Vogelman, L. (1991) Some Psychological Factors to Consider in Strikes, Collective Violence and the Killing of Non-strikers. In Violence in Contemporary South Africa, SAB conference proceedings, Johannesburg, September.


Lloyd Vogelman

In Violence in Contemporary South Africa, South African Breweries (SAB) conference proceedings (1991).

Lloyd Vogelman is a founder and former Director of the Centre for the Study of Violence and Reconciliation.

This paper has two broad aims. The first is to consider some of the psychological factors leading up to strikes. The second is to attempt to explore some of the psychological variables which contribute to collective violence and the killing of non-strikers in particular. To illustrate these arguments some issues relating to the killing of four non-strikers by SAHRWU members in 1987, during what became known as the SATS strike, will be described. The murder of the non-strikers involved repeated stabbing, throwing a rock on the victims' heads and then burning of their bodies in a deserted veld in Prolecon, There were at least four active participants within a larger support group. The incident took place after the non-strikers had been detained and beaten for a number of hours by a crowd of workers in Cosatu House. The incident took place on the 28th of April, a month and a half after the strike began.

There are numerous social, political and economic factors which help to explain the actions of those who pleaded and were found guilty of murder in the trial that followed. This paper, however, will briefly focus on a few of the psychological variables which contributed to both the strike and the subsequent killings.

An examination of violence and conflict in strike-related situations often indicates that hostile and violent reactions develop as a result of inadequate industrial relations and political mechanisms which make instituting and implementing change difficult. This may be the result of an intransigent management style, as in the case of the SATS strike, and Apartheid, which limits black political participation. On a social and economic level the potential for violence is also increased when people's options are limited - for instance, the ability to move to another job. This results in individuals becoming desperate to preserve their work, particularly in a climate of unemployment. From a social psychological perspective, there are a number of violence-promoting factors which can be identified in strikes.

Violence-Promoting Factors

To comprehend the violence of strikers, a central question will be addressed. Why do strikers, like those who admitted to the murder of four non-strikers in the SATS strike, who do not have anti-social (psycho-pathic) tendencies or histories of violence as indicated by psychological assessments (done by the author), engage in impulsive and brutal collective aggressive conduct?

The roots of group violence, like individual violence, can often be found in historical relationships which provide a reservoir for hostility. This is then drawn upon to mobilise, reinforce and justify violent behaviour. Having stated this, why when a negative historical relationship exists between some workers and their employers, is hostility not expressed earlier? The answer to this is partly gained in exploring the issues of deprivation, frustration and aggression.

Deprivation, Frustration and Aggression

Frustration is a basic prerequisite for anger. It is associated with, but not always linked to, the expression of aggression and violence. Under certain circumstances frustration may lead to aggression. To understand this it is useful to focus on the concept of deprivation. Berkowitz (1972) defines deprivation and frustration in the following manner:

I would say a person is deprived if he lacks a goal object people generally regard as attractive and desirable, but is frustrated only when he had been anticipating the pleasure to be gotten from this object and then cannot fulfill this expectation.

Very often in South Africa striking workers are deprived in a number of areas. They are economically deprived - their wages are low and it is sometimes difficult to ensure their own and their families' physical survival. Apartheid deprives them, because they are black, of basic political rights and political power. On a day-to-day level they are deprived of human dignity through racism. Their living conditions, particularly if they are living in old compounds, as many SATS workers did, deprives them of privacy, adequate sleeping conditions and living space. It may even inhibit the development of serious heterosexual relationships.

The intensity of deprivation at the work-place leads to a desire to improve work conditions. This desire and attempt to overcome occupational frustration may be expressed in workers joining unions and resisting what they perceive as unfair treatment of workers. The latter can manifest itself in strike action.

Different levels of satisfaction are linked to different expectations. Those in deprived situations who have few expectations often feel less disappointed by unfulfilled hopes. While labour experts, historians and other social scientists are probably best placed to provide answers for why workers' frustrations become manifest at the time they do, from a psychological perspective, part of the answer can be attained by investigating the changing nature of expectations.

Prior to 1987, SATS workers experienced depriving work conditions but had little hope and few expectations of correcting them. The climate of uprising in the years 1984-87, the government's reform initiatives, as well as the development of SARWHU in the railway sector led to a belief that their work and living conditions would improve. Their expectations were raised. The greater the expectation, the greater the potential for disappointment and frustration. When the workers went on strike they believed the power gained through the united mass withdrawal of their labour would ensure SATS heeded their minimum demand for negotiations around better work conditions. This did not occur and as the strike progressed it became clear their assessment of their strength was misplaced. One consequence was an increasing sense of powerlessness, frustration, and disappointment as a result of unmet expectations. This helped to lay the foundation for the use of extreme behaviour in order to restore lost power.

Frustration and Reference Groups

Individuals tend to evaluate their psychological well-being in relative rather than absolute terms [Hyman, 1942; Merton and Kitt, 1950; Festinger, 1954]. Comparative evaluation frequently determines whether an individual feels satisfied or not. On the basis of this social comparison theory, individuals often select others for appraisal according to three primary factors:

  • the frequency and intimacy of their association with others;
  • the degree of similarity; and
  • the individuals' attraction to them.
    [Berkowitz, 1972; Festinger, 1954.]

These factors may operate exclusively or be inter-related. Taking the first factor within the context of the workplace, comparisons are likely to have been made with black and white workers as well as white supervisors. With reference to white workers there was cause for disenchantment because of racial discrimination in wages and white workers receiving greater promotion possibilities. The above should not imply that black workers do not, for example, compare themselves against managers, most of whom are white. Individuals are, however, sometimes inhibited from yearning for things they believe they will never achieve.

The individuals' tendency towards comparison with those similar may have led the SATS workers to evaluate themselves in relation to other black workers in other companies. Such comparisons would have provoked dissatisfaction if those workers received higher wages and had better work conditions. This would have contributed towards the decision to strike, especially if SATS workers believed their work position to be stagnant while the conditions of workers in other companies were improving.

When is Frustration Likely to Lead to Aggression?

Social psychologists [Berkowitz, 1972; Tyson, 1987] suggest that frustration is most likely to heighten aggression when it is intense and the source of the frustration is perceived as arbitrary and unfair. This is important, for instance in relation to the police who tend to intervene in strikes involving a mass of workers. In the SATS strike police action on the 22nd of April, six days before the killings of the non-strikers, resulted in the death and injury of a number of workers. Police behaviour was and is often seen as arbitrary by victimised striking workers. What is crucial in understanding violence is not only the objective conditions and the actual behaviour but also the perception of, in this case, the police's conduct.

Aggressive cues in the environment, for example weapons, also help to promote the relationship between frustration and aggression. In prolonged strike situations workers often carry weapons, and the police are present armed with guns, tear gas or batons. Weaponry has an arousing effect. Berkowitz, cited in Tyson [1987], in talking about how aggressive cues increase the strength of aggressive responses, states "Guns not only permit violence, they can stimulate it as well. The finger pulls the trigger but the trigger may also be pulling the finger".

The Target of Aggression

To understand who becomes the target of aggression, the work of Dollard et al [1939] is useful. They argue that victims are selected in the following manner:

  • the source of the frustration;
  • other persons;
  • fantasy objects; and
  • the aggressor himself.

Ideally, an individual should express his aggression at those persons or structures who are the source of his/her frustration. However, aggression is not always directed at the source because of misperceptions due to ideology or the source's superior power and the great risks involved. At SATS some workers had been employed for a long period - many for more than twenty years - but little direct anger was expressed at the source of their frustration - their supervisors and the SATS management. The power of both was their ability to easily dismiss workers. Job loss provoked great fear because it was a means to both personal and family survival.

In a strike situation, direct expression of anger is further inhibited if employers refuse, as the SATS management did, to negotiate. Expressive avenues were further limited by legal restrictions on public protest. Physical proximity and access to the source of the frustration are therefore two other contributing factors in determining who becomes the target of aggression.

Where the police are seen to be siding with management in a strike, they become an additional source of frustration. Voicing direct protest about police conduct is not only inhibited by legal parameters, but also by fear of further harassment.

A third primary potential target for aggression is non-strikers. They are seen as undermining the primary power workers have - the withdrawal of their labour. Non-strikers tend to be viewed as traitors. As workers, they were expected to identify with the workers' cause. The intensity of this expectation creates the potential for disappointment and aggression.

Although management and the police are sources of frustration, it is often non-strikers who become victims of extreme violence. This occurs for a variety of reasons.

  • The factors previously discussed which inhibit direct expression of anger towards management and the police.
  • Non-strikers are accessible and strikers know who they are.
  • Relative to the mass of workers, they tend to have little physical power. They are therefore more vulnerable to attack.
  • Aggression towards them does not bring many immediate negative economic and social consequences.
  • Non-strikers, by working, defined themselves out of the group. They provoke violence not only because of what they have done, but because of what they are - a different category of people who are in opposition to the group.

In conclusion, although objective conditions may help to explain violence towards non-strikers, the psychological process of displacement also assists in accounting for this occurrence.

Three Factors Which Weaken the Moral Restraints against Violence

For frustration to lead to aggression and for it to be directed at a group of human beings, various moral constraints against violence need to be weakened. Kelman [1973] has identified three inter-related factors which help to do this. They are: authorisation; routinisation; and dehumanisation.

In the SATS strike, the killing of non-strikers took place within a context of authorisation. The perceived authorities, in this case the crowd, were seen to provide a legitimising framework for violent behaviour. The collective of workers was perceived as having the right to control individual behaviour. The work of Milgram [1977] indicates behaviour, when authorised, is more often than not carried out and further individuals feel that because acts are authorised, there is an automatic justification for them. Accepting that people generally comply with the authorisation process, what are some of the factors that motivate them to do so?

Anxiety and the Authorisation Process

In ambiguous and anxiety-provoking situations like strikes, particularly where workers have little experience of labour disputes, individuals may feel helpless and confused. Uncertainty exists about the attitudes and conduct of both management and the police as well as the outcome of the dispute. This occurred in the SATS strike when the workers realised their conditions for returning to work would not easily be met.

One defence mechanism for overcoming anxiety, moral conflicts and the necessity of making decisions in ambiguous social settings is complying with the authorisation process. A point to stress is that authority is not accepted without question, it must have legitimacy. This helps to explain why many workers who are strongly socialised to accept authority do not always go along with the wishes of management when they insist workers return to work before negotiations can begin. At a certain point in the events leading up to a dispute, the trust of workers is betrayed or further lost. This results in management's authority becoming less credible and legitimate. While this is a major cognitive change for workers, the underlying psychic structure which emphasises the acceptance of authority may not shift substantially. Rather, allegiance is shifted to another authority which is perceived to be acting in the interests of workers - the mass of workers or the union.

Denial of Responsibility

Insofar as the individuals see themselves as part of a large group and as having little choice in the authorisation process, they hold themselves less responsible for their conduct. To maintain individual responsibility in mass group aggressive situations is not impossible but is extremely difficult. The tendency to follow the group is even stronger when there is an absence of even a small grouping of leaders adopting a contradictory view. Thus in situations where the union is new, officials are overloaded with work and are inexperienced, it may be difficult for union representatives to offer ideas different to the emergent norms and violent practices of the worker crowd.

Bystander Apathy

The work on bystander apathy (Latane and Dailey, 1976; Seedman and Hellman 1974) may also help to account for why individuals choose to abdicate moral responsibility when they witness acts they do not approve of fully. A belief may exist that it is the role of authority (the union or the crowd's leaders) to intervene and prevent the aggressive conduct from continuing. This belief contributes to bystander apathy.


The second major feature Kelman [1973] has identified which breaks down moral restraints against violence is routinisation. This refers to the process where repeated authorisation for violent conduct provides continual justification for the act. Linked to this process is the "foot in the door" or "slippery slide" syndrome. Once individuals agree to involve themselves in small, often unrelated events, they set the scene for broader and more compromising compliance in other activities [Fraser, 1988]. Thus individuals may be asked on an ad hoc level to find materials necessary for a murder of non-strikers, e.g. to find rope. Once they agree with this request, they are more likely to comply with requests demanding even more active participation in murder.

Routinisation has the effect of ensuring that individuals become more concerned about the mechanics and accomplishment of tasks rather than their meaning. This can be especially manifest in the organisational sphere where a division of labour exists and each individual is responsible for a certain job. Two consequences of this are a diffusion of responsibility and decisions becoming operational rather than moral ones.


In the authorisation and the routinisation processes, previous moral considerations are forsaken. However for killing to occur, victims still need to be deprived of their human status. Dehumanisation involves not perceiving the individual as: independent; different to others; and part of a global community of people who are concerned for each others' welfare and rights [Kelman, 1973]. In situations of conflict the potential for dehumanisation increases. In strikes non-strikers are often dehumanised. By placing themselves outside the community morality of the strikers, non-strikers become a category of people. An understanding of the non-strikers' individual circumstances is lost. They are seen to be wholly destructive. The language of the strikers' may further deprive non-strikers of their humanity. Words like "mpimpi" help to define them as enemy and subhuman.

In becoming objects non-strikers are easier to kill. One is not killing a breathing, thinking, feeling individual, one is killing an object destructive to the interests of the individual worker and his/her family and community.

Psychological Factors that Operate in Crowds

Since the violence perpetrated by strikers often occurs in crowds, aspects of crowd psychology will be briefly explored. Only three factors will be looked at - arousal, deindividuation and conformity. Other significant variables such as anonymity, modelling of behaviour and emotional contagion will not be touched on.


Amongst a large group of people, it is likely they will become aroused as a consequence of overcrowding, emotional expression, sloganeering, dancing, singing and the breakdown of meeting procedure. These factors are often present in strikes. Irrational thought and behaviour are two consequences of arousal. Research [Sergent, 1973] suggests when "individuals become extremely aroused for a period of time they become hyper-suggestive" [Tyson, 1987, p5]. Proposals to kill or engage in violence often become more palatable in an atmosphere of arousal.

The relationship between arousal and violence is further underpinned by individuals tending to understand their arousal states in terms of situational cues [Tyson, 1987]. For instance, arousal experienced at a soccer match may be experienced as excitement, whereas arousal in the midst of a strike where one's meeting place is surrounded by armed police is likely to be interpreted as anger.


Many workers found guilty of murder in strike-related situations have no history of violence. In fact, much of their personal histories indicate tolerance, non- violence, caution, and restraint. Yet as strikes progress, these qualities are diminished. Self-monitoring restraints and consideration of the implications of actions are absent. This psychological shift can be partially explained by the psychological mechanisms of deindividuation and disinhibition.

Deindividuation refers to the process whereby there is a reduction in the self-monitoring of behaviour. Together with situational pressures, this increased disinhibition leads the individual to do things he/she would not normally do [Diener, 1979]. While noting that a number of factors give rise to deindividuation, including arousal, anonymity and defusion of responsibility, only its characteristics, based on the work of Diener [1976, 1979] will be focused upon.

Characteristics of Deindividuation

Decreased Self Awareness and Self Regulation

People tend to regulate their behaviour on the basis of internal psychological controls, fear of punishment, and by comparing it to others. Self-monitoring normally takes place quickly and easily with the individual generally attending to either the environment, others in the environment or himself. There cannot be an intense focus on the self and the environment at the same time [Tyson, 1987). If attention is directed towards the environment or a group of people, there is less focus on the self and the potential for deindividuation is increased.

Since the scenario in strikes often involves, amongst other things, overcrowding in inadequate halls, police repression, strategies and plans to both intensify the strength of the strike and ensure economic survival, attention is directed externally. This diminishes the normal influence internal standards have on behaviour. In the killings which took place during the SATS strike, it was, in part, the perpetrators' lack of self-awareness and regulation which led them to behave discordantly with their past social and personal standards.

No Long-Term Planning

A second characteristic of deindividuation is no long-term planning. There is little consideration of the future. This was evident in the SATS strike when on the night of the murders, those actively involved did little to conceal their identity. One also sees this process amongst some who have participated in 'necklacings' - they permit themselves to be filmed without thinking how such film could result in their arrest. Linked to non-consideration of consequences is another feature of deindividuation - irrationality.


It may be argued that, considering the social and economic ramifications of the SATS strikers losing their jobs (after being dismissed during the dispute), the decision to place moral pressure and\or punish the non-strikers was rational. It would, however, be more difficult to posit that the nature of the killings was rational. The evidence for this lies in part in the destruction of the victims' bodies beyond functional utility.

Responding to Immediate Cues

Poor self-monitoring and an inadequate consideration of the consequences of actions not only allows for irrationality but it encourages the individual to respond to immediate cues. For instance, it was the discovery of a work ticket on a kidnapped non-striker who, when captured, denied working which intensified the rage of the crowd and propelled the process of murder.


The work particularly of Asch [1952] indicates that conformity is a strong determining force in behaviour. Underlying conformity is the desire to be more secure. In insecurity- provoking situations such as strikes, it is likely individuals will become more reliant on others. The more unsure another person is, the more likely he is to conform. Thus those who have no experience of labour disputes may be more vulnerable to conformity pressures, especially when things are not going as they had envisioned.

Another reason for people conforming even to aggression is their fear of being ostracised, rejected and possibly even physically hurt. To avoid this they define themselves as part of the group and comply with the group's norms.


The non-strikers' weak structural position, their accessibility and the strikers' perception of their behaviour as antagonistic to the cause of the collective makes them likely targets for aggression during strikes. The expression of aggression is made easier by the psychological factors operating in crowds and the number of processes which weaken the moral restraints against violence.

In the SATS strike, like in other crowd settings, the variables and processes discussed acted in an inter-related fashion. No one variable was and can be the sole cause of violent behaviour. Depending on the nature of situational pressures and the individual's personality, some factors will be more determining. Nevertheless, what cannot be underestimated is the individual's vulnerability to situational pressures. While this paper has not addressed the whole question of individual differences, what hopefully has emerged is that psychological pathology is not a prerequisite for engaging in group violence.


Asch, S.E. (1952). Effects of group pressure upon the modification and distortion of judgement. In G. E. Swanson, T.M. Newcomb and E. L. Hartley (Eds.), Readings in Social Psychology. New York: Holt.

Berkowitz, L. (1972). Frustrations, comparisons, and other sources of emotion aroused as contributors to social unrest. Journal of Social Issues, 28, 77-92.

Diener, E. (1976). Effects of prior destructive behaviour, anonymity, and group presence on deindividuation and aggression. Journal of Personality and Social Psychology, 33, 497-507.

Diener, B. (1979). Deindividuation, self-awareness and disinhibition. Journal of Personality and Social Psychology, 37, 1160-1171.

Dollard, I., Doob, L., Miller, N. Mowrer, O., and Sears, R. (1939). Frustration and Aggression. New Haven: Yale University Press.

Festinger, L. (1954). A theory of social comparison processes. Human Relations. 7, 117-140.

Fraser, S (1988). Evidence in mitigation in State v. Sibisi.

Hyman, H. (1942). The Psychology of Status. Archives of Psychology, 269. Cited in Berkowitz, L (1972).

Kellman, H. C. (1973). Violence without moral constraints. Journal of Social Issues. 29, no.4.

Latane, B. and Darley, J.M. (1976). The unresponsive bystander. Why doesn't he help? New York: Appleton-Century-Crofts.

Merton, R and Kitt, A. (1950). Contributions to the theory of reference group behaviour. In Merton, R and Lazarsfeld, P (Eds), Continuities in Social Research: Studies in the scope and method of "the American Soldier". Glencoe, 1ll.: Free Press.

Milgram, S. (1977). The Individual in a Social World. London: Addison-Weseley.

Sargant, W.W. (1973). The Mind Possessed: A physiology of possession, mysticism and faith healing. London: Heinemann. Cited in Tyson (1987).

Seedman, A. and Hellman, P. (July 29,1974) Why Kitty Genovese haunts New York?: The untold story.

Tyson, G. (1987). Report in evidence in mitigation in State v Ngidi.

© Centre for the Study of Violence and Reconciliation

Copyright © 2021 CSVR. All Rights Reserved.