Brief Summary

Criminal Law — Murder — Common purpose — Appellants convicted of four counts of murder during a strike — Appellants initially pleaded not guilty but later changed pleas to guilty, admitting to participation in the murders — The deceased were non-strikers taken by force and killed as an act of intimidation against non-participating workers — Legal issue of whether the death sentence was appropriate in light of mitigating and aggravating factors — Court held that the brutal and premeditated nature of the murders warranted the death penalty for four appellants, while one appellant received a lesser sentence due to found extenuating circumstances.

Comprehensive Summary

Summary of Judgment

1. Introduction

This matter was a sentence appeal in the Supreme Court of South Africa (Appellate Division), arising from convictions on, inter alia, four counts of murder committed during a period of labour unrest. The appeal was confined to sentence and was brought with the leave of the trial judge.

The parties were Wilson Matshili (first appellant), Patrick Molefe (second appellant), Phineas Netshitungulwane (third appellant), Takalani David Mamphaga (fourth appellant), and George Maungedzo (fifth appellant) as appellants, and the State as respondent. In the trial court the appellants were designated accused 2, 5, 6, 7 and 9 respectively, and the judgment on appeal primarily referred to them by those designations.

The appellants (together with other accused) were tried in the Witwatersrand Local Division before SPOELSTRA J sitting with assessors. They were convicted on four murder counts. In respect of accused 2, 5, 7 and 9, the trial court found no extenuating circumstances and imposed the death sentence. In respect of accused 6, extenuating circumstances were found and a sentence of 12 years’ imprisonment (on the four murder counts taken together) was imposed. The present appeal challenged these sentences.

The general subject matter concerned the appropriate punishment for multiple, exceptionally violent murders committed in the context of a strike, and the extent to which psychological “mob” or crowd influences could operate as mitigating factors reducing moral blameworthiness. A further significant feature was the coming into operation, after trial, of the Criminal Law Amendment Act 107 of 1990, which altered the approach to the death sentence and required the appellate court to reconsider sentence afresh.

2. Material Facts

The murders were committed on the night of 28 April 1987 against the background of a strike by a large number of black employees of the South African Transport Services (SATS). The appellants were among the strikers. The four deceased were SATS workers who were not participating in the strike.

On that day, the four deceased and a fifth person, Albert Phuluwa, were taken by strikers from workplaces in Johannesburg to a building called Cosatu House, where hundreds of strikers had gathered. During the afternoon, a call went up from members of the crowd that the five workers should be killed. The court treated the events at Cosatu House and the subsequent attacks as central to both aggravation (brutality, intimidation) and mitigation (the atmosphere and group influence).

At about 20h00, three of the workers were taken by car, accompanied by a number of strikers, to a secluded area outside Johannesburg called Prolecon. There one worker was repeatedly stabbed with a large bread knife, and a rock weighing about 31 kg was repeatedly dropped onto his head. The other two, whose hands and feet were bound, were also struck repeatedly on their heads with the rock. The car returned to Cosatu House to fetch the other two non-strikers. One was Phuluwa, who escaped when they arrived at Prolecon; the other was taken into the bush, tied, stabbed (the knife broke), and then repeatedly struck with the rock. Petrol was poured over the bodies and they were set alight, although the deceased had already died. The district surgeon recorded the cause of death in each case as “multiple injuries” including skull fractures and brain damage.

A key factual feature for sentencing was that the appellants initially pleaded not guilty, but changed their pleas at the end of the State case and tendered pleas of guilty, together with admissions describing their participation. The admissions established that they were at Cosatu House, became aware of the decision to kill the non-strikers, agreed with it, and performed acts which, on the basis of common purpose, rendered them guilty of the four murders.

The admissions delineated each appellant’s role. Accused 2, 5, 6 and 9 went with the first three deceased to Prolecon. Accused 5 stabbed two of the deceased and assisted in holding the victims while they were bound. Accused 9 dropped the rock on the head of one of the first three deceased. Accused 6 did not take part in the attack itself and left after the first killing. Accused 7 arrived later with the fourth deceased (and Phuluwa), threw the rock several times on the fourth deceased’s head, and then struck the other three bodies with the rock. Accused 2 had purchased petrol earlier that day and poured it over the corpses.

As to disputed or contested matters relevant to mitigation, the State challenged aspects of the appellants’ accounts of their subjective states and the expert conclusions drawn from them, contending that the murders were calculated and that time away from Cosatu House (including the drive to Prolecon and the interval between groups of killings) would have dissipated any crowd influence. The court, however, treated the defence evidence on these matters as uncontradicted for purposes of assessing whether the State had displaced a reasonable possibility of diminished responsibility.

3. Legal Issues

The central legal questions concerned sentencing, specifically whether (in light of the Criminal Law Amendment Act 107 of 1990) the death sentence remained the only proper sentence for accused 2, 5, 7 and 9, and whether accused 6’s 12-year imprisonment sentence warranted interference on appeal.

The dispute required the court to engage in the application of legal principles to facts and in a value judgment about appropriate punishment, balancing aggravating features against mitigating considerations and the purposes of punishment. Central to that exercise was whether the evidence of “mob psychology” and situational pressures (including alleged de-individuation, conformity pressure, and obedience pressure) reasonably established diminished moral blameworthiness sufficient to avoid the death penalty.

In addition, the court had to determine the consequence of the new sentencing regime under the 1990 Act for an appeal where the trial court had approached sentence through the then-operative framework of extenuating circumstances. This included determining how to treat onus-related considerations in assessing mitigating factors once the appellate court was required to reconsider sentence afresh.

4. Court’s Reasoning

The Appellate Division held that, since the coming into operation of the Criminal Law Amendment Act 107 of 1990, its task was to consider sentence afresh and to determine whether, having regard to aggravating and mitigating factors and the purposes of punishment, the death sentence was the only proper sentence. Under this approach it was no longer necessary for an accused to prove extenuating circumstances in order to avoid the death penalty.

In evaluating aggravation, the court emphasised the nature and manner of the killings. Four persons were killed in a manner described (in agreement with the trial court) as brutal and gruesome, involving stabbing, binding, crushing head injuries with a large rock, and burning of bodies. The deceased were defenceless, and there was evidence of dolus directus and a lack of impulsivity, with a period of time between the decision to kill and the implementation, and (for some appellants) an interval before the fourth deceased was killed. The court also treated certain acts as indicative of preparation and planning, including accused 2’s purchase of petrol and accused 5’s role in fetching the car used to transport victims.

The court also regarded the motive as significantly aggravating. The murders were not only directed at punishing particular non-strikers but were committed to coerce other workers and pressure SATS, amounting to intimidation and terror. The burning of bodies was viewed as reinforcing the message intended by the perpetrators. The victims were characterised as innocent persons exercising a right to work, given neither an opportunity to explain nor a chance to cease working, and shown no mercy.

On mitigation, the court accepted that the strike context generated severe stressors, including frustration at deadlocked negotiations, termination of employment, lack of income, inability to find alternative work, food shortages, and confrontations with police. The court treated this context as explaining an intense hostility toward non-strikers, who were viewed (unreasonably, but subjectively) as “scabs” or “verraaiers/mpimpis” responsible for the strikers’ predicament. This receptiveness to violence was treated as a mitigating factor in explaining why the appellants were drawn into the idea of killing, even though it was not presented as their own original idea.

A substantial portion of the reasoning addressed the psychological evidence about crowd influence at Cosatu House. The trial court had described the atmosphere as hot, dirty, noisy, marked by singing and dancing, with inflammatory songs and speeches, ululation, and “wild” dancing, with strikers also airing grievances publicly. Expert testimony by psychologists and an anthropology professor was led to the effect that such circumstances can produce crowd violence through pressures of conformity, modelling, arousal, obedience to authority figures, and “de-individuation,” potentially diminishing responsibility. The appellate court noted that such principles had featured in S v Safatsa and Others 1988 (1) SA 868 (A) and S v Thabetha and Others 1988 (4) SA 272 (T), and accepted that mob psychology could, in principle, operate as a mitigating factor.

The court, however, stressed that whether group influence actually diminished an accused’s responsibility was a question of fact in each case. It observed that the trial court had accepted the phenomena in principle but was not persuaded that they sufficiently diminished moral responsibility for accused 2, 5, 7 and 9 under the extenuation test then applied, though it had found them influenced to varying degrees.

In reassessing the evidence, the appellate court took into account the appellants’ own descriptions of anger, confusion, and loss of control, and the experts’ opinions that certain appellants were intensely de-individuated, with possible obedience pressures (particularly in relation to evidence that some were told to stab or strike). The State attacked the expert evidence on multiple bases, including the contention that the experts did not interview appellants, that the trial court had found them poor witnesses and rejected parts of their evidence, that the appellants minimised their roles, and that time away from Cosatu House should have permitted reflection and dissipated the crowd influence. The State argued that the murders were calculated and wicked rather than crowd-driven.

Despite acknowledging that the State’s argument had some merit, the appellate court concluded it could not prevail. It held that the matter had to be approached on the basis that it was for the State to disprove the mitigating factors under consideration, and found that this onus had not been discharged. The appellants’ evidence on their subjective states stood uncontradicted, as did the expert testimony. The court therefore held there was a reasonable possibility that the duration of exposure at Cosatu House was sufficient to affect the appellants, that the influence did not cease upon departure, that although it might have waned it continued to determine conduct to a substantial degree at Prolecon, and that conformity, obedience, and solidarity continued to play a role even where there was some time for reflection.

From this, the court drew the conclusion that the appellants suffered from a lack of self-restraint they would ordinarily have exercised, and therefore acted with diminished responsibility, reducing moral guilt despite the brutality and reprehensibility of the crimes. This inference was reinforced by personal circumstances, including that accused 5, 7 and 9 were first offenders, accused 2’s prior assault convictions were dated and not treated as negating the conclusion, and that all had been employed for years, had family responsibilities, and expressed remorse.

Balancing aggravating and mitigating factors with the purposes of punishment, the court accepted that societal outrage and retribution weighed heavily and that ordinarily the murders would merit the utmost severity. Nonetheless, it held that the cumulative mitigating factors meant that the death sentence was not imperatively called for, that rehabilitation was not excluded, and that deterrence could be adequately served by substantial imprisonment.

In fixing substituted sentences, the court considered differences in participation but held that the roles of accused 5, 7 and 9 did not justify differential punishment among them. It imposed 21 years’ imprisonment (treating the four murder counts as one for sentence) for accused 5, 7 and 9, and 18 years’ imprisonment for accused 2, whose role was treated as somewhat lesser on the basis that his direct participation was confined to bringing and pouring petrol. The court also directed concurrency in relation to sentences imposed on the appellants for certain other related crimes.

In respect of accused 6 (third appellant), the court upheld the trial court’s finding of extenuation and sentence of 12 years’ imprisonment, rejecting the contention that it was unduly severe. It distinguished the sentence imposed on a co-accused (accused 8) on the basis that accused 8’s role was lesser. It also held that the trial judge had been generous in describing accused 6 as merely an approving bystander, pointing to his association with the decision to kill, his escorting of the first three deceased to Prolecon, and his guarding role while victims were tied, notwithstanding that he left after the first killing.

5. Outcome and Relief

The appeal by accused 2, 5, 7 and 9 against sentence succeeded. Their death sentences were set aside and substituted with terms of imprisonment, with the four murder counts treated as one for purposes of sentence. Accused 5, 7 and 9 were each sentenced to 21 years’ imprisonment, and accused 2 was sentenced to 18 years’ imprisonment. The court further ordered that sentences imposed on these appellants in respect of certain other counts were to run concurrently with the new sentences.

The appeal by accused 6 against his sentence of 12 years’ imprisonment was dismissed, and his sentence was left undisturbed.

No separate costs order was recorded in the judgment.

Cases Cited

S v Safatsa and Others 1988 (1) SA 868 (A)

S v Thabetha and Others 1988 (4) SA 272 (T)

Legislation Cited

Criminal Law Amendment Act 107 of 1990

Rules of Court Cited

No rules of court were cited in the judgment.

Held

The Appellate Division held that, under the sentencing approach required by the Criminal Law Amendment Act 107 of 1990, it had to reconsider sentence afresh and decide whether the death penalty was the only proper sentence. Although the murders were exceptionally brutal, planned to a degree, and motivated by intimidation associated with strike enforcement, the court held that there remained a reasonable possibility that powerful crowd influences and situational pressures at Cosatu House diminished the appellants’ self-restraint and reduced their moral blameworthiness. On that basis, the death sentences imposed on accused 2, 5, 7 and 9 were not the only proper sentences and were replaced with substantial terms of imprisonment.

The court further held that accused 6’s 12-year sentence, imposed on the basis of a lesser role and accepted diminished responsibility, was not disturbable on appeal.

LEGAL PRINCIPLES

The judgment applied the principle that, following the commencement of the Criminal Law Amendment Act 107 of 1990, an appellate court must approach sentencing (including the death penalty) by reconsidering the matter afresh, determining whether the death sentence is the only proper sentence after weighing aggravating and mitigating factors and the purposes of punishment, rather than applying a framework that requires proof of extenuating circumstances to avoid death.

It reaffirmed that mob or crowd psychology may, depending on the facts, constitute a mitigating consideration in sentencing, because pressures of conformity, obedience, heightened arousal, and “de-individuation” may diminish an individual’s capacity for self-regulation and reduce moral blameworthiness, even where intention to kill is present.

It further reflected that whether crowd influence reduced responsibility is a fact-based inquiry; where the State fails to displace a reasonable possibility that such influence materially affected the accused’s conduct, the court may treat responsibility as diminished for purposes of sentence.

The judgment also illustrated that, in fixing sentence among co-perpetrators convicted on common purpose, the court may differentiate according to relative participation, but may still impose equal terms where differences do not justify materially different punishment, and may order concurrency where appropriate in relation to sentences on related counts.

S v Matshili and Others (613/89) [1991] ZASCA 48; 1991 (3) SA 264 (AD); (6 May 1991)