A bi-annual update complementing the

Commentary on the Criminal Procedure Act


NO 2 OF 2015


Andrew Paizes, Author (Editor)

Steph van der Merwe (Author)



The feature article in this edition examines a decision of the Supreme Court of Appeal, S v Nkosi
[2015] ZASCA 125 (unreported, SCA case no 20727/14, 22 September 2015), in which this question
arose for consideration: if A, B and C set out to rob D, and A shoots at D who returns fire at the
robbers in justifiable self-defence and shoots and kills C, can B be liable for the murder of C? The
court held that, as long as B had the necessary intention to kill (at least in the form of dolus
eventualis), he should be liable for murdering his fellow robber in these circumstances. The feature
article questions the correctness of this decision. It is argued that, although the decision at first blush
seems to square with the general principles of our criminal law, a more searching analysis suggests that
this may not be so.

The legislation feature contains a summary of some of the more important provisions of an Act that
comprehensively deals with a major world-wide problem: human trafficking. The Prevention and
Combating of Trafficking in Persons Act 7 of 2013, which came into operation on 9 August 2015, was
passed in order to give effect to South Africa’s obligations as a party to significant international
instruments. The summary includes commentary on some of the more important provisions dealing with
the offences and special procedural measures created by the Act.

Some very interesting questions have arisen in the cases that are considered in this issue of CJR. In the
context of the substantive criminal law, they include these: in what circumstances may setting fire to one’s own property constitute the offence of arson? Is the offence of criminal defamation constitutionally compliant?

Some of the questions concerning criminal procedure that are considered include these: where A
and B are tried together and it emerges that A is represented by someone without the right of appearance, what is the impact of this on B’s trial? Who has to put the charge to an accused person and may his or her legal representative enter a plea to the charge on the accused’s behalf? Does the failure to swear in an intermediary vitiate the proceedings? Does a person against whom a warrant to search has been issued have a right to documents or information that led to the warrant being issued? Some of the evidentiary questions discussed concern the question of entrapment and the constitutional validity of the whole of s 252A; the compellability as a witness of the estranged spouse of the accused; and the vexed
question of the status and admissibility of a particular type of evidence: information in digital form
taken from the cell phone of a person following the alleged commission by him or her of an offence.

Andrew Paizes



  • Can a party to a common purpose to rob X be liable for the murder of a fellow robber who has been killed in self-defence by X?


  • The Prevention and Combating of Trafficking in Persons Act 7 of 2013


(a) Criminal Law

  • Arson: Setting fire to one’s own immovable property
  • s 242: Criminal defamation: Is it constitutional?
  • Offences against children: Who is a care-giver for the purposes of s 305(3) of the Children’s Act?

(b) Criminal Procedure and Evidence

(i) Pre-sentence

  • Prosecutorial misconduct: Interference with defence witness
  • s 19: Does a person against whom a warrant to search has been issued have a right to documents or information that led to the warrant being issued?
  • ss 20, 205, 225: The status and admissibility of evidence obtained from seized cell phones following the commission of offences allegedly committed by possessors of those phones
  • s 73(1): Representation of co-accused by someone without right of appearance: Impact on trial proceedings
  • s 105: Accused to plead in response to charge put by prosecutor
  • s 112(1)(b): The ambit of judicial questioning to test a plea of guilty to drunken driving
  • s 165: Administration of the oath by interpreter
  • s 170A: Does the failure to swear in an intermediary vitiate the proceedings?
  • s 195(1): Compellability of estranged wife
  • s 252A: Entrapment and the constitutional validity of s 252A

(ii) Sentencing

  • Remorse as a factor in sentencing

(iii) Appeal and Review

  • Review of a decision not to divert a prosecution: A note on relevant considerations
Can a party to a common purpose to rob X be liable for the murder of a fellow robber who has been killed in self-defence by X?

Yes, said the Supreme Court of Appeal in S v Nkosi [2015] ZASCA 125 (unreported, SCA case no 20727/14, 22 September 2015), where the appellant, Thabo Macbeth Nkosi, appealed unsuccessfully against a conviction for murder in circumstances in which the court, in appropriately Shakespearian style, considered the ‘vexed question’ of ‘[w]hat is fair and what is foul in these circumstances with regard to the appellant’s culpability for his fellow-robber’s death at the hands of the victim’ (at [1]).

Majiedt JA (with whom Mpati P and Shongwe JA agreed) found it had been proved that ‘the appellant and his fellow robbers reasonably foresaw the likelihood of resistance and a shootout, hence the need to arm themselves with loaded firearms’. A shootout did take place between the five robbers and X, the owner of the coal yard that was the target of the robbery. The court described it as a ‘wild shootout in [a] small office’, and it left the deceased, one of the five robbers, fatally wounded. One of the other robbers had been shot in the pelvis and X had sustained a gunshot wound in the leg.

The court relied on the views of Snyman (Criminal Law 5 ed (2008) at 201) who, Majiedt JA said, ‘correctly points out ... [that] our courts have consistently held accused persons who engage in a wild shootout with others, in the course of an armed robbery, criminally liable on the basis of dolus eventualis for the unexpected deaths that may result’ (at [5]). Counsel for the appellant had relied on cases including S v Molimi & another 2006 (2) SACR 8 (SCA), S v Dube & others 2010 (1) SACR 65 (KZP) and S v Mkhwanazi & others 1988 (4) SA 30 (W). The first two were, correctly, found to be distinguishable. More in point, said the court, was S v Lungile & another 1999 (2) SACR 597 (SCA) where, in the course of a robbery at a store, a policeman had arrived at the scene and exchanged gunfire with the robbers, resulting in the death of one of the store’s employees, a result that was shown, on the facts, to have been foreseen by the appellants.

This left the decision in Mkhwanazi to be considered. In that case a neighbouring shopkeeper went to the assistance of a staff member of a Post Office branch that was being robbed by three men, one of whom had a firearm. A shootout ensued between the three robbers and the shopkeeper, who also had a firearm, and one of the robbers was mortally wounded by a shot fired by the shopkeeper. The other two robbers were charged with the murder of their fellow robber. In discharging the accused at the end of the State’s case, Van Schalkwyk J relied on three propositions. First, that dolus eventualis had not been proved since there was no evidence to show that the accused foresaw and were indifferent to the possibility that one of their members might be killed in the course of the robbery. Second, that (as Majiedt JA put it at [9]) ‘the prosecution’s proposition that each of the gang members should be guilty of murder in the event of one of their member’s being killed by a third party, in defence of life or property, was untenable and found no support in the authorities’. And, finally, that the State had failed to prove the causal requirement of the actus reus, since the proximate cause of the death of the deceased robber was the lawful conduct of the shopkeeper.

The first proposition did not apply in Nkosi since the court found that dolus eventualis clearly existed in this case. Majiedt JA then turned to the last two propositions and found them to be ‘contrary to authorities in this court’ (at [11]). The argument on causation was correctly rejected: in Lungile’s case it was correctly held that the second appellant could not rely on the lawfulness of the acts of the policeman, who had acted out of necessity, since, although the policeman’s conduct was lawful, that of the second appellant, who was a party to the execution of an armed robbery, was not. Neither could it be said in Lungile or Nkosi that the conduct of the person doing the shooting was unforeseeable so as to render it a novus actus interveniens.

The second reason given by Van Schalkwyk J is not as easy to dismiss. Majiedt JA turned to the decisions in S v Nkombani & another 1963 (4) SA 877 (A) and S v Nhlapo & another 1981 (2) SA 744 (A), the latter of which he found to have been wrongly distinguished by Van Schalkwyk J in Mkhwanazi. In Nkombani one of the robbers was killed by a gunshot fired by another of the robbers at the intended victim of an attempted hold-up. The majority upheld the conviction of both the gang member who fired the fatal shot and the co-conspirator who had supplied one of the guns and who had not even been at the scene of the attempted robbery and shooting. The latter’s conviction was upheld because ‘the State proved beyond reasonable doubt that he foresaw the possibility of a shooting affray in which one of the henchmen might be hit by a bullet fired by the other’. It was, in other words, ‘an envisaged incident or episode in the crime to which he was a party’ (per Holmes JA at 896A–B).

In Nhlapo a security guard had been shot and killed in the course of a shootout at a large store between three armed security guards and three armed robbers in circumstances in which it was reasonably possible that the shot had been fired by one of the other security guards. Van Heerden JA considered that the appellants had ‘planned and executed the robbery with dolus indeterminatus in the sense that they foresaw the possibility that anybody involved in the robbers’ attack, or in the immediate vicinity of the scene, could be killed by cross-fire’. Majiedt JA could not ‘understand how Nhlapo [was] distinguishable from the factual scenario in Mkhwanazi, as Van Schalkwyk J found’. He concluded (at [13]): ‘on the facts of this case the appellant was well aware that the fact of him and his fellow robbers being armed with loaded firearms may result in a shootout or, as it was referred to in Bergstedt and in Dube, that they may encounter ‘‘dangerous resistance’’. He reasonably foresaw subjectively that, in the course of encountering such ‘‘dangerous resistance’’, the firearms may be used with possible fatal consequences. He was thus correctly convicted of murder and the appeal must fail.’

On the face of it, the court was right since all the elements of the offence (murder) would appear to have been satisfied. There was voluntary conduct, imputed to the appellant by reason of the doctrine of common purpose, which caused the death of the victim, in circumstances in which it is clear that the appellant must have foreseen the real possibility of that conduct causing that result. And yet I am confident that I am not alone in feeling a sense of disquiet about this conclusion. It is, in my view, counter-intuitive and offensive to one’s sense of justice to conclude that the appellant was guilty of murdering his fellow robber. If this were so, why are our law reports not replete with cases in which robbers who fire at policemen or security guards who return fire at them, are charged with attempted murder in respect of each other in circumstances where none of them is killed in the cross-fire? There must be a very large number of cases where this happens, and yet we do not see instances where robbers are charged with—let alone convicted of—attempted murder in such circumstances. The reason is, in my view, that it is wrong in principle to hold persons liable for attempted murder in such cases.

The true reason for this proposition is, however, not self-evident. In CJR 2013 (1) I wrote a feature article entitled ‘The conundrum of the Marikana miners: Can there be liability for murder in such cases?’, in which I set out these reasons, which may be summarised in the following set of propositions:

(1) If a single wrongdoer, A, sets out to rob a bank which is protected by armed guards, and fires at the guards, who return fire and wound A, there can be no basis for holding A liable for attempted murder, since murder is the intentional and unlawful killing of another human being.

(2) If A and B set out to do the same, and A fires at the guards, who fire back and shoot and kill B, is there now a basis for holding A liable for B’s death? On the face of it, one would seem entitled to take the position that, since it was A’s act that caused B’s death, there can be no obstacle to finding him liable for murder once mens rea has been established in the form of, at least, dolus eventualis.

(3) A deeper look at the problem suggests, however, that this position may be an over-simplification. Once A and B have formed a common purpose to rob, any conduct performed by one of them in furtherance of the common purpose will be imputed to the other if it formed (expressly or impliedly) a part of the mandate or agreement between them. In such circumstances, the act of one becomes the act of each or both of them.

(4) One may be forgiven for thinking that it makes little or no difference which of these two formulations one chooses to adopt: what difference can it make whether we view the relevant conduct in Nkosi’s case as the conduct of each of the robbers or the conduct of all the robbers? In my view the choice is material. If, to return to our hypothetical example, the act of shooting at the guards is regarded as the conduct of each robber, it is possible to focus on A’s act alone and defend a position that, since it was A’s act that caused B’s death, A should be held liable for the murder of B. But if the act is that of both A and B, one is alerted to what I believe to be the real problem in cases like these. For, if the act is as much B’s as it is A’s, it may be that one runs into the same obstacle that one encountered in (1) above. To hold A liable for murder in circumstances where the causal act was as much the victim’s as his would make it, at the very least, unclear whether the requirement that ‘another human being’ was killed by the ‘act of the accused’ has been satisfied.

(5) Which of these two ways of describing the situation is to be preferred? I argued, in the previous article, that it was the second that in this context squared more with our sense of justice and public policy. Once an act is committed by one of the parties to a common purpose, and that act falls within the mandate given to that party by the other or others, the law ceases to have its normal concern for determining precisely who did what. Non-actors, in the strict sense, ‘become’ actors, and identity is, for the purpose of determining who did what, lost to the collective. When one of those parties himself turns out to be a ‘victim’ as a result of such conduct, the law faces the difficult task of matching harm which is experienced individually, with conduct which has been attributed collectively. It may, in particular cases, become inappropriate and, even, unjust to ‘unscramble the eggs’, as it were, and attempt to identify individuals within the collective for the purpose of attributing specific criminal responsibility to a particular individual. A disjunction, in other words, may arise when the smooth flow of general principles is disturbed by the creation of a (justifiable) legal fiction. The result is that it may be dangerous to apply those principles mechanically without thought to how and to what extent these distortions may lead to injustice.

Nkosi and Mkhwanazi are, in my view, situations of this kind. Van Schalkwyk J in the latter case sensed it, and he thought it to be ‘inherently untenable’ to convict a person of murder in such circumstances. Such a result, he added, was one ‘surely to be avoided’, whether on the grounds stated by him ‘or the moralistic idea that it is against public policy to convict on a charge of murder in circumstances such as this’. He found no support for it in the authorities and, despite what the court said in Nkosi, I think he was right. Lungile dealt with the shooting of a third party, an employee of the store that was robbed; Nhlapo with the shooting of a security guard. Neither considered in any way the kind of situation that was in issue in Nkosi or Mkhwanazi where the victim was one of the robbers.

It is true that Nkombani did concern the shooting of one party to the common purpose to rob, by another such party, but the context was different. In that case A and B had conspired with X and Y to commit a robbery at a petrol station. X provided a motor car and a pistol with ammunition; Y, who was A’s employer, provided a similar pistol. At the petrol station, a struggle took place in which the attendant, Z, held the wrist of A. A then fired a shot at Z, intending to kill him. The bullet missed and struck B in the head, killing him. The majority upheld the convictions of both A and Y of the murder of B. It was held that A had foreseen the possibility that the shot might strike and kill B and that Y ‘must have foreseen the possibility of bullets flying in a mobile shooting affray, and a confused and desperate scene in which anybody present could be hit by either of the robbers, with their six-shooting automatic pistols’ (per Holmes JA at 895–6).

Is Nkombani, then, authority against the proposition I have made? I think not. As far as A is concerned, he was held liable for his own conduct, and no reliance was placed on the doctrine of common purpose, which was not relevant to his liability. To argue that A’s act of shooting at Z in circumstances where the possibility of killing B was not only foreseeable but actually foreseen as a real possibility, was also the act of B himself is untenable. What of Y, however? The court likened his position to ‘that of a person who reprehensibly hires another to commit a crime, and furnishes him with the required equipment’ (at 896). Thus, although the court used the doctrine of common purpose to uphold his conviction, it would, perhaps, have been more accurate to invoke the principle of instrumentality: A and B were Y’s instruments in performing the attempted robbery. But, even if we treat Y as a party to the common purpose to rob, there would be no warrant for regarding the ambit of the mandate or agreement any differently. If the act in question was not part of the mandate as far as B was concerned, it cannot become so when Y’s liability is considered.

In short, it is my submission that the appellant in Nkosi should not have been convicted of the murder of his fellow robber. Had the shot aimed at the deceased missed its target, I have little doubt that the appellant would not even have been charged with (let alone convicted of) attempted murder. Usually, when it just ‘seems wrong’ to convict in certain circumstances, there are good reasons, in the end, for that instinctive feeling. The reasons are, however, sometimes difficult to articulate. This, in my view, is one such case. 

Andrew Paizes

The Prevention and Combating of Trafficking in Persons Act 7 of 2013

The above Act (hereafter the ‘PCTP Act’) came into operation on 9 August 2015, except for ss 15, 16 and 31(2)(b)(ii) which must still be put into operation. See Proc R32 in GG 39078 of 7 August 2015.

Objects of the PCTP Act
The purpose of the PCTP Act is to give proper effect to South Africa’s obligations as a party to international instruments, such as the United Nations Protocol to Prevent, Suppress and Punish Trafficking in Persons, especially Women and Children. See s 3(a) of the PCTP Act as read with paragraph 1 of the Memorandum which accompanied the Bill [B 7B—2010] that preceded the Act.

The PCTP Act seeks to ‘combat trafficking in persons in a co-ordinated manner’ (s 3(g)) by providing for ‘the co-ordinated implementation, application and administration of this Act, including the development of a draft national policy framework ...’ (s 3(f)).

In the preamble to the PCTP Act, it is stated that South African common law and legislation ‘do not deal with the problem of trafficking in persons adequately’. The PCTP Act does so in a comprehensive manner. For present purposes the emphasis will be on the offences and special procedural measures created by this Act. The extent to which the Criminal Procedure Act 51 of 1977 has been amended and supplemented, will also be identified. It should be noted, however, that the PCTP Act also deals extensively with non-criminal matters.

Chapter 5, for example, deals with the accreditation of organisations to provide services to adult victims of trafficking, whilst Chapter 7 regulates the return and repatriation of victims of trafficking. Section 1 of the PCTP Act states that ‘victim of trafficking’ means a child who is found to be a victim of trafficking after an assessment in terms of s 18(6), or an adult person who has been issued with a letter of recognition as provided for in s 19(10). Section 1 also states that ‘trafficking in persons’ has ‘the meaning assigned to it in s 4(1)’.

The offence ‘trafficking in persons’ in contravention of s 4(1): Elements and possible sentence

The above offence is the core offence in the PCTP Act. Section 4(1) provides as follows:

‘Any person who delivers, recruits, transports, transfers, harbours, sells, exchanges, leases or receives another person within or across the borders of the Republic, by means of—
(a) a threat of harm;
(b) the threat or use of force or other forms of coercion;
(c) the abuse of vulnerability;
(d) fraud;
(e) deception;
(f) abduction;
(g) kidnapping;
(h) the abuse of power;
(i) the direct or indirect giving or receiving of payments or benefits to obtain the consent of a person having control or authority over another person; or
(j) the direct or indirect giving or receiving of payments, compensation, rewards, benefits or any other advantage,

aimed at either the person or an immediate family member of that person or any other person in close relationship to that person, for the purpose of any form or manner of exploitation, is guilty of the offence of trafficking in persons.’

The concept ‘exploitation’ in s 4(1) has an extended and special meaning given to it by s 1 of the PCTP Act. In terms of this section ‘exploitation’ includes—but is not limited to—one or more of the following seven activities, circumstances or phenomena:

(a) All forms of slavery or practices similar to slavery are covered by the term ‘exploitation’; and in terms of s 1 ‘slavery’ means reducing a person by any means to a state of submitting to another person’s control as if the latter were ‘the owner’ of the person concerned.

(b) Sexual exploitation is included in the concept ‘exploitation’; and ‘sexual exploitation’ is, for purposes of the PCTP Act, defined as the commission of any sexual offence referred to in the Criminal Law (Sexual Offences and Related Matters) Amendment Act 32 of 2007 or any offence of a sexual nature in any other law.

(c) Another manifestation of ‘exploitation’ is ‘servitude’ which is, in terms of s 1 of the PCTP Act, a condition in which a person’s labour or services are secured ‘through threats of harm to that person or another person, or through any scheme, plan or pattern intended to cause the person to believe that, if the person does not perform the labour or services in question, that person or another person would suffer harm ...’

(d) Another phenomenon that is included in ‘exploitation’, is ‘forced labour’ which is defined in s 1 as a person’s labour or services obtained or maintained ‘without the consent of that person ... and ... through threats of harm, the use of force, intimidation or other forms of coercion, or physical restraint to that person or another person ...’.

(e) Child labour as defined in s 1 of the Children’s Act 38 of 2005, also falls within the ambit of ‘exploitation’ as defined in s 1 of the PCTP Act. In terms of s 1 of the Children’s Act ‘child labour’ means ‘work [done] by a child which ... is exploitative, hazardous or otherwise inappropriate for a person of that age ... and [which] places at risk the child’s well-being, education, physical or mental health, or spiritual, moral, emotional or social development ...’.

(f) The removal of body parts also constitutes ‘exploitation’. In terms of s 1 of the PCTP Act ‘removal of body parts’ means ‘the removal of or trade in any body part in contravention of any law ...’; and the same section also states that, for purposes of the PCTP Act, ‘body part’ means ‘any blood product, embryo, gamete, gonad, oocyte, zygote, organ or tissue’ as defined in the National Health Act 61 of 2003.

(g) The final method or means of exploitation identified in s 1 of the PCTP Act is the non-consensual impregnation of a person for the purpose of selling her child when the child is born.

Section 1 of the PCTP Act states that the words ‘abuse of vulnerability’ (used in s 4(1)(c)) mean ‘any abuse that leads a person to believe that he or she has no reasonable alternative but to submit to exploitation ...’. In terms of s 1 ‘abuse of vulnerability’ would also include—but would not be limited to—‘taking advantage of the vulnerabilities’ resulting from one or more of the following facts regarding the person concerned: illegal entry into or stay in South Africa; pregnancy; any disability; drug-addiction; being a child; social or economic circumstances. Indeed, it might very well be correct to say that ‘abuse of vulnerability’ as identified in the PCTP Act probably constitutes the most prevalent—but at the same time the least visible—form of trafficking in persons: the perpetrator takes control of the body and freedom of another person in circumstances where the latter is insecure, weak and unprotected.

A person convicted of the offence of trafficking in persons in contravention of s 4(1) as set out above, is in terms of s 13(a) of the PCTP Act—but subject to the provisions of s 51 of the Criminal Law Amendment Act 105 of 1997—liable to a fine not exceeding R100 million or imprisonment, including imprisonment for life, or such imprisonment without the option of a fine or both.

The offence ‘trafficking in persons’ in contravention of s 4(2): Elements and possible sentence
In terms of the above section a person is guilty of an offence if he or she should—within or beyond the borders of South Africa—adopt a child (s 4(2)(a)) or conclude a forced marriage with another person (s 4(2)(b)) for ‘the purpose of the exploitation of that child or other person in any form or manner ...’. The concept ‘exploitation’ must be understood in the context of the extended meaning given to it in s 1 of the PCTP Act.

This meaning was set out above, sv The offence ‘trafficking in persons’ in contravention of s 4(1): Elements and possible sentence. For purposes of the PCTP Act, a ‘child’ is ‘a person under the age of 18 years’ (s 1); and a ‘forced marriage’ means ‘a marriage concluded without the consent of each of the parties to the marriage’ (s 1).
The fact that the child concerned was adopted legally, is no defence if the adoption was for the purpose of exploitation. This much is clear from the fact that s 4(2)(a) refers to an adoption ‘facilitated or secured through legal or illegal means’.

A person convicted of contravening s 4(2) is liable to a fine not exceeding R100 million or imprisonment, including imprisonment for life, or such imprisonment without the option of a fine or both. See s 13(b) of the PCTP Act.

The offence ‘debt bondage’ in contravention of s 5: Elements and possible sentence
In terms of the above section any person who intentionally engages in conduct that causes another person to enter into debt bondage, is guilty of an offence. Section 1 states that ‘debt bondage’ means the following:

‘[T]he involuntary status or condition that arises from a pledge by a person of—

(a) his or her personal services; or
(b) the personal services of another person under his or her control,

as security for a debt owed, or claimed to be owed, including any debt incurred or claimed to be incurred after the pledge is given, by that person if the—
(i) debt owed or claimed to be owed, as reasonably assessed, is manifestly excessive;
(ii) length and nature of those services are not respectively limited and defined; or
(iii) value of those services as reasonably assessed is not applied towards the liquidation of the debt or purported debt;’

The possible sentence for contravention of s 5 is a fine or imprisonment for a period not exceeding 15 years or both. See s 13(c) of the PCTP Act.

Identification and travel documents: Contravening s 6
The above section provides that any person who—in facilitating or promoting trafficking in persons—possesses (or intentionally destroys, confiscates, conceals or tampers with) any actual or purported identification document, passport or other travel document of a victim of trafficking, commits an offence.

A person convicted of contravening s 6 is liable to a fine or imprisonment for a period not exceeding 10 years or both (s 13(d)).

The offences created by ss 7 and 8
The offences created by the above sections are aimed at all those persons or agents who facilitate, promote, encourage or benefit from trafficking in persons.
Section 7 criminalises making use of the services of victims of trafficking, whereas s 8 criminalises certain conduct which facilitates trafficking in persons. Section 8(1)(c) states, for example, that a person commits an offence if he or she intentionally leases premises or subleases premises for facilitating or promoting trafficking in persons; and s 8(1)(d) makes it an offence to finance, control or organise the commission of an offence under Chapter 2 of the PCTP Act.

Legal duties and criminal liability of electronic communications service provider: s 8(2) to (5)
Section 8(1)(c) states that a person commits an offence if he or she intentionally advertises, publishes, prints, broadcasts, distributes (or causes the advertisement, publication, printing, broadcast or distribution of) information that facilitates or promotes trafficking in persons by any means, including the use of the Internet or other information technology. In terms of s 8(2)(a) an electronic communications provider operating in South Africa is required to take all reasonable steps to prevent the use of its service for the hosting of information referred to in s 8(1)(c). A service provider who fails to do so commits an offence (s 8(3)). Furthermore s 8(2)(b) states that an electronic communications service provider that ‘is aware or becomes aware of any electronic communications which contain information’ referred to in s 8(1)(c), and which is stored upon or transmitted over its system, must without delay convey certain information to the South African Police Service (s 8(2)(b)(i)), must ‘take such reasonable steps as are necessary to preserve evidence as may be required by the relevant investigation and prosecuting authorities’ (s 8(2)(b)(ii)) and must ‘without delay take such reasonable steps as are necessary to prevent continued access’ to the electronic communications concerned (s 8(2)(b)(iii)). An electronic communications service provider that fails to comply with the provisions of s 8(2)(b) is guilty of an offence (s 8(3)).

However, it should also be noted that the liability of an electronic communications service provider is qualified in two important respects by the provisions of the PCTP Act: s 8(4) provides that nothing in s 8 places a general obligation on a service provider to monitor data transmitted or stored by it (s 8(4)(a)) or to actively seek facts or circumstances indicating an unlawful activity (s 8(4)(b)).

In terms of s 1 of the PCTP Act an ‘electronic communications provider’ means ‘a person who is licensed or exempted from being licensed in terms of Chapter 3 of the Electronic Communications Act ... 36 of 2005 ... to provide an electronic communications service ...’

Liability of carriers for certain matters relating to trafficking in persons
Section 1 of the PCTP Act states that a ‘carrier’ includes ‘a person who is the owner or employee of the owner, an agent, an operator, a lessor, a driver, a charterer or a master, of any means of transport’.
Trafficking of persons, by its very nature, often involves the movement or transport of the victim, whether it be within South Africa or across its borders. The PCTP Act therefore deals specifically with the liability of ‘carriers’ as defined in s 1.

Sections 9(1) provides that a carrier who transports a person within or across South Africa’s borders—and who knows that the person concerned is a victim of trafficking or ought reasonably to have known that the person concerned is such a victim—is guilty of an offence. The possible sentence is a fine or imprisonment for a period not exceeding five years (s 13(e)).

Where a carrier suspects on reasonable grounds that any of its passengers is a victim of trafficking, it must immediately report this suspicion to a police official for investigation (s 9(2)). A carrier’s failure to do so constitutes an offence (s 9(3)).

Section 9(4) contains a provision which seeks to ensure that—in certain circumstances—the state (or the South African taxpayer for that matter) will not incur expenses for problems created by carriers involved, directly or indirectly, in trafficking of persons. Section 9(4) states:

‘A carrier is liable to pay the expenses incurred or reasonably expected to be incurred in connection with the care, accommodation, transportation and repatriation or return of the victim to his or her country of origin or country or place from where he or she was trafficked, if the court finds, on a balance of probabilities, that the carrier has knowingly transported a victim of trafficking or ought reasonably to have known or suspected that it was transporting a victim of trafficking.’

Trafficking of a child by parent or guardian: s 36
The above section provides that if a children’s court (that is, a court as referred to in s 42 of the Children’s Act 38 of 2005) has reason to believe that a child’s parent or guardian—or any other person who has parental responsibilities and rights in respect of a child—has trafficked the child, the court concerned may suspend all parental responsibilities and rights of that parent, guardian or other person (s 36(1)(a)) and place the child concerned in temporary safe care in terms of s 152 of the Children’s Act (s 36(1)(b)), pending an inquiry by a children’s court.

However, in terms of s 36(2) any action taken by a children’s court in terms of s 36(1) ‘does not exclude a person’s liability for committing any offence under Chapter 2’ of the PCTP Act.

Factors to be considered in sentencing: s 14
Section 14 states that a court that must sentence a person convicted of any offence under Chapter 2 of the PCTP Act, must consider—but is not limited to—‘aggravating factors’ identified in s 14(a) to (l). It is not entirely clear why the legislator chose to refer to ‘aggravating factors’ when it is clear that the factors in the list (a) to (l) are really open-ended factors which still require or call for specific factual findings before they can be identified as ‘aggravating factors’ (or, it would appear, mitigating factors, depending on the factual finding and the context of each case).

Be that as it may, the court is required to consider the following: the significance of the convicted person’s role in the trafficking process (s 14(a)) and the nature of the relationship between the convicted person and the victim (s 14(j)) as well as the question whether the victim’s drug addiction was caused by the convicted person (s 14(c)) and whether the latter has any previous convictions relating to the offence of trafficking in persons or related offences (s 14(b)). As far as the victim is concerned, the court must take into account whether the victim was held captive for any period (s 14(e)), in what conditions he or she was so held (s 14(d)) and whether the victim was a child (s 14(i)) or had any physical disability (s 14(l)). The court is also required to consider the state of the victim’s mental health (s 14(k)) and the extent of the abuse, if any, suffered by the victim (s 14(f)) as well as the physical and psychological effects the abuse had on the victim (s 14(g)).

In the preamble to the PCTP Act reference is made to the increase of trafficking in persons and the role played by organised criminal networks in the trafficking of persons globally. It is therefore necessary that the sentencing court should also consider ‘whether the offence formed part of organised crime’ (s 14 (h)).

Extra-territorial jurisdiction: s 12
There is a need for extra-territorial criminal jurisdiction when national legislation deals with crimes of international or transnational concern. See the discussion of s 110A in Commentary, sv Extra-territorial jurisdiction and universal jurisdiction: Some statutory developments. Examples of such statutory provisions are s 15 of the Protection of Constitutional Democracy against Terrorist and Related Activities Act 33 of 2004; s 6 of the Prevention and Combating of Torture of Persons Act 13 of 2013 and s 35 of the Prevention and Combating of Corrupt Activities Act 12 of 2004.

Indeed, trafficking in persons is not only a crime of national concern, but also one of transnational and international concern; and the PCTP Act falls into the category of statutes referred to in the previous paragraph.

Section 12 of the PCTP Act therefore provides for extra-territorial jurisdiction. In terms of s 12(1) a South African court has jurisdiction in respect of an act committed outside South Africa ‘which would have constituted an offence’ under Chapter 2 of the PCTP Act had it been committed in South Africa, ‘regardless of whether or not the act constitutes an offence at the place of its commission ...’. However, one or more of the following connecting factors must be present, namely that the person to be charged

‘(a) is a citizen of the Republic;
(b) is ordinarily resident in the Republic;
(c) has committed the offence against a citizen of the Republic or a person who is ordinarily resident in the Republic;
(d) is, after the commission of the offence, present in the territory of the Republic, or in its territorial waters or on board a ship, vessel, off-shore installation, a fixed platform or aircraft registered or required to be registered in the Republic;
(e) is, for any reason, not extradited by the Republic or if there is no application to extradite that person; or
(f) is a juristic person or a partnership registered in terms of any law in the Republic.’
In terms of s 12(2) only a High Court has jurisdiction in respect of an accused referred to in s 12(1)(d).

Extra-territorial jurisdiction and the powers of the National Director of Public Prosecutions (NDPP): s 12(4) and (5)
In terms of s 12(4)(a) the NDPP must—in writing and subject to s 12(4)(b)—designate an appropriate court in which to prosecute a person accused of committing an offence under Chapter 2 of the PCTP Act outside South Africa as provided for in s 12(1).

Section 12(4)(b) states that for purposes of determining a court’s jurisdiction to try the offence, the offence is deemed to have been committed either ‘at the place where the accused is ordinarily resident’ (s 12(4)(b)(i)) or at the accused’s ‘principal place of business’ (s 12 (4)(b)(ii)).

The institution of a prosecution in terms of s 12 must be authorised in writing by the NDPP (s 12(5)). A statutory requirement that a written authorisation (whether it be by the NDPP or a Director of Public Prosecutions) is necessary must be understood in the context of what Gorven J said in Booysen v Acting National Director of Public Prosecutions & others 2014 (2) SACR 556 (KZD) at [20]: ‘The purpose is ... to facilitate an ability to prove that the requisite empowered person has in fact made the decision in question. The existence of writing is a jurisdictional fact required to be in place before a prosecution can proceed.’ This was said with reference to s 2(4) of the Prevention of Organised Crime Act 121 of 1998. However, the same would be true of s 12(5) of the PCTP Act. For further examples of statutes requiring the written authorisation of the NDPP for purposes of a prosecution, see Chapter 1 in Commentary, sv Written authorisation of NDPP required for prosecution of certain offences.

Legal duty to report trafficking in persons: ss 18 and 19
There is no general legal duty (as opposed to a possible moral duty) on members of the public to report crime. See the discussion of this matter in Chapter 1 in Commentary, sv The prosecution, the police, the public and the reporting and investigation of crime, where various examples of a statutory duty to report certain crimes, are identified. The PCTP Act can now be added to the list of statutes creating a duty to report certain criminal activities.

Section 18(1)(a) provides that—despite any law, policy or code of conduct prohibiting the disclosure of personal information—any person who knows (or ought reasonably to have known or suspected) that a child is a victim of trafficking, must immediately report that knowledge to a police official for investigation. A designated child protection organisation—as defined in s 1 of the Children’s Act 38 of 2005—has a similar duty. See s 18(1)(b) of the PCTP Act. There is also a duty to report in respect of an adult victim of trafficking, but here the duty to report arises only if the person required to report has ‘during the execution of his or her duties’ come ‘into contact’ with the adult concerned (s 19(1)(a)). An accredited organisation is also required to make a report in respect of an adult victim of trafficking (s 19(1)(b)). An ‘accredited organisation’ is an organisation, including a government institution, accredited in terms of s 24 of the PCTP Act to provide services to adult victims of trafficking. See s 1 of the PCTP Act.

A failure to report (as required by s 18 in respect of children or by s 19 in respect of adults) constitutes an offence and can attract a sentence of a fine or imprisonment for a period not exceeding five years or both (ss 18(9) and 19(13) as read with s 13(e)).

Section 19(2) provides that any person—other than those referred to in s 19(1)(a)—‘who on reasonable grounds suspects that an adult person is a victim of trafficking, may report that suspicion to a police official for investigation’. It would appear that the main purpose of this provision—which is somewhat superfluous—is to make it clear that a voluntary report remains possible if the duty to report as envisaged in s 19(1) does not arise. Another purpose of s 19(2) would be to bring the person who takes the initiative to report within the ambit and protection of s 19(4) as discussed below.

The duty to report and legal professional privilege: ss 18(2) and 19(3)
A legal practitioner is not obliged to report as required by ss 18(1) and 19(1) if his or her report would be based on information that came to light as a result of communications falling within the ambit of legal professional privilege or litigation privilege. This much is made clear by ss 18(2) and 19(3). Section 18(2) provides, for example, that s 18(1) does not apply to the situation where a lawyer’s information is based upon communications made between him and his client for purposes of legal advice or pending or contemplated litigation or litigation which has commenced (s 18(2)(a)) or communications between the lawyer and a third party for purposes of litigation which is pending or contemplated or which has commenced (s 18(2)(b)). Section 19(3) contains similar provisions as regards the duty to report that an adult is a victim of trafficking.

Reporting of suspected trafficking and the provisions of ss 18(3) and 19(4): Protection of identity
The person who is under a legal duty to report or who has made a voluntary report in terms of s 19(2), is required to provide the police with reasons for his or her knowledge or suspicion (ss 18(3)(a) and 19(4)(a)).

The person who reports is entitled to have his or her identity kept confidential if his or her safety is at risk as a result of the report, unless the interests of justice require otherwise (ss 18(3)(c) and 19(4)(c)). This right to confidentiality of identity should be read in the context of the common law’s ‘informer’s privilege’ which is examined in Commentary in the discussion of s 202, sv Information given for the detection of crime. In Swanepoel v Minister van Veiligheid en Sekuriteit 1999 (2) SACR 284 (T) it was held that an informer has a substantive right to the non-divulgement of his identity and that a summons in which delictual damages were claimed for the unlawful, malicious and intentional divulgement by the police of the identity of the plaintiff, who was an informer, to the suspects concerned, disclosed a cause of action. It can, accordingly, be argued that the provisions of ss 18(3)(c) and 19(4)(c) place a legal duty on the police, or anybody else, to respect the confidentiality of the identity of the person who is obliged to report or who has made a report in terms of s 19(2). When will ‘the interests of justice’, as stipulated in ss 18(3)(c) and 19(4)(c), require disclosure of identity? Here, too, one can probably compare the position with the ‘informer’s privilege’ which must be relaxed, for example, when disclosure is material to the ends of justice. See the discussion of Roviaro v United States 353 US 53 (1957) in Schwikkard & Van der Merwe Principles of Evidence 3 ed (2009) at 169. In this case the protection as regards the informer’s identity was lifted because the informer could, on the facts of the case, have been a crucial defence witness on the issue whether the accused had knowingly transported illegal drugs as charged.

There will always be competing interests—and the constitutional right to a fair trial must inevitably be the decisive factor. But in the ordinary course of events, a fair trial is not placed in jeopardy simply because the accused does not know who reported him. In fact, it must also be appreciated that protection of the identity of people who report crime—whether they are statutorily obliged to do so or whether they do so in their capacity as informers—contributes enormously to the proper functioning of the criminal justice system. ‘The informer system’ said Kriek JP in Els v Minister of Safety and Security 1998 (2) SACR 93 (NC) 100b, ‘is one of the cornerstones of the battle against organised crime, and when the identity of one informer is made known, other informers ... will desist from ‘‘informing’’ ...’.

Sections 18(3)(c) and 19(4)(c) are not the only statutory provisions which seek to protect the identity of people who report crime. Other examples can be found in s 38(3) of the Financial Intelligence Centre Act 38 of 2001 and s 17(9) of the Protection of Constitutional Democracy against Terrorist and Related Activities Act 33 of 2004.

The duty to report and exemption from liability: ss 18(3)(b) and 19(4)(b)
In terms of the above sections the person who makes his or her report in good faith, is not liable to civil or disciplinary action on the basis of the report, despite any law, policy or code of conduct prohibiting the disclosure of personal information.

Criminal prosecution of victim of trafficking and duties of prosecutors: s 22
The above section is an important one in that it sets in motion and creates a process which may result in the diversion of a criminal trial where the accused happens to be a victim of trafficking. This provision fits in with one of the broad themes of the PCTP Act, namely protection of the human dignity of the victims of trafficking.

Section 22(1) states that when deciding whether to prosecute a victim of trafficking, the prosecutor must give due consideration to the question whether the offence was committed as a direct result of the accused’s position as a victim of trafficking. This provision is not in conflict with principles governing the prosecutor’s discretion to prosecute; it is, furthermore, entirely consistent with the broad considerations contained in the Prosecution Policy issued by the NDPP in terms of s 21(1)(a) of the National Prosecuting Authority Act 32 of 1998. See generally the discussion of the prosecutor’s discretion and the Prosecution Policy in Chapter 1 in Commentary, sv The discretion to prosecute and sv The discretion to prosecute and the prosecution policy issued by the National Director of Public Prosecutions.

Section 22(2) contains the following procedural novelty: if a prosecutor—in the course of a criminal prosecution—on reasonable grounds suspects that the accused is a victim of trafficking, he must apply to the court for postponement (s 22(2)(a)) and must, in the prescribed manner, refer the accused concerned to the provincial department of social development, which must conduct an assessment in terms of s 18(6) if the accused is a child or s 19(8) if the accused is an adult. The ‘prescribed manner’ in which the prosecutor is required to make his referral as provided for in s 22(2)(b) is found in regulations called the Prevention and Combating of Trafficking in Persons Regulations Relating to Prosecutor’s Referral of Suspected Victims of Trafficking in Persons, 2015. These regulations were made by the Minister of Justice and Correctional Services under s 43(1)(a) of the PCTP Act and came into operation on 21 August 2015. See GN R737 in GG 39119 of 21 August 2015.

Section 22(3) must be followed once the required assessment concerning the possible status of the accused as a victim of trafficking has been made:

‘A letter of recognition that an adult person is a victim of trafficking or a finding by the provincial department of social development after an assessment referred to in section 18(6) that a child is a victim of trafficking serves as a ground for the withdrawal of the criminal prosecution or the discharge of the victim of trafficking if the prosecutor is satisfied that the offence was committed as a direct result of the person’s position as a victim of trafficking.’

Confirmation that the accused is a victim of trafficking does not result in the automatic withdrawal of the charges or discontinuance of the trial: the crucial test is whether the prosecutor is satisfied that the commission of the offence was a ‘direct result’ of the accused’s position as a victim of trafficking. But even if the prosecutor is so satisfied, all the other principles and factors governing the decision to prosecute must also ultimately be taken into account.

It should be noted that in terms of s 22(4) no criminal prosecution may be instituted against a person referred to in s 22(1), and no prosecution may proceed against a person referred to in s 22(2), without the written authority of the Director of Public Prosecutions (DPP) having jurisdiction. The DPP’s written authorisation is a jurisdictional fact and must be in place before the prosecution or continuation of the prosecution, as the case may be, can proceed. A prosecution without this authorisation would be a nullity, whereas a prosecution instituted without written authorisation required in terms of policy directives (as opposed to a specific statutory provision) need not necessarily be treated as a nullity. See Masinga v National Director of Public Prosecutions & another (unreported, KZP case no AR 517/2013, 7 May 2015) which is discussed in Chapter 1 in Commentary, sv The irregularity of a prosecution instituted without the written authorisation required in terms of policy directives: Nature and consequences.

Compensation to victim of trafficking: s 29
The above section makes provision for compensation to a victim of trafficking in addition to any sentence a court may impose in respect of an offence under Chapter 2 of the PCTP Act. In terms of s 29(1)(a) a court may—of its own accord or upon the request of the victim of trafficking or the prosecutor—order the convicted person to pay appropriate compensation to any victim who, as a result of the offence, suffered one or more of the following: damage to or the loss or destruction of property, including money (s 29(1)(a)(i)); physical, psychological or other injury (s 29(1)(a)(ii)); being infected with a life-threatening disease (s 29(1)(a)(iii)); or loss of income or support (s 29(1)(a)(iv)). Section 29(1)(b) stipulates that ‘appropriate compensation’ as used in s 29(1)(a) includes expenses reasonably expected to be incurred in relation to the matters identified in s 29(1)(a)(i) to (iv).

Section 29(1) is considerably wider than s 300 of the Criminal Procedure Act 51 of 1977. Section 300(1) limits compensation to offences which caused damage to or loss of property (including money) belonging to another person and does not include the matters identified in s 29(1)(a)(ii), (iii) and (iv) of the PCTP Act.

Section 29(2) determines that where the amount of the damage, loss or injury exceeds an order for compensation which can be made by a magistrate’s court, namely R300 000, the victim may institute a civil action for the recovery of the excess. See also ss 29(1) of the PCTP Act as read with s 300(1)(a) of the Criminal Procedure Act and GN R62 in GG 36111 of 30 January 2013.

Section 29(1)(a) of the PCTP Act also provides that the provisions of s 300(1)(a), (2), (3) and (4) of the Criminal Procedure Act apply—with the necessary changes as required by the context—to an order made under s 29(1)(a). One result of this incorporation of certain provisions in s 300 is that a compensatory order in terms of s 29 has the effect of a civil judgment.

It should be noted that upon a prosecutor’s application a court may—quite apart from any sentence and compensation order as regards the victim—also order the convicted person, or a carrier as provided for in s 9(4), to compensate the state (s 30(1)). Here, too, the order has the effect of a civil judgment (s 30(2)).

Prescription of the right to prosecute: s 48
The above section amended s 18 of the Criminal Procedure Act so that there shall be no prescription period for the prosecution of the offences as provided for in ss 4, 5 and 7 (and involvement in these offences as provided in s 10) of the PCTP Act.

Competent verdicts: s 48
The above section inserted s 261A into the Criminal Procedure Act. This new section stipulates no less than 16 offences which can serve as possible competent verdicts where a person is charged with trafficking in persons in contravention of s 4 of the PCTP Act, or where the charge is the offence of involvement in trafficking as provided for in s 10 of the same Act.

Right of complainant to make representations in certain matters regarding the convicted offender’s placement on parole or under correctional supervision: s 48
Section 299A(1) of the Criminal Procedure Act lists the offences in respect of which a complainant has the right as described above. Section 48 of the PCTP Act has now added contraventions of the following sections in the PCTP Act to the list in s 299A: 4, 5, 7 and 10.

Transitional arrangements: s 49(1)
In terms of s 49(1) of the PCTP Act criminal prosecutions ‘instituted in terms of any law, in respect of conduct which constitutes an offence provided for in Chapter 2, but which were instituted prior to the commencement of this Act, must be continued and concluded as if this Act had not been passed’.

Amendment of schedules to the Criminal Procedure Act: s 48
Schedules 1, 2, 5 and 6 to the Criminal Procedure Act have been amended by s 48 of the PCTP Act which added certain contraventions of the PCTP Act to these schedules.

Remarks in conclusion
The PCTP Act is an elaborate Act which can make an important contribution to the prevention and combating of trafficking in persons within or across the borders of South Africa. But its ultimate success will depend on the extent to which prosecutors and law enforcement officers have the knowledge, means and commitment to secure its objects as set out in s 3.

 (a) Criminal Law

Arson: Setting fire to one’s own immovable property

Can an accused person who sets fire to his own immovable property be convicted of arson; and if so, in what circumstances? This was one of many questions the Supreme Court of Appeal had to consider in S v Dalindyebo [2015] ZASCA 144 (unreported, SCA case no 090/2015, 1 October 2015). In that case the appellant, who held the land in question as hereditary monarch for the benefit of his tribe and subjects, was not able to alienate the land without the approval of the state, so he could not claim that the property was his to set fire to at will. But even if he was, the court held, the development of our law was contrary to his submission that he could, with impunity, set fire to the immovable property merely because he was the owner.

The court found different definitions of the crime in various academic commentaries. According to JRL Milton (South African Criminal Law and Procedure Volume 2: Common Law Crimes 3 ed (1996) at 777), arson ‘consists in unlawfully setting an immovable structure on fire with intent to injure another’. CR Snyman (Criminal Law 5 ed (2008) at 548), on the other hand, takes the view that a person commits arson if he ‘unlawfully and intentionally sets fire to:

(a) immovable property belonging to another; or

(b) his own immovable insured property, in order to claim the value of the property from the insurer’.

According to Snyman, then, there is a very limited sphere within which setting fire to one’s own immovable property can lead to criminal liability: where, that is, the intention is to defraud an insurer. Milton, however, submits that the better view is that the crime exists to protect, amongst other things, economic interest in property, and that there are no compelling reasons to restrict the ambit of the crime to the burning of habitation. He finds Roman-Dutch authority for the proposition that an owner who burns his own property with intent to injure another commits arson. In R v Mavros 1921 AD 19, the court seemed to accept that arson in modern South African law was equivalent to the ‘brandstichting’ of the Roman-Dutch law. And Innes CJ in that case held (at 22) that ‘the crime of brandstichting is committed by a man who sets fire to his own house wrongfully, maliciously and with intent to injure or defraud another person’.

The court in Dalindyebo considered that the word ‘or’ in this passage was disjunctive, and that Snyman had interpreted the decision in Mavros too narrowly. ‘Simply put’, said the court at [64], ‘the effect of the decision is that one can be guilty of arson when one wrongfully and/or maliciously sets fire to one’s own immovable property either with the intention to injure another person or to defraud another person’. In S v Van Zyl 1987 (1) SA 497 (O) the court accepted that arson could be committed where a person sets fire to his own immovable property with the intention to prejudice the property interests of another person (in that case the person who had inhabited the house owned by the appellant, who was a builder). (See, too, Shannon Hoctor ‘The nature of the crime of arson in South African Law’ (2013) 19 (2) Fundamina 321).
The fact that the act of destroying one’s own property is not, ordinarily, unlawful, is no bar to a conviction since ‘the intention with which the accused acts will serve to convert an ostensibly lawful act into an unlawful one’ (at [66]).

s 242: Criminal defamation: Is it constitutional?
Yes, said the court in S v Motsepe 2015 (2) SACR 125 (GP), a case in which fourteen institutions applied to intervene in the appeal as amici curiae in the interests of the media out of a concern as to the effect of criminal defamation on the freedom of the media. It was argued that the common-law crime of defamation was inconsistent with the Constitution; that it amounted to an unjustifiable limitation on the right of freedom of the media; that the court should develop the common law to limit the crime to the publication of defamatory statements by persons who were not members of the media; and that the civil remedy for defamation sufficed to deter and prevent defamation by the media. These arguments did not persuade the court, and the common-law crime of defamation, defined recently by the Supreme Court of Appeal in S v Hoho 2009 (1) SACR 276 (SCA) at [23] as consisting of ‘the unlawful and intentional publication of matter concerning another which tends to injure his reputation’, survives.

It was argued on behalf of the amici that the decision in Hoho, in which it was held that the crime of defamation was consistent with the Constitution, was incorrect and that it failed properly to apply the provisions of the Constitution. The amici conceded that the court in Motsepe was bound to follow the decision of the Supreme Court of Appeal but contended that Hoho posed no bar to prevent the court from finding the crime to be unconstitutional in so far as it applies to members of the media, since Hoho did not address this specific concern and paid no attention to the differences between defamatory statements by the media and those by members of the public.

The court accepted that the right to freedom of the media was of critical importance and that the media stood ‘in a distinct position relative to the general right to freedom of expression’ given their special position as a key facilitator and guarantor of the right (at [32]). Counsel for the amici pointed to various international instruments and international case law to support their argument. The European Court of Human Rights, in Cumpana and Mazare v Romania [2004] ECHR 692, (2005) EHRR 14, spoke of the ‘chilling effect’ that the fear of criminal sanctions posed to the exercise of journalistic freedom of expression (at [113]–[114]). The civil remedy was, argued the amici, an adequate means to deter and prevent defamation by the media.

The court, however, agreed with the respondent’s counsel that Hoho, which did refer to the media in making ‘a crucial point that criminal defamation is constitutional because the publication of false statements in a national newspaper would be more harmful than publication anywhere else’ (at [35]), had to be interpreted in the light of other findings by our highest courts to the effect that the right to freedom of expression of the press was not unrestrained. It had to yield to the individual’s right not to be unlawfully defamed as well as the right to dignity. The right to freedom of expression did not have a superior status to other rights, and the Constitutional Court in Dikoko v Mokhatla 2006 (6) SA 235 (CC) accepted that it had, sometimes, to ‘take a back seat’ when it intersected with the ‘foundational’ constitutional value of dignity.

Almost all the international instruments and cases referred to by the amici involved extreme situations of governmental abuse of journalists which did ‘not find application in South Africa where journalists and citizens enjoy the benefits of the law and the Constitution’ at [39]). The court accepted that freedom of expression was ‘fundamental to our democratic society’, but stressed that it was ‘not a paramount value’ (at [44]). It had to be construed in the context of other constitutional values, ‘in particular, the values of human dignity, freedom and equality’, and, even though the crime of defamation undoubtedly limited the right to freedom of expression, the court found such limitation to be reasonable and justifiable and consistent with the criteria laid down in s 36 of the Constitution. The court agreed that the criminal sanction was a more drastic remedy than that provided by the civil law, but found this disparity to be ‘counterbalanced by the fact that the requirements for succeeding in a criminal defamation matter are much more onerous than in a civil matter’, and that there was a heavier burden of proof resting on the State in a crime of defamation.

It is interesting to note that the State, on the facts in Motsepe, failed to prove its case. The appellant had relied on the truth of the statement, which was based on information he had received from an attorney and had negligently failed to verify. He had acted recklessly, but recklessness does not equate to intention. His belief that the published words were covered by one of the recognised defences meant that he lacked the necessary intention, so his conviction by the trial court could not stand.

Offences against children: Who is a care-giver for the purposes of s 305(3) of the Children’s Act?
Section 305(3) of the Children’s Act 38 of 2005 provides as follows:

‘A parent, guardian, other person who has parental responsibilities and rights in respect of a child, care-giver or person who has no parental responsibilities and rights in respect of a child but who voluntarily cares for the child either indefinitely, temporarily or partially, is guilty of an offence if that parent or care-giver or other person—

(a) abuses or deliberately neglects the child; or
(b) abandons the child.’ (Emphasis added)

The meaning and ambit of the words emphasised above came to be considered by the court in S v JR & another 2015 (2) SACR 162 (GP). Ranchod J (Mngqibisa-Thusi J concurring) pointed out that the section created three different offences: abusing a child, deliberately neglecting a child and abandoning a child. The prosecution in this case had relied on the second of these in that the appellants had failed to give adequate attention to the child’s injuries after she had been assaulted. Both appellants had failed to seek immediate medical treatment for her after the various assaults and, when they did, they falsely attributed the blue marks on her body and the vaginal bleeding to a blood disorder. The question, in respect of the second appellant, was whether he satisfied the conditions in the phrase emphasised above.

The court found the meaning of s 305(3) to be, in this respect, clear: even a person who voluntarily cares for a child temporarily or partially may be guilty of the offence of deliberately neglecting a child. It was clear from the wording that the legislature sought to spread the net as widely as possible in determining who was deemed to be a caregiver. It would seem, said Ranchod J, that ‘even if a person is a guest at the house of another who has a small child and the guest voluntarily cares for the child for a few minutes while the parent absents him- or herself, that guest falls within the ambit of that section’ (at [37]). Ranchod J did not find this strange given the imperative in s 28(2) of the Constitution that ‘[a] child’s best interests are of paramount importance in every matter concerning the child’.

It was plain that the second appellant satisfied the conditions in s 305(3): he admitted caring for the child on many other occasions when her mother was unavailable; he planned to marry the child’s mother, and regarded the child as his own; on the occasion in question he had bathed the child and noticed that she was bleeding vaginally. On several occasions he could see she was in pain and, not only did he fail to inform her mother but he failed, too, to take her to a doctor for medical attention. His excuse that it was ‘not his duty to do so’ could not avail him.

 (b) Criminal Procedure and Evidence
 (i) Pre-sentence

Prosecutorial misconduct: Interference with defence witness

Alkema J had occasion to censure the conduct of the prosecutor in S v Masoka & another 2015 (2) SACR 268 (ECP). The accused had indicated, in his plea explanation on a charge of robbery, that he had an alibi, and he disclosed the name and address of the witness. His attorney then consulted with the witness on the following day. The prosecutor instructed the investigating officer to obtain a witness statement from the alibi witness. This was done but not mentioned at the trial until the statement was produced after the accused had completed his examination-in-chief and the prosecutor had started to cross-examine him.

The court considered that the prosecutor was guilty of serious misconduct and had to be duly censured. The rule was that no legal representative was allowed to consult or influence or interfere in any way whatever with the opposing parties’ witnesses. It was ‘a rule of ethics applicable to prosecutors and private practitioners alike, and it operate[d] in both criminal and civil matters’. The principles governing the duties of prosecutors in such circumstances were, said the court, not to be found in either the common law or statutes, and there was no provision in the Criminal Procedure Act to deal with them. There were, instead, ‘mostly unwritten rules having their origin in concepts of justice, fairness, morality and equity’, which had evolved all over the world and had been ‘shaped by the legal convictions of the societies in which they were used’. In South Africa, Alkema J added, many of these rules had been formulated in The Code of Conduct for Members of the National Prosecuting Authority, promulgated under s 22(6) of the National Prosecuting Authority Act 32 of 1998 and published in Government Gazette 33907 of 29 December 2010.

It had been said in S v M 2002 (2) SACR 411 (SCA) at [28] that the freedom of witnesses from interference, whatever side they may take, is a ‘keystone in the temple of justice’, without which ‘the structure would disintegrate’ (see, too, R v Manda 1951 (3) SA 158 (A) at 166–7 and S v Hassim & others 1972 (1) SA 200 (N) at 203H). Interference of the kind displayed in Masoka compromised the integrity of the witness. And, because it was a criminal trial, the right of the accused to a fair trial was also compromised. The evidence of the witness in such cases becomes worthless, the true facts cannot then be ascertained, and the entire system of justice is undermined.

s 19: Does a person against whom a warrant to search has been issued have a right to documents or information that led to the warrant being issued?
Yes, said Sutherland J, in HO t/a Betxchange & another v Minister of Police & others 2015 (2) SACR 147 (GJ). In that case a search warrant had been issued in terms of the Counterfeit Goods Act 37 of 1997, and the applicant had applied for copies of the documents or statements that had led to the warrant being issued (‘founding papers’). Sutherland J considered that the very purpose of requiring judicial oversight over the issue of a warrant to enter, search and seize was to protect a (natural or juristic) person’s rights to privacy and to subordinate to judicial scrutiny and oversight a belief by the police, however bona fide, that they really have a need to invade a person’s privacy and that there is a cogent basis for a lawful invasion to be authorised. Not every alleged crime, Sutherland J observed, justified the issue of a search warrant to procure evidence. Such considerations, in his view, pointed to the existence of a ‘right of access to the founding papers in respect of a search warrant, as part and parcel of the broader right of privacy and freedom from arbitrary state action’ (at [29]).

Sutherland J found these to be ‘values which permeate the Constitution’, and he referred to s 14, which concerned the invasion of privacy and the misappropriation of private information, as well as other sections of the Constitution which addressed the absence of arbitrariness (including ss 12, 20, 21, 25, 33 and 35). In his view, it was improper to resist disclosure in any exercise to assert the right to privacy and freedom from arbitrary state power, where a procedure that is orderly and conclusive to expeditious litigation is selected. He found the respondents’ insistence on the use of a procedure under rule 53 of the Uniform Rules to be ‘inappropriate’, and the idea of awaiting a prosecution for further relief to be ‘misdirected’ (at [30]).

ss 20, 205, 225: The status and admissibility of evidence obtained from seized cell phones following the commission of offences allegedly committed by possessors of those phones
This very important question was considered by the Western Cape High Court in two very recent decisions in which the court had to consider issues relating to s 20, s 205, the vexed problem of improperly obtained evidence and the ambit of s 35(5) of the Constitution, as well as the reach and purpose of two important pieces of legislation: the Electronic Communications and Transactions Act 25 of 2002 (ECTA) and the Regulation of Interception of Communications and Provision of Communication-related Information Act 70 of 2002 (RICA).

These cases are S v Brown (unreported, WCC case no CC 54/2014, 17 August 2015) and S v Miller & others (unreported, WCC case no SS13/2012, 2 September 2015). Brown was a murder trial in which an innocent bystander had been killed in a gang shooting, in which the State sought to adduce evidence of three photographs downloaded from a cell phone dropped at the scene of the shooting by the assailant, tendered in order to show the identity of the accused as the person who did the shooting. The defence unsuccessfully raised a number of objections to the admissibility of this evidence: that the integrity of the chain of evidence as far as the safekeeping of the phone had not been proved, and that the authenticity of the images had not been established; that the evidence had not been covered by a subpoena issued by a magistrate in terms of s 205; that the evidence was hearsay; and that the downloading of the images without authorisation by a judicial officer was an unlawful invasion of the right to privacy.

Bozalek J summarily dismissed the hearsay objection. The evidence was not hearsay as it did not depend for its probative value on the credibility of another person. It was akin to real evidence used to show nothing more than the identity of the possessor of the cell phone.

The other objections, although they warranted more attention, also failed. Bozalek J wrestled with the question whether the images fell to be treated as real evidence or as documentary evidence. He approved the views of Prof J Hofman ((2006) SACJ 257 at 268) that electronic evidence such as graphics, audio and video material ‘now resemble documents more than the knife and bullet that are the traditional examples of real evidence’. They were, in data form, ‘susceptible to error and falsification’ and their evidential value depended ‘on witnesses who can both interpret them and establish their relevance’. He concluded, then, that given the ‘potential mutability and transient nature of images ... generated, stored and transmitted by an electronic device’, it was more appropriate to deal with them as documentary rather than real evidence (at [20]). It was, accordingly, necessary for the State to show both the originality and the authenticity of the images. Authenticity had not been contested, but it was necessary to consider the effect of s 14 of ECTA, which provides that a data message satisfies the requirements of ‘original form’ if it meets the conditions set out in that section. These are that (a) the integrity of the information, from the time it was first generated in its final form as a data message, has passed assessment in terms of s 14(2); and (b) the information is capable of being displayed or produced to the person to whom it is to be presented. As to (a), s 14(2) gives pointers as to how ‘integrity’ must be assessed. All relevant circumstances must be considered, but mention is made of ‘whether the information has remained complete and unaltered’ and ‘the purpose for which the information was generated’.

In concluding that the requirement of ‘original form’ had been satisfied, Bozalek J took into account these considerations: there had been no suggestion of tampering on the part of the police officer who had downloaded the images; the software programme he used excluded any possibility of tampering; the images had been transferred from phone to court exhibit in no more than just a minute or two; the phone had been in lay hands, after being found on the crime scene, for no more than four hours, making it improbable that any tampering had occurred then; and, significantly, the data had been transmitted to the phone two days before the shooting, when the phone was in the possession of its original possessor. In any event, said the learned judge, s 15(1)(b) of ECTA provided exemption from the requirements of ‘original form’ if a data message was ‘the best evidence that the person adducing it could reasonably be expected to obtain’, a requirement satisfied on the facts in Brown’s case.

Bozalek J turned, next, to the claim that the evidence should be excluded because it had been illegally obtained in violation of privacy rights. He dismissed this on the grounds that: the police were entitled to seize the phone in terms of s 20 of the Criminal Procedure Act when it has been presented to them by members of a neighbourhood watch with the explanation that it had been found on the crime scene; no authorisation from a judicial officer was necessary to download the images, and there was no statutory provision to prohibit such action; and the right to privacy was ‘misconceived’ (at [28]), since the accused denied that the phone was his and so could not, at the same time, claim privacy in respect of the images on it. It was, moreover, neither unfair nor detrimental to the administration of justice (as set out in s 35(5) of the Constitution) to receive the evidence.

The final argument was that the downloading of the images fell outside the scope of a subpoena issued by a magistrate in terms of s 205 of the Act in respect of the phone’s details and call records. This argument, it was held, misconstrued the purpose of s 205 and the role it played in this case. Section 205 was not in point here as the phone was already in the lawful possession of the police. It had the capacity to download the data using its own software programme, and the State was under no obligation to seek further information or data from the service provider using s 205 or any other procedure. Although statutory intervention might be necessary in future to ‘hold the balance between the privacy of private electronic communications or data and the interests of justice’ by, for instance, requiring the prior authorisation of a judicial officer before the analysis of the contents of an electronic device, Bozalek J could not find ‘any provisions in our law which would preclude the SAPS from subjecting the phone to analysis and downloading information where that was objectively necessary for the purposes of a criminal investigation’ (at [31]).

The second of the two cases decided by the court was S v Miller & others. The accused had been charged under the Prevention of Organised Crime Act (POCA) for activities related to the alleged unlawful stripping of abalone off the coast. Subpoenas had been issued in terms of s 205 of the Criminal Procedure Act directing the cellular network service provider to hand over to a police officer ‘itemised billing’ documents for the cell phones of the accused. These were fed into a computer equipped with a software programme which would furnish details as to who had phoned whom, when, and how often. The evidence of this police officer, who presented the output of this exercise, was challenged on two grounds: first, that the cell phone records procured by the State under s 205 had been unlawfully obtained since the subpoenas presented to the issuing magistrate in terms of that section were fatally defective; second, that the acquisition of the evidence violated certain statutory provisions and, moreover, infringed the accused’s right to privacy under s 14 of the Constitution.

The argument on s 205 was based on the claim that the magistrate had not properly applied his mind to the matter, as shown by the fact that the date for appearing to resist or attack the subpoena was listed as ‘31 November 2006’, a non-existent date on the calendar. Gamble J agreed with the cases that stressed that the function of the issuing magistrate was not merely that of ‘a rubberstamp’ (at [23]), and that the subpoena had to be drawn up in such a fashion that it was ‘as narrowly tailored as possible to meet the legitimate State interest of investigating and prosecuting crime’. He found, nevertheless, that the magistrate’s error in this case was an understandable one. An incorrect date, moreover, would not have allowed the person subpoenaed simply to ignore the direction.

The learned judge turned next to the argument that the evidence had been improperly obtained. The defence referred to both ECTA and RICA to support its claim, but Gamble J found neither sufficient to exclude the evidence. Since s 205(1) of the Criminal Procedure Act contains a cross-reference to s 15 of RICA, the prohibition in s 12 of RICA (which prohibits a service provider from intentionally providing any real-time or archived communication-related information to any person other than the customer) is, said Gamble J, ameliorated by s 15, which allows the obtaining of such information ‘in respect of any person in accordance with a procedure prescribed in any other Act’. The use of s 205 for this purpose was, then, unobjectionable.

Was the seizure of the actual cell phones and SIM cards (contained in so-called ‘starter packs’) from persons on the scenes in question lawful? Yes, said Gamble J. The police acted in terms of s 20 of the Criminal Procedure Act, which gave the police the general power to seize ‘anything’ for the purpose set out in that section. This allowed them to have access, too, to the contents of the articles seized in the same way that the seizure of, say, a diary or photograph album or even a locked safe would. The defence argued that specific provisions of ECTA required additional steps since modern cell phones are akin to mini computers with capacities relating to the sending of email, word processing, search programmes as well as storing of information, music and photographs. This made it necessary, it was argued, to consider s 86 of ECTA which, subject to the provisions of Act 127 of 1992, made it an offence for anyone intentionally to access or interpret any data ‘without authority or permission to do so’. The ‘authority’, it was argued, meant the ‘cyber inspector’ described in s 80, which, in the 13 years since ECTA came into effect, had yet to be appointed.

This argument did not convince Gamble J, who took a purposive approach to interpreting ECTA. He pointed out that s 81 provides that the SAPS ‘may’, and not ‘must’, apply for assistance from a cyber inspector to assist it in an investigation. He considered it, in any event, given the importance of speedy investigation in the fight against crime, counter-productive to require the police to follow a bureaucratic procedure to access digital information as opposed to non-digital evidentiary material. He was, in short, ‘not persuaded that it was the intention of the Legislature when passing [ECTA] to criminalise or prescribe the accessing of cell phone data by the police in circumstances where the instrument had been lawfully taken into possession during the course of a criminal investigation’ (at [56]).

Gamble J considered, next, the argument of the defence that the evidence should be excluded as it was the ‘fruits of the poisoned tree’, which was a doctrine invoked in the United States to the effect that ‘once the procurement of evidence is in any way tainted by illegality, such evidence must be excluded without more’ (at [59]). The defence drew attention, too, to the recent decision of the United States Supreme Court in Riley v California 573 US (2014) for its argument that the evidence was inadmissible. But Gamble J (at [60]) expressly declined to follow the majority approach in Riley, saying that our courts had not endorsed the doctrine of the fruits of the poisoned tree (see, for instance, S v De Vries & others 2009 (1) SACR 613 (C) at [28], S v Melani & others 1996 (1) SACR 335 (E), S v Marx & another 1996 (2) SACR 140 (W) and S v Malefo en andere 1998 (1) SACR 127 (W)). In Malefo it was said that the doctrine was ‘a rigid exclusionary rule that was at odds with our law (and in particular the provisions of section 35(5) of the Constitution)’, and that it ‘had been roundly criticised in America whence it came’ (at [60]). Gamble J observed that Alito J in the minority in Riley bemoaned the natural consequences of the majority opinion and pointed to anomalies that it might produce. It would, for instance, favour digital over non-digital information, in the sense that the police would be able to seize a phone bill or photograph in the pocket of a suspect but not one in digital form on a cell phone belonging to and found on that suspect. Alito J thought it necessary to find ‘a new balance of law enforcement and privacy interests’.

Gamble J maintained that, because of s 35(5), our law is not hit by these anomalies. Section 35(5) allows our law to achieve the necessary balance. If, then, the accessing of stored data on a cell phone was, prima facie, an invasion of the right to privacy, the question that would arise under s 35(5) was whether the admission of that evidence would either render the trial unfair or be otherwise detrimental to the administration of justice. He found, on the facts, that neither was the case.

As to the fairness of the trial, arguments relating to the injustice caused by the late emergence of the evidence or the prolonging of the trial were rejected on the facts. So, too, the argument relating to the effect of admitting the evidence on the administration of justice. If wrongdoers made extensive use of cell phones—instruments ‘ideally suited to the commission of crime’—the interests of justice, said Gamble J, demand that the State should be afforded the reasonable opportunity to present that evidence. No one, he said, obliged the accused to make use of cellular communication, and those who did ‘willingly ran the risk that those communications [might] later be detected by the authorities’. The conduct of the police in regard to the acquisition of the evidence was, moreover, bona fide throughout the investigation. Warrants were applied for and obtained prior to the search and seizure, and the officer who analysed the contents of the records genuinely believed he was entitled to act as he did. He took steps to ensure that the evidence so obtained was used restrictively and not made available to all and sundry.

s 73(1): Representation of co-accused by someone without right of appearance: Impact on trial proceedings
S v Swapi & others (unreported, ECB case no 14/14, 1 September 2015)
Our courts have on several occasions held that proceedings are irregular and the trial a nullity where the accused was represented by someone who had no right of appearance and was therefore legally not permitted to represent an accused as envisaged in s 73 of the Criminal Procedure Act. See S v Khan 1993 (2) SACR 118 (N); S v Stevens en ’n ander 2003 (2) SACR 95 (T); S v Nghondzweni 2013 (1) SACR 272 (FB). See also S v Dlamini en ’n ander 2008 (2) SACR 202 (T) where it was held that lack of the right to appear during a portion of the trial (the attorney was suspended during the course of the trial but continued representing the accused) tainted the trial to the extent that it had to be set aside in its entirety and not just from the point at which the attorney was suspended.

However, the irregularity that occurs where an accused was represented by someone who had no right of appearance is not necessarily in each and every instance a fatal irregularity nullifying the trial. See, for example, S v Chukwu & another 2010 (2) SACR 29 (GNP) which is discussed—and also compared with Nghondzweni (supra)—in the notes on s 73 in Commentary, sv Representation by candidate attorney (or former candidate attorney) with no right of appearance.

The cases referred to above all dealt with the situation where there was only one accused, or the situation where there were co-accused who were all represented by one ‘lawyer’ who had no right of appearance.

In S v Swapi & others (unreported, ECB case no 14/14, 1 September 2015) the High Court was, on review, required to address a somewhat different question: must the proceedings in a part-heard criminal trial be set aside in their entirety if accused number one and two had a qualified lawyer but accused number three was represented by someone who, in the words of Streicher J at [10], ‘was not entitled to appear as an attorney at this trial’?

At [13] Stretch J (Cossie AJ concurring) pointed out, most pertinently, that accused number three was ‘not the only person on trial’ and that the two co-accused, represented by a qualified lawyer, had—like the prosecution—‘a direct and substantial interest in the future conduct of the ... proceedings’. On review it was pointed out at [14] that the presiding magistrate, accused number one and two, the prosecutor and the new attorneys of accused number three, acting on his instructions, all favoured a separation of trials as provided for in s 157(2) of the Criminal Procedure Act: the argument was that the review court should order a separation of trials, permitting the proceedings against the first two accused to carry on where they left off when the matter was sent on review and for the prosecution of accused number three to start de novo in a separate trial (if, indeed, the prosecution should wish to institute such a prosecution again). In fact, in a statement submitted to the review court the senior prosecutor stated that after having perused the trial record and the evidence which the prosecution still wished to present, he was of the view that the proposed separation would be in the interests of justice and that upon completion of the trial against the first two accused ‘the prosecuting authority would be better equipped to consider whether to prosecute accused no 3 afresh’ (at [15]).

It seems clear that the route proposed by all would forthwith have been made an order of court but for the fact that an earlier decision, S v Gwantshu & another 1995 (2) SACR 384 (E), stood in the way. The decision in Gwantshu was by a review court (two judges) of the same division as the review court (also two judges) in Swapi; and the court in Swapi, said Stretch J at [20], was therefore bound to follow Gwantshu unless it could be found that Gwantshu was ‘clearly wrong, and/or ... distinguishable on the facts’.

In Gwantshu two accused stood trial in the regional court. Both were represented. But during the trial it came to the attention of the presiding magistrate that the representative of accused number two was a candidate attorney who had no right of appearance in either the district or regional court, such right having expired some time prior to the trial. The magistrate submitted the matter for review, with the request that the proceedings be set aside to enable accused number two ‘to appoint a qualified legal representative in a hearing de novo’ (Gwantshu at 384i). Relying on cases such as S v Khan (supra) where the entire proceedings were set aside in the trial of a single accused represented by a candidate who had no right of appearance, Mullins J (Lang AJ concurring) decided in Gwantshu to set aside the proceedings against both accused irrespective of the wishes of the affected parties, without perusing the record of the part-heard matter and regardless of the fact that a co-accused was represented by a qualified lawyer (at 386ad).

In Swapi at [22] Stretch J took the view that an approach more nuanced than the one in Gwantshu was required: in trials of co-accused the proceedings in respect of an accused who was represented by a qualified attorney need not necessarily be set aside, particularly not (a) where the trial record ‘does not call for such a course of conduct’; (b) where all the accused, the prosecutor and the presiding judicial officer do not consider ‘such an approach necessary, convenient or in the interests of justice’; (c) where it appears to be in the interests of justice to commence afresh against the affected accused only; (d) where ‘a separation of trials with appropriate measures is unlikely to prejudice the accused or the administration of justice’.

At [23] the review court in Swapi made the following observations (emphasis added):

‘At the end of the day the main test in deciding whether the entire trial should start afresh (in other words without separating the affected accused from the others) is whether any of the accused will suffer prejudice, or are likely to suffer prejudice if this course of conduct is to be preferred. The views of the prosecution should also be thrown into the balance. Ideally, matters such as these should be dealt with on a case by case basis, and each matter should be considered on its own merits. To my mind, the court in Gwantshu applied a procedure (which had been followed in very different circumstances and which was the only option in those circumstances), to the circumstances of the matter which it was seized with, without giving any consideration to relevant factors such as the views of the parties and the nature and extent of the evidence already led.’

The review court in Swapi accordingly concluded that the decision in Gwantshu ‘was clearly wrong’: the rule requiring the setting aside of the entire proceedings where a single accused was on trial and represented by someone without the right of appearance does not mean that trial proceedings in respect of co-accused, who had properly qualified lawyers, must necessarily also be set aside. The issue is a fact-sensitive one and must be decided with reference to the presence or likelihood of prejudice. The absence of prejudice would require that the trial against the unaffected co-accused should proceed without the affected accused who will have to be charged de novo on account of the fact that the criminal trial becomes a nullity in respect of the affected accused only.

After having examined the transcript of the proceedings of the court a quo, Stretch J agreed with the senior prosecutor’s view that it would be in the interests of justice to separate the trial of accused number one and two from that of accused number three. Given this finding and also having concluded that Gwantshu was wrongly decided and therefore not binding in terms of the doctrine of precedent, Stretch J (Cossie AJ concurring) made the following order (at [27]): (a) the proceedings against accused number three ‘must be stopped, and any proposed hearing with him as an accused shall commence de novo and be held separately from the present proceedings’; (b) the proceedings against accused number one and two ‘shall continue and be finalised before the regional magistrate seized with this trial’, provided that the questions put by the representative of accused number three—and all answers given in response to these questions—‘shall be ignored by the presiding officer and shall be expunged from the trial record before the proceedings continue’.

It is submitted that preference should be given to the decision in Swapi as opposed to the one in Gwantshu. In Swapi the emphasis was put on what was feasible in all the circumstances in order to promote the interests of justice. But in Gwantshu, it would appear, there was a rather mechanical application of earlier precedents in which the trial proceedings were set aside in their entirety because a tainted representative appeared for the single accused on trial. A new dimension to the problem is added where there are co-accused who had qualified lawyers. It is this dimension that was appreciated in Swapi but ignored in Gwantshu. Another factor in favour of Swapi is the constitutional right of accused persons to have their trial begin and conclude without unreasonable delay. See s 35(3)(d) of the Constitution. The Constitutional Court has said that although this right on its own is not determinative of the interests of justice, it may be a factor relevant to the determination of the interests of justice. See S v Basson 2004 (1) SACR 285 (CC) at [71]. By permitting the trial of the co-accused to proceed separately if circumstances so permit, ‘unreasonable delay’ in the conclusion of the trial of the co-accused is avoided—and such avoidance is, indisputably, in the interests of justice. Indeed, a court should also take great care to ensure that the broad general right to finality of the co-accused who did have qualified lawyers is not eroded. See ss 106(4) and 108 of the Criminal Procedure Act 51 of 1977. Section 106(4) contains the general principle that an accused who has pleaded to a charge ‘shall ... be entitled to demand that he be acquitted or convicted’. See the discussion of s 106 in Commentary, sv Section 106(4). Section 108, in turn, provides that where an accused ‘pleads a plea other than a plea of guilty’, he shall—subject to certain statutory provisions—‘by such plea be deemed to demand that the issues raised by the plea be tried’.

s 105: Accused to plead in response to charge put by prosecutor
S v ZW 2015 (2) SACR 483 (ECG); S v Motlhaping (unreported, NWM case no CAF 17/15, 17 September 2015); S v Negondeni [2015] ZASCA 132 (unreported, SCA case no 00093/2015, 29 September 2015)
Section 105 of the Criminal Procedure Act 51 of 1977 is headed ‘Accused to plead to charge’. It states:

‘The charge shall be put to the accused by the prosecutor before the trial of the accused is commenced, and the accused shall, subject to the provisions of sections 77, 85 and 105A, be required by the court forthwith to plead thereto in accordance with section 106.’

The reference in s 105 to ss 77, 85 and 105A is not material for purposes of the discussion of the issues that surfaced and observations that were made in the three cases referred to above and discussed below.

In S v ZW 2015 (2) SACR 483 (ECG) it became clear on appeal that at the trial the prosecutor had not put the two charges of rape to the appellant. The magistrate also did not require the prosecutor to do so. It was left to the legal representative of the appellant to confirm in open court at the commencement of the trial that he and his client ‘had perused the charge-sheet’ and, furthermore, that the client understood the charges as explained to him by the representative (at [41(a)]). Stretch J (Nepgen J concurring) found this unacceptable (at 41(c)): ‘The provisions of s 105 are peremptory, not only with respect to the stating of the charges in open court, but also particularly with respect to the party seized with the duty to do so, being the prosecutor who after all is the official representative of the state, being the accused’s accuser.’

During the course of the appeal, all counsel concerned informed the court that in the lower courts this irregular circumvention of s 105 has become standard practice. ‘If this is indeed so,’ said Stretch J at [38], ‘presiding officers are invited not only to be vigilant in discouraging and reprimanding such sloppy prosecution, but also to resist becoming a part of what can only be described as a series of unfortunate irregularities.’

In S v ZW (supra) it was concluded that the prosecutor’s failure to comply with the mandatory provisions of s 105 constituted an irregularity which could have been cured by the magistrate’s intervention; and the absence of such intervention amounted to a misdirection (at [40(e)]).

The full bench decision in S v Motlhaping (unreported, NWM case no CAF 17/15, 17 September 2015) dealt with a slightly different factual situation but also involved the provisions of s 105. In this case the trial judge—when the prosecutor was about to read the charges to the accused—addressed counsel for the defence directly by asking her whether her client ‘was familiar with the charges against him’ (at [5]). Upon receiving an affirmative answer, the trial judge asked defence counsel whether her client had instructed her to plead. She replied: ‘Not guilty’. The trial judge then enquired whether the accused himself could confirm that he was pleading not guilty to all the charges. The accused replied: ‘Not guilty’.

On appeal Landman J (Gura and Chwaro JJ concurring) said that the trial judge’s approach ‘in requiring or permitting counsel to plead on behalf of the appellant is not in accordance with the provisions of the [Criminal Procedure Act] ... In my view it is best to follow the letter of the law and have the indictment read to an accused and have the accused plead to it’ (at [7]). Reference was made to s 105 (at [4]). However, it was held that the fairness of the trial was not affected by the unacceptable procedure adopted by the trial judge (at [8]).

S v Negondeni [2015] ZASCA 132 (unreported, SCA case no 00093/2015, 29 September 2015) indicates why the rule that an accused should plead personally is sensible. In this case the accused, shortly before being called upon to plead, stated that he and his defence lawyer had ‘not yet finished a consultation’. In response to a question by the trial judge, he again said: ‘We have not consulted sufficiently’. The trial judge then, most surprisingly, ruled that the trial should commence and that the appellant could consult with his defence counsel during adjournments of the court (at [4]).

When the first charge was put to the accused, he pleaded guilty but immediately thereafter, when asked by the trial judge to confirm his plea, said: ‘Maybe I did not understand well’. In response to a further question from the bench, the accused said: ‘I do understand but when I am asked to plead on the charge of murder I am not so sure as to whether I should plead not guilty or ... explain the circumstances’. It was only at this stage that the trial judge remanded the case to the next day so that the accused could consult more fully with his legal representative (who, as it turned out, had to be replaced by another defence counsel because of non-availability on the next day).

On appeal Willis JA (Leach and Mathopo JJA concurring) stated that in view of the accused’s ‘patently concerned and hesitant stance at the commencement of the trial, the court a quo was at the outset of the proceedings wrong in insisting that the trial proceed as it did’ (at [16]).

The above incident in Negondeni illustrates that there is merit in requiring an accused to plead personally. It is distinctly possible that if the accused’s counsel had been asked to plead on behalf of the accused, the plea tendered by him (he was, after all, satisfied that they had had a proper consultation) would probably simply have been confirmed by the accused if the latter were required to confirm or contradict the plea tendered by counsel on his behalf. However, it will be shown below that there are situations where an accused should not be required to plead personally.

It is respectfully submitted that ZW (supra) and Motlhaping (supra) should not be interpreted and applied too strictly. It is submitted that the prosecutor’s duty to put the charge is based on the accused’s right to know what the charge is. Is it then, strictly speaking, absolutely necessary for the charge(s) to be read out in circumstances where defence counsel, after proper consultation with the accused, takes the initiative by informing the court and the prosecutor that his client is aware of all the charges against him and in a position to plead? In so doing there is a proper waiver by the accused of the right to have the charges put to him.

Consider the case where the accused, an accountant, is charged with four dozen charges of fraud and is defended by senior counsel who is assisted by two fairly experienced juniors. No court in South Africa would insist, or should require, that in this type of case ‘the letter of the law’ should be followed by requiring that the charges with all their detail should nevertheless be put to the accused by the prosecutor. The procedural objective should be to allow a situation where an informed plea in respect of each count can be received. The plea determines the ambit of the lis between the defence and prosecution; and this, it is submitted, can also be achieved in cases where defence counsel has indicated that his client is ‘familiar’ with the charges and ready to plead to each numbered count in the charge-sheet or indictment. Accuracy remains important, and that is why it is standard practice for defence counsel to indicate clearly to the court, and before any evidence is led, whether the pleas tendered by his client are in accordance with instructions to counsel.

It can be argued that, in the above circumstances, putting the bare essence of each count would meet the ‘open court’ requirement stated in ZW (supra) at 41(c). In any event, the charge-sheet or indictment, as the case may be, is a public document.

The decision in Motlhaping (supra) also raises the following question: is an accused in all circumstances required to plead personally or may his counsel do so on his behalf? Here, too, the ‘letter of the law’ (as contained in s 105) might seem to require that an accused should always plead personally. It would appear that Motlhaping favoured such an approach. However, s 105 should also be read with s 73(2) which states: ‘An accused shall be entitled to be represented by his legal adviser at criminal proceedings ...’. In 1980 the (then) Appellate Division in S v Mpongoshe & another 1980 (4) SA 593 (A) relied on s 73(2) for a finding that juvenile accused could in certain circumstances plead vicariously through their legal representatives. Kotzé JA held as follows (at 603B–C, emphasis added):

‘Sub-section (2) of s 73 of the Act, unlike its predecessor (s 158 of Act 56 of 1955), which limited the expressed function of a legal adviser to the examination and cross-examination of witnesses, confers upon an accused the wider right to have a legal adviser to represent him (ie to stand in his place—see Shorter Oxford English Dictionary sv ‘‘represent’’) at his trial ... In the circumstances of the present case where the accused are children who have barely progressed beyond the stage where the criminal law presumes them to be doli incapaces and who face serious and complicated statutory charges, I have no hesitation in holding that the right to be represented by a legal adviser embraces the right of tendering the plea vicariously through such legal adviser provided that he is duly instructed and not prohibited by law from appearing. To hold differently would render the right ineffectual.’

It follows that a presiding judicial officer should not insist that a juvenile accused should plead personally to a complicated and serious offence if defence counsel concerned has taken the initiative to plead on behalf of the accused. See also Van der Merwe, Barton & Kemp Plea Procedures in Summary Criminal Trials (1983) at 147. See further the discussion of s 73 in Commentary, sv Section 73(2): Meaning of ‘to be represented’.

In Mpongoshe (supra) at 603F Kotzé JA expressly refrained from deciding whether the right to tender a plea vicariously is an unlimited one. However, on the basis of the wide and correct interpretation that Mpongoshe gave to the words ‘to be represented’ in s 73(2), it would be entirely acceptable to permit an accused who is an unsophisticated adult of limited mental ability to plead vicariously through his legal adviser where the charges happen to be serious and complicated. See also the discussion of s 105 in Commentary, sv Sections 105 and 73(2): Accused need not plead personally. After all, the procedural objective should be to ensure that an informed and accurate plea is tendered. This is precisely also why there is—as was noted earlier—a rule of practice that in the ordinary course of events when the accused is required to plead personally, the legal representative concerned should indicate to the court whether the plea as tendered by his client is consistent with the instructions received.

s 112(1)(b): The ambit of judicial questioning to test a plea of guilty to drunken driving
S v Funani (unreported, ECB case no 4/2015, 17 April 2015)
An accused who has pleaded guilty to the offence of drunken driving in contravention of s 65(1)(a) of the National Road Traffic Act 93 of 1996 cannot be convicted of this offence unless his admissions in response to judicial questioning in terms of s 112(1)(b) of the Criminal Procedure Act also include an admission that at the relevant time his driving was impaired as a result of alcohol consumed by him. See the discussion of s 112 in Commentary, sv Purpose of questioning. See also S v Mzimba 2012 (2) SACR 233 (KZP) at [6] where it was explained that this must be so because the relevant substantive criminal law element is that the skill and judgment required for driving a motor vehicle was impaired or diminished as a result of alcohol consumption.

The position as set out above was confirmed in S v Funani (unreported, ECB case no 4/2015, 17 April 2015) at [7]–[9]. However, in Funani Hartle J also dealt with two further important matters concerning s 112(1)(b):

(i) If the accused should, for sentencing purposes, ‘co-incidentally’ disclose information that could possibly constitute the admission that should in the first place have been present for purposes of the conviction after section-112(1)(b) questioning, ‘the situation is not alleviated at all’ because the magistrate ‘ought in the first place to have elicited [the relevant admission] during his questioning of the accused in terms of s 112(1)(b) ...’ (at [12]). This approach is supported: the incorrect conviction because of inadequate judicial questioning cannot ex post facto be justified on the basis of factual matters that emerged for purposes of sentencing. To put the matter differently: an accused’s conviction can only be based on factual admissions made prior to conviction and in response to judicial questioning in terms of s 112(1)(b).

(ii) If an accused—like the one in Funani—were involved in a collision ‘the mere fact of the collision cannot in itself be regarded as proof that the accused was under the influence of liquor’ (at [14]). This approach is also supported. The fact of the matter is that the element of impairment of driving skills must be covered by judicial questioning and admitted by the accused—it cannot be inferred. See further the discussion of s 112 in Commentary, sv Factual information to be elicited.

s 165: Administration of the oath by interpreter
In S v Maloma (unreported, GP review case no 209/15, 27 May 2015) the court considered a question it considered to be ‘of cardinal importance to all criminal courts’ (at [2]): can an interpreter administer the oath to a witness? The answer to this question seems to be plainly provided by s 165, which states clearly enough that ‘[w]here the person concerned is to give his evidence through an interpreter or an intermediary ... the oath, affirmation or admonition ... shall be administered by the presiding judge or judicial officer or the registrar of the court, as the case may be, through the interpreter or intermediary or [emphasis added] by the interpreter or intermediary in the presence or under the eyes of the presiding judge or judicial officer, as the case may be’.

The oath in this case was administered by the interpreter in the presence of the judicial officer, so there should have been no difficulty. The presiding regional court magistrate, however, in view of the decision in S v Pilane (unreported, NW case no CA 10/2014, 5 March 2015), which was the only High Court decision in point that could be located, had requested the full bench to rule upon the question. The court in Pilane found that if the oath was not administered by the presiding judicial officer in compliance with s 162, the evidence of the witness in question was inadmissible and that there would be an irregularity that would vitiate the entire proceedings.

Bam J (with whom Mlambo JP and Potterill J agreed) found that the court in Pilane had failed, simply, to take account of s 165. The fact that the Supreme Court of Appeal, in S v Machaba & another [2015] ZASCA 60, 2 All SA 552 (SCA) referred to Pilane in confirming that it was peremptory, in terms of s 162, for the judge or registrar to administer the oath, did not signify since the question whether an interpreter was empowered to do so was not addressed, and the court was clearly not called upon to consider s 165 at all.

Pilane’s case, the court concluded, could thus not be supported.

s 170A: Does the failure to swear in an intermediary vitiate the proceedings?
The court in S v Mahlangu (unreported, GP case no A382/2014, 17 July 2015) rejected an argument that it did. Although it was ‘clearly preferable’ that the intermediary be sworn in as a ‘precaution to alert a mediator to the grave repercussions of misinterpreting or misrepresenting questions posed’ (at [4]), it was not a necessity, since he or she does not furnish any evidence. The court disagreed with what was said to the contrary in S v Booi & another 2005 (1) SACR 599 (B), where the intermediary was ‘placed on an even keel with a district surgeon, a pathologist or a police officer’.

Jansen J pointed out that although the court in S v Motaung 2007 (1) SACR 476 (SE) had treated the failure to swear in an intermediary as an irregularity, it was held that this did not per se render the witness’s evidence inadmissible, because it did not necessarily mean that the proceedings were not in accordance with justice. In S v QN 2012 (1) SACR 380 (KZP), moreover, it was held not to be an irregularity where the intermediary was not sworn in. The court there said that it was not apposite to liken an intermediary to an interpreter and that, although the best course was to swear in an intermediary, the failure to do so did not constitute an irregularity (see Commentary on s 170A sv The intermediary and the oath or affirmation).

In explaining why the approach in QN was to be preferred to that in Booi or Motaung, Jansen J said, simply (at [7]): ‘A mediator does exactly what her epithet depicts—she mediates the questions put, not the answers’. See, further, the views of Banoobhai and Whitear-Nel (2013) Obiter 359 and 365 where the authors endorse the view of Gorven J in QN that it is a salutary practice to require an intermediary to discharge his or her duties under oath. They exhort the legislature to make provision for the swearing in of intermediaries subject to the proviso set out in Motaung that the mere failure to swear in, or to swear in properly, an intermediary should not, in itself, render the witness’s evidence inadmissible.

s 195(1): Compellability of estranged wife
Can the fact that a marriage relationship has been severely damaged negate, in some way, the immunity set out in s 195(1) which renders one spouse a competent but non-compellable witness against another in criminal proceedings? No, said the court in S v Mgcwabe 2015 (2) SACR 517 (ECG). One of the witnesses against the appellant in that case was his estranged wife, who was not living with him at the time. She had not initially been informed of her right to refuse to testify, and it was clear that she would not have testified had she been so informed. However, as she was already in court, she stated that she wanted to proceed with her testimony. This, said Nepgen J (with whom Alkema J agreed), was a misdirection. Although the rationale for the privilege is the preservation of the sanctity of marriage, or ‘the consideration that the marital relationship should be protected’ (at [12]), there was no justification for calling the appellant’s wife without informing her of her right not to give evidence.

Section 195(1) gives the spouse ‘an absolute right to make an election not to testify’; it does not provide that a spouse shall not be compellable to give evidence ‘only if this is necessary to preserve the marriage relationship’. It did not help that she had elected to continue with her evidence after she was made aware of the provisions of s 195(1) since the decision to do so was taken because, as she put it, she was already in court. It was not a decision taken after proper consideration before she had been called as a witness.

If a witness is competent but not compellable she should, said Nepgen J, be informed by the prosecutor of her rights prior to being called as a witness, and this fact should be placed on the record at the outset of the proceedings. The judicial officer should ascertain, in any event, if the witness was aware of the provisions of s 195(1) in cases such as this. If it is brought to the attention of the witness for the first time when she is called to testify, she should be afforded an opportunity to come to a decision after proper consideration.

In this case, then, the magistrate should, at least, have adjourned the matter to give the appellant’s wife an opportunity to make a considered decision. The magistrate had committed a serious misdirection in failing to investigate the circumstances under which she had come to testify and by permitting her to continue giving evidence after she had indicated that she would not have testified at all had she been aware of her rights.

The court pointed out, too, that the privilege in s 195(1) was that of the spouse, not that of the accused. The accused, then, has no role to play in the making of the decision by the spouse. But what had to be considered was whether, in circumstances such as these, the appellant could be said to have been prejudiced by the misdirection. The court accepted that there was such prejudice. It arose by there being evidence before the court, upon which the prosecution relied, which would not have been before the court had the witness been told at the appropriate time of the privilege which she had.

s 252A: Entrapment and the constitutional validity of s 252A
In Myoli & another v Director of Public Prosecutions, Eastern Cape & others (unreported, ECB case no 593/2014, 22 September 2015) counsel for the applicants asked the court to declare s 252A unconstitutional on the grounds that the section per se renders the criminal trial unfair by its very terms. Alkema J, however, declined to do so. The learned judge found that the exclusionary nature of the terms and of the proper operation of the section could not be said to be unfair or to infringe any constitutional right. There was always tension between the use of traps and the right to a fair trial, but it had to be understood that the use of a trap is sometimes the only method to combat crime, and corrupt police officials could, very often, be successfully prosecuted only if a trap was used.

If the use of a trap was the only practical method of successfully combating crime and the evidence was obtained in a fair manner, the public interest demanded that the evidence of a trap be admitted. The potential of such evidence to render a trial unfair had, however, to be considered. In this regard, the section itself specifically excluded any evidence which would render the trial unfair or be detrimental to the interests of justice. And, if the conduct of the police did not go beyond merely providing an opportunity to commit an offence, ‘the very mischief which arises from the use of a trap or undercover agent is prevented and the trial cannot be said to be unfair’ (at [24]).

Section 252A, said Alkema J, ‘retains the tension between the public interest to use a trap in certain circumstances on the one hand, and the right to a fair trial on the other hand’ (at [25]). Even if there was an infringement of the right to a fair trial, Alkema J maintained that, having regard to the public interest, such infringement was justified as a reasonable limitation of that right under s 36 of the Constitution.

It was argued, in the alternative, that s 252A lacked clarity, was, in certain respects, incomprehensible, and had attracted academic and judicial criticism (see Commentary sv Purpose and scope of s 252A; S v Odugo 2001 (1) SACR 560 (W); S v Kotzè 2010 (1) SACR 100 (SCA) at [20] and [26]). The criticism in these authorities, said Alkema J, included the view that the section may, in certain respects, be unconstitutional. The case of the applicants was, however, that the entire s 252A was, per se, unconstitutional. It did not follow from the criticism expressed of certain aspects of s 252A that the entire section was unconstitutional. In any event, the section had to be interpreted subject to the Constitution, as Wallis AJA insisted in Kotzè, and such an interpretation was, in the court’s view, able to render the section ‘constitutional in its entirety’ (at [31]).

 (ii) Sentencing

Remorse as a factor in sentencing 

S v Ntozini (unreported, ECG case no CA&R 46/2014, 7 September 2015)
An offender’s true remorse for what he did is a mitigating factor relevant to sentencing; and ‘... the absence of remorse can count, in a proper case, against the accused’. See S v Mabena 2012 (2) SACR 287 (GNP) at [24.3]. It should be borne in mind that remorse is not the equivalent of regret and that the presence or absence of genuine remorse is a factual question. See the remarks made by Ponnan JA in S v Matyityi 2011 (1) SACR 40 (SCA) at [19].

In S v Ntozini (unreported ECG case no CA&R 46/2014, 7 September 2015) the trial court concluded that the convicted rapist of a ten-year-old child had no genuine remorse in that his defence (which was ultimately rejected) was inconsistent with an expression of remorse. The accused in this case claimed that on the evening in question he was so drunk that he could not remember anything and that the rape allegation was really an orchestrated effort by the complainant and her mother to incriminate him.

There was overwhelming evidence that the offender had raped the victim and the trial court rejected as false his version that he did not know what he was doing. But rejection of the accused’s version, it was said on appeal against sentence, did not mean ‘that the trial court was obliged to ignore the evidence in so far as it concerned the assessment of an appropriate sentence’ (at [11]).

Goosen J (Tshiki J and Cossie AJ concurring) stated as follows (at [12], emphasis added):

‘There was no evidence to gainsay the allegation that the appellant was deeply shocked when he realised that he was been accused of sexually assaulting the complainant. It was not disputed that he had attempted to commit suicide in police custody. These facts point to an appreciation of the effect of the offence committed and suggest that the appellant is indeed remorseful. The trial court appears to have had scant regard to these factors in coming to the conclusion that it did.’

The above approach is in line with S v Matyityi (supra) at [13] where it was indicated that a court must have regard to all the circumstances pertaining to the accused’s conduct, and not only to what was said in court, in determining whether remorse is present or not. See also the discussion of s 274 in Commentary, sv Mitigating factors: The plea of guilty and remorse and sv The plea of not guilty and remorse: Acceptance of responsibility versus absence of insight. In Ntozini at [14] Goosen J also pointed out that ‘acceptance of responsibility for one’s conduct may be gleaned from the surrounding circumstances and conduct of an accused ... and not exclusively from what the accused ... says in court’.

A finding as regards the presence or absence of remorse is very often a rather tricky one. Remorse expressed in court often lacks credibility, simply because the accused was caught red-handed, faced overwhelming evidence against him and really had no option but to plead guilty. See generally S v Mathe 2014 (2) SACR 298 (KZD) at [27] and S v Mashinini & another 2012 (1) SACR 604 (SCA) at [24]. But in Ntozini the two facts which were not in dispute provided a sufficient basis for an inference that remorse was present: the ‘deeply shocked’ condition of the accused when he was sober and informed of the allegation; and the fact that he had later in custody attempted to kill himself. At [21] Goosen J accordingly also concluded that the appellant did take responsibility for his criminal conduct and the consequences of his actions.

 (iii) Appeal and Review

Review of a decision not to divert a prosecution: A note on relevant considerations

Van Deventer v National Director of Public Prosecutions NO (unreported, GP case no 64268/2013, 7 August 2015)
In the above matter the applicant sought a review and setting aside of the decision of the respondent (hereafter the NDPP) not to divert his criminal trial on charges of drunken driving and driving a motor vehicle without a driver’s licence. These alleged offences were committed when the applicant was twenty years old. The applicant alleged that the NDPP’s decision not to divert was not in accordance with the principle of legality in that it was arbitrary, irrational, capricious and inconsistent with s 179(2) of the Constitution read with ss 20(1)(c), 21(1)(b) and 22(2)(a)(c) of the National Prosecuting Authority Act 32 of 1998. It was also alleged that the NDPP had failed to comply ‘with its own policy directives for the diversion of criminal cases’ (at [10]).
At [17] Matojane J pointed out that in Part 7 of the Policy Directives of the National Prosecuting Authority of South Africa it is stated: ‘By ‘‘diversion’’ is understood the election—in suitable and deserving cases—of a manner of disposal of a criminal case other than through normal court proceedings’.

Matojane J had no difficulty in concluding that the NDPP ‘had due regard to the guidelines and policy directives in existence’ and had considered all relevant circumstances in the exercise of the discretion to prosecute and not divert (at [21]).

What makes Van Deventer noteworthy though, is that it alerts practitioners to the kind of case or factual situation where representations to the prosecuting authority for a criminal matter to be diverted, would really be futile. It is true that South Africa does not have a system of compulsory prosecution; and the discretion to prosecute is a wide one. See the discussion in the introduction in Chapter 1 in Commentary, sv The discretion to prosecute. But the bottom line is correctly stated in paragraph 3C of the public document Prosecution Policy issued by the NDPP in terms of s 21(1)(a) of the National Prosecuting Authority Act 32 of 1998: ‘Once a prosecutor is satisfied that there is sufficient evidence to provide a reasonable prospect of a conviction, a prosecution should normally follow, unless public interest demands otherwise’. Section 179(5)(a) of the Constitution states that this Prosecution Policy ‘must be observed in the prosecution process’.

Having regard to all the facts referred to or implied in the judgment, it would appear that the applicant in Van Deventer failed to advance any weighty reason or consideration favouring diversion. It would appear that he relied on the fact that he was twenty years old at the time of the commission of the offences; and three years later when his series of representations to the prosecuting authority finally ran out, the argument seemed to have been that as a 23-year-old he was likely to end up with a criminal record should the prosecution go ahead. There were, in contrast, some very formidable considerations which, in the public interest, could not be ignored and in fact required that there should be no diversion as requested. At [18] Matojane J pointed out that the NDPP had

‘noted that [there] were no substantive facts set out in [the] applicant’s representations, and on this application for that matter, as to why [the] applicant is entitled to have his criminal charges diverted. The seriousness and prevalence of the offences as well as the circumstances under which they were committed were taken into account. [The NDPP] stated that drunken driving, driving without a driver’s licence and driving recklessly or negligently through a red traffic light accounted for deaths of many innocent drivers, passengers and pedestrians on a far too regular basis. He concluded that failure to prosecute [the] applicant effectively will not be in the public interest and will send a wrong message to the public in general including but not limited to [the applicant] and victims of such crimes.’

In the light of the above and the fact that diversion is in terms of policy directives in principle confined to ‘suitable and deserving cases’, the NDPP had no doubt made the correct decision. A decision to divert, it is submitted, would have been in conflict with the provisions of s 179(5)(a) and (b) of the Constitution, which require that prosecution policy as well as policy directives ‘must be observed in the prosecution process’. 

 Table of Cases

Click here to see the full Table of Cases of the Criminal Justice Review No 2 of 2015 


Click here for a printable .pdf version of this Review


Regulations in terms of the Prevention and Combating of Trafficking in Persons Act 7 of 2013
Expert Evidence in Clinical Negligence: A Practitioner's Guide
Crime Scene Investigation
Criminal Procedure Legislative Guide 2e
Criminal Procedure Act 51 of 1977 & Regulations 15e


Click here to register a profile and opt-in to receive these newsletters.

Click here to opt-out from this communication (please note that users are still required to have a profile in order to opt-out from receiving marketing communications from this website).


> Jutastat e-publications user helpdesk

> Juta Customer Services

> Juta Law Marketing


Jutastat electronic subscribers and print subscribers may receive important service-related information relating to their Juta publications from time to time.

Visit the Juta Law Website

© Juta and Company (PTY) LTD  2019