ARTICLES

 

Deindividuation and criminal responsibility

 

 

Shannon Hoctor

BA LLB LLM DJuris PG Dip (Latin); Professor of Law Stellenbosch University

 

 

---

SUMMARY

This contribution assesses the social-psychological concept of deindividuation in the particular context of its use in instances of crowd violence in South African criminal law. Well-established as a mitigating factor, and yet not used over the past couple of decades, the question is posed as to whether this concept still finds application in contemporary South African law and society.

---

 

 

1 INTRODUCTION

Crowd violence is an all-too-commonly encountered phenomenon in the South African context. Given that crowd violence may inflict more harm than where an individual is acting alone, potentially concomitantly violating public peace and order as well as threatening or infringing the rights of other people, it is required that any such violence be consistently and comprehensively dealt with by the authorities. The response to such violence is typically the application of criminal law.¹ In such factual scenarios, the State has often made use of the doctrine of common purpose in the criminal law context to facilitate the process of proving individual guilt, where such an individual was a member of the crowd engaging in violent behaviour. The common purpose doctrine is discussed below in more detail, but prior to doing so, we turn to the question as to which psychological factors are at play when a person is a member of a crowd.²

 

2 DEFINING DEINDIVIDUATION

Deindividuation has been described as "one of the most widely cited effects of social groups", which seeks to provide an explanation for various expressions of antinormative collective behaviour.³ Such behaviour is typically associated with a state in which persons are "blocked from awareness of themselves as distinct individuals, fail to monitor their actions, and can behave impulsively".⁴ Social psychologists have used the term "deindividuation" to describe the situation where, being part of a crowd caught up in strong emotion, the attention of the individual concerned is directed away from personal self-awareness and the usual methods of self-control and towards the shared emotion.⁵ The apparent experience of loss of individuality inherent in deindividuation has been described as "a psychological state characterised by reduced self-awareness and personal identity salience, brought on by external conditions such as being an anonymous member of a large crowd".⁶ The consequences of deindividuation typically include the weakening of control over one's own behaviour, and less concern about "normative standards, self-presentation and later consequences of one's behaviour".⁷ Zimbardo, one of the leading deindividuation theorists, summarises the effect of deindividuation within the context of criminal conduct as follows:⁸

"Deindividuation makes the perpetrator anonymous, thereby reducing personal accountability, responsibility, and self-monitoring. This allows perpetrators to act without conscience-inhibiting limits."

Factors which contribute towards deindividuation include anonymity, diffusion of responsibility, the presence of a group, and a shortened time perspective.⁹ These factors, along with the physiological arousal usually linked with noise, excitement and stimulation, are typically associated with crowds.¹⁰ While the notion of deindividuation has roots dating back to the 19^th century and is still a part of the social psychological perspectives explaining the causes of antisocial behaviour to this day, it remains a somewhat contested notion in theoretical terms.

 

3 DEINDIVIDUATION THEORY

The roots of the modern concept of deindividuation are generally acknowledged to lie in the theory of social contagion as outlined by Le Bon in 1896. Drawing on the impact of psychological mechanisms of anonymity, suggestibility, and contagion, Le Bon postulated that "in a mob, the emotions of one person spread through the group ... [leading to] ... a breakdown of normal control mechanisms ... [which] ... result in violent, immoral acts ..."¹¹ Hence, the crowd, in psychological terms, "is a group of individuals who, in specific circumstances, acquire new characteristics that are very different from the characteristics of the individuals that constitute it".¹² As a result of the individual conscious personality submitting to a group unconscious personality,¹³ the crowd consequently constitutes "a single collective being which is guided by a mental unity and a collective soul" that makes individuals feel, think and act differently - even automatically - than they would independently.¹⁴ Though Le Bon's ideas have been criticised, they have formed the basis for virtually all modern deindividuation theorists,¹⁵ and in particular the influential concept of "group mind", the collective consciousness that manipulates the homogenous and highly emotional mass that constitutes the "mob".¹⁶

Festinger, Pepitone, and Newcombe¹⁷ reintroduced a more scientific version of Le Bon's theory into social psychology in 1952. They argued that certain conditions often present in groups can lead individuals to feel as if they have lost their personal identities and merged anonymously into the group, in other words, that such individuals become deindividuated.¹⁸Anonymity is a key factor in deindividuation. The more anonymous the group members are, the less they feel that they have an identity of their own, and the less likely they are to be held accountable for their acts, the more irresponsibly they may behave.¹⁹ Festinger, Pepitone, and Newcombe stated that "under conditions where the member is not individuated in the group, there is likely to occur for the member a reduction of inner restraints in doing certain things".²⁰ However, they did not hypothesise that the loss of individuality is replaced by a collective mind that guides the crowd's actions, but rather that the loss of individuality removes individual controls, which releases a person from individualised moral restraints.²¹ Being "submerged" in the group makes the group more attractive to the individual, who may sometimes seek out such an experience.²²

Further theoretical refinement of deindividuation theory can be credited to Zimbardo,²³ who presented a theoretical framework including the variables leading to deindividuation and the behaviour which results therefrom. Hence, as Vilanova et al summarise Zimbardo's approach:²⁴

"[I]nternal or external variables related to the subject (anonymity, sense of shared or diffused responsibility, numerous groups, altered time perspective, arousal, overload of sensory input, trust that there will not be cognitive interactions, physical involvement in group actions, or altered states of consciousness) cause a state of deindividuation. This state is characterised by changes in the perception of oneself and others, such that self-observation and concern for social evaluation are reduced."

As a result, there is "a weakening of controls based on guilt, shame, fear and commitment, which in turn leads to lowered thresholds for the expression of inhibited behaviour."²⁵ Anonymity as the cause of diminished concern for self-evaluation, which enables individuals to disregard social norms of behaviour, is central to Zimbardo's thinking.²⁶ Zimbardo's proposed model developed the thinking about deindividuation by including both internal variables and variables external to the subject that may cause deindividuation, and that the resulting behaviours could also be prosocial, although his focus was overwhelmingly on antisocial behaviour.²⁷

While certain aspects of Zimbardo's deindividuation theory have been contested, others were subsequently corroborated by Diener and his collaborators,²⁸ including the direct relationship between the diffusion of responsibility and aggressive behaviour,²⁹ the relationship between the reduction of self-consciousness and antinormative behaviour,³⁰ and the direct relationship between arousal and oppositional behaviour.³¹ Diener sought to refine Zimbardo's theory by focusing on the psychological mechanism causing deindividuation, rather than external variables.³² In this regard, the particular intrapsychic aspect of the deindividuation process for Diener is that "when conscious attention is not focused on oneself, the decision of whether to elicit a behaviour is undermined", which in turn "reduces inner restrictions and makes room for antinormative behaviour".³³

Postmes and Spears point out that each of the researchers mentioned above emphasised different aspects of deindividuation: Festinger et al associated deindividuation with the feeling of not being scrutinised or accountable when submerged in the group; Zimbardo focused on reduced self-observation, thus emphasising anonymity; while Diener viewed reduced self-awareness as the defining feature of the state.³⁴ However, the work of these theorists may together be classified as classical deindividuation theory.³⁵

In an effort to further refine deindividuation theory, Prentice-Dunn and Rogers³⁶ sought to distinguish between a decrease in public self-awareness and a decrease in private self-awareness. A decrease in public self-awareness, resulting from accountability cues such as anonymity and diffusion of responsibility, gives rise to a lack of concern with evaluations and a belief that negative consequences will not ensue from behaviour. However, a decrease in private self-awareness results from "attentional cues" such as group cohesiveness and physiological arousal and draws attention away from oneself and one's own behaviour. Prentice-Dunn and Rogers argue that whilst antinormative and disinhibited behaviour can result from both processes, only the reduced private self-awareness route should be defined as deindividuation.³⁷ Postmes and Spears describe this position as contemporary deindividuation theory. Both classical and contemporary views nevertheless agree on the main thrust of the deindividuation hypothesis: The psychological state of deindividuation brings about antinormative and disinhibited behaviour.³⁸

Whilst empirical findings may give general support to this conclusion, unconditional support is, however, lacking.³⁹ The experiments confound a number of different variables (e.g., the effects of anonymity with the effects of being part of a group).⁴⁰ Foster points out that there are a number of questions and limitations which can be identified with regard to deindividuation theory: the methodologically flawed nature of some studies, the fact that there are still questions regarding external validity or generalisation of findings, and the question of whether the purported inner state implies a loss of control, or merely a reduction in control.⁴¹ Moreover, it has been argued, aggression is not the inevitable result of deindividuation in that other behaviours can result,⁴² particularly if the social context provides strong cues.⁴³ In this regard, the Social Identity model of Deindividuation Effects (SIDE) has been developed, which focuses on the positive nature of deindividuation to seek to explain crowd behaviour by an individual's conformity to salient group norms.⁴⁴

According to the SIDE model, rather than leading to a loss of personal identity, deindividuating settings can "facilitate a transition from a personal to a more social or collective identity" and so what appears "antinormative" in fact reflects what is normative in the crowd.⁴⁵ The transition, in terms of this model, is therefore from a personal to a group identity in deindividuating circumstances,⁴⁶ amounting to a change in identity, rather than a loss of identity. The acceptance of a common social identity would also help to explain a group targeting its violence against a particular, opposing, target.⁴⁷

Whilst a discussion of these aspects, unfortunately, falls outside of the ambit of this article, the question has been raised in the context of international criminal law as to how involvement in these grave crimes by ordinary citizens can be explained. While these questions cannot be dealt with in this article, it is perhaps useful to very briefly allude to some of the research done in this particular context, to provide a broader understanding of the attempts to explain participation in group criminality.

In his overview of criminological theories,⁴⁸ Neubacher explores Milgram's experiments on obedience⁴⁹ and associated doctrines of authorisation, routinisation and dehumanisation.⁵⁰ In addition, he examines the application of the theory of neutralisation, in terms of which while the legitimacy of the social order is accepted, the perpetrator learns to neutralise the prevailing social norms in certain situations, which allows him to violate such norms.⁵¹In respect of international crimes committed by otherwise law-abiding State officials, neutralisation by the State in the context of labelling the adversary a political enemy is a particularly powerful means of justifying and explaining macro-crimes.⁵²

Waller seeks to interrogate the same question of how ordinary, rank-and-file individuals come to participate in the most serious crimes and does so by adopting a psychological explanation of the problem.⁵³ In his analysis Waller utilises the following categories to explain conduct giving rise to mass killing and genocide amongst this group of people:

(i) cultural construction of worldview,⁵⁴ which incorporates collectivistic values in terms of which group identity shapes worldview, authority orientation which relates to how the social world may be ordered hierarchically, and social dominance, which deals with how some individuals may dominate within a hierarchical system;

(ii) psychological construction of the "other",⁵⁵ characterised by us-them thinking which reinforces the superiority of the group that the individual is in, moral disengagement which results in certain individuals or groups being placed outside of certain values, rules and considerations of fairness - leading to the dehumanisation of victims, and the tendency to blame victims for their own victimisation; and

(iii) social construction of cruelty,⁵⁶ which may arise and be sustained by professional socialisation through organisational structures, group identification as a result of which "outside" values are excluded and locally generated values dominate, and binding factors of the group, which keep people within an organisation or hierarchy.

All these processes can function in association with the deindividuative processes discussed above to enhance the effect of deindividuation upon the individual.

 

4 DEINDIVIDUATION IN THE SOUTH AFRICAN CRIMINAL LAW

The concept of deindividuation has arisen in the context of group criminal activity, and thus inevitably involves an assessment of whether common purpose liability is applicable. The common purpose doctrine, while it retains a controversial aspect, given its predominance in criminal trials involving group criminal activity in the pre-democratic era, has been increasingly regarded as an important and necessary part of the inquiry into criminal liability in such cases. While the common purpose doctrine considerably predates the period in which it was applied in trials with a political context in the 1980s and early 1990s,⁵⁷ the cases decided in this period were extremely influential in the development of the doctrine, none more so than the seminal case of S v Safatsa.⁵⁸

It cannot be denied that the common purpose doctrine fulfils a very useful role where, using the context of murder, as a result of an attack by a group of perpetrators, a victim has been killed. While the unlawfulness of each member of the group, along with the intention to kill on the part of each member of the group, may readily be established, it may be very difficult to establish who of the group should be held causally liable for the death of the victim. If the ordinary principles of causation were applied, this may indeed be an impossible quest, given the joint conduct of the group members in killing the deceased. Would it be just to simply then acquit all members of the group on a charge of murder? It is clear that such a result would be unacceptable to the legal convictions of the community. This is where the common purpose doctrine comes into its own, neatly avoiding the need to establish a causal link between the act of each individual accused in the group and the harmful result. The imputation of the acts of each member of the group to every other member of the group ensures that the element of causation is not required. The common purpose doctrine may be defined as follows:

"Where two or more people agree to commit a crime or actively associate in a joint unlawful enterprise, each will be responsible for the specific criminal conduct committed by one of their number which falls within their common design."

This definition, as framed by Burchell,⁵⁹ was approved by the Constitutional Court in the case of S v Thebus,⁶⁰ in which case the Court held that the doctrine of common purpose was constitutional, despite its operation dispensing with the element of causation in proving the guilt of each member of the group. It is noteworthy that the form of common purpose under review in the Thebus case was that form of the doctrine which is commonly regarded to be the manifestation that most militates against the sacrosanct principles of criminal guilt, active association common purpose, which incorporates a common purpose formed on the spur of the moment. The other form of the common purpose doctrine,⁶¹ the common purpose predicated upon a prior agreement between the parties, is relatively uncontroversial and resonates with the well-established rules relating to liability for conspiracy.⁶² If there is proof of a prior agreement between the members of the group acting together, then it usually presents no problem to establish that each person in the group associated himself with the others. However, as is frequently the case, no such prior agreement can be established, then the active association form of common purpose may be established where the following five requirements⁶³ are met (again, expressed in the context of murder): (i) presence at the scene of the crime; (ii) awareness of the assault on the victim by another; (iii) intention to make common cause with the person or persons committing the assault; (iv) performance of some act of association with the conduct of the others in the group (indicating common purpose); and (v) intention to kill the victim.

Deindividuation may conceivably arise in either form of conduct flowing from a common purpose, although in practice it has been associated with active association common purpose. Despite the earlier academic criticism of the common purpose doctrine,⁶⁴ since the decision in Thebus, the courts have felt free to apply the doctrine to numerous cases involving criminal conduct committed by a group of persons. Most recently, the Constitutional Court in S v Tshabalala⁶⁵ has swept away any previous doubt regarding the utility of the doctrine, and the appropriateness of its use. The Court held that the doctrine could apply even to autographic crimes (crimes that can only be committed through the instrumentality of a person's own body),⁶⁶ and further held that the dispensing with the need to prove the causation requirement in respect of each member of the group,⁶⁷ which has attracted some academic opposition, was justifiable. As regards the latter aspect, the Court held that the object and purpose of the common purpose doctrine is to overcome an otherwise unjust result which offends the legal convictions of the community, by removing the element of causation from criminal liability and replacing it, as appropriate, with imputing the act which caused the unlawful harm to all the co-perpetrators.⁶⁸

Having set out the current state of the common purpose doctrine, let us turn our attention to those cases in which the question of deindividuation has been canvassed in respect of unlawful activity in the context of a group of persons acting together. Prior to doing so, it may be useful to take into account the idea that even though deindividuation has been raised in cases based on terrible brutality in the mob context, there is an argument that despite the atrocity committed, those involved in its commission are normal people, and that anyone placed "in certain situations, under certain conditions, and subjected to certain pressures and constraints [such as deindividuation] is capable of committing acts of extreme atrocity, cruelty, cupidity ,.."⁶⁹ It should also be noted that where the presence of deindividuation has been indicated by psychiatric or psychological testimony, the South African courts have not regarded this phenomenon as a substantive defence, but instead have regarded it as a factor relevant to mitigation of sentence.

The first case in which the presence of deindividuation was raised as extenuating circumstances was that of Safatsa.⁷⁰ In this case, a crowd of approximately one hundred people attacked the deceased while he was in his house by pelting the house with stones and throwing petrol bombs through the windows. When the deceased attempted to flee, he was caught and stoned with rocks, before petrol was poured over him, and he was set alight. On behalf of those convicted of the murder in Safatsa, it was argued by Professor Tyson, who was described as a highly qualified and experienced psychologist, that they were deindividuated. The court a quo, per Human AJ having considered Professor Tyson's testimony, stated

"I consider, on the basis of my assessment of the psychological literature, that it is highly probable that an individual in a mob situation will experience deindividuation and that this de-individuation will lead to diminished responsibility in much the same way as do the consumption of too much alcohol or great emotional stress."⁷¹

However, despite this affirmation of the potential of deindividuation to found a verdict of diminished responsibility, the court a quo nevertheless did not find that there were extenuating circumstances on the facts of the case. This finding was challenged on appeal, but the Appellate Division did not deviate from the finding of the trial court, holding as follows:

"The views expressed by the witness were of a wholly generalised nature, and unrelated to the individual accused. The generalisation of the probability referred to by the witness cannot be specifically related to any individual accused in the absence of any evidence at all regarding the actual motivation and state of mind of such individual accused. No such evidence was placed before the trial Court."⁷²

As a result, the death sentences imposed on the appellants were confirmed. Shortly thereafter some of the key role-players in the introduction of the concept of deindividuation into South African jurisprudence were reunited in the case of S v Thabetha.⁷³ Human AJ was once again the judge in this matter. Once again, the factual scenario underlying this case related to the killing of an alleged political opponent, after the deceased had been chased, attacked and killed by stoning by a group of persons, among them the accused. The events took place after the funeral of the chairman of a community civic association, who was believed to have been killed by members of a "vigilante" group. The deceased was believed to be a vigilante by the members of the group. While the accused in question all pleaded guilty to murder, it was contended on their behalf that there were extenuating circumstances that militated against the imposition of the death penalty, including youthfulness, deprived family and social backgrounds, provocation (by the vigilantes),⁷⁴ and their psychological state at the time of the crime. In respect of the psychological state of the accused, Professor Tyson was again called as an expert witness to testify about the effects of deindividuation on the accused. As in the Safatsa case, Professor Tyson's qualifications and the correctness of his evidence regarding the phenomenon of mob violence giving rise to the deindividuation of the members of the group were accepted.⁷⁵ The content of Professor Tyson's testimony⁷⁶ was clearly and accurately summarised by Professors Paizes and Skeen in a passage later cited with approval by the Appellate Division,⁷⁷as follows:

"It is not uncommon for people without a violent predisposition to act differently in crowds and to engage in atypical violent behaviour. This is occasioned by a number of factors. First... there are strong pressures on an individual in such circumstances to conform, both because the aggressive conduct of the crowd comes to be perceived as normative and appropriate and because of the fear of disapproval, rejection or even physical harm. There is, too, the question of obedience to authority figures which must be considered in these cases. A third factor is what is referred to by psychologists as 'modelling': a number of studies have shown that people who observe aggressive models are likely to be far more aggressive... as people who observe non-aggressive models. Then, fourthly, there is the question of psychological arousal caused by shouting, singing, dancing or other kinds of physical exertion, which may deprive members of the crowd of rational thought and lead to heightened aggression.

Where all or some of these reactions occur, the result is frequently what is called 'deindividuation', in which a person loses his self-awareness and focuses all his attention on his environment. This state induces behaviour similar to that of people who are hypnotised or intoxicated. It interferes with one's cognitive abilities and hampers one's ability to regulate one's conduct. External cues replace internal standards of behavioural direction and one becomes emotional, impulsive and irrational. And, if additional factors such as provocation and endemic political frustration are added to this already combustible mix, the result may well be diminished responsibility."

Significantly, the Court accepted the conclusion of Professor Tyson that the accused in the "charged atmosphere prevailing that day ... became completely deindividuated, that there was the factor of arousal and there are numerous mitigating circumstances".⁷⁸ In addition to accepting that the psychological processes giving rise to deindividuation constituted extenuating circumstances, the Court held that the youthfulness of the accused was in itself an extenuating circumstance.⁷⁹

The acceptance of deindividuation as an extenuating circumstance was welcomed by Skeen,⁸⁰ with the qualifications that such application occurs in "appropriate and proven" cases, and that such a finding must not be established as a generality, but "in relation to the accused in person", as was indeed the case in Thabetha. The acceptance of deindividuation as a mitigating factor was confirmed in the unreported Eastern Cape case of S v Gqeba,⁸¹ following the "necklacing" (where the deceased was doused with petrol, a tyre was placed around her neck, and she was set alight and burnt to death) of a young woman who had been accused by a crowd of having an affair with a Zulu policeman. In sentencing, the Court took into account the social conditions in which the accused lived, along with factors such as relative deprivation and frustration leading to aggression and, notably, deindividuation. Notwithstanding the brutality of the crime, the Court specifically accepted that all the accused were to some extent deindividuated at the time of the killing and that this constituted an important mitigating factor.⁸² Although the Appellate Division in S v Ncaphayi⁸³ did not specifically discuss the phenomenon of deindividuation, it is evident that it was alive to the influence of a crowd on the mind of an accused, in particular where the accused is youthful.⁸⁴

In the cases which followed that of Ncaphayi, while the acceptance of deindividuation as a possible ground for diminished capacity, and consequently mitigation, had clearly been accepted at this point, some caveats were expressed. It was noted by the Appellate Division in S v Motaung⁸⁵ that even if the court is happy to accept the correctness of the theoretical underpinnings of deindividuation - in this case, provided by the eminent expert witness Professor Diener, an acknowledged expert on this phenomenon⁸⁶ - the application of this notion is dependent on the reliability of the evidence on which it is based.⁸⁷ The Court accepted that deindividuation "is characterised, inter alia, by a lack of self-awareness on the part of the person subjected to the process".⁸⁸ Moreover, as is evident from the case of S v Machasa, the courts' acknowledgement of deindividuation is tempered by the consideration that where it is held that the influence of the crowd did not affect the accused such that he could be regarded as acting with diminished responsibility, there can be no question of deindividuation as such operating to mitigate punishment.⁸⁹

Nevertheless, it is evident that where it was held appropriate, the application of deindividuation to criminality arising out of crowd violence, even where this gave rise to brutality, had come to be accepted by the courts. In S v Matshili,⁹⁰ deindividuation was contended where the appellants, who were striking workers of the then South African Transport Services (SATS) embittered by the fact that those who continued to work allowed SATS to continue its operations without settling the strike, mercilessly executed four non-strikers. Although the Court regarded the killing as "a particularly serious case", it accepted that on the basis of the expert evidence led on behalf of the defence there were strong mitigating factors inherent in the fact that the appellants "were subject to certain powerful, situational forces or influences which caused them to behave in an uncharacteristically violent manner".⁹¹ Noting the reliance on "psychological phenomena" amounting to deindividuation in Safatsa and Thabetha, which evidence was accepted by the court a quo, the Appellate Division affirmed this approach:⁹²

"There is no problem about this. They accord with age-old descriptions of the mob as 'our supreme governors' and 'that great enemy of reason'... And there is no dispute that this mob psychology was, in principle, capable of constituting a mitigating factor (as it did in Thabetha's case). Whether it does is in each case a question of fact, namely did what I call the group influence result in the accused's responsibility being diminished to an extent sufficient to reduce his moral guilt?"

Despite the court a quo holding otherwise, the Appellate Division accepted that in the circumstances of the case the appellants "suffered from a lack of self-restraint, which it is fair to assume they would otherwise have exercised", and that given that they could therefore be regarded as acting with diminished responsibility at the time of the killing, their moral guilt must be held to be reduced, however brutal and reprehensible their conduct.⁹³Therefore, despite the seriousness of the crimes, which would normally merit "the utmost rigour of the law", it was held that the appellants "were subjected to psychological forces which caused them to act in an uncharacteristically violent manner towards persons against whom they had an intense resentment", and that these crimes were consequently committed "under abnormal circumstances".⁹⁴ The death sentence imposed on four of the appellants was therefore set aside, and all received substantial terms of imprisonment.

The Appellate Division adopted a similar approach in the case of S v Khumalo,⁹⁵ where despite describing the actions of the crowd in murdering a municipal policeman as barbaric and repugnant, the Court accepted the testimony of the expert witnesses for the defence, including Professor Tyson, that all the accused were to a greater or lesser extent "deindividualised" at the time of the killing. The Court's understanding of this phenomenon was that it is a temporary state of mind resulting from a combination of external factors, which is characterised by the person in question being so identified with the group in which he finds himself, his attention so completely focused on external factors rather than himself, and being so emotionally swept up and carried away, that he loses his inhibitions, his self-consciousness and his self-control, and acts differently -more impulsively, more aggressively, less rationally - than what he ordinarily would.⁹⁶ The imposed death penalty was accordingly set aside on appeal in favour of imprisonment for all convicted offenders.

The established status of the phenomenon of deindividuation as a mitigating factor was further confirmed in S v Matala,⁹⁷ despite the Court finding that the appellants, who killed a woman as a supposed witch by setting her alight, had not been deindividuated.⁹⁸ It was held by the Appellate Division that even in the absence of expert evidence the possibility of deindividuation being present could be considered by a court.⁹⁹ Similarly, in S v Matela,¹⁰⁰ while the Court found that there was no evidence indicating that the appellants had been deindividuated, the possibility of deindividuation was clearly accepted by the court, based on extensive precedent.

There do not appear to be any further High Court or Supreme Court of Appeal cases where the phenomenon of deindividuation has been raised. It has more recently been suggested that deindividuation ought to have been considered in a case to avail the accused,¹⁰¹ and it is clearly listed and discussed as a mitigating factor associated with diminished responsibility.¹⁰²Although it appears that deindividuation has not been directly pleaded as a mitigating factor for the past couple of decades, at least not in reported cases, there is no question that it is open to defence counsel to do so.

Could the presence of deindividuation go further than merely providing grounds for mitigation, and serve as the basis for a substantive defence, excluding criminal liability? This question has yet to be tested in South African courts, but this possibility is clear from the analogy drawn between the conduct of a deindividuated individual and the conduct of a person who is drunk or has been hypnotised.¹⁰³ It has been postulated that as a result of being in a state of deindividuation (and dependent on the level or intensity of the process of deindividuation), a person may be unable to make rational and moral decisions about his actions,¹⁰⁴ or may be unable to foresee the consequences of his actions.¹⁰⁵ It is submitted that it is highly unlikely that deindividuation may in itself constitute the basis for a substantive defence (it may be different if the deindividuation is also associated with provocation and/or intoxication). As regards its possible effect on capacity, it is well established that the presence of deindividuation can result in a verdict of diminished capacity. However, the evidence of the cases in which deindividuation was successfully pleaded in mitigation indicates that the courts will not look past the dehumanisation¹⁰⁶ inherent in the vengeful and targeted killings in the crowd context, and will find that the accused did not lack self-control, however strong the internal and external influences on their conduct. Similarly, there is no evidence that a defence based on lack of intention would prevail, as in all the cases there was no question that the accused believed themselves to be acting within the law, and were clearly acting with the direct intent to kill in each case.

It should always be remembered in ascertaining the effect of deindividuation, in the usual context of the operation of the common purpose doctrine, that with regard to criminal liability (whatever the position is in psychology or sociology or philosophy)¹⁰⁷ a crowd cannot have a (relevant) intent, and thus "the intent of the crowd" actually refers to the intent of every member of the crowd.¹⁰⁸ The Court in the Khumalo case pointed out that there is no logical objection to a syllogism which states: every member of the crowd (or in brief, the crowd) had the intent to kill; the accused were members of the crowd; therefore the accused had the intent to kill.¹⁰⁹However it must always be borne in mind that what is required to be proved is the intent of each accused, and any attempt to infer such intent from the mental state of a larger group of persons can only be justified where the court has no reasonable doubt that all the members of such group indeed had such a singular intent, and that the accused was a member of such group in the sense that he shared such intent.¹¹⁰ It follows that just as intent must be proven for each accused, proof of deindividuation must be established for each individual accused in order for it to be effectively taken into account as a factor mitigating sentence.

 

5 CONCLUDING REMARKS

It is evident that deindividuation remains one of social psychology's most influential concepts.¹¹¹ Deindividuation theory has been used to explain social atrocities such as genocide¹¹² and has found application in respect of a number of contemporary issues,¹¹³ notably in the context of computer-mediated communication,¹¹⁴ where it has been applied to a wide-ranging set of scenarios, including video games,¹¹⁵ the interaction between Jews and Arabs,¹¹⁶ and gender perspectives.¹¹⁷ Even the role of the presence of other people in an individual's shopping behaviour has been assessed in terms of deindividuation theory. It is evident that the factor of anonymity remains a central consideration in the deindividuation process.¹¹⁸ Vilanova et al identify some further promising areas of research relating to deindividuation:¹¹⁹ the insights that social cognitive neuroscience can bring to understanding the process; the use of mathematical models to better predict the conditions in which deindividuation may occur; seeking ways to prevent transgressive behaviour in deindividuation situations; age differences in relation to complying with group norms; and understanding othering (viewing or treating others as different or alien to oneself).¹²⁰ As a model for understanding group phenomena, deindividuation theory therefore still finds application in modern practice,¹²¹ and its theoretical influence may grow further as it draws from other theories, and is employed to explain essential social issues such as the reduction of prejudice and public order and the enhanced construction of social identity.

What then of the use of deindividuation in the criminal legal context? As suggested above, it is extremely unlikely that deindividuation would be accepted as the foundation for a defence in its own right, although it may conceivably be a factor in establishing an existing defence such as provocation or emotional stress. However, it may be questioned whether the strengthening and validating of the functioning of the common purpose doctrine has effectively put paid to the leading of evidence in mitigation based on deindividuation. This is Kistner's view:¹²²

"With the grounding of the common purpose rule in the Bill of Rights, the scenario evinced in the prominent late apartheid-era cases tried under the common purpose rule, in which expert evidence on crowd psychology was admitted for consideration of mitigation of sentence due to extenuating circumstances, is highly unlikely in common law cases under the 1996 Constitution, dampening the excitement of the social psychologist who discovered a wider public role, greater importance, enhanced scope, and diversification of his client base with the high profile trials of the late 1980s."

Does the fact that deindividuation has not been argued in mob violence cases for some years mean that it cannot be so employed today? There can be no reason in principle why this should be, and indeed, the discussion of deindividuation as a mitigating factor in the relevant literature¹²³ would lend credence to the conclusion that deindividuation may certainly be pleaded in the sentencing context. The practical question that arises in the context of sentencing accused who have been convicted on the basis of the common purpose doctrine is whether it matters - is it not true that courts simply impose a blanket sentence for all found guilty on the basis of common purpose? Would it not be required that all the accused were deindividuated for the sentence to be mitigated?

Kemp et al refer to a number of criticisms of the common purpose doctrine,¹²⁴ including that when "meting out sentence in [common purpose] cases, courts seldom draw a sufficient distinction between the individual participants based on their actual role in and contribution towards the commission of the crime". Is it true that the courts routinely disregard the personal circumstances of those found guilty on the basis of common purpose? The authors refer to the part of the general critique of Burchell of the common purpose doctrine¹²⁵ as support for this criticism. However, Burchell does not deal with sentencing in any detail in this discussion, simply making a brief link between fair labelling and fair sentencing, without elaboration. In any event, the court ultimately has discretion over the sentence imposed on an individual accused, and so, on a conspectus of all the significant factors impacting on sentence, the court has the task of handing down a sentence which is fair and just for each and every accused.¹²⁶ It follows that a court could (and should) apply differing sentences to the accused in the same common purpose, finding grounds for mitigation concerning any members of the common purpose who were deindividuated at the time of acting, as opposed to those members of the common purpose who were not deindividuated.

The lack of reference or application of the phenomenon of reported case law in recent times by no means excludes the courts from considering this issue in relation to contemporary cases. If one takes account of the comments of Milne JA in the Matala case,¹²⁷ it may be that the courts are so convinced of the effect of deindividuation and associated phenomena ("mob psychology") that judicial notice may be taken of these processes. In any event, given the clear connection between violence and factors such as anonymity¹²⁸ and reduced self-awareness, the questions around the effect of being in a crowd remain very pertinent to the courts, particularly in the sentencing process, where a finding of diminished capacity may be based on deindividuation.

 

 

1 The civil law may also be transgressed - see the ruling of the Constitutional Court in SATAWU v Garvas 2013 (1) SA 83 (CC) regarding harm caused to non-protestors by protestors or striking workers, discussed by Khumalo "Developing the Crime of Public Violence as a Remedy to the Violation of the Rights of Non-Protestors During Violent Protests and Strikes - A Critical Analysis of the South African Jurisprudence" 2015 36 Obiter 578.
2 The author first explored these issues in a brief note some two decades ago, in "Crowd Violence and Criminal Behaviour: Dissecting Deindividuation" 2000 21(1) Obiter 161-166. This is an expanded and updated assessment of this topic.
3 Postmes and Spears 1998 123(3) Psychological Bulletin 238.
4 Hogg and Vaughan Social Psychology 7ed (2014) 119.         [ Links ]
5 Louw and Edwards Psychology: An Introduction 2ed (1997) 764; Hogg and Vaughan Social Psychology 434. Aronson, Wilson and Akert Social Psychology 6ed (2007) 283 put it in these terms: "In other words, getting lost in a crowd can lead to an unleashing of behaviours that we would never dream of doing by ourselves".
6 Baron, Branscombe and Byrne Social Psychology 12ed (2009) 399.
7 Mummendey "Aggressive Behaviour" in Hewstone, Stroebe, Codol & Stephenson Introduction to Social Psychology - A European Perspective (1988) 285.
8 Zimbardo The Lucifer Effect (2007) 295.
9 Mummendey in Hewstone et al Introduction to Social Psychology 285; Aronson, Wilson and Akert Social Psychology 284.
10 Middlebrook Social Psychology and Modern Life (1974) 528; Hogg and Vaughan Social Psychology 434.
11 Taylor, Peplau and Sears Social Psychology 12ed (2006) 305; see also Smith and Mackie Social Psychology 2ed (2000) 382.
12 Vilanova, Machado Beria, Brandelli Costa and Koller "Deindividuation: From Le Bon to the Social Identity Model of Deindividuation Effects" 2017 4(1) Cogent Psychology 3.
13 This process serves to insulate the individuals who constitute the group from feelings of social responsibility and fear of reprisal - Wilson and Brewer "Individuals and Groups Dealing With Conflict: Findings From Police on Patrol" 1993 14(1) Basic and Applied Psychology 55.
14 Vilanova et al 2017 Cogent Psychology 3.
15 Foster "Social Influence III: Crowds and Collective Violence" in Foster and Louw-Potgieter Social Psychology in South Africa (1991) 446-7; Hogg and Vaughan Social Psychology 432.
16 Middlebrook Social Psychology and Modern Life 528; Brown "Intergroup Relations" in Hewstone et al Introduction to Social Psychology 384. Though the crowd is made up of people who tend to exhibit automated behaviour, in terms of Le Bon's theory every crowd has a "conductor", an individual who decisively influences the opinion of the crowd -Vilanova et al 2017 Cogent Psychology 3.
17 Festinger, Pepitone and Newcomb "Some Consequences of Deindividuation in a Group" 1952 47 Journal of Abnormal and Social Psychology 382.
18 Atkinson, Atkinson and Hilgard Introduction to Psychology 8ed (1983) 564; Smith and Mackie Social Psychology 382-3.
19 Taylor, Peplau and Sears Social Psychology 305; Smith and Mackie Social Psychology 383.
20 Festinger et al Journal of Abnormal and Social Psychology 382.
21 Postmes and Spears Psychological Bulletin 239.
22 Sabini Social Psychology 2ed (1995) 439.
23 Zimbardo "The Human Choice: Individuation, Reason and Order vs Deindividuation, Impulse and Chaos" in Arnold and Levine (eds) Nebraska Symposium on Motivation (1969) 237.
24 Vilanova et al 2017 Cogent Psychology 4.
25 Postmes and Spears Psychological Bulletin 239.
26 Chang "The Role of Anonymity in Deindividuated Behavior: A Comparison of Deindividuation Theory and the Social Identity Model of Deindividuation Effects (SIDE)" 2008 6(1) The Pulse 2.
27 Vilanova et al 2017 Cogent Psychology 6; Postmes and Spears Psychological Bulletin 239.
28 Vilanova et al 2017 Cogent Psychology 7. See Diener's involvement in the South African case of S v Motaung 1990 (4) SA 485 (A) as an expert witness for deindividuation.
29 Diener, Dineen, Endresen, Beaman and Fraser "Effects of Altered Responsibility, Cognitive Set, and Modeling on Physical Aggression and Deindividuation" 1975 31 Journal of Personality and Social Psychology 328.
30 Diener and Wallbom "Effects of Self-Awareness on Antinormative Behavior" 1976 10 Journal of Research in Personality 107.         [ Links ]
31 Diener, Westford, Diener and Beaman "Deindividuating Effects of Group Presence and Arousal on Stealing by Halloween Trick-or-Treaters" 1973 8 Proceedings of the 81st Annual Convention of the American Psychological Association 219.
32 Vilanova et al 2017 Cogent Psychology 7.
33 Ibid.
34 Postmes and Spears Psychological Bulletin 240ff.
35 Ibid.
36 Prentice-Dunn and Rogers "Effects of Public and Private Self-Awareness on Deindividuation and Aggression" 1982 43 Journal of Personality and Social Psychology 503; Prentice-Dunn and Rogers "Deindividuation and the Self-Regulation of Behavior" in Paulus (ed) The Psychology of Group Influence 2ed (1989) 86.
37 See discussion in Postmes and Spears (fn 2 above) 240.
38 Ibid.
39 Mummendey in Hewstone et al Introduction to Social Psychology 285; Taylor, Peplau and Sears Social Psychology 306.
40 Atkinson et al Introduction to Psychology 566; Smith and Mackie Social Psychology 383.
41 Foster in Foster and Louw-Potgieter Social Psychology in South Africa 450.
42 Atkinson et al Introduction to Psychology 566; Aronson, Wilson and Akert Social Psychology 285; Hogg and Vaughan Social Psychology 434.
43 Taylor, Peplau and Sears Social Psychology 306.
44 Chang 2008 The Pulse 3-4, see Postmes and Spears Psychological Bulletin 241-242, and Kugihara "Effects of Aggressive Behaviour and Group Size on Collective Escape in an Emergency: A Test Between a Social Identity Model and Deindividuation Theory" 2001 40 British Journal of Social Psychology 575. As indicated above, this is consistent with Zimbardo's theory.
45 Postmes and Spears Psychological Bulletin 254.
46 Postmes and Spears Psychological Bulletin 254; Hogg and Vaughan Social Psychology 435.
47 Smith and Mackie Social Psychology 383.
48 Neubacher "How Can It Happen That Horrendous State Crimes Are Perpetrated?" 2006 4 JICJ 787.
49 Neubacher describes Milgram's understanding of the essence of obedience as where a person comes to the point where he "sees himself as a tool that carries out the will of others and is thus no longer responsible for his own actions" (2006 JICJ 789).
50 Kelman built on Milgram's work and identified these aspects. Authorisation occurs where persons see themselves in a "no-choice" situation, as they have to be obedient; routinisation results in opportunities for the questioning of moral responsibility to be minimised through adherence to routine; and dehumanisation, where the perpetrator excludes the victim from the general protection afforded to a member of the community (Neubacher 2006 JICJ 791-792). See fn106 below for Zimbardo's definition of dehumanisation.
51 Neubacher 2006 JICJ 792-794.
52 Neubacher 2006 JICJ 794ff.
53 Waller "The Ordinariness of Extraordinary Evil: The Making of Perpetrators of Collective Violence" in Smeulers (ed) Collective Violence and International Criminal Justice: An Interdisciplinary Approach (2010) 19.
54 Waller in Smeulers Collective Violence and International Criminal Justice: An Interdisciplinary Approach 26-28.
55 Waller in Smeulers Collective Violence and International Criminal Justice: An Interdisciplinary Approach 28-31.
56 Waller in Smeulers Collective Violence and International Criminal Justice: An Interdisciplinary Approach 31 -35.
57 For the historical development of the doctrine see Rabie "The Doctrine of Common Purpose in Criminal Law" 1971 88 SALJ 227.
58 1988 (1) SA 868 (A). In this case the Appellate Division specifically rejected the argument that a murder conviction could only be established if a causal connection could be established between the individual conduct of the accused and the death of the deceased.
59 Burchell Principles of Criminal Law 5ed (2016) 477.
60 s v Thebus 2003 (2) SACR 319 (CC) par 18.
61 It seems that Burchell suggests a third, hybrid form of the common purpose doctrine -Principles of Criminal Law 477ff.
62 As described, for example, in Kemp, Walker, Palmer, Baqwa, Gevers, Leslie and Steynberg Criminal Law in South Africa 3ed (2018) 306.
63 As authoritatively set out in the Appellate Division case of s v Mgedezi 1989 (1) SA 687 (A) 705I-706C, and followed in a number of cases, including: s v Jama 1989 (3) SA 427 (a) 436; s v Barnes 1990 (2) SACR 485 (N) 492; s v Nooroodien 1998 (2) SACR 510 (NC) 517-518; and s v Mbanyaru 2009 (1) SACR 631 (C) par 14.
64 De Wet Strafreg 4ed (1985) 193, Burchell Principles of Criminal Law 485ff, and see Kemp et al Criminal Law in South Africa 286.
65 2020 (2) SACR 38 (CC).
66 On the facts, the autographic crime of common-law rape.
67 To be clear, the causation requirement is not entirely dispensed with, it is simply that there must be a causal link between the conduct of any of those acting in the common purpose and the harmful result. Establishing this causal link suffices for liability for all other members of the common purpose.
68 Par 56.
69 Fattah "Is Punishment the Appropriate Response to Gross Human Rights Violations? Is a Non-Punitive Justice System Feasible?" 2007 Acta Juridica 215-216.
70 S v Safatsa supra. Prior to 1991, the death penalty was mandatory for murder, except where extenuating circumstances were found to be present. Terblanche A Guide to Sentencing in South Africa 3ed (2016) 209 defines extenuating circumstances as circumstances "which influenced subjectively, thus reducing her moral blameworthiness" (original emphasis).
71 S v Safatsa supra 904E-F.
72 S v Safatsa supra 904G-H.
73 1988 (4) SA 272 (T).
74 A psychiatrist, Dr Shevel testified regarding the aforementioned factors in respect of the accused in the Thabetha case supra. The court accepted that Dr Shevel had interviewed all the accused, and that his evidence and recommendations for extenuating circumstances was sincerely and honestly given (285I-J).
75 279D-G.
76 Which may be found at 279G-281A of the Thabetha case supra.
77 The passage appears in 1988 Annual Survey of South African Law 417-418, discussing the Thabetha case. It was cited with approval by the Appellate Division in S v Matshili 1991 (3) SA 264 (A) 271A-F, and is also used to describe the effect of deindividuation by Terblanche A Guide to Sentencing in South Africa 228. It follows that this passage is particularly important for the South African courts' understanding of the content of the phenomenon of deindividuation.
78 286B-C.
79 286A-B.
80 Skeen "Deindividuation and Extenuating Circumstances" 1989 5 SAJHR 81. Skeen regards the need to prove the existence of deindividuation as a sufficient safeguard against spurious claims in respect of extenuating circumstances. In the current context, there is a duty on defence counsel to provide all relevant information to substantiate a plea of mitigation (see Terblanche A Guide to Sentencing in South Africa 99).
81 S v Gqeba ECD Case No 53/89, discussed by Keightley "Mob Violence and Judicial Discretion in Sentencing" 1990 6 SAJHR 296.
82 Keightley 1990 SAJHR 300.
83 1990 (1) SACR 472 (A). Once again, the facts involve the pursuit and killing of a political opponent by a mob, by stoning and stabbing.
84 495D-G. See also the testimony of the expert witness Prof Manganyi which was delivered in this case, and which inter alia sets out deindividuation in crowd situations as a condition favourable to the perpetration of atrocities by otherwise moral individuals (497I-J).
85 Supra.
86 As the court is evidently prepared to do - see 506B-507A, as well as the court's statement at 526I that in "weighing the moral (as opposed to the legal) culpability" of the appellants, the evidence of Prof Diener is directly relevant and helpful.
87 507B-H.
88 506E-F.
89 S v Machasa 1991 (2) SACR 308 (A) 316c-h.
90 Supra.
91 269B-C.
92 271F-H.
93 274A-B.
94 274F-G.
95 1991 (4) SA 310 (A). The deceased, who was apparently targeted as a representative of the erstwhile regime, was chased and severely assaulted before being set alight.
96 360I-361B.
97 1993 (1) SACR 531 (A).
98 537H-J.
99 537F-H, in respect of which the court quotes Hiemstra Suid-Afrikaanse Strafproses 4ed (1987) 625, where the author points out that ordinary people who are not by nature violent or murderers, can develop a sudden tendency to violence when they are emotionally swept up in a group context, and that this can be regarded as a mitigating factor. The court comments that this statement of Hiemstra is "presumably on the basis of common judicial experience".
100 1994 (1) SACR 236 (A). The deceased in this case were four white persons who were pursued, and then assaulted and killed after their vehicle came to an abrupt halt in a township. The court accepted that there was a general tension between black and white communities in the area at the time, as a result of ongoing industrial action, a disparity between resources and opportunities available to the communities, and as a result of attacks by white right-wing extremists on innocent black people in the area. The court held that it was not however possible to link any of these factors directly with the motivation for the brutal killing of the deceased.
101 Whitear-Nel "Recent Cases: Evidence" 2011 24 SACJ 83, discussing the case of S v Ramabokela 2011 (1) SACR 122 (GNP).
102 Terblanche A Guide to Sentencing in South Africa 228.
103 S v Thabetha supra 280E; S v Motaung supra 506C.
104 Drawing into question whether the accused could contend that at the time of the harm being inflicted, he lacked capacity as a result of non-pathological causes. On this defence, see Hoctor Snyman's Criminal Law 7ed (2020) 139ff.
105 In which case the accused would lack the mens rea form of intention. On this element, see Hoctor Snyman's Criminal Law 159ff.
106 Zimbardo The Lucifer Effect 307 describes dehumanisation as follows: "Dehumanisation occurs whenever some human beings consider other human beings to be excluded from the moral order of being a human person. The objects of this psychological process lose their human status in the eyes of their dehumanisers. By identifying certain individuals or groups as being outside the sphere of humanity, dehumanising agents suspend the morality that might typically govern reasoned actions towards their fellows."
107 See in this regard Kistner "'Common Purpose': The Crowd and the Public" www.repository.up.ac.za/handle/2263/51032 (accessed 2022-10-30) (also published in 2015 26(1) Law & Critique 27).
108 S v Khumalo supra 343H.
109 343H-I. The court notes that there are examples of this use of language in the cases of Safatsa 901H-J and Motaung 525D-J in particular.
110 S v Khumalo supra 343I-344A.
111 Postmes, Spears and Lea "Social Identity, Normative Content and 'Deindividuation' in Computer-mediated Groups" in Ellemers, Spears and Doosje (eds) Social Identity: Context, Commitment, Content (1999) 168.
112 Postmes and Spears Psychological Bulletin 238, citing the work of Staub "Cultural-Societal Roots of Violence: The Examples of Genocidal Violence and of Contemporary Youth Violence in the United States" 1996 51 American Psychologist 117, and Staub and Rosenthal "Mob Violence: Cultural-Societal Sources, Instigators, Group Processes, and Participants" in Eron, Gentsy and Schlegel (eds) Reason to Hope: A Psychosocial Perspective on Violence and Youth (1994) 281.
113 Vilanova et al 2017 Cogent Psychology 14-17 discuss the research in some detail, the individual research contributions are listed in the footnotes which follow.
114 See Postmes, Spears and Lea in Ellemers, Spears and Doosje Social Identity: Context, Commitment, Content 168.
115 Webb and Soh "Cheating in Networked Computer Games: A Review" in Proceedings of the 2nd International Conference on Digital Interactive Media in Entertainment and Arts (2007) 105; Chen and Wu "Group Identification as a Mediator of the Effect of Players' Anonymity on Cheating in Online Games" 2015 34 Behaviour and Information Technology 658.
116 Walther, Hoter, Ganayem and Shonfeld "Computer-Mediated Communication and the Reduction of Prejudice: A Controlled Longitudinal Field Experiment Among Jews and Arabs in Israel" 2015 52 Computers in Human Behavior 550.
117 Guegan, Moliner and Milland "Social Asymmetries and Anonymity in Dyadic ComputerMediated Communication" 2016 75 Swiss Journal of Psychology 15.
118 Chang 2008 The Pulse 3, who refers to research relating to the impact of anonymity on violence (Silke "Deindividuation, Anonymity and Violence: Findings from Northern Ireland" 2003 143 Journal of Social Psychology 493), aggression (Douglas and McGarty "Identifiability and Self-Presentation: Computer-Mediated Communication and Intergroup Interaction" 2001 40 British Journal of Social Psychology 399) and the tendency for adolescents to disclose sexual information to others over the Internet (Chiou "Adolescents' Sexual Self-Disclosure on the Internet: Deindividuation and Self-Impression" 2006 41 Adolescence 547).
119 Vilanova et al 2017 Cogent Psychology 17-18.
120 Vilanova et al 2017 Cogent Psychology 18, citing Klein, Spears and Reicher "Social Identity Performance: Extending the Strategic Side of SIDE" 2007 11 Personality and Social Psychology Review 28.
121 See its application to police on patrol in Wilson and Brewer 1993 Basic and Applied Psychology 55.
122 Kistner www.repository.up.ac.za/handle/2263/51032 13.
123 Terblanche A Guide to Sentencing in South Africa 228-229.
124 Kemp et al Criminal Law in South Africa 284
125 Burchell Principles of Criminal Law 485-491.
126 For a case where the court distinguished between offenders who had committed an offence with a common purpose, see S v Mambo 2006 (2) SACR 563 (SCA), in relation to the robbery conviction.
127 S v Matala supra 537E-H.
128 See Silke Journal of Social Psychology 493. Aside from the issue of deindividuation, does the link between anonymity and violence indicate a basis for the retention of the controversial Prohibition of Disguises Act 16 of 1969? For a discussion of this statute see Hoctor "The Offence of Being Found in Disguise in Suspicious Circumstances" 2013 34(2) Obiter 316.