themselves, stress and frustration may be heightened by their beliefs that they are not meeting their obligations as sons or husbands as a result of their failure to provide financial assistance. In such circumstances, actual dismissal may have a dramatic effect.

Furthermore, for many workers in South Africa, access to a job goes hand-in-glove with access to housing. This is especially true of thousands of dwellers in single-sex hostels or compounds, such as mine workers or many of the SATS employees on the railways. For these workers, the implication is that not only does a striker stand to lose his job if he is dismissed, but that he loses his source of accommodation as well.

Familial pressures on strikers who live at home have the potential to be extremely stress-inducing. This is especially the case where strikers' families put pressure on strikers to return to work. This pressure is in direct contrast to the demands made on the individual worker by his fellow strikers who are dependent on the unity and solidarity of the strikers if the strike is to succeed.

Not only do familial and financial obligations of the sort described here impact on the stress levels of workers in a strike situation, but the relationship between the home and the workplace is a reciprocal one. That is to say that the stresses and strains of industrial disputes may also have manifestations within the family and home environment. This is particularly important in the context of violence resulting from industrial conflict situations, as it is evident that there are victims of such violence who remain hidden from view because they reside in the private arena of striking workers' lives, rather than in the workplace (Simpson & Vogelman, 1990a & 1990b).

Figure 1: Process of a non-violent strike


Process Analysis of Violent and Non-violent Strikes

It is argued that, where an effective picket exists, workers develop a sense of controlling their participation in a strike. This greater degree of control helps to facilitate:

  • the maintenance of unity within the workforce;
  • the maintenance of discipline, regular communication between union leadership and members and, therefore, a high degree of leadership control;
  • the ability to canvass moral support in the community which may in turn assist in the management of familial conflict and tensions in the home;
  • the possible constraint of police action;
  • the development of the negotiation process itself; and
  • the possible shortening of the duration of the dispute.

These combined factors contribute to the institutionalisation of the conflict, place constraints on the use of violence and, therefore, offer the greatest likelihood of a negotiated compromise. That is why the picket has become a legitimate devise in most Western countries. The denial of the right to picket leads to frustration among the participants of the strike. This frustration is heightened by a number of other possible factors, some of which are linked directly to the absence of any moral sanction in the strike in the form of the picket. Frustration is heightened by:

  • The division of the workforce through the employment of non-strikers. One of the most fundamental causes of frustration is the role of strike breakers in undermining the power of the striking workers by going to work and thereby making the strikers redundant. This leave workers vulnerable and exposed to the likelihood of dismissal.

  • Militancy of the workforce and the inability of the union to exert leadership because of inadequate access to members. This militancy is often accompanies by disappointment at the failure of the union to resolve the dispute in the favour of the workers which may lead to increased frustration. This is especially the case where workers are new to trade union organisation and may be naive about the extent of their own power, or lack thereof.

  • Disappointment with the lack of concrete support from other unions.

  • Perceived co-operation between employers and the police, usually manifested by the direct involvement of the police which leaves workers feeling frustrated at their apparent powerlessness.

  • Apparent intransigence of management, especially a refusal to negotiate or to negotiate in good faith. This source of frustration will be exacerbated if workers perceive management to be using delaying tactics, or where workers are convinced that they have been treated unfairly.

  • The prolongation of the strike, whatever the cause, is an enormous stimulus to frustration. Every day that workers are out on strike increases the levels of frustration, tension and stress due to the workers' inability to meet their financial and familial commitments. The duration of the strike coupled with the individual worker's obligations may easily lead to the provocation of violence in such a situation.

All these factors increase the strikers' frustration, contributing to their loss of faith in the negotiating process. Faced by the possibility of losing the strike, individuals attempt to 'police' the strike by assaulting non-strikers, leading to violent confrontation rather than a negotiated compromise.

Figure 2: Process analysis of the escalation of violence


The Afcol Strike

It has been argued that the right to an effective picket is denied in South Africa and that, partly as a consequence, a culture of covert coercion in strike situations has developed in recent years among black workers. It is argued here that absence of the right to picket played an important role in the dispute at Afcol.

The origins of the dispute at Afcol can be tranced back to Star Furnishers, an Afcol subsidiary, on 14 April 1988, when conflict between members of the Paper, Printing, Wood and Allied Workers' Union (PPWAWU) and the National Union of Furniture and Allied Workers (NUFAW) resulted in a fight breaking out. As a result of the subsequent disciplinary hearings, four workers, all members of PPWAWU, were dismissed. Union members then went out on an 'illegal' strike over what they perceived to be unfair dismissals.

On Wednesday 20 April, the striking workers were informed by management via the union organiser that the appeal against the dismissals of the four had failed. Furthermore, management issued an ultimatum to the rest of the striking workforce to return to work by 1:00 p.m. that day. After some confusion, workers decided to return to work but were informed by management that the entire striking workforce had been dismissed.

On the following day when workers once again gathered at the factory to return to work, they were informed that only certain workers would be re-employed. On that same day, 12 of the Star Furnishers workers were arrested by the police.

By Friday 22 April, workers at the other Afcol plants had also come out on an in illegal strike in solidarity with the Star Furnishers workers. In response to a management warning to interdict the striking workers, workers returned to work on the following Monday. However, a legal dispute was declared by the union over the dismissals at Star Furnishers.

On 13 May, a strike ballot was held at all the Afcol plants and 80% of the total membership of 2 000 elected to go out on strike from 30 May. On Wednesday 1 June, the employers sought an interdict declaring the strike illegal and initiated a lock-out denying striking workers any access to the affected plants.

The following day, workers gathering in an overcrowded and highly emotion-charged meeting at Khotso House, were informed that management were hiring 'scab labour'. Following this meeting, the first assaults against non-striking workers occurred at Croesus Station, resulting in the eventual death of a non-striker from Afcol's Edblo factory. Further incidents were to ensue the following day after strikers were informed that management was supplying transport to non-striking workers.

Finally, on 9 June 1988, in the wake of the three day national stay-away in protest against the proposed Labour Relations Amendment Bill, 2,000 Afcol workers gathered in Regina Mundi Church in Soweto. There, workers were informed that non-strikers were being used to train 'scabs' and that there had been no progress towards winning their demands. Some of the strikers then took the decision to attempt to break the lock-out at one of the affected Afcol plants. In the following week, the five 'scabs' who were the subject of the subsequent murder trial were assaulted and killed.

Instead of establishing a picket at the factory once their legal strike commenced on 30 May 1988, the union organisers were forced to advise their members on 2 June to relocate to the union offices. When these offices were found to be too small to accommodate the workers, they were once again re-directed to Khotso House in central Johannesburg. It is argued here that these denials of the right to picket and the right to access to the affected plant frustrated the strikers' attempts to persuade their colleagues to join them. It meant that, for the duration of the strike, discipline was exercised in a covert manner away from the workplace and beyond the realm of collective control.

The already high levels of frustration which resulted amongst the Afcol strikers was further compounded by six aggravating factors:

Division of the Workforce

The lack of workforce unity contributed to the frustration among the strikers. A major target of this frustration was the membership of the rival, mainly 'coloured' union, NUFAW. These workers often held the more skilled and supervisory jobs. Their union was perceived by the strikers to be a 'sweetheart' union that enjoyed special status with management through its closed shop agreement. It was seen to be unsympathetic to the demands and aspirations of the majority of Afcol workers. Indeed, this union was seen to be actively working with management to divide the workforce by training workers brought in to replace striking workers (Heffer, 1990).

COSATU General Secretary, Jay Naidoo (Kock, 1989), stated publicly that there were widespread rumours among workers at the time, that various managements were replacing African workers with 'coloured' and Indian workers, for the reason that African workers 'were always going on strike'. The perceived role played by these 'coloured' workers is likely to have strengthened certain stereotypical views held by these workers, and this undoubtably helped to exacerbate the conflict.

Militance of the Workforce and Inability of the Union to Exercise Discipline Over its Members

It has already been noted that the 1980s was a period of growing militancy among black workers. Management and the state responded to this militancy by attempting to 'roll back' the power of labour. The amendments to the Labour Relations Act passed in 1988 were perceived by workers to be part of an anti-union strategy in this respect. On 6, 7 and 8 June 1988, during the course of the Afcol strike, a three day nation-wide stay-away took place and an estimated three million workers registered their opposition to the proposed amendments to the Labour Relations Act. As affiliates of COSATU, the two largest unions organising in the Afcol plants – PPWAWU and the National Union of Metalworkers of South Africa (NUMSA) – reflected this militancy. However, unlike other Western countries, workplace behaviour was not subject to the restraining influence of an enfranchised community most likely to disapproving of violence.

In this context, having embarked on a legal strike, the need for the union leadership to meet regularly with its membership to report back on negotiations with management became crucial. However, a management instituted lock-out prevented this ease of communication. Combined with the denial of the right to picket, the lock-out effectively eliminated the opportunity for containing the conflict and entrenching the discipline of the union leadership. Instead of strike discipline being exercised openly by the union, it was enforced by frustrated militants away from the workplace and without collective control.

Lack of Solidarity from Other Unions

The strikers had expected to obtain the support of the South African Breweries Shop Steward Council and COSATU affiliates (Afcol being a subsidiary company of South African Breweries). This support of solidarity action was not forthcoming and this both frustrated the striking Afcol workers and left them feeling vulnerable and exposed.

Perceived Co-operation between Police and Employers

The arrest of twelve Start Furnishers (a subsidiary of Afcol) workers in the course of the dispute was likely to have reinforced perceptions among workers that the police were acting aggressively in the interests of the employers. It was rumoured amongst the workers that a member of management accompanied the police when arrests were made during the strike. This perception would have contributed to the increased anger and frustration among the strikers.

Apparent Management Intransigence

The mass dismissal of Star Furnishers employees on Wednesday 20 April, the decision by management not to re-employ these workers (van Holdt, 1988), an attempt to interdict the strike, the lockout of Afcol employees on 2 June and the failure of management to resolve the dispute, all reinforced the strikers' feelings that management was becoming increasingly intransigent. This intransigence was indicated clearly in a management memo which stated:

We must be sure that we do not re-remploy any of the workers who have now been discharged … . By re-employing we will be breaking down the advantage we now hold … . The benefits that we will have gained from the rather traumatic experiences of last week will be worth it in the medium to long term.(van Holdt, 1988: 5-6)

This obviously increased the strikers' sense of desperation and frustration with the negotiation process.

Possibly of central importance here is the resort to mass dismissals by management. Dismissal in a strike situation is possibly the greatest source of escalating conflict with the greatest potential for resultant violence on the part of striking workers. The all too common propensity to dismiss immediately shifts the terrain during a strike to an 'all or nothing' struggle. The strike becomes an isolated incident which more often than not holds out the prospect of no common ground and no common future, instead of an event within a continuing relationship, which for both sides has a future as well as a past.

In this way, industrial relations disputes about wages and working conditions may easily be escalated into conflicts of principle – trade union recognition or dismissals – where the space for compromise is more limited and where resolution is more difficult by virtue of its lying outside of the arena of collective bargaining. The dismissed workers are excluded from the processes to decide their fate and effectively lose all control in this regard. The psychological effects of this may lead to disinhibition in relation to acts of violence which otherwise may have been avoided.

Prolongation of the Strike

Frustration is less likely in a short demonstration stoppage in which the strikers' grievances are dealt with quickly by a responsive management and a negotiated compromise is reached. When the strike continues over a long period of time, as happened in the case of Afcol, it becomes a trial of strength in which both sides begin to perceive it as a battle. Fears of dismissal are exacerbated, as are familial and financial pressures. Strikers and their dependents are subject to increasing material deprivation as they deplete their meagre savings and are forced to depend on hand-outs from the union or from neighbours and relatives. Pressures build up to end the strike and strikers become increasingly frustrated at their inability to persuade management or to maintain the support of the majority of the workforce.


The argument presented here may seem to be flawed in the case of the Afcol strike by virtue of the fact that no attempt was made by the strike participants to organise a picket. However, the argument does not rest only on the fact that these participants were denied the right to picket; it rests on the fact that the present structures of dispute resolution lead to frustration and that this frustration is exacerbated and compounded by a range of additional factors. The end result is a loss of faith in the negotiation process which in turn gives rise to the need for individual strikers to 'police' strikes.

However, a number of qualifications are necessary. While an effective picket is likely to reduce the frustrations experienced by the strikers, it will not, on its own, eliminate violence. As suggested previously, the decline of violence in advanced capitalist countries was accompanied by the 'constitutionalisation of the working class'. In the South African situation this notion must be interpreted very broadly. Reference has been made to the inseparable inter-relationship between the factory floor and the wider community, and this begs the question of whether conflict in the industrial relations arena can actually be 'institutionalised' in the absence of broader political transformation within the country.

To some extent this question is answered by Johan Liebenberg (1988: 648), industrial relations consultant to the Chamber of Mines, who wrote:

Given these background environmental causes (of violence in industrial conflict) it is clear that we cannot solve the problem of violence solely through industrial relations. Collective bargaining may provide insights into the means or the methods for tackling these problems, but we cannot expect our industrial relations structures to be the means for the elimination of these wider problems. We would be overworking collective bargaining.

It is evident that the only lasting solutions to the problem of violence in industrial conflict must also be political solutions; from the provision of meaningful channels for political expression, to the provision of a comprehensive welfare net for the unemployed. At the level of the enterprise, for the right to picket to be effective, it must be accompanied by increased participation by workers in the enterprise itself. It should be noted here that the argument for the right to picket as a means of limiting violence in strike situations is promised on the assumption of an ongoing adversarial relationship between management and labour, in which the trade union is in opposition to management. This is a tradition in industrial relations which has been inherited by South Africa from the United Kingdom. There are alternative arguments promised on a more co-operative dispensation such as that characteristic of the system of co-determination in West Germany. One such argument is implied by Albertyn (1989) who suggested that the existence of a pre-entry closed shop agreement (in which membership of a trade union is a prerequisite for employment), would automatically mean that production stops if the union goes on strike. The implication is that there would be no 'scabbing' and therefore no need for a picket whether this is an appropriate or realistic solution in the South African case with the multiplicity of unions which exist in any one factory, and with the historical antagonism which has developed, is a matter for discussion elsewhere. Perhaps, even more important is the search for solutions to the problem of industrial violence appropriate to the special circumstances of a post-apartheid society, which must prioritise some fundamental re-distribution of wealth and resources.

Essentially, what is suggested is the long term need to begin developing a democratic culture within the industrial relations arena as well as in the wider community. The starting point in this process would appear to be the emergence of codes of conduct, negotiated between workers and management which have the potential to regulate strike situations by democratically defining what is legitimate behaviour in strikes. Of even greater significance is the fact that recent evidence suggests that the right to picket is becoming an increasingly common demand in strike situations in South Africa.

This embryonic democratic culture was nowhere more visible than in the recent Hextex strike in Boland. The fifth longest strike in 1989, involving 1,020 workers, witnessed no violence whatsoever. Before the strike, the company and union agreed to the following (Patel, 1990):

  • access for union officials during the strike;
  • on-site picket facilities for strikers;
  • access for strikers to the company canteen during the strike;
  • the right to strike without fear of dismissal; and, limitations on the number of strikers to have access and picket rights.

As SACTWU (1990: 4), the organising union at Hextex, concluded in a broader context:

Obviously, such negotiations will benefit our society. But only if it produces a fair, workable system of industrial relations to which all parties are generally committed. If unions are powerless and right-less, then industrial conflict will not go away. It simply becomes less controllable and predictable.

The institutionalisation of conflict in the industrial relations arena, especially the entrenchment of the right to picket and the right of access to the factory during a strike, would also make for a more formal and orderly policing of strike situations. However, if a democratic culture is to be developed within the industrial relations arena, then this must of necessity also entail the development of alternative forms of public order policing as well. This, too, would seem to imply the need for broad political solutions, as the only means through which the very process of law enforcement itself can be democratised.


Albertyn, C. (1989) Freedom of association and the morality of the closed shop, Industrial Law Journal, Vol. 10(4), 681-1005.

Cock, J. & Nathan, L. (1989) War and Society: The militarisation of South Africa, Cape Town: David Phillip.

Dahrendorf, R (1959) Class and Conflict in Industrial Society, London: Routledge & Kegan Paul.

Diener, E. (1979) Deindividuation, self-awareness and disinhibition, Journal of Personality and Social Psychology, Vol. 37(7), 1160-71.

Dison, D. (1989) Violence and the Law: An examination of some recent trials. Paper presented at the Centre for the Study of Violence and Reconciliation, Seminar No. 9, 26 October.

Geary, R. (1985) Policing Industrial Disputes, 1893-1985. London: Metheun.

Heffer, T. (1990) Report submitted to the Supreme Court of South Africa (Witwatersrand Local Division) in the matter between State and Elias Phasha and Four Others.

Jacobus, R. (1989) Ancillaries to the right to strike. In Benjamin, P., Jacobus, R. & Albertyn, C. (eds) Strikes, Lock-outs and Arbitration in South African Labour Law. Cape Town: Juta, 53-62.

Koch, E. Interview with Jay Naidoo, Weekly Mail, Johannesburg, 20-26 January 1989.

Lambert, R. & Webster, E. (1988) The re-emergence of political unionism in contemporary South Africa. In Cobbett, W. & Cohen, R. (eds) Popular Struggles in South Africa. London: Africa World Press, 20-41.

Labour Monitoring Group (1985) The November stay-away, South African Labour Bulletin, Vol. 10(6), 74-100.

Liebenberg, J. (1988) Militance and violence on the mines. In Bendix, W. (ed.) South African Industrial Relations of the Eighties. Cape Town: IPC.

Napier, C. J. & McBride, S. W. (1986) Industrial picketing in South Africa: Does it have a role? Industrial Relations Journal, 2nd quarter, p. 51.

IG Metall (1988) Minimum Standards for Labour Relations and Labour Conflicts for South African subsidiaries of German Companies. Joint demands of the Industriegewerkschaft Metall of the Federal Republic of Germany and the South African Council of the International Metalworkers Federation, Johannesburg and Frankfurt.

Patel, I. (1990) Negotiating strike rules and rights: The Hextex strike South African Labour Bulletin, Vol. 14(8), 75-6.

Pelling, H. (1976) A Short History of the Labour Party. London: Macmillan.

SACTWU (1990) Vol. 2 (6), 2-4.

Simpson, G. & Vogelman, L. (1990a) Industrial relations, strikes and stress. Unpublished paper. Presented to the Professional Society of Black Occupational Health Nurses seminar, Occupational Health Service and Worker Acceptability in a Changing Society.

Simpson, G. & Vogelman, L. (1990b) Strikes and social stress: Some effects on the domestic arena. Unpublished paper. Presented to the National Council for Mental Health Conference, Family Mental Health is Priceless.

South African Labour Bulletin, 1977, Vol. 3(7), 65.

South African Labour Bulletin, 1985, Vol. 11(2), 3.

South African Labour Bulletin, 1989, We are not always exposed to violence, Vol. 14(3), 30.

The Star, 21 June 1988.

Vogelman, L. (1989) Strike violence, South African Labour Bulletin, Vol. 14(3), 47-56.

Von Holdt, K. (1988) Union bashing fails at Afcol, South African Labour Bulletin, Vol. 13(6), 5-7.

Webster, E. (1989) Report submitted to the Supreme Court of South Africa (Witwatersrand Local Division) in the matter between State and Elias Phasha and Four Others.

Webster, E. (1979) Evidence submitted to Wiehahn Commission on the settlement of industrial disputes, South African Labour Bulletin, Vol. 5(1), 83-93.

+ posts

CSVR is a multi-disciplinary institute that seeks to understand and prevent violence, heal its effects and build sustainable peace at the community, national and regional levels.

Leave a Reply

Your email address will not be published. Required fields are marked *

Translate »