A bi-annual update complementing the

Commentary on the Criminal Procedure Act


NO 1 OF 2015


Andrew Paizes, Author (Editor)

Steph van der Merwe (Author)



The first of the feature articles in this edition raises an interesting and important question: what should happen if an accused has been convicted of assault with intent to cause grievous bodily harm, and, before sentence has been imposed, the victim of the assault dies of his injuries and the prosecution wishes to prosecute the accused, in a second trial, on the more serious charge of murder?

The second feature article considers the question whether a tenet of customary law—in this case ‘ukuthwala’, a practice leading up to a customary marriage—may be relied on as a defence to negate criminal liability in crimes as serious as child trafficking, rape and assault with intent to cause grievous bodily harm.

Outside of the feature articles, a number of important issues are considered. In S v Maarohanye & another 2015 (1) SACR 337 (GJ) the full bench of the High Court considered, once again, the nature, definition and scope of dolus eventualis, and cited with approval criticisms of the conventional approach submitted in feature articles in previous editions of this Review. There is a summary of the principal provisions of the Legal Aid South Africa Act 39 of 2014, as well as a review of the important regulations dealing with forensic DNA material made under the South African Police Service Act 68 of 1995.

Some of the other questions raised and considered in recent cases include these: (i) Can the non-payment of value-added tax constitute theft? (ii) Is s 38 of the National Prosecuting Authority Act 32 of 1998, which allows for the appointment of ‘outside prosecutors’, unconstitutional for falling short of the imperative that the prosecuting authority has to exercise its functions without fear, favour, or prejudice? (iii) What is a court to do when a prosecution has been instituted without the written authorisation required by policy directives (as opposed to legislation)? (iv) What should a court of appeal do when an appellant has pleaded to two charges but where the trial court has delivered a verdict on only one of these? (v) When should there be a separation of trials where not all of the accused are facing the same charges? (vi) Can the accuracy of a gas chromatograph be proved, under s 212 of the Criminal Procedure Act, by means of a certificate under s 212(4), or must there be an affidavit in terms of s 212(10)? (viii) What, precisely, is the status at a trial of evidence given by an accused at a bail application? (viii) How should the ‘multiple rape’ provisions be applied in sentence proceedings?

Three important Supreme Court of Appeal cases on improperly obtained evidence are also considered: the first deals with intercepted cell phone conversations; the second with the admissibility of a pointing out and confession made by an accused in circumstances where the warning to him of his rights was ‘woefully inadequate’; and the third with how to deal with confessions made by suspects in custody.

Andrew Paizes




  • Death of complainant: May the High Court set aside a conviction of assault so that murder charge can follow?

  • Customary law as a possible criminal defence


  • The Legal Aid South Africa Act 39 of 2014
  • The ‘Forensic DNA Regulations, 2015’ made under s 15AD of the South African Police Service Act 68 of 1995.


(a) Criminal Law

  • Dolus eventualis once more
  • Fraud: Potential prejudice and intent to deceive
  • Theft and the non-payment of VAT

(b) Criminal Procedure and Evidence

(i) Pre-sentence

  • Constitutionality of s 38 of the National Prosecuting Authority Act 32 of 1998: ‘Outside prosecutors’, the oath (or affirmation) and the fair trial right
  • The irregularity of a prosecution instituted without the written authorisation required in terms of policy directives
  • s 73: Legal representation and legal professional ethics
  • s 106(4): The right of an accused to a verdict on a charge to which he has pleaded
  • ss 155 and 156: Separation of trials: Accused not all facing same charges
  • s 212(4) and (10): Proving accuracy of a gas chromatograph by means of a certificate instead of affidavit; status of certificate as prima facie proof
  • s 225: Improperly obtained evidence: Three Supreme Court of Appeal cases
  • s 235 and s 60(11B)(c): Status at trial of evidence given by an accused at a bail application

       (ii) Sentencing

  • Sentencing and multiple rapes: Section 51(1)—as read with Part 1 of Schedule 2—of the General Law Amendment Act 105 of 1997
  • s 298: Correction of sentence by the sentencing court

       (iii) Forfeiture and Confiscation

  • s 48 of Prevention of Organised Crime Act (POCA) of 1998: Application of proportionality test
  • s 50 of POCA: Proof that person knew or had reasonable grounds to suspect property was an instrumentality of an offence
    Confiscation in terms of s 18 of POCA: Requirement of a ‘benefit’

      (iv) Appeal and Review

  • Bail: Abolition of the automatic right of appeal to the Supreme Court of Appeal against refusal of bail by a High Court sitting as court of first instance
Death of complainant: May the High Court set aside a conviction of assault so that murder charge can follow?

On 14 February 2013 the accused in S v Lelaka (unreported, NWM case no CAF 10/2014, 10 October 2014) had appeared in the lower court where he was charged with, pleaded guilty to and was convicted of assault with intent to do grievous bodily harm committed four days earlier. The accused had legal representation and the magistrate was satisfied that the accused had admitted all the elements of assault with intent to do grievous bodily harm (hereafter referred to as ‘assault GBH’). Bail was withdrawn and the matter was remanded to 28 February 2013 for purposes of proving previous convictions, if any (at [4]).

However, on 28 February 2013 the prosecutor informed the magistrate that the complainant in the assault GBH charge had in the interim died (at [5]). The prosecutor then successfully applied for a further adjournment so that the relevant post-mortem report could be obtained ‘and to enable the state to weigh its options with regard to the new developments’ (at [5]).

On 27 May 2013 when the case recommenced, it was disclosed that in the post-mortem report the cause of death was given as ‘severe blunt force head trauma’ (at [6]). On this day the prosecutor sought a further postponement so that the case could be submitted to the Director of Public Prosecutions (‘DPP’) ‘for instructions on what to do in light of this new development’ (at [6]).

The matter was then once again remanded to 13 June 2013. On that day the accused appeared in court with a new legal representative who argued that the convicted accused was entitled to have his matter finalised by the imposition of sentence (at [7]). The prosecutor, however, took the view that ‘the state . . . [could not] . . . proceed with the trial in the light of the new information that had emerged, namely, the death of the complainant’ (at [7]).

The presiding magistrate came to the following conclusion: first, she was seized of the matter and could not wait for the response of the DPP concerned; second, the factual situation was not covered by the Criminal Procedure Act 51 of 1977; third, if she were to proceed, ‘she would be compelled to sentence the accused for assault GBH when the facts point to murder having been committed’, which ‘would lead to injustice’ (at [8]). Having come to these conclusions, the magistrate decided to recuse herself ‘and pointedly stated that the reason for doing so, was to enable the case to be taken on review with the hope that the review court may decide that the matter be commenced de novo’ (at [8]).

On review a full bench held (correctly, it is submitted) that the magistrate had erred in recusing herself (at [9]). On the principles governing recusal by presiding judicial officers, see the notes on s 145 in Commentary, sv Guidelines for the recusal of judicial officers (including assessors).

However, at [9] Matlapeng AJ (Gura J and Djaje AJ concurring) also stated that the magistrate should have referred the matter to the review court for special review without having resorted to recusal. ‘The question that arises’, said Matlapeng AJ at [13], ‘is whether it is permissible for this Court to use its inherent power, based on the interests of justice to set the proceedings aside, to enable the state to prefer a suitable charge against the accused’.

The court in Lelaka (supra) remarked that ‘the facts of this matter are both unique and novel’ (at [14]). It was noted (at [12]) that the case was not reviewable on the grounds set out in s 24(1) of the Supreme Court Act 59 of 1959 (as replaced by s 22(1) of the Superior Courts Act 10 of 2013). The court furthermore noted that review procedures provided for in the Criminal Procedure Act also did not cover the situation. In reviewing the matter, Matlapeng AJ relied on the High Court’s ‘inherent power to protect and regulate . . . [its] . . . own process, and to develop the common law, taking into account the interest of justice’ (s 173 of the Constitution). For a discussion of this inherent power and its invocation in Lelaka, see the notes in Chapter 30 in Commentary, sv Introduction. At [18] Matlapeng AJ observed as follows:

The interests of justice demand that a person should be charged and if convicted, punished for a crime he/she committed. I am of the view that it will be a monumental failure of justice and an anomaly for a person to be charged, convicted and sentenced for assault GBH, whilst there are allegations that the complainant has died as a result of such assault.

On review it was argued on behalf of the accused that setting aside of the proceedings for the trial to start de novo on a murder charge would be in violation of not only the fair trial right in s 35(3)(m) of the Constitution, which, as part of the right to a fair trial, gives an accused person the right ‘not to be tried for an offence in respect of an act or omission for which that person has previously been either acquitted or convicted’, but also the established principle that there should be finality to litigation. These arguments were rejected as ‘fallacious’ (at [19]). Section 35(3)(m) of the Constitution, it was said at [20], ‘does not bring about a new concept in our law . . . [and] . . . merely constitutionalises the well-known maxim in our law, nemo debet bis vexari pro una et eadem causa [nobody ought to be troubled twice on account of one and the same cause]’. It is respectfully submitted that this approach—even if one were to accept it as correct—made no contribution to solving the problem in Lelaka. At [20] it was stated that no injustice could follow from the mere fact that the prosecution is given the opportunity of a new trial in order to replace the assault GBH charge with the more serious charge of murder. Matlapeng AJ also observed as follows (at [21]):

It is common cause that at the time of the institution of the proceedings against the accused, the state was not in possession of the information that the deceased was about to die. The deceased was still in the hospital. The accused pleaded guilty to the lesser charge of assault GBH, which entitles him to a lesser sentence. However, the deceased has since died, which changes the charge to murder. It is only fair that the accused be charged with murder, and if convicted, be sentenced for murder.

The proceedings in the lower court were accordingly set aside (at [25]). However, can it be argued that the review court should have returned the matter to the lower court for finalisation of the trial, that is, for the completion of the sentencing proceedings which were interrupted by the magistrate’s recusal and referral of the matter to the High Court for purposes of a review?

The eagerness of the prosecutor and magistrate to get the matter to a court of review possibly stemmed from a belief that completion of the proceedings in the lower court in respect of the conviction on assault GBH would have provided the accused with a successful plea of prior conviction in the event of a subsequent prosecution on a charge of murder. The review court, too, noted that if the prosecutor had remained passive, the accused would have got away ‘with murder both literally and figuratively’ (at [22]). However, finalisation of the lower court proceedings would, in terms of our law governing the plea of prior conviction, not necessarily have precluded the prosecution from subsequently charging the accused with the murder of the person who was the complainant in the assault GBH conviction. Lansdown & Campbell South African Criminal Law and Procedure (formerly Gardiner & Lansdown) vol V (1982) at 439 explain as follows:

The plea of autrefois convict is not . . . available where it was impossible at the previous trial to prefer the more serious charge now presented. Thus a conviction for assault is no bar to a prosecution for murder or culpable homicide where the victim has died since the conviction, for the more comprehensive offence could not have been proved at the former trial, and the fact of death has altered the essential nature of the crime.

The above statement of the law is best illustrated with reference to S v Gabriel 1971 (1) SA 646 (RA) where the accused was at his first trial convicted of the attempted murder of his wife and sentenced to four years’ imprisonment. After this conviction and two years after the assault, the accused’s wife died as a result of the injuries inflicted during the assault. He was then charged with and convicted of her murder. On appeal his reliance on prior conviction was rejected on the basis that at his earlier trial for attempted murder of his wife, it was not possible to have preferred the more serious charge of murder. Likewise, in Lelaka (supra) it was common cause that at the time of the institution of the proceedings against the accused, the victim was still alive—and her death also occurred only after conviction.

In Gabriel (supra) the Rhodesian Court of Appeal relied on not only English case law, but also a passage in Voet 48.2.12 (Gane’s translation), where it is said that when someone has been punished for inflicting a wound which then at a later stage becomes a ‘lethal wound’, a further charge of ‘homicide’ would be in order. See also generally the discussion of Gabriel in the notes on s 106 in Commentary, sv Plea of autrefois acquit or convict: The requirement: Same offence or offence substantially identical. According to Kruger Hiemstra’s Criminal Procedure (2014) at 15-15 a ‘good example of facts which arise later’ can also be found in R v Dayzell 1932 WLD 157. In this case the accused, who had been convicted of the statutory offence of negligent driving was later successfully prosecuted on a charge of culpable homicide when the victim involved in the accident caused by the accused’s negligent driving died.

In Lelaka no reference was made to Gabriel and Dayzell. The review court therefore did not take into account that finalisation of the lower court proceedings might not necessarily have precluded the State from charging the accused with murder. The review court in Lelaka foresaw that if the proceedings in the lower court were not set aside, there would be a ‘monumental failure of justice’ (at [18]). However, if decisions such as Gabriel and Dayzell were followed, the accused in Lelaka could later have been charged with murder despite his earlier conviction and sentence for assault GBH.

Of course, a subsequent trial on a charge of murder would also produce its own set of issues and problems relating to matters such as competent verdicts, sentencing and duplication of convictions for criminal record purposes. In Gabriel, for example, it was stated that at the second trial the prosecution should, before plea, inform the court that it would ask for a conviction of murder only and, if murder is not proved, no other verdict than a verdict of not guilty (at 660). In the event of a conviction on the murder charge, the sentence imposed at the first trial must be taken into account (at 651).

Be that as it may, the fact remains that in Lelaka the review court had to address a difficult problem. It should also be pointed out that Lelaka can in one important respect be distinguished from cases such as Gabriel and Dayzell: in Lelaka the lower court proceedings were still in progress and, by setting these proceedings aside, the review court could pre-empt the problematical second trial situations that the courts in Gabriel and Dayzell had to address. But can it be said that in Lelaka there were really exceptional circumstances that required the review court to invoke its inherent jurisdiction to interfere in unterminated lower court proceedings to prevent a grave injustice which might otherwise have arisen? The matter is indeed debatable.

Steph van der Merwe

 Customary law as a possible criminal defence

The issue raised in S v Jezile (unreported, WCC case no A 127/2014, 23 March 2015) was whether and to what extent a tenet of customary law may be relied upon as a defence to negate criminal liability. The appellant in that case had been convicted in a regional court on one count of child trafficking in violation of ss 284(1) and 305(1)(s) of the Children’s Act 38 of 2005, three counts of rape, one count of assault with intent to cause grievous bodily harm, and one count of common assault. A sentence of what amounted effectively to 22 years’ imprisonment was imposed. The main thrust of his appeal was that the ‘trial court had misdirected itself in not proceeding from the premise that the merits should have been determined within the context of the practice of ukuthwala, or customary marriage’.

All the convictions related to a series of events which occurred between January and March 2010, mostly in a remote rural area of the Eastern Cape. The appellant, who was 28 years old at the time, left his residence in Philippi in the Western Cape for his home village in the rural Eastern Cape with the specific intention of finding a young woman there to conclude a marriage in accordance with his custom. He wanted a virgin, ideally around 16 years of age. He noticed the complainant, then 14 years old, and decided she would make a suitable wife. Although he had not been introduced to her or even spoken to her, he asked his family to start the traditional lobola negotiations with her family. These were concluded over the course of a single day. The complainant was called the next morning to a gathering of male members of both families to be told that she was to be married in another village. She was removed from her home and taken to the appellant's home, introduced to him and informed that he was to be her husband. She was dressed in traditional attire and instructed to partake in various traditional ceremonies and household duties. She thus became the appellant’s customary wife. Lobola of R8 000 was paid by the appellant to her maternal grandmother, who gave it to the complainant’s mother.

The complainant, however, was unhappy. She left her new home a few days into the marriage, hiding first in a nearby forest and then, on her mother’s instruction, at another house. She was found and promptly returned to the appellant by her own male family members two to three days later. Shortly thereafter, the appellant told her that he would be returning to Cape Town with her. This move was sanctioned by her male family members. In Philippi, sexual intercourse took place between them on various occasions, all of which were against her will. They argued much and during one such argument she sustained an open wound to her leg. Shortly thereafter, she fled and reported the matter to the police.

In the appeal it was argued that the trial court had demonstrated a lack of understanding of the practice of ukuthwala. It was submitted that ‘consent’ within this practice was ‘a concept that must be determined in accordance with the rightful place which customary law has in our constitutional dispensation, because it is an integral part of ukuthwala that the “bride” may not only be coerced, but will invariably pretend to object (in various ways) since it is required, or at least expected, of her to do so’ (at [52]). This, as a result, informs the intention of the male since, ‘depending on the permutation the consent of the female [is] irrelevant’.

With the assistance of a vast array of amici curiae, the court (Yekiso, Saldanha and Cloete JJ) examined the practice of ukuthwala. It took notice of the fact that the practice was the subject of much public attention and debate and that ‘its current practice is regarded as an abuse of traditional custom and a cloak for the commission of violent acts of assault, abduction and rape of not only women, but children as young as eleven years old by older men’ (at [56]). Several organisations regarded it as a ‘harmful cultural practice’.

The court set out the relevant constitutional and legislative provisions as well as conventions and protocols to which South Africa is a signatory. It pointed, in particular, to s 211(3) of the Constitution, which provides that the courts ‘must apply customary law when that law is applicable, subject to the Constitution and any legislation that specifically deals with customary law’. It pointed, too, to s 28(2) of the Constitution, which provides that a child’s best interests are of paramount importance in every matter concerning the child, as well as the Recognition of Customary Marriages Act 120 of 1998. The latter sets out three requirements for a valid customary marriage: both parties must be over 18 years of age; both must consent to be married under customary law; and the marriage must be negotiated and entered into or celebrated in accordance with customary law. In the case of minors, there must be consent from both parents (or legal guardians) of the parties.

A renowned expert on customary law, Professor RT Nhlapo, explained that it was ‘critical to understand that customary law posits both regular and irregular means of initiating and concluding a customary marriage’ and that ‘ukuthwala is one such irregular method which would, if the precepts of the custom were correctly followed, eventually lead to the conclusion of a valid marriage under customary law’. It was not a marriage in itself, but was merely a method to initiate marriage negotiations by the respective families. He maintained that ukuthwala was not properly performed in this case: the young age of the complainant, her lack of consent and the fact that lobola was paid before the ukuthwala occurred, indicated that this was not a true instance of ukuthwala, which was sometimes abused to justify ‘patently offensive behaviour such as rape, violence and similar criminal misconduct under the guise of ukuthwala’ (at [75]). These ‘misapplied forms’ of ukuthwala were a ‘perversion of the custom’ which was prevalent in rural areas because of widespread poverty and was ‘no more than sexual slavery under the guise of a customary practice’ (at [78]).

The court found that it ‘appeared that the appellant conceived of ukuthwala as a form of marriage contrary to the opinions of the experts and the amici (that it is no more than a portal that commenced the process of marriage negotiations)’ (at [85]). He claimed that he had no reason to think that she was an unwilling wife, and that there had been compliance with the traditional practices of his custom as he understood it since (i) her uncles and his family had entered into the negotiations; (ii) they had arrived at an agreement as to the amount of lobola to be paid; and (iii) her family had willingly handed her over to him in marriage and returned her to him when she had run away. He relied, too, on the ‘customary practice that a female would not explicitly consent to the removal by the man when conducting the ukuthwala and would pretend to resist as a sign of her modesty’ (at [87]). This, he claimed, ‘created an ambiguity about consent’, and her acts of running away were ‘no more than in accordance with such long standing and recognised practice’.

The court, however, rejected the appellant’s account of how he viewed her act of running away. The act of ‘walking through a forest at night, alone, sleeping in the open and exposed to the elements was nothing more than a desperate attempt on her part at escaping from the appellant’ (at [88]), and it was ‘nothing more than a cynical attempt on the appellant’s part to claim that he harboured a belief that when the complainant ran away, she did so out of sheer modesty’. His reliance on the customary practice was, thus, ‘entirely misplaced’. The trial court had not found, correctly in the view of the full bench, that she had consented to the customary marriage, the sexual intercourse or the relocation to Cape Town. Moreover, the incidents complained of took place after a ‘traditional’ ukuthwala would have occurred, so that he ‘could not in any event have placed reliance on the practice of ukuthwala (in the traditional sense) as justification for his conduct’ (at [90]). What he did attempt to do, said the court, was ‘to rely on the aberrant form of ukuthwala as being the living form of customary law to justify his conduct’. The trial court had found (correctly, in the view of the appeal court) that ‘the appellant had not asserted any customary law precept to have justified his conduct, or that he had acted in the belief that he had entered into a customary marriage that permitted sexual coercion’ (at [92]; emphasis added). On appeal, however, the appellant ‘re-asserted a reliance on the practice of ukuthwala, albeit in its aberrant form, which was permissive of coercion in respect of the sexual assaults to subdue her’. Academic articles were cited in the initial heads of argument to the effect that ‘violence was always a part of the traditional form of ukuthwala’, and that the act of sexual penetration, ‘violently enacted or not’, was ‘one crucial part of the process of turning a girl into a wife, and thus enabled her attainment of an adult status’.

The court found his reliance on this ‘insightful piece of research in order to justify his conduct’ to be ‘misplaced’. In its view ‘it [could] not be countenanced that the practices associated with the aberrant form of ukuthwala could secure protection under our law’ (at [95]). It added: ‘We cannot therefore, even on the rather precarious ground of the assertion by the appellant of a belief in the aberrant form of ukuthwala as constituting the “traditional’’ customs of his community, which led to a “putative customary marriage”, find that he had neither trafficked the complainant for sexual purposes (as defined) nor committed the rapes without the necessary intention’ (at [95]; emphasis added).

This conclusion contains two propositions. The first is that there was nothing to justify the appellant’s conduct. In particular, the aberrant form of ukuthwala asserted by him did not negate the unlawfulness of his conduct. It is impossible to quarrel with this conclusion. Any assertion that what the appellant did in Jezile could be justified on the basis of an ‘aberrant’ form of a customary practice, one severely criticised by experts in the field as a ‘perversion of the custom’, would surely bring our system of criminal justice into disrepute.

The second is somewhat more complex and nuanced. It is that the appellant could not be said to have lacked mens rea (or fault) in the form of intention (or dolus) by reason of his ‘belief in the aberrant form of ukuthwala as constituting the “traditional” customs of his community, which led to a “putative customary marriage”’. The requisite form of fault in the offences with which the appellant was charged was intention (dolus). To establish this form of fault, it must be shown, in this context, that the appellant at least foresaw the real possibility that his conduct was unlawful. A genuine belief that his conduct was lawful would, if it were shown to be reasonably possible, negate this element.

And this is where things become a little tricky for the State, since one would expect, in cases like this, that the defence would call attention to both the confusion surrounding the various forms of the cultural practice (traditional and aberrant) and the fact that the custom seemed to indicate that the bride’s ostensible resistance and unwillingness should not be taken seriously. Both factors conduce to creating reasonable doubt as to whether an accused might have genuinely believed that he was acting lawfully.

The court in Jezile, however, rejected such a defence out of hand. No reasons were given other than that the defence rested on the ‘rather precarious ground of the assertion by the appellant’ of such a belief (emphasis added). However, if the earlier assessment by the court of the appellant as a poor witness is considered, especially its description of his explanation of her attempt to run away as an act of sheer modesty as ‘cynical’, the court’s rejection of this defence makes good sense. For if, as the trial court found and the court of appeal accepted, consent had not been shown in respect of the customary marriage, the act of intercourse or the relocation to Cape Town, it is very difficult to accept that it was even reasonably possible that the appellant genuinely believed that his conduct was lawful. This conclusion is fortified by the finding of the trial court—noted by the court of appeal at [49]—that ‘even on the appellant’s version he knew that he had no right to force the complainant into anything against her will, which effectively put paid to any doubt being cast on his mens rea given the court’s acceptance of the complainant’s version’.

Andrew Paizes

 The Legal Aid South Africa Act 39 of 2014

The above Act came into operation on 1 March 2015 (Proc R7 in GG 38512 of 27 February 2015). This Act—hereafter referred to as the ‘LASA Act’—repealed the whole of the Legal Aid Act 22 of 1969 and contains several transitional provisions which secure the uninterrupted availability and management of legal aid and advice. See ss 25 and 26 of the LASA Act. Section 26(3) provides, for example, that anything done in terms of legislation repealed by s 25 and which could have been done in terms of the LASA Act, ‘is regarded as having been done in terms of this Act’.

The LASA Act regulates legal aid (and especially legal aid in criminal matters) in greater detail than the now-repealed Legal Aid Act which had to be amended repeatedly. See, for example, the summary of s 22 below, sv ‘Chapter 5 (ss 22 to 27): General provisions’.

The LASA Act consists of five chapters:

Chapter 1 (s 1): ‘Definitions’

‘Legal Aid South Africa’ (LASA) is defined as ‘the national public entity, that is established under section 2(1), in order to give effect to the objects provided for in section 3’.

Chapter 2 (ss 2 to 14): ‘Legal Aid South Africa’

Section 2 establishes LASA as ‘a public national entity’ (as provided for in the Public Finance Management Act 1 of 1999) which is governed by a Board of Directors.

Section 3 identifies the objects of LASA: to render or make available legal aid and legal advice (s 3(a)); to provide legal representation to persons at state expense (s 3(b)); and to provide education and information concerning legal rights and obligations, as envisaged in the Constitution and the LASA Act (s 3(c)).

The powers, functions and duties of the Board are set out in s 4. One of the duties identified is the duty to provide legal representation at state expense, as envisaged in the Constitution and the LASA Act, ‘where substantial injustice would otherwise result and render or make legal aid and advice available’ (s 4 (1)(f)).

Section 5 provides that the directors, employees and agents of LASA ‘must serve impartially and independently and exercise powers and perform their functions in good faith and without fear, favour, bias or prejudice’.

The rest of the sections in Chapter 2 deal with various aspects relating to the Board, such as its composition (s 6) and termination of membership (s 10).

Chapter 3 (ss 15 and 16): ‘Performance of administrative work of Legal Aid South Africa’

The Board is required to appoint ‘a fit and proper person who has applicable knowledge and experience’ as the chief executive officer of LASA (s 15), who is, amongst other duties, tasked with giving effect to the objects of LASA (s 16(1)(b)).

Chapter 4 (ss 17 to 21): ‘Support structure of Legal Aid South Africa’

Sections 17 and 18 deal with the employees and agents of LASA as well as the terms of employment.

Section 19 contains an important provision as regards legal professional privilege. This section is headed ‘Protection of client privilege in certain circumstances’. Section 19(1) states that a private legal practitioner instructed by LASA to represent a person who qualifies for legal aid under the LASA Act must, when required by LASA to do so, ‘grant access to the information and documents contained in the file relating to the person in question for the sole purpose of conducting a quality assessment of the work done by the legal practitioner’. It should be noted that the client—as holder of the privilege—cannot prevent access in these circumstances. It is the lawyer’s duty to grant access, and this duty only arises where LASA requests access for quality assessment purposes. Section 19(2) determines, furthermore, that the documents and information referred to in s 19(1) remain privileged against any other party as information falling within legal professional privilege, despite having been made available to LASA. It is submitted that LASA can, in turn, also be prevented from using the information for any other purpose than quality assessment.

Section 20 regulates recovery of costs by LASA. Section 21 deals with the finances of LASA: the funds of LASA consist of money appropriated by Parliament (s 21(a)) and money received from any other source (s 21(b)) and must be budgeted for, managed and accounted for, in terms of the Public Finance Management Act 1 of 1999.

Chapter 5 (ss 22 to 27): ‘General provisions’

Section 73 of the Criminal Procedure Act 51 of 1977 deals with the right of an accused to legal representation after arrest and at criminal proceedings. See the notes on s 73 in Commentary, sv An introduction and historical background to the right, opportunity and means to obtain legal representation. Section 73 must be read with s 22 of the LASA Act. Section 22 supplements s 73(2A) in that it provides for legal aid by direction of courts in criminal matters.

In terms of s 22(1)(a) a court in criminal proceedings is required to consider the following factors in deciding whether it should direct that a person be provided with legal representation at state expense: the personal circumstances of the accused concerned (s 22(1)(a)(i)); the nature and gravity of the charge on which the person is to be tried or of which he has been convicted, as the case may be (s 22(1)(a)(ii)); whether any other legal representation at state expense is available or has been provided (s 22(1)(a)(iii)); and any other factor which in the opinion of the court should be taken into account (s 22(1)(a)(iv)). Subject to s 22(3), the court may direct that legal representation at state expense be provided only if it has referred the matter, together with any report the court may consider necessary, for LASA’s evaluation and LASA has made a recommendation whether or not the person concerned qualifies for legal representation (s 22(1)(b)).

If a court has referred a matter in terms of s 22(1)(b), LASA must evaluate and report on the matter in accordance with the regulations made under s 23(1) and the Legal Aid Manual as provided for in s 24(1) as read with s 26(6)(a). See s 22(2)(a). LASA’s report must be in writing and must be submitted to the court’s registrar or clerk, as the case may be, who must make copies available to the court as well as the person concerned (s 22(2)(b)).

In terms of s 22(2)(c) LASA’s report must include the following: a recommendation whether or not the person concerned qualifies for legal representation; particulars relating to the factors referred to in s 22(1)(a)(i) and (iii); and any factor which, in LASA’s opinion, should be taken into account.

It was pointed out above that a court’s referral in terms of s 22(1)(b) is subject to the provisions of s 22(3). Section 22(3) provides that a court may refer a matter in terms of s 22(1)(b) only if the person concerned

(a) (i)  has applied to Legal Aid South Africa for legal representation at state expense;

     (ii)  has been refused legal representation at state expense by Legal Aid South Africa; and

     (iii) has exhausted his or her internal right to appeal within the structures of Legal Aid South Africa against the refusal;

(b)     has applied for legal representation and has not received any response to the application within a reasonable time; or

(c)    has been refused legal representation at state expense by Legal Aid South Africa and the court is of the opinion that there are particular circumstances that need to be brought to the attention of Legal  Aid South Africa by the court in a report referred to in subsection (1)(a)(ii).

A decision by LASA in any criminal proceedings relating to the following matters is subject to review by the High Court at the instance of the person affected by such decision: the particular legal practitioner assigned to any person (s 22(4)(a)(i)); the fee to be paid by LASA to a particular person (s 22(4)(a)(ii)); the number of legal practitioners to be assigned to a particular person or group of persons (s 22(4)(a)(iii)); or the contribution, if any, to be paid to LASA by the persons in question, and when and the manner in which the fee is to be paid (s 22(4)(a)(iv)). LASA may, in any review proceedings referred to in s 22(4)(a)(ii), not be required to pay more than the maximum amounts determined in the Legal Aid Manual in terms of s 24(1)(c). It is, furthermore, provided that only a court in review proceedings may make an order relating to the matters referred to in s 22(4). See s 22(5).

Section 22(6) provides that in determining whether any person is entitled to legal representation at state expense and before any court orders such representation, the legal aid applicant bears the onus of showing, on a balance of probabilities, that he is unable to afford the cost of his own legal representation (s 22(6)(a)); that he has made a full disclosure of all relevant facts and documents pertaining to his inability to pay for his own legal representation (s 22(6)(b)); that he has a lifestyle that is consistent with his alleged inability to afford the cost of his own legal representation (s 22(6)(c)); and that he has cooperated fully with any investigation conducted by LASA (s 22(6)(d)). The decision in Legal Aid Board v The State & others 2011 (1) SACR 166 (SCA) must now be read subject to s 22(6), which makes it clear that the burden of proof is on the applicant who must meet the civil standard of proof in respect of the issues as identified.

Section 22(7) contains the sensible provision that no accused may receive legal representation at state expense if that accused has applied for the release of an amount for reasonable legal expenses in terms of s 44(1)(b) of the Prevention of Organised Crime Act 121 of 1998, and where the court has turned down the application due to a lack of a full disclosure as required in terms of s 44(2)(b) of that Act.

The rest of Chapter 5 deals with matters such as regulations (s 23) the Legal Aid Manual (s 24) and transitional arrangements (s 26).

 The ‘Forensic DNA Regulations, 2015’ made under s 15AD of the South African Police Service Act 68 of 1995


The ‘Forensic DNA Regulations, 2015’ (hereafter the ‘DNA Regulations’) were made by the Minister of Police under s 15AD of the above Act (hereafter the ‘SAPS Act’) and came into operation on 13 March 2015. See GN R207 in GG 38561 of 13 March 2015.

General background

Section 6 of the Criminal Law (Forensic Procedures) Amendment Act 37 of 2013 (hereafter the ‘DNA Act’) inserted Chapter 5B in the SAPS Act. This chapter is headed ‘Establishment, Administration and Maintenance of National Forensic DNA Database of South Africa’ and consists of ss 15E to 15AD. The full text of Chapter 5B is reproduced as Appendix E in the Supplementary binder of Commentary.

Chapter 5B regulates the establishment, administration and maintenance of the National Forensic DNA Database of South Africa (the ‘NFDD’). Section 15F of the SAPS Act states that the objective of Chapter 5B is to establish and maintain a national forensic DNA database in order to perform comparative searches for the following purposes: (a) to serve as a criminal investigative tool in the fight against crime; (b) to identify persons who might have been involved in the commission of offences, including those committed before the coming into operation of Chapter 5B; (c) to prove the innocence or guilt of an accused person in the defence or prosecution of that person; (d) to exonerate a person convicted of an offence; or (e) to assist with the identification of missing persons or unidentified human remains.

Chapter 5B of the SAPS Act must be read with Chapter 3 of the Criminal Procedure Act 51 of 1977. The latter chapter consists of ss 36A to 37. It is headed ‘Ascertainment of Bodily Features of Persons’ and regulates police powers in this regard. Some bodily samples so obtained are destined for forensic DNA analysis and then become matters to be dealt with in terms of the provisions of Chapter 5B of the SAPS Act. Section 15Q of this Act contains, for example, rules governing the analysis, retention, storage, destruction and disposal of samples.

The interaction between Chapter 5B of the SAPS Act and Chapter 3 of the Criminal Procedure Act is further demonstrated by the provisions of s 36A(5)(a) of the latter Act. In terms of this section an authorised person taking a buccal sample in terms of Chapter 3, or any other law, is required to do so in accordance with the requirements of any regulation made by the Minister of Police. See also the notes on s 36A in Commentary, sv Taking of a buccal sample: Section 36A(3), (4) and (5).

Section 15AD(1) of Chapter 5B provides that the Minister of Police must—in order to achieve the objects of this chapter—make regulations regarding all matters which are reasonably necessary or expedient to be provided for and which must be followed by all police officials or members of the Independent Police Investigative Directorate (hereafter ‘IPID’). Some of the matters specifically included are: the manner in which to secure a crime scene for the purposes of collecting crime scene samples (s 15AD(1)(a)); the manner in which to preserve safely and ensure timely transfer of collected samples to the forensic science laboratories (s 15AD(1)(c)); the application process for access to the forensic DNA profile and crime scene samples for exoneration purposes (s 15AD(1)(g)); and the requirements for the taking of a buccal sample in a designated area (s 15AD(1)(j)).

The DNA Regulations referred to above were the Minister’s response to s 15AD(1). It is an elaborate set of regulations. What follows is merely a selection of some of these regulations, especially those which can protect and enhance the integrity of the samples and so-called chain of identification and, ultimately, the reliability of the outcomes.

The taking of a DNA sample

Regulation 2(1) of the DNA Regulations determines that a ‘DNA reference (buccal) collection kit’ must be employed in collecting a buccal sample. In terms of s 15E(e) of the SAPS Act a ‘buccal sample’ means ‘a sample of cellular material taken from the inside of a person’s mouth’. A similar definition is found in s 36A(aA) of the Criminal Procedure Act. In the Afrikaans translation of regulation 2(1), a ‘buccal sample’ is referred to as a ‘wangholtemonster’. The Supreme Court of the United States has accepted that buccal cell collection by wiping a cotton swab against the inside of a person’s mouth to collect some skin cells is a quick and painless procedure posing no threat to the health or safety of the person concerned. See Maryland v King 569 US (2013); 133 S Ct 1958 (2013).

Regulation 2(4) provides for the use of protective clothing:

The personnel protective clothing provided in the DNA reference (buccal) collection kit must be worn by the authorised person when a buccal sample is collected from any person. The personnel protective clothing provided in the DNA reference (buccal) collection kit must be disposed of by placing these items in the original packaging (pouch) of the kit, which in turn must be attached to the evidence sealing bag containing the DNA reference sample.

The buccal sample must be taken by the authorised person immediately after the fingerprints have been taken (reg 2(8)) and in the presence of a person who is required to countersign the collection form (reg 2(7)). The ‘authorised person’ referred to in the regulations is defined in s 15E(b) of the SAPS Act and means ‘with reference to buccal samples . . . a police official or member of the Independent Police Investigative Directorate . . . who is not the crime scene examiner of the particular case and who has successfully completed the training prescribed by the Minister of Health . . . in respect of the taking of a buccal sample’. See also generally s 36A(1)(b)(ii) of the Criminal Procedure Act where a similar definition of ‘authorised person’ is given.

The re-taking of a buccal sample

An authorised person may in certain circumstances re-take or supervise the re-taking of a buccal sample ‘if the buccal sample taken . . . was either not suitable or insufficient for forensic DNA analysis’ (s 35D(5)(a) of the Criminal Procedure Act). Regulation 2(9) states that the re-taking of a buccal sample must take place within thirty days after receiving such a request from the Forensic Science Laboratory.

The keeping of records in respect of collected buccal samples and crime scene samples

Regulation 3 deals with the above matter and distinguishes between samples taken in three different situations: samples from arrested persons (reg 3(1)); samples for investigative purposes (reg 3(2)); and samples collected by the IPID (reg 3(3)). Regulation 3(1)(b) provides as follows as regards arrested persons:

The unique barcode form reference number of the DNA Reference (buccal) Collection kit must be recorded on form SAPS 76 and on the collection form (provided with the DNA Reference (buccal) Collection kit). The original collection form must be filed in the docket and the copy of the form, together with the buccal sample must be placed in the evidence sealing bag.

The collection form which accompanies the kit referred to in reg 3(1)(b), must be completed before the sample is taken (reg 3(1)(c)).

The circumstances in which samples may be taken for investigative purposes are set out in s 36E of the Criminal Procedure Act. In this regard regulation 3(2)(a) provides as follows: ‘When a buccal sample is collected for investigative purposes from a person who is not arrested, the DNA Reference (buccal) Collection kit must be utilised and his or her fingerprints must be taken on form SAPS 192.’ The authorised person must clearly indicate on the collection form that the sample was taken for investigative purposes (reg 3(2)(b)).

A DNA reference (buccal) collection kit must also be used by an authorised person who collects a buccal sample on behalf of the IPID (reg 3(3)(a)). Regulation 3(3)(b) requires that the words ‘for investigative purposes’ must be recorded on the evidence sealing bag.

Preservation of samples

Reasonable steps must be taken to ensure that samples are not exposed to heat degradation (reg 4(5)).

Packaging of buccal samples

A buccal sample may not be packaged with other exhibits (reg 4(6)(a)) and must be packaged in a separate evidence sealing bag and submitted to the Forensic Science Laboratory (reg 4(6)(b)).

Buccal sample and covering letter

A buccal sample must be accompanied by a covering letter clearly requesting that it ‘must be compared with crime scene samples that have previously been or will be submitted to the Forensic Science Laboratory’ (reg 4(6)(c)). The information that must be recorded in the covering letter is set out in regulation 4(7) and includes the bar code number of the buccal sample (reg 4(7)(a)).

Reporting of forensic DNA findings

The Forensic Science Laboratory must perform forensic DNA examinations on all buccal samples received at the laboratory (reg 7(2)). A forensic analyst attached to the laboratory must ensure that forensic DNA profiles (as defined in s 15E(m) of the SAPS Act) are derived from a crime scene sample, bodily sample and buccal sample within thirty days from the receipt thereof at the laboratory (reg 6(4)). Regulation 6(4) also requires the forensic analyst to convey the result of the analysis to the relevant investigating officer who is, in turn, required to file the report of the result in the police docket. Regulation 6(5) provides as follows:

The forensic analyst must report to the investigating officer the outcome of the examination and the results of the tests for purposes of section 212(6)(a) and (b) of the Criminal Procedure Act if—

(a)  the person under investigation or the DNA of a suspect matches the DNA found in the crime scene sample;

(b)  an identification of human remains has been made;

(c)  preliminary tests on the exhibit material in the case are negative or no DNA could be found in the crime scene sample relevant to the case;

(d)  a person under investigation or a suspect may be excluded by the DNA found in the crime scene sample; and

(e)  DNA was found in crime scene samples in the case, but no match could be made as no buccal sample was received by the Forensic Science Laboratory for comparison.

It should be noted that s 212(6) was amended with effect from 31 January 2015 to include the words ‘bodily sample or crime scene sample’. See s 3(a) of the Criminal Law (Forensic Procedures) Amendment Act 37 of 2013 which came into operation on 31 January 2015 (Proc R89 in GG 38374 of 30 December 2014). Section 212 provides for the admissibility of affidavits and certificates as prima facie proof of their contents. See the notes on s 212 in Commentary, sv Meaning of prima facie proof in s 212 and, sv What may be proved by way of affidavit or certificate in terms of s 212?

Access to the forensic DNA profile and crime scene sample for exoneration purposes

Regulation 26 deals with the above matter which was specifically identified in s 15AD of the SAPS Act as a matter requiring regulation by the Minister of Police.

Regulation 26(1) states that any person who believes he was wrongfully convicted, may submit a written request to the ‘authorised officer’ to have access to the forensic DNA profile derived from a crime scene sample that was collected and submitted to the Forensic Science Laboratory in a particular case. Section 15E(a) of the SAPS Act defines ‘authorised officer’ as ‘the police officer commanding the Division responsible for forensic services within the Service or his or her delegate’. See also s 36A(aA) of the Criminal Procedure Act.

The written request referred to in regulation 26(1) must identify ‘the relevant station and case number in respect of which it is alleged that there was a wrongful conviction’ and must, furthermore, advance reasons why the forensic DNA profile is required (reg 26(2)).

In terms of regulation 26(3) the authorised officer or his delegate must consider the request within thirty days after having received it at the Forensic Science Laboratory. The same regulation also determines that if a request is refused, written reasons must be provided.


It would seem that the DNA Regulations not only provide adequate support to the processes of gathering samples as provided for in Chapter 3 of the Criminal Procedure Act, but also ensure that the NFDD as established in terms of Chapter 5B of the SAPS Act can make an important contribution to the perennial problem of separating the guilty from the innocent.

 (a) Criminal Law

Dolus eventualis once more

S v Maarohanye & another 2015 (1) SACR 337 (GJ)

Dolus eventualis has been the focus of our attention in three feature articles in Criminal Justice Review (see ‘Dolus eventualis revisited: S v Humphreys 2013 (2) SACR 1 (SCA) in CJR 1 of 2013; ‘Dolus eventualis again’ in CJR 1 of 2014; and ‘The trial of Oscar Pistorius—dolus eventualis once again’ in CJR 2 of 2014). The first of these has been cited with approval and relied upon by the full bench of the High Court in Maarohanye.

The trial court had convicted the appellants of murder after four children had been killed and two others injured when the appellants, while racing their motor vehicles in a public road in a built-up area, had lost control of their vehicles and ploughed into a group of schoolchildren walking on the pavement. The court (Mlambo JP, Maluleke J and Pretorius J), following what was said in the CJR article, accepted that ‘[d]olus eventualis . . . is not amenable to containment within a simple formula, the facts of the matter having a lot to do with the ultimate conclusion’ (at [21]). The court continued (again relying on the above article):

All we can say is that the strongest case for dolus eventualis is likely to be found where there is foresight of a substantial possibility of causing the result in question; where the activity is part of an overtly dangerous and unlawful enterprise; and where the accused is uncaring about whether the victim lives or dies as a result of his conduct. On the other hand, the weakest case will tend to be where there is foresight of only a slight possibility of death, and where the accused strongly hopes that life will not be lost in consequence of his conduct, and has taken considerable care to ensure that the risk is eliminated.

The court added that it ‘is, obviously, not easy to state with certainty precisely where the dividing point on this line will be; the facts of each case should provide the key’. In Humphreys, for instance, it was clear, since there ‘the reason for taking the risk was not callous indifference to human life but, rather, impatience and foolish impetuosity. Humphreys was not indifferent to the fate of his passengers but, based on previous conduct under similar circumstances, he was unrealistic in his assessment of the degree of risk on that occasion, hence the finding of culpa.’

This pronouncement is in line with all that we have submitted in the three articles on the topic in CJR and is to be welcomed for what, we submit, is its flexibility, universality of application, and sensitivity to the nuances of each individual case. Several previous pronouncements of our courts on this subject have, in our view, lacked these qualities.

On the facts of this case, the court was of the view that dolus eventualis had not been shown to exist, and that the appellants should, instead, have been convicted of culpable homicide. It had been shown that the appellants were, at the time, under the influence of drugs which had ‘induced a sense of euphoria in them, giving them a sense that they would not cause any collision and that other road users would give way to them’ (at [22]). The full bench of the High Court accepted, then, that ‘the appellants’ faculties were affected, in the sense that their judgment was impaired, thus inducing “an exuberant or over optimistic frame of mind” which caused them to take risks and drive as they did on the day in question’. This ‘should have left the trial court with substantial doubt regarding an appreciation of, and, importantly, a reconciliation, by the appellants to,  the consequences of their driving conduct causing death or serious injury’ (at [23]). This state of mind, said the court, was ‘completely at variance with that required to establish an appreciation of the consequences of one’s conduct and further reconciling oneself to such consequences taking place’. In this context, there could ‘be no suggestion that the appellants had the foresight that their escapade could result in death and serious injury to pedestrians and that they reconciled themselves with that eventuality’ (at [24]). ‘Their mental make-up must . . .  have been the opposite of that causing death or injury: their disposition was that no collision, let alone a life-threatening one, would take place’; there was ‘no foresight of the possibility of a collision causing death or serious injury, coupled with reconciliation to that eventuality and proceeding with it despite that reconciliation’, and, therefore, ‘no dolus eventualis’.

One cannot fault this conclusion if one accepts, as the court did, that the effect of the drugs on the appellant’s minds was to render it reasonably possible that there was no foresight of death at all, or, even, no foresight of the real or substantial possibility of death. A more interesting question would have been whether dolus eventualis would have been found to exist if this was not so. If, then, the effect of the drugs had been to embolden the appellants without displacing the consciousness of the risk of death, should they have been convicted of murder? If one turns to the approach endorsed by the court, it would be in the appellant’s favour that they were engaged in an act (driving) which, if properly carried out, is not only lawful and useful, but necessary for the proper running of the economy. In addition, they would clearly be hoping and expecting that death or serious injury would not ensue, since the prospect of an accident would imperil their own lives as well as others’. On the other hand, however, would be the fact that the activity (driving) had been undertaken in a clearly unlawful and dangerous manner, and it could not be said that racing vehicles in a public road was anything other than an abuse of the act of driving, having no public or economic utility and constituting an inherently dangerous and life-threatening activity.

It would have been interesting to learn where on the line the court would have placed the matter. It is, we submit, ordinarily inappropriate to convict a person of murder in deaths arising out of the negligent driving of a motor vehicle, even if it can be established that the accused did foresee the real possibility of causing death. This is so even where the accused is grossly negligent, as in Humphreys, since the act of driving a car is not to be placed in the same category as, say, an assault, but should be treated in the same way as other socially and economically useful and important activities such as mining or running a factory. However, once the activity is undertaken in a manner that is, at its outset, clearly unlawful and lacking in any kind of social or economic value, it may be  argued that it falls out of this category and that it is no longer inappropriate to consider a conviction of murder.

The value of the pronouncement in Maarohanye is that it invites the courts to engage in evaluations of this kind and moves away from the ‘one size fits all’ approach to dolus eventualis which has characterised many previous judgments on this question. We hope the courts accept this invitation.

Fraud: Potential prejudice and intent to deceive

S v Ndwambi [2015] ZASCA 59 (unreported, SCA case no 611/2013, 31 March 2015)

Can an accused who makes a misrepresentation to someone who knows all along that the representation is false and who was never going to rely on it to his prejudice, be guilty of fraud? This was the situation in Ndwambi, where the appellant sold a fake rhinoceros horn to a police trap for R350 000. He was convicted by the trial court, his appeal to the Free State High Court failed, and he appealed to the Supreme Court of Appeal.

It was argued that the State had not proved the intent to deceive, which was part of the intent to defraud, and that it had failed to prove the element of prejudice. The intent to defraud, said Meyer AJA, had two elements: the intention to deceive and the intention to induce the person to whom the representation had been made thereby to alter or to abstain from altering his or her legal position. Dolus eventualis would suffice in respect of each of these elements. In this case, had it been proved that the appellant had made the false representation knowingly or, at least, ‘without knowledge whether it was true or false but knowingly exposing . . . the State to a risk that it may be false and deceitfully leaving him ignorant of the exposure’? The appellant had given an account which was disbelieved and rejected. This left the court without the benefit of credible evidence from him and with only the State evidence. This prima facie evidence was able to sustain an inference that the necessary intention was present, and any suggestion that he did not know that the representation was false, said the court, ‘lack[ed] a factual foundation and would therefore amount to impermissible speculation or conjecture’ (at [17]). It ‘lay exclusively within the power of the appellant . . . to show what the true facts were but [he] failed to give an acceptable explanation’, so that the ‘prima facie inference became conclusive in the absence of rebuttal’.

The second argument was that, because the police had no intention to pay for the rhino horn, there could be no prejudice. This contention, said Meyer AJA (at [18]), ignored ‘the longstanding principle that the law looks at the matter from the point of view of the deceiver and not the deceived’. It was held in R v Dyonta & another 1935 AD 52 that even if the representee at no stage intended to act on the representation, there would still be fraud if the misrepresentation was ‘one which in the ordinary course [was] capable of deceiving a person and thus enabling the accused to achieve his object’. It was sufficient if the prejudice was only potential and it was immaterial if the person was not actually deceived because he had ‘knowledge or a special state of mind which effectively protect[ed] him from all danger of prejudice’.

In the present case the misrepresentation was ‘calculated to prejudice’. The prejudice need not be financial or proprietary, and it was not required that the prejudice should relate to the person to whom the misrepresentation was addressed (see R v Heyne & others 1956 (3) SA 604 (A) at 622F). Even though the transaction involved fake rhino horn, it must, said the court at [22], ‘indubitably be so that transactions of this kind contribute to the illegal trade in rhino horn, which we as a country must all be concerned about’.

Theft and the non-payment of VAT

Director of Public Prosecutions, Western Cape v Parker [2015] 1 All SA 525 (SCA)

In this case the following question of law was reserved: ‘Whether a VAT vendor who has misappropriated an amount of VAT which it has collected on behalf of SARS can be charged with the common law crime of theft.’ Although a failure to pay VAT is a statutory offence under s 28(1)(b) read with s 58 of the Value-Added Tax Act 89 of 1991, punishable by a sentence of two years’ imprisonment, the State was of the view that this provision was too lenient for certain cases of misappropriation of VAT, and sought sterner sanctions within the common-law crime of theft.

It was argued that a VAT vendor acts as an agent for SARS and that, if the vendor uses the VAT money for purposes other than to pay the Commissioner, he or she misappropriates those funds and is guilty of theft even if the vendor is the owner of that money, since he or she is in a position of trust vis-à-vis SARS with regard to that money.

Support for this contention, it was argued, could be found in the wording of s 7(1) of the VAT Act, which, it was argued, either expressly or impliedly created this relationship of trust. Further support, it was claimed, could be found in the judgment of the Constitutional Court in Metcash Trading Ltd v Commissioner, South African Revenue Service & another 2001 (1) SA 1109 (CC), where it was said that ‘vendors are entrusted with a number of important duties in relation to VAT’ and that ‘vendors are in a sense involuntary tax-collectors’.

Pillay JA (with whom Brand, Leach, Shongwe JJA and Willis AJA agreed), however, rejected these arguments. There was, he said, nothing in the wording of the Act that expressly or impliedly created such a relationship of trust. On the contrary, it was one of debtor and creditor. The relationship was sui generis and did not confer on the vendor the status of a trustee or an agent of SARS. If it did, the vendor would be required either to keep separate books of account or to keep a liquid fund sufficient, at any given time, to cover outstanding VAT obligations. The Act contained no such provisions, and to read the Act as creating a trust relationship would necessitate an ‘innovative approach’ which would violate the nullum crimen rule.

Reliance on Metcash, too, was misplaced. The words ‘entrusted with’ merely conveyed the idea that the vendor was required to comply with various sections of the Act which safeguarded its operation; they might as well be replaced by the words ‘burdened with’. And the description of vendors as being ‘in a sense involuntary tax collectors’ was not intended to classify vendors as official tax collectors. It signified, only, that it was left to each vendor to ensure compliance with VAT obligations, which was ‘quite different from imposing the status of a formal tax-collector or a trustee of SARS on a registered vendor’ (at [10]).

As a result, the question of law as formulated by the State was answered in the negative.

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

Constitutionality of s 38 of the National Prosecuting Authority Act 32 of 1998: ‘Outside prosecutors’, the oath (or affirmation) and the fair trial right

Moussa v S [2015] 2 All SA 565 (SCA)

The national prosecuting authority can, in the circumstances as provided for in s 38(1) of the above Act (the NPA Act), ‘engage, under agreements in writing, persons having suitable qualifications and experience to perform services in specific cases’. For purposes of s 38, ‘services’ would ‘include the conducting of a prosecution under the control and direction’ of the National Director of Public Prosecutions (NDPP), or his or her Deputy (DNDPP), or a Director of Public Prosecutions (DPP), as the case may be (s 38(4)).

Section 38 must be understood in the context of the following observations by Hartzenberg J in S v Tshotshoza & others 2010 (2) SACR 274 (GNP) at [19]: ‘All over the world, outside prosecutors are engaged to prosecute on behalf of the State. There cannot be objection in this country to the engagement of outside prosecutors in specific cases. There are many reasons why it may become necessary for the NPA to engage outsiders. One thinks of a shortage of staff or of staff with the necessary expertise and experience to prosecute in particular cases’. For a discussion of Tshotshoza, see Chapter 1 in Commentary, sv Professional independence and the fair trial risk where there is private funding of prosecutions.

In S v Delport & others (unreported, GNP case no A458/2012, 13 June 2013) the outside prosecutors appointed in terms of s 38 of the NPA Act were an advocate at the Pretoria bar and an advocate from the South African Revenue Service. According to Makgoba and Van der Byl JJ these two prosecutors were not, in addition to their ‘engagement’ in terms of s 38, required to take the oath (or make an affirmation) as prescribed by s 32 of the NPA Act for members of the prosecuting authority (at 74(c)). The oath or affirmation of the two outside prosecutors when they were admitted as advocates, was considered sufficient (at 73(b)). For a brief summary of Delport, see Chapter 1 in Commentary, sv Appointment of prosecutors in terms of s 38 of Act 32 of 1998.

In Moussa v S [2015] 2 All SA 565 (SCA) the court a quo had dismissed the appellant’s application for an order declaring s 38 of the NPA Act unconstitutional on the basis that it fell short of the constitutional principle enshrined in s 179(4) of the Constitution, namely that national legislation had to ensure that the prosecuting authority exercised its functions without fear, favour or prejudice. On appeal to the Supreme Court of Appeal, it was submitted in the appellant’s heads of argument that s 38 was inconsistent with the Constitution in that it failed to require that outside prosecutors should conduct themselves without fear, favour or prejudice. In the course of oral argument, counsel for the appellant ‘when pressed about the precise ambit of the appellant’s argument, stated that he was contending that s 38 . . . was unconstitutional because it did not compel “outside counsel” to take the oath as prescribed for permanent members of the prosecution services by s 32 . . .’. It was also contended that ‘it is the taking of the oath that is foundational to the independence of the prosecutor’ (at [12]) and, so the argument continued, the absence of the oath ultimately failed to secure the independence required for purposes of an accused’s constitutional fair trial right (at [26] and [29]). It should be noted that in Moussa the outside prosecutor concerned—unlike the two outside prosecutors in Delport (supra)—had actually taken the oath as envisaged in s 32 despite the fact that s 38 contains no such requirement. The appellant’s argument effectively placed ‘form above substance’ (at [29]).

At [18] Navsa ADP, writing for a full bench, agreed with the court below that the inquiry had to commence with reference to s 179(4) of the Constitution. This section provides as follows: ‘National legislation must ensure that the prosecuting authority exercises its functions without fear, favour or prejudice’. The NPA Act, it was noted, ‘is that legislation’ (at [18]) and provides a statutory scheme which is ‘in line with the constitutional imperative of ensuring independence, impartiality and prosecutions without fear, favour or prejudice’ (at [25]). Navsa ADP referred to the fact that s 38(4) requires that outside prosecutors conduct their prosecutorial duties ‘under the control and direction’ of the NDPP, a DNDPP or a DPP, as the case may be (at [25]):

This must mean that when persons from “outside” are engaged as prosecutors, they do so after consideration at the highest level and that the prosecutions that they are involved in are subject to the control and direction of the highest ranking officials within the NPA, who themselves have taken the oath of office prescribed by s 32. This translates into ensuring that the decision and basis of the prosecution are within the control of those officials. All of this is to ensure that constitutional imperatives are met.

Navsa ADP also confirmed the correctness of the conclusion of the court below that outside prosecutors, properly appointed in terms of s 38, are statutorily required to perform their functions as a part of the national prosecuting authority in the manner dictated by s 32(1)(a): ‘The structure of the NPA Act is such that control and supervision are in place to ensure compliance with s 32(1)(a) and constitutional norms’ (at [28]). Section 32(1)(a) provides as follows: ‘A member of the prosecuting authority shall serve impartially and exercise, carry out or perform his or her powers, duties and functions in good faith and without fear, favour or prejudice and subject only to the Constitution and the law.’ At [29] it was concluded that ‘the taking of the oath by itself’ does not secure a fair trial and also does not guarantee prosecutorial independence and impartiality: ‘It is the manner in which prosecutions are initiated and conducted that is the test of prosecutorial independence’.

A comparative survey of legislation and case law in countries such as England, Canada, the United States of America, India and Australia made it clear that the appointment of ‘outside prosecutors’ was not unique to South Africa; and in only one jurisdiction (the Canadian province of Quebec) were prosecutors required to take the same oath as required from a state prosecutor in permanent employment (at [31]). It would therefore seem that the absence of a requirement in s 38 that outside prosecutors must take the oath which permanent prosecutors are in terms of s 32(2)(a) required to take, is in line with foreign prosecutorial systems similar to the South African system.

Even though no reference was made to Delport (supra), Navsa ADP also took into account in Moussa that the outside prosecutor was a member of the bar, had taken an oath upon admission and was ultimately an officer of the court (at [27]). Outside counsel is not a free agent, given his own oath when admitted and the fact that he remains subject to the control and direction of the senior officials of the national prosecuting authority as provided for in s 38(4). The fact of the matter is that the fair trial right as entrenched in s 35(3) of the Constitution governs the situation and is not at risk by the mere fact that s 38(1) of the NPA Act does not require that an outside prosecutor must take an oath like permanent prosecutors. Section 38(1) was held not to be unconstitutional (at [30]). The appeal was dismissed with costs (at [43]).

Moussa (supra) should be read with Porritt & another v National Director of Public Prosecutions & others 2015 (1) SACR 533 (SCA) at [19] where Tshiqi JA said that in the South African criminal justice system there are ‘sufficient structural guarantees . . . to ensure that an accused’s right to a fair trial is protected, irrespective of whether the prosecutor concerned is an employee of the [national prosecuting authority] or an outside counsel . . .’ Porritt is discussed in Chapter 1 in Commentary, sv Recusal or removal of prosecutor: Test to be applied.

The irregularity of a prosecution instituted without the written authorisation required in terms of policy directives

Masinga v National Director of Public Prosecutions & another (unreported, KZP case no AR 517/2013, 7 May 2015)

Section 179(5) of the Constitution states that the National Director of Public Prosecutions (NDPP) ‘must determine . . . prosecution policy’ and ‘must issue policy directives’. The Constitution also provides that the prosecution policy and policy directives must be observed in the prosecution process (s 79(5)(a) and (b)) and that the NDPP ‘may intervene in the prosecution process when policy directives are not complied with . . .’ (s 79(5)(c)).

The constitutional provisions referred to above are also captured in ss 21 and 22 of the National Prosecuting Authority Act 32 of 1998 (NPA Act). All prosecutions must comply with the prosecution policy and policy directives (s 21(1) of the NPA Act). See also generally Freedom Under Law v National Director of Public Prosecutions & others 2014 (1) SA 254 (GNP) at [147]–[148] and [182] and Chapter 1 in Commentary, sv Prosecution policy and issuing of policy directives.

In terms of Part 8 of the policy directives issued by the NDPP, certain categories of persons may not be prosecuted without the written authorisation or instruction of the Director of Public Prosecutions (DPP) concerned or a person authorised thereto in writing by the NDPP or DPP (either in general terms or in any particular case or category of cases). Prosecutors, magistrates and judges are identified as persons in respect of whom written authorisation is required. See paras 1 and 2(f) of Part 8 and see generally S v Thenga 2012 (2) SACR 628 (NCK) at [36].

In Masinga v National Director of Public Prosecutions & another (unreported, KZP case no AR 517/2013, 7 May 2015) the applicant sought a review of criminal proceedings in the regional court where he had been convicted of attempted murder and sentenced to ten years’ imprisonment. The applicant, who was a magistrate at the relevant time, complained that his prosecution took place without the required written authorisation from the DPP.

It was common cause that the DPP’s written authorisation had not been obtained (at [6]). However, Ms D—the prosecutor in the matter—stated in her affidavit that she had obtained the oral authorisation of the then acting DPP after having discussed the contents of the docket with him and after he had agreed with her that the appropriate charge was attempted murder (at [6]–[7]). Ms D also claimed that she had not ignored the relevant policy directive and knew that she had to obtain the DPP’s authorisation. This claim drew the following response from Ploos van Amstel J at [8]: ‘She apparently overlooked the requirement that the authorisation had to be in writing. It is somewhat surprising that the acting DPP also overlooked that requirement.’ It was nevertheless concluded that Ms D had indeed obtained the DPP’s oral authorisation (at [13]).

The failure to obtain the DPP’s written authorisation, held Ploos van Amstel J at [10], constituted an irregularity: ‘Written authorisation was required by the policy directives, and both section 179(5) of the Constitution and section 21(1) of the NPA Act provide that the policy directives must be observed in the prosecution process.’ The true question was therefore whether this irregularity was of such a nature that it automatically resulted in a failure of justice, vitiating the trial. At [11] reference was made to the provisions of s 322(1) of the Criminal Procedure Act 51 of 1977 in terms of which no conviction or sentence shall be set aside or altered by reason of any irregularity or defect in the proceedings, unless it appears that a failure of justice has in fact resulted from such irregularity or defect. Referring to case law, Ploos van Amstel J noted that in each instance the inquiry is whether the relevant irregularity is of such a fundamental and serious nature that the proper administration of justice and the dictates of public policy require it to be fatal to the proceedings, in which event the court would not even consider whether the irregularity caused prejudice to the accused. The position, said Ploos van Amstel J at [12], is not altered by the fact that compliance with the policy directives is required by the Constitution: the Supreme Court of Appeal has accepted the common-law test as adequate for purposes of constitutional as well as non-constitutional errors. At [14] he held:

I do not consider that the irregularity in this case was of such a nature that it per se amounted to a failure of justice. There was oral authorisation by the acting DPP, who was informed of the evidence against the applicant and agreed that he should be charged with the attempted murder of his wife. The applicant did not protest before or during the trial that the prosecution had not been authorised in writing. He raised the point for the first time on appeal. To hold that the absence of written authorisation in those circumstances per se amounted to a failure of justice, irrespective of whether the applicant was prejudiced thereby, would be contrary to the public interest and will bring the administration of justice in disrepute. The position may be different where a prosecution against a magistrate was instituted and proceeded . . . without the knowledge or consent of the DPP, or contrary to his instructions.

The next issue was whether the applicant had been prejudiced by the irregularity. The applicant argued that if the DPP’s written authorisation had been sought, the latter might have preferred a charge of assault with intent to do grievous bodily harm instead of attempted murder. Ploos van Amstel J found that there was ‘no merit in this complaint’ (at [18]): first, the information in the docket clearly supported a charge of attempted murder; second, in his discussion with Ms D, the acting DPP not only agreed with her that on the available facts the appropriate charge was attempted murder, but also authorised her to proceed with the prosecution; third, the regional court had, indeed, convicted the applicant of attempted murder; fourth, this was not a case where a disgruntled individual had laid a frivolous charge against a magistrate and the DPP had not authorised the prosecution; fifth, it could safely be accepted that the DPP who authorised the prosecution orally, would also have done so in writing.

It was accordingly concluded that the applicant was not prejudiced by the irregularity and that the irregularity caused no failure of justice. The application was dismissed (at [19]).

The decision in Masinga (supra) must of course be distinguished from those situations where legislation (as opposed to policy directives) requires that no prosecution for a specific offence may be instituted without the DPP’s written authorisation. In S v Molefe 2012 (2) SACR 574 (GNP) a woman was prosecuted for contravention of s 113(1) of the General Law Amendment Act 46 of 1935, that is, unlawfully disposing of the body of a newly born child with intent to conceal its birth. On review it was held that in the absence of the relevant DPP’s written authorisation as required by s 113(3) for purposes of instituting a prosecution for contravention of s 113(1), the accused could not have been prosecuted because the written authorisation was a mandatory prerequisite for the prosecution (at [6]) and verbal authorisation could not remedy the situation (at [5]–[7]). A statutory requirement that a written authorisation (whether it be by the DPP or NDPP) is necessary must be understood in the context of the following observations in Booysen v Acting National Director of Public Prosecutions & others 2014 (2) SACR 556 (KZD) by Gorven J (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. See also Chapter 1 in Commentary, sv Prosecuting authority: Validity of authorisation issued in terms of s 2(4) of the Prevention of Organised Crime Act 121 of 1998 (POCA). For various examples of statutes requiring the written authorisation of the NDPP, as opposed to a DPP, see Chapter 1 in Commentary, sv Written authorisation of NDPP required for prosecution of certain offences.

s 73: Legal representation and legal professional ethics

S v DD 2015 (1) SACR 165 (NCK)

In S v DD (supra) the accused, a 17-year-old minor, was convicted of the murder and rape of his sister M, the murder of his father D, the murder of his mother C and, furthermore, defeating the ends of justice. He had pleaded not guilty and offered no explanation of plea. He testified in his own defence, claiming that at the time when his parents and sister were shot in the residence on the farm, he was in a barn some 34 metres away.

In written submissions on the merits, the defence stated that they had been instructed to make two specific submissions pertaining to the rape of the sister, M.

The first submission was that if there had been an incestuous relationship between the accused and M, his 14-year-old sister, there would not have been any motive for M to report the prior sexual activity to their parents, D and C. At [54.1] Kgomo JP described this submission as ‘baseless’: the prosecution never suggested that there was an ‘incestuous love relationship’ between the accused and M; and, more fundamentally, it was never part of the case for the defence that such a relationship existed and that consensual sexual intercourse took place. ‘Besides, said Kgomo JP, ‘nothing approximating this waffle was put to any of the witnesses, nor did the minor broach the subject in his testimony’. In this regard reference was made to President of the Republic of South Africa & others v South African Rugby Football Union & others 2000 (1) SA 1 CC at [61]–[63].

The second submission—which seems to be inconsistent with the first—was that the evidence failed to establish that the only reasonable inference was that the accused caused the injury to the private parts of M, his sister. The written submission, as noted and translated by Kgomo JP at [54.2], was that the injury ‘could as well, for example, also have been caused by the deceased Mr D [the minor’s father]. It could also have been him [Mr D] who was responsible for the pre-existing rupture of Ms M’s hymen.’ 

After having referred to case law and legal professional ethics governing the duties and responsibilities of a legal representative, Kgomo JP concluded as follows (at [55]): ‘To impugn the character, dignity and the memory of the deceased Mr D at such a late stage and so unfairly is inappropriate, as it is hurtful to his family and friends.

One of the ethical rules referred to by Kgomo JP at [55] is contained in para 3.4 of the Code of Conduct: Uniform Rules of Professional Ethics of the General Bar Council of South Africa. In terms of this paragraph counsel may not attribute ‘wantonly or recklessly’ to another person, the crime with which his client is charged, ‘unless the facts or circumstances given in the evidence, or rational inferences drawn from them, raise at the least a not unreasonable suspicion that the crime may have been committed by the person to whom the guilt is so imputed’. It seems fairly clear that in S v DD the second submission referred to above fell short of what is ethically required and permitted. On the facts as reported in S v DD, it would appear that during the course of the trial nothing was advanced in cross-examination and evidence to provide some reasonable suspicion that D, the father, was responsible for the injury to the private parts of M, his daughter, who was the victim in the rape charge against the accused. The mere fact that counsel was instructed to make the submission can hardly mean that there was ‘a not unreasonable suspicion’ as stated in paragraph 3.4 of the ethical rule as cited above. At any rate, counsel is and should remain in control of the defence case. See S v DD (supra) at [54.2] where reference was made to R v Matonsi 1958 (2) SA 450 (A) 456A–D.

Of course, the remarks made by Kgomo JP in S v DD do not in any way whatsoever place improper limitations upon counsel’s duty to defend and represent his client in a fearless manner. It remains the duty of counsel to ‘fearlessly argue his client’s case’ (per Louw AJ in S v Mofokeng 2004 (1) SACR 349 (W) 357g–h). However, fearless representation remains subject to legal professional ethics and the client cannot be permitted to require his lawyer to act contrary to the latter’s professional rules.

There will always be some tension between the lawyer’s duty fearlessly to represent his client and the ethical rule that a lawyer may only in limited circumstances attribute to another person the crime with which his client is charged. Obviously, the ethical rule was never meant to defeat counsel’s duty to defend his client, and ultimately the right to a fair trial should control the situation on the basis of all the facts, circumstances and probabilities of the case. Suni ‘Who Stole the Cookie from the Cookie Jar? The Law and Ethics of Shifting Blame in Criminal Cases’ 2000 (68) Fordham Law Review 1643 at 1659 explains that 

while there may be cases where blame-shifting behaviour by criminal defence lawyers simply goes too far, in most cases the adversary system of criminal justice permits, and in some cases even requires, attorneys to engage in such behaviour on behalf of their clients. While there are those who believe a prohibition on such conduct may be appropriate where a guilty client is involved and the attorney has no basis to believe the truth of the blame-shifting allegations, as one moves along a sliding scale of increasing doubt regarding client guilt and increasing reason to believe the blame-shifting evidence, the justifications, both institutional and situational, increase to where, at some point, they likely render what would be permissible questioning and argument mandatory based on the attorney’s duty of zeal.

s 106(4): The right of an accused to a verdict on a charge to which he has pleaded  

S v Fongoqa & others (unreported, WCC case no A317/14, 13 May 2015)

In Fongoqa & others (supra) Henney J addressed the following issue: what is a court of appeal required to do where the appellants were asked to plead to two charges, but the trial court delivered a verdict on only one of those charges?

In Fongoqa the appellants had in fact pleaded not guilty to both charges. They were convicted on one charge. The absence of a verdict in respect of the other charge inevitably meant that the trial court had failed to accommodate the right of the appellants to a verdict as set out in s 106(4) of the Criminal Procedure Act 51 of 1977: ‘An accused who pleads to a charge, other than a plea that the court has no jurisdiction to try the offence . . . shall, save as is otherwise expressly provided by this Act or any other law, be entitled to demand that he be acquitted or be convicted’. Section 108 also determines that by a plea of not guilty, ‘an accused . . . shall . . . be deemed to demand that the issues raised by the plea be tried’.

After having referred to s 106(4) and S v Sithole & others 1999 (1) SACR 227 (T), Henney J took the view that the absence of a verdict in respect of a charge to which an appellant has pleaded, need not necessarily result in an acquittal on appeal. At [67] it was said: ‘[T]he correct approach on appeal would be that the court determines whether an accused’s right to a fair trial was infringed where he . . . pleaded on a charge, but no verdict was delivered on that charge.’

According to Henney J an acquittal on appeal would not be in the interests of justice if the trial court had for some reason or another failed to deliver a verdict in response to an accused’s plea of guilty as adequately and properly supported by statements or admissions in terms of s 112(1)(b), 112(2) or 220 (at [68]). In these circumstances the fact that an accused is entitled to a verdict does not mean that he is in the absence of a verdict entitled to an acquittal.

However, in Fongoqa the appellants had pleaded not guilty to a second charge put to them; they had disputed the State’s allegations; the State had not lodged a cross-appeal on the basis that on the available evidence the appellants should have been convicted on the charge in respect of which no verdict was given; and, finally, there were ‘no good reasons emanating from the record why their plea of not guilty should not be upheld’ (at [69]). The appellants were accordingly acquitted on the charge in respect of which the court below had not pronounced a verdict (at [70] and [77]).

ss 155 and 156: Separation of trials: Accused not all facing same charges

S v Maringa & another [2015] ZASCA 28 (unreported, SCA case no 20116/2014, 23 March 2015)

The decision in S v Maringa & another (unreported, GNP case no A127/2013, 17 September 2013), discussed in Commentary in the notes to s 156, was upheld by the Supreme Court of Appeal. Seven accused had faced 399 charges in a regional court, including fraud, forgery, uttering and corruption. The first appellant (accused no 1) and the second appellant (accused no 4) objected to being tried together with the other accused. The first appellant had not been charged with any of the corruption counts and the second appellant had been charged only with some (but not all) of the counts of fraud. It was argued that it was contrary to ss 155 and 156 of the Criminal Procedure Act to charge them together with all the other co-accused where they did not face the same charges. This argument failed in both the regional and the High Court, and was rejected, too, by the Supreme Court of Appeal.

Section 156 provides that ‘[a]ny number of persons charged in respect of separate offences committed at the same place and at the same time or at about the same time, may be charged and tried together in respect of such offences if the prosecutor informs the court that evidence admissible at the trial of one of such persons will, in his opinion, also be admissible as evidence at the trial of any other such person or such persons.’

Schoeman AJA (with whom Navsa ADP, Leach JA, Willis JA and Meyer AJA agreed) found (at [14]) that the purpose of ss 155 and 156 was ‘to avoid a multiplicity of trials where there are a number of accused’ where ‘essentially the same evidence on behalf of the prosecution is led on charges faced by all the accused’, and thereby ‘to avoid prejudice to both the accused and the prosecution’. The trial court, he said, exercised a discretion to decide whether to allow a trial to proceed or to order a separation of trials. The way this discretion had to be exercised was set out in S v Ntuli & others 1978 (2) SA 69 (A) at 73: the court had to weigh the likelihood of prejudice to the accused applicant against the likelihood of prejudice to the other accused or the State and decide if it was in the interests of justice to order a separation of trials. The court had to determine the weight of each relevant factor in deciding what possible injustice could be caused to each of the parties, including the State.

There was, said Schoeman AJA, no authority that each accused had to be charged with every offence in the indictment (at [17]). In S v Naidoo 2009 (2) SACR 674 (GSJ) the accused’s claim to a misjoinder was rejected because the ‘evidence relied on by the prosecution relate[d] to the ongoing, continuing or repeated participation of each of the accused . . . in the illegal rackets in which they [were] all participants’. Despite the fact that the nature of the part played by each differed, the evidence remained the same to prove the conspiracy between them or the individual counts on which the applicant accused had been charged in the alternative.

In Maringa it was clear from the charge sheet that the alleged offences had been committed at about the same time and place and had, furthermore, been committed in furtherance of a common purpose. What was alleged was that there was a scheme designed fraudulently to sell properties belonging to the Johannesburg Metropolitan Municipality and to transfer them to buyers in order for the accused to collect the proceeds of the sales. It was necessary for officials in the South African Revenue Service and the Deeds Office to co-operate to allow the scheme to succeed, since the properties could not otherwise be transferred or registered. These officials were bribed, so the corruption charges were ‘part and parcel of the overall design of the scheme’. There was, said the court (at [19]), ‘a whole mosaic of evidence that [would] be necessary to prove the scheme and the participation of the various accused in its different facets'.

The only prejudice to the appellants was that they would have to sit through a trial while evidence was presented which did not relate to the charges they faced. In the court’s view, the prejudice to them was ‘exaggerated’ in that the corruption and other charges were but a part of the scheme that would be proved. If, however, separation were ordered, the State would suffer considerable prejudice. There would have to be three separate trials, since the two appellants could not then be tried together, where the same witnesses would have to testify about the same facts. This was ‘inimical to the interests of the State and against the principle that there should not be a multiplicity of trials relating to essentially the same facts and body of evidence’ (at [20]). Against this, the prejudice asserted by the appellants was, ‘in the greater scheme of things, minimal’.

s 212(4) and (10): Proving accuracy of a gas chromatograph by means of a certificate instead of affidavit; status of certificate as prima facie proof

S v Eke (unreported, ECG case no CA&R 163/14, 26 March 2015)

In an appeal against the appellant’s conviction of driving with impermissible blood alcohol levels, two issues arose: (i) the accuracy of the gas chromatograph used to measure the blood sample and, in particular, whether the State could prove the accuracy of that device by way of a certificate issued in terms of s 212(4) of the Criminal Procedure Act or whether an affidavit in terms of s 212(10) was required; and (ii) the status of a certificate as ‘prima facie proof’ in terms of s 212(4).

In respect of the first issue the court (per Plasket J, with Makaula and Lowe JJ concurring) considered two conflicting decisions: S v Ross 2013 (1) SACR 77 (WCC) and S v Van der Sandt 1997 (2) SACR 116 (W), both discussed in Commentary in the notes to s 212(4) and (10). It preferred the latter decision, which was to the effect that a certificate could be used for this purpose. In that case Van Dijkhorst J said that s 212(4) did not contain any indication that the requirements of proof of trustworthiness and correctness had been jettisoned, and there was no reason to do so. The purpose of the section was merely to obviate the need for viva voce evidence in certain cases, not to introduce a new type of evidence such as ‘expert factual evidence of a result without explanation or clarification’. The expert was required to explain the operation of the instrument of measurement and say why it was trustworthy, unless there was a high degree of likelihood that it was accurate or had been tested (such as in the case of a yardstick): see S v Mthimkulu 1975 (4) SA 759 (A), discussed in the notes to s 212(10) in Commentary. If the expert did not set out why the instrument was reliable, there would be no fair trial.

As to whether the certificate should deal with the calibration of the instrument, Van Dijkhorst J in Van der Sandt suggested that ‘a court should be practical’: if it could take judicial notice of hearsay evidence about the assized scales, as explained in Mthimkulu, there could be no serious objection to taking judicial notice of the fact that there was ‘a high likelihood that scientists in designated government laboratories when calibrating their instruments [would] do so against correct standards’ for the purpose of s 212(4).

The court in Eke took the view that the decision in Ross, which held that a certificate could not be so used, would, if followed, ‘render s 212(4) certificates both meaningless and redundant’, since the certificate would, in every case, contain only the result and would have to be supplemented by an affidavit dealing with the calibration and accuracy of the instrument (at [20]). The legislature, said Plasket J, could not have intended such a result. The certificate procedure, on the contrary, while not intending to reduce the burden of proof resting on the State, was designed to facilitate the procurement of certain evidence of an expert nature to ‘avoid undue wastage of mainly official manpower by court attendances for the purpose of frequently undisputed evidence on matters nearly always uncontrovertible’ (at [21]). The court in Ross did not consider the decision in Van der Sandt. In any event, s 212(10) applied only to situations where the Minister had prescribed conditions and requirements in respect of the measuring instrument concerned.

The second issue concerned the status of the certificate as ‘prima facie proof’ in s 212(4). The certificate in this case had been handed in ‘by agreement’. The defence had closed its case without leading any evidence, but the appellant had, in her plea explanation, placed in issue the accuracy of the result of the measurement. It was argued that a plea explanation was ‘evidence’, and that the State had, as a result, to do more than merely hand in the certificate. This argument did not succeed. Although a plea explanation was regarded as ‘evidential material’ in the sense that it was an unsworn statement made by an accused to disclose what was in issue between the accused and the State, it did not constitute ‘evidence’. It was not sufficient to affect the prima facie value of the certificate and only a challenge by way of evidence could do so (see S v Britz 1994 (2) SACR 687 (W) and R v Chizah 1960 (1) SA 435 (A) at 442. In the absence of such evidence, the prima facie proof became conclusive proof and the State’s onus was discharged (see Ex parte the Minister of Justice: In re R v Jacobson & Levy 1931 AD 466 at 478–9, S v Veldthuizen 1982 (3) SA 413(A) at 416G–H).

What could the appellant in Eke have done to rebut the correctness of the certificate? Plasket J said there were three possible options: first, she could have applied to the trial court to exercise its discretion in terms of s 212(12) to have the analyst subpoenaed to give oral evidence; second, she could have subpoenaed the analyst herself to testify; and third, she could have testified herself or called witnesses if she had a factual basis to cast doubt on the accuracy of the result—such as showing that she had consumed no alcohol at the relevant time.

s 225: Improperly obtained evidence: Three Supreme Court of Appeal cases

S v Jwara & others [2015] ZASCA 33 (unreported, SCA case no 916/13, 25 March 2015)

Magwaza v S [2015] 2 All SA 280 (SCA)

S v Gcam-Gcam [2015] ZASCA 42 (unreported, SCA case no 1034/13, 25 March 2015)

In three separate decisions, the Supreme Court of Appeal considered various aspects of the principles relating to improperly obtained evidence, which are discussed at length in Commentary in the notes to s 225.

In S v Jwara & others the issue before the court was the admissibility of intercepted cellphone conversations involving the three appellants, who were all members of the South African Police Service and who had been suspected of being involved in a number of offences. Direction had been granted under the Interception and Monitoring Prohibition Act 127 of 1992 to monitor cellphone calls to and from the cellphones of all three appellants, who were, on the strength of that evidence, convicted by the trial court of several offences, including contraventions of s 2(1)(d) and (f) of POCA, dealing in drugs, defeating or obstructing the course of justice, theft, fraud and attempted theft. The trial court had admitted the evidence of the intercepted calls on the ground that this evidence fell within the borders of the Interception Act but that, even if it did not, the evidence would nevertheless be admissible in terms of s 35(5) of the Constitution.

The outcome of the appeal hinged largely on the question whether the evidence had been correctly admitted. Gorven AJA (with whom Brand, Ponnan, Willis JJA and Dambuza AJA agreed) noted that the Interception Act had not been attacked for being unconstitutional, even though it did limit the right to privacy, since there was ‘an adequate and objective basis to justify the infringement’ of that important right. The Act, in his view, struck ‘a balance between the need for search and seizure powers and the right to privacy of individuals’ (at [11]).

The appellants launched three attacks on the admissibility of the evidence. The first was that the application for the direction in the case of the first appellant had not complied in all respects with the elaborate procedure set out in the Interception Act. This was rejected on the ground that the infractions were only minor, were technical in nature and, unlike the objections in S v Pillay & others 2004 (2) SACR 419 (SCA) (discussed in Commentary), ‘did not go to the foundation of the application’ (at [13]). The second was that the Interception Act did not provide for the interception of communications from cellphones, which were not yet in use in South Africa when the Act was promulgated. This objection, said the court, had been correctly rejected in S v Cwele & another 2011 (1) SACR 409 (KZP).

The third attack was on the trial court’s finding that the evidence would, even without the Interception Act, have been admissible since it was not excluded under s 35(5) of the Constitution. Gorven AJA considered, however, that the trial court had correctly exercised its discretion not to exclude the evidence in terms of s 35(5) because: (1) the deficiencies were of a purely technical nature; (2) there was, unlike what had happened in Pillay’s case, nothing misleading said in the application for the direction; (3) the procedure in the Interception Act had been followed as closely as possible; and (4) monitoring the conversations was the ‘only means to investigate’: since the suspects were all members of the police force, and because of the ‘endemic corruption therein’, no other investigative tool could have been used without jeopardising the investigation. To have excluded the evidence under s 35(5), the court concluded, would have led to a failure of justice.  

The second of the cases, Magwaza v S, concerned the admissibility of a pointing out and a confession, on the strength of which the appellant had been convicted by the trial court of murder and robbery with aggravating circumstances, a decision upheld by a full bench of the High Court. Both the trial court and the full court focussed solely on the voluntariness of the appellant’s conduct, and, as Ponnan JA (with whom Maya, Mhlantla, Zondi JJA and Meyer AJA agreed) observed, ‘[n]either touched, even tangentially, on the Constitution’s exclusionary provision—s 35(5)’(at [21]). There was certainly reason to do so since, as Ponnan JA remarked (at [17]), even if it were to be accepted that the cumulative effect of the evidence given by the investigating officers—which was very unclear as to what had been conveyed to the appellant—was that there had been ‘a warning of sorts’, that warning was ‘woefully inadequate’. For, while there was some reference in their evidence to the rights to silence and legal representation, there was ‘no indication that the appellant was warned of the consequences of not remaining silent (the logical corollary of the right to silence) or of his entitlement to the services of a legal representative at State expense’. The suggestion that this deficiency may have been cured by the detailed warning given to the appellant subsequently by the police captain who recorded his statement was without merit in the court’s view, since the appellant had already confessed to the robbery by that time. 

It was important, said the court at [17], ‘to appreciate that a constitutional right is not to be regarded as satisfied simply by some incantation which a detainee may not understand’. The ‘purpose of making a suspect aware of his rights is so that he may make a decision whether to exercise them and plainly he cannot do that if he does not understand what those rights are’. It had to follow, therefore, that ‘the failure to properly inform a detainee of his constitutional rights renders them illusory’, since ‘[w]hat must govern is the substance of what the suspect can reasonably be supposed to have understood, rather than the formalism of the precise words used’ (see R v Evans (1991) 4 CR (4th) at [144], [160] and [162]). 

Ponnan JA cited what Froneman J had said in S v Melani & others 1996 (1) SACR 335 (E) at 347e–h—that the right to consult with a legal practitioner and especially the right to be informed of this right is closely connected to the presumption of innocence, the right of silence and the proscription of compelled admissions and confessions, which were all recognised as ‘basic principles of our law’. Froneman J stressed that the failure to recognise the importance of informing an accused of his right to consult with a legal adviser during the pre-trial stage had the effect of depriving persons, especially the uneducated, the unsophisticated and the poor, of the protection of their right to remain silent and not to incriminate themselves. This, he said, offended ‘not only the concept of substantive fairness which now informs the right to a fair trial in this country but also the right to equality before the law’. These rights, said Ponnan JA, existed ‘from the inception of the criminal trial, that is from arrest, until its culmination (up to and during the trial itself)’. 

The failure to warn the appellant properly of his constitutional rights caused him a ‘high degree of prejudice . . . because of the close causal connection between the violation and the conscriptive evidence’ (at [18]), since ‘plainly, the rights infringement resulted in the creation of evidence which otherwise would not have existed’. The court examined the position in Canada under the Charter of Rights (which is discussed at length in Commentary in the notes to s 225), and applied what was said in R v Ross (1989) 37 CRR 369 at 379: that ‘the use of any evidence that could not have been obtained but for the participation of the accused in the construction of the evidence for the purpose of the trial would tend to render the trial process unfair’. 

Ponnan JA endorsed the view expressed by De Villiers JP in R v Ndoyana & another 1958 (2) SA 562 (E) at 563 that the circumstances leading up to an accused’s appearance before a magistrate or justice of the peace to make a confession are ‘not less important than the circumstances surrounding the actual making of the confession’. So, too, the warning issued by Harcourt J in S v Majozi & others 1964 (1) SA 68 (N) at 71E–G that one ‘must not permit the proceedings before the magistrate or justice to draw a veil between the preceding events and the completed confession’. The trial court and full bench, in his view, had failed to appreciate the observation of SE van der Merwe (Schwikkard & van der Merwe Principles of Evidence 3 ed (2009) at para 12.9.7)  that ‘[i]f an accused was not prior to custodial police questioning informed by the police of his constitutional right to silence, the court might in the exercise of its discretion conclude that even though the accused had responded voluntarily, all admissions made by the accused to the police should be excluded in order to secure a fair trial’. 

All these considerations prompted the court to conclude (at [21]) that ‘those factors which justify exclusion materially outweigh[ed] those which call for admission’. After giving the matter ‘anxious consideration and not without some hesitation’, it came to the decision that the evidence should have been excluded. Although accepting that the ‘public reaction to the exclusion of such evidence [was] likely to be one of outrage’, it was mindful of the commitment made in S v Tandwa & others 2008 (1) SACR 613 (SCA) at [121] to winning the ‘struggle for a just order’, which could be achieved only ‘through means that have moral authority’, an authority which would be lost ‘if we condone coercion and violence and other corrupt means in sustaining order’.

One aspect of the judgment calls for comment. The court relied heavily, in its approach to this evidence, on the position of the Canadian courts regarding conscriptive evidence. It pointed out, correctly, that Canadian jurisprudence has rejected a strict distinction between real and testimonial evidence, holding that the distinction forged in R v Collins [1987] 1 SCR 265 was unfounded (see R v Burlingham (1995) 28 CRR (2d) 244), and that more recent decisions such as R v Ross (1989) 37 CRR 369 at 379 ‘emphasised that the admissibility of evidence under s 24(2) depended ultimately not on its nature as real or testimonial, but on whether or not it would only have been found with the compelled assistance of the accused’ (at [13]). That is true. But there is, curiously, no mention of the more recent decision of the Supreme Court in R v Grant 2009 SCC 32, [2009] 2 SCR 353, which marks a further point of departure for  Canadian jurisprudence. In Grant the court set itself against previous decisions which had been read as creating ‘an all-but-automatic exclusionary rule for non-discoverable conscriptive evidence . . . and increasing its importance to the ultimate decision on admissibility’ (at [64]). Such an approach, said the majority of the court, seemed to go against the requirement in s 24(2) of the Charter that the court had to consider ‘all the circumstances’. The underlying assumption that the use of conscriptive evidence always or almost always rendered the trial unfair was, in its view, also open to challenge. The ‘near-automatic presumption that the admission of a broad class of evidence would render the trial unfair, regardless of the circumstances in which it was obtained’ (at [65]) was, said the majority, difficult to reconcile with ‘trial fairness as a multi-faceted conceptual concept’. Instead, the majority set out  ‘three avenues of inquiry’ for determining what might bring the administration of justice into disrepute. These are listed and discussed in Commentary in the notes to s 225 sv The Canadian position on illegally or improperly obtained evidence. It is probable, however, that an evaluation of each of these avenues of inquiry would have left the decision of the Supreme Court of Appeal intact.

The third of the cases on improperly obtained evidence, S v Gcam-Gcam, contains salutary reminders relating to the incidence of the onus of proof and helpful advice on how to determine whether it has been discharged. The trial judge in that case had, said Cachalia JA (with whom Shongwe JA and Gorven AJA agreed), ‘misdirected himself by approaching the evidence of the appellant on the basis that he (and his co-accused) needed to “put up credible versions” to refute the “overwhelmingly strong and convincing evidence” of the police regarding the admissibility of confessions’. All that was required of the appellant, said the court (at [48]), ‘was to present a version that was reasonably possibly true, even if it contained demonstrable falsehoods’.

The court set out the following warning and reservations about confessions made by suspects in custody (at [49]):

When confronted with confessions made by suspects to police officers whilst in custody—even when those officers are said to be performing their duties independently of the investigating team—courts must be especially vigilant. For such people are subject to the authority of the police, are vulnerable to the abuse of such authority and are often not able to exercise their constitutional rights before implicating themselves in crimes. Experience of courts with police investigations of serious crimes has shown that police officers are sometimes known to succumb to the temptation to extract confessions from suspects through physical violence or threats of violence rather than engage in the painstaking task of thoroughly investigating a case. This is why the law provides safeguards against compelling an accused to make admissions and confessions that can be used against him in a trial.

Further (at [50]):

In addition, courts must be sceptical when the State seeks to use a confession against an accused where he repudiates it at the first opportunity he is given. Because ordinary human experience shows that it is counter-intuitive for a person facing serious charges to voluntarily be conscripted against himself. Often it is said that the accused confessed because he was overcome with remorse and penitence; “a desire which vanishes as soon as he appears in a court of justice”. That is sometimes true, but is usually not.

In the present case, said Cachalia JA (at [51]), not even that explanation was advanced for why the appellant had confessed:

It was simply said that the appellant was asked during his questioning whether he wished to make a statement, and he agreed. The statement was taken from him and reduced to writing. And when he was asked whether and why he wished to make a second statement, (which the State used against him in the trial) having already made one, the answer appearing on the police record of what was said was that he wanted it to be “written down”. This nonsensical answer should have caused the court to approach the matter with heightened scepticism.

There were, said Cachalia JA, several reasons why the appellant’s complaint that his confession was improperly obtained from him rang true. First, three of the accused at the trial said that they had been severely assaulted before making confessions. Second, all three accused who contested their confessions in the trial-within-a-trial had said that they were not warned of their constitutional rights. All three, said the police, upon being asked whether they wanted legal representation, responded in exactly the same manner, saying that they would only require an attorney when they appeared in court. This evidence, said Cachalia JA, seemed ‘contrived’. And, third, it was unlikely that the appellant would have asked for his statement to be written down before one official after already having had a statement written down by another.

His version was, thus, reasonably possibly true, so that the onus resting on the State had not been discharged.

s 235 and s 60(11B)(c): Status at trial of evidence given by an accused at a bail application

Machaba & another v S [2015] 2 All SA 552 (SCA)

The only evidence against the second appellant in his trial for murder and robbery with aggravating circumstances was that he was the owner of the firearm with which the deceased had been shot and killed and that this firearm was in his possession in 2004, two years after the deceased had been killed. However, during the cross-examination of the appellant in his bail application, he had testified that he had not been in possession of the firearm in 2002, as he had lost it, and that his younger brother had taken it. He started by saying that he did not report the loss; then said that he did report the loss, but not in 2002; and ended by saying, again, that he did not report the loss at all. This question was put by the court: what was the importance of his evidence in the bail application, and what weight had to be given to it?

Schoeman AJA (with whom Mpati P and Majiedt JA agreed) referred to the decision in DPP, Transvaal v Viljoen 2005 (1) SACR 505 (SCA) at [33], where Streicher JA held that it did not follow from the fact that the record of bail proceedings formed part of the record of the trial that evidence adduced during the bail proceedings must be treated as if that evidence had been adduced and received at the trial. ‘The record of the bail proceedings’, Streicher JA pointed out, ‘remains what it is, namely a record of what transpired during the bail application’.

Handing in the bail application in terms of s 60(11B)(c) was, said Schoeman AJA in Machaba at [27], ‘a shortcut to achieving the same object as provided for in s 235 of the CPA’, and this ‘ha[d] the effect that the record [was] prima facie proof that any matter recorded on the record was properly recorded’. He stressed, however, what was asserted in Commentary in the notes to s 235: that the record did ‘not, however, constitute prima facie proof of any fact it contain[ed]’ (emphasis added).

The court then considered a further question: can the ‘Valachia principle’ be applied to the record of bail proceedings? In R v Valachia & another 1945 AD 826 at 835, it was held that ‘when the State proves that an accused made an admission in a statement, the whole statement must be assessed including the exculpatory portions’ (per Schoeman AJA at [29]). The court then had to give such portions such weight as, in its opinion, they deserved. The fact that the statement was not made under oath or subject to cross-examination detracted very much from its weight but, said the court in Valachia, the accused was still entitled to have the exculpatory portions taken into consideration, ‘to be accepted or rejected according to the Court’s view of their cogency’.

Did this principle apply to the record of bail proceedings? Schoeman AJA referred to the decision in S v Cloete 1994 (1) SACR 420 (A) at 428a–g, where EM Grosskopf JA held that it did apply to a plea explanation made in terms of s 115 of the Criminal Procedure Act. He pointed out that there was a ‘practical difference’ between an extra-curial statement and a plea explanation, in that the State could choose not to introduce the former into evidence if it was of the view that it would weaken its case whereas, in the case of plea explanation, it was the accused who decided what to say, with the result that he ‘may more readily place self-serving exculpatory material before the court’. However, Grosskopf JA countered, if an accused attempted to abuse this procedure in this way, the court ‘should ensure that such an attempt does not succeed by refusing to attach any value to statements which are purely self-serving and, generally, by determining what weight to accord to the statement as a whole and to its separate parts’.

Section 60(11B)(c), said Schoeman AJA (at [31]), was ‘in the same vein’. It was part of the record of trial subject to the qualification that it was essential that the accused had to be warned of the consequences of testifying in the bail application prior to its acceptance as part of the record. As with s 115, it was a ‘procedure whereby material can be placed before court’, and the Valachia principle was applicable in this context as well.

The weight to be placed on the exculpatory material had, said the court, to be determined in the light of the fact that the accused had been subjected, in the bail proceedings, to only perfunctory cross-examination, since the prosecution had not, at that stage, been aiming at a conviction. It could, accordingly, not be equated to testimony given during the trial.

 (ii) Sentencing

Sentencing and multiple rapes: Section 51(1)—as read with Part 1 of Schedule 2—of the General Law Amendment Act 105 of 1997

Cock v S; Manuel v S [2015] 2 All SA 178 (ECG) and S v Motshesane (unreported, FB case no A244/2014, 12 March 2015).

In terms of the above provisions an accused convicted of rape must—in the absence of substantial and compelling circumstances—be sentenced to life imprisonment ‘where the victim was raped more than once whether by the accused or by any co-perpetrator or accomplice’, or where the victim was raped ‘by more than one person where such persons acted in the execution or furtherance of a common purpose or conspiracy’. These provisions are for the sake of brevity hereafter referred to as the ‘multiple rape provisions’.

Cock v S; Manuel v S [2015] 2 All SA 178 (ECG) involved two separate appeals dealt with in one judgment by Pickering J (Plasket and Smith JJ concurring) in which a dictum in S v Mahlase 2013 JDR 2714 (SCA) was reluctantly followed and also severely criticised.

In Mahlase (supra) at [9] the Supreme Court of Appeal took the view that an accused convicted of rape in the multiple rape circumstances as envisaged in Part 2 could not receive the mandatory minimum sentence of life imprisonment if at his trial as sole accused his co-perpetrators or accomplices had as yet not been apprehended and convicted. In Cock; Manuel (supra) Pickering J found this approach illogical and artificial because it disregards the requirement that a court must sentence an accused on the basis of the facts found proved (at [26]):

The Mahlase dictum . . . gives rise, with respect, to the illogical situation that a trial court, having found beyond reasonable doubt that the complainant was raped more than once by two men and having convicted the accused accordingly, must, for purposes of the Act, disregard that finding and proceed to sentence the accused on the basis that it was not in fact proven that she was raped more than once; that the provisions of the Act relating to the imposition of the prescribed minimum sentence of life imprisonment are therefore not applicable; and that the minimum sentence applicable in terms of the Act is one of only ten years imprisonment.

Pickering J also noted the arbitrary nature of an approach that an accused who happened to be the first of a gang of rapists to have been arrested and convicted would, for purposes of the multiple rape provisions, be immune from the mandatory minimum sentence. It makes no sense to have a situation where the credible and cogent evidence of the victim that she was raped by two or more men, one of whom she identified as the accused, ‘should be disregarded, not only to the prejudice of the victim and of the State, but also, by way of contrast, to the benefit of the accused . . .’ (at [27]).

It must also be pointed out that the interpretation of the multiple rape provisions in Mahlase (supra) is also hardly acceptable when applied to the situation where A, the accused, is convicted upon his plea of guilty as supported by his statement under s 112(2) confirming his guilt and admitting how he had raped the victim immediately after his absent and on the run (or since deceased) co-perpetrator B had done so. To ignore the multiple rape provisions in these circumstances would be rather absurd: surely, the uncontested factual details which were accepted for purposes of determining the criminal liability of the individual accused, A, remain unaffected by the absence of B or the fact that B has not been (or never will be) convicted of the rape described by A.

Another and completely different issue concerning the multiple rape provisions arose in S v Motshesane (unreported, FB case no A244/2014, 12 March 2015). In this appeal the court below had convicted the appellant on a single charge of rape. However, for purposes of sentencing, the court below took into account that on the night in question the accused had raped the victim more than once. In accordance with the multiple rape provisions and in the absence of substantial and compelling circumstances, the court below had sentenced the appellant to life imprisonment. 

On appeal it was submitted that there was no evidence that the appellant had raped the victim more than once. The trial record showed that when asked by the prosecutor how many times the accused had intercourse with her on the night concerned, the victim said: ‘Dit was die hele aand gewees . . . hy het my seergemaak . . . ons het nooit geslaap daardie aand nie. Hy was besig met my die hele aand. [It was the whole evening . . . he hurt me . . . we never slept that night. He was busy with me the whole evening]’.

At [5] Reinders AJ (Mocumie J concurring) concluded that on the evidence as quoted above, the complainant never testified that the accused had raped her more than once. Nor, held Reinders AJ, could it by way of inferential reasoning be concluded that more than one rape had occurred. It was also pointed out that if the prosecutor wanted to prove more than one rape, he should have dealt with the matter in more detail in the course of his examination-in-chief.

It was accordingly held that the trial court had erred in relying on the multiple rape provisions (at [7]).

For further case law as regards the question whether an accused has raped his victim more than once, see the discussion in Commentary under the repealed s 277, sv ‘Rape’ and substantial and compelling circumstances.

s 298: Correction of sentence by the sentencing court

S v Hendricks (unreported, WCC case no A420/14, 18 February 2015)

In Hendricks (supra) the appellant’s appeal against his various sentences on five different counts was upheld in part; and the court of appeal specifically recorded that the effective sentence was to be 25 years’ imprisonment. Nine days later the Department of Correctional Services alerted the registrar of the court to an inconsistency in the sentence, namely that ‘by ordering two years of the sentence on count 3 to run concurrently with the remaining sentence, the effective period of imprisonment was in fact 28 years and not 25 years’ (at [2]).

In responding to this inconsistency, the court concerned stated that it was ‘at all material times’ the intention of all three judges that the appellant’s effective sentence should be reduced to 25 years and that the inconsistency was due to ‘a typographical error which eluded all 3 members of the court’ (at [3]).

Section 298 of the Criminal Procedure Act 51 of 1977 provides for the correction of a sentence: ‘When by mistake a wrong sentence is passed, the court may, before or immediately after it is recorded, amend the sentence’. However, in Hendricks, s 298 could not be applied because the time period of more than a week that had elapsed since the incorrect effective sentence was imposed, had rendered the ‘corrective action’ contemplated in s 298 ‘not sufficiently immediate’ (at [4]). For case law dealing with the requirement ‘before or immediately’ as stated in s 298, see the discussion of s 298 in Commentary, sv Prerequisites. The court in Hendricks was absolutely correct in not relying on s 298 in order to correct its mistake. However, at [4] the court held, with reference to S v Wells 1990 (1) SA 816 (A) at 820C–D, that it was ‘entitled, under the common law, to correct an order or sentence provided the tenor of the earlier order is preserved.’ Gamble J (Blignault and Mantame JJ concurring) took the view that the tenor of the earlier order was that the appellant’s sentence should be reduced from an effective sentence of 44 years to one of 25 years (at [5]). The court accordingly corrected the earlier typographical error to ensure that the effective sentence would be 25 years’ imprisonment (at [6]). It is submitted that should a lower court find itself in a position where it cannot correct its own sentence on the basis of s 298, the appropriate route would be to submit the matter for special review in terms of s 304(4) of Act 51 of 1977. See generally S v Lottering 2009 (2) SACR 560 (ECG).

(iii) Forfeiture and confiscation

s 48 of Prevention of Organised Crime Act (POCA) of 1998: Application of proportionality test

National Director of Public Prosecutions v Salie & another 2015 (1) SACR 121 (WCC)

The applicant applied in terms of ss 48(1), 50(1)(a) and (b) and 53(1)(a) of POCA for an order declaring forfeit to the State certain property owned by the respondents, who were mother and daughter. The property consisted of three immovable properties and a motor vehicle. It appeared that the respondents had been running brothels at three venues, one owned by the first respondent and the other two rented from their owners. The applicant contended that the respondents had committed offences created by ss 2 and 20(1)(a) of the Sexual Offences Act 23 of 1957: keeping a brothel and living off the proceeds of prostitution. It was argued that the properties in respect of which orders were sought were the proceeds of the respondents’ unlawful activities and were therefore liable to forfeiture in terms of s 48 of POCA.

It was argued by the first respondent that forfeiture of any of the properties would be disproportionate and would infringe the right not to be arbitrarily dispossessed of property in s 25(1) of the Constitution. The court accordingly proceeded to examine whether proportionality was a requirement, not only for instrumentality in relation to the offence, but also for the proceeds of unlawful activities. In this regard Breitenbach AJ held as follows:

(1)    The definition of ‘proceeds of unlawful activities’ in POCA made it clear that the connection between the proceeds and the unlawful activities did not have to be direct. The proceeds included, for instance, benefits which someone could legitimately have acquired, but retained by or as a result of his or her offences (at [102]).

(2)    The three properties in question and the motor vehicle were the proceeds of unlawful activities as defined in POCA because they were assets which the respondents were able to retain by using the money which they made in connection with the brothels and the contraventions of the Sexual Offences Act.

(3)    The purposes of Chapter 6 of POCA included removing the incentives to crime which related not to the instrumentalities of the offences, but to the proceeds of unlawful activities (at [113]–[114]).

(4)    There were equivalences between s 18(1) of POCA (in Chapter 5) and s 50(1)(b) of POCA (in Chapter 6) which, together, supported the proposition that proportionality was a requirement of not just instrumentalities, but also forfeiture to the state of proceeds. As a result, proportionality was a requirement for forfeiture.

As to whether the forfeiture of the properties in question in this case would be justified on the proportionality test, the court took into account the following considerations:

(1)    The offences in question were, although serious, less serious than many other offences.

(2)    Prostitution was degrading to women. In the present case, however, the brothels did not constitute a major public nuisance and there was no evidence that they presented some of the more pressing social ills associated in many cases with such institutions.

(3)    Although none of the property derived directly from the commission of the offences, all of it comprised assets which the respondents were able to retain using money made from the commission of the offences. There was, thus, a relatively close connection between the property and the commission of the offences.

(4)    The total value of the respondents’ interest in the property was less than the proven brothel income and the income was a significant percentage (58%) of the value of the property.

These considerations led Breitenbach AJ to conclude that forfeiture was justified; there was no arbitrary deprivation of property in terms of s 25(1) of the Constitution.

s 50 of POCA: Proof that person knew or had reasonable grounds to suspect property was an instrumentality of an offence

National Director of Public Prosecutions v Skapu (unreported, ECG case no 2028/2013, 29 January 2015)

This case concerned an application for an order in terms of s 50 of POCA declaring forfeited to the state a motor vehicle belonging to the respondents which had been used in transporting dagga unlawfully. The vehicle was registered in his name and it was common cause that it had been used as an ‘instrumentality’ of an offence.

The respondent relied on s 52 of POCA to oppose the application, arguing that the property should be excluded from the order since he neither knew nor had reasonable grounds to suspect that the property was an instrumentality of an offence. He claimed that the driver had offered him money to allow him to use the vehicle, and that, although he had allowed him to use it, the driver had transported dagga without his knowledge or consent.

Evidence was adduced by the State that the respondent had been involved in other drug-related cases and that a vehicle belonging to him had been forfeited to the state a few years before. Was this similar fact evidence admissible? Yes, said the court. It showed ‘a modus operandi on his part, where he would apparently use “mules” or “runners” to convey dagga and then claim that he was not aware that they would use his vehicle to transport dagga’ (at [11]): it ‘seems that respondent would be so unlucky or is it mere co-incidence that people with whom he entrusted his motor vehicles so happen to choose to use same to transport dagga’. The evidence was both logically and legally relevant in showing it ‘to be highly improbable that this [was] a mere coincidence or ill-luck on the part of the respondent’.

Confiscation in terms of s 18 of POCA: requirement of a ‘benefit’

York Timbers (Pty) Ltd v NDPP 2015 (1) SACR 384 (GP)

The appellant, which owned and operated a sawmill, was convicted in a regional magistrates’ court of contravening s 24 of the National Environmental Management Act 107 of 1998 in that he had built a road over a part of his property without environmental authorisation. The grading of the road had been done by an overzealous forester who had been requested only to survey and mark out a proposed new ramp road, which was planned specifically to avoid environmental harm to the residents of the neighbouring town. The grading took place after the appellant had engaged the services of an environmental affairs practitioner and after the relevant environmental impact assessment had been lodged with the authorities. After conviction, but before sentence, the respondent had filed an application for a confiscation order in terms of s 18 of POCA. The appellant was sentenced to a fine of R180 000 in respect of the offence, and a confiscation order was granted in the amount of R450 000. It appealed against the confiscation order.

It was held that it was clear from the provisions of s 18(1) that the granting of a confiscation order was not mandatory but discretionary. The onus was on the respondent to make out a case for the order. However, the evidence of the appellant was undisputed that it had derived no benefit whatsoever from the grading operation, which had been done for the ‘noble purpose’ of accommodating the neighbouring community. In addition, the respondent had failed to prove that the appellant had the necessary mens rea to commit an offence, let alone an offence that fell within the ambit of Chapter 3 of POCA.

Furthermore, the court considered that what the appellant had done could not resort under the ills which the legislature sought to control and eliminate when it enacted POCA. The premature road grading activity could not be compared to the ‘offences relating to the proceeds of unlawful activities’ defined in Chapter 3 of POCA. The court took ‘respectful note’ of the judgment in National Director of Public Prosecutions v RO Cook Properties (Pty) Ltd; National Director of Public Prosecutions v 37 Gillespie Street Durban (Pty) Ltd & another; National Director of Public Prosecutions v Seevnarayan 2004 (2) SACR 208 (SCA). The court a quo in that case, it observed, had ‘also relied, as we did, on the long title (and, in our case, the preamble) of the Act when finding that evasion of personal income tax . . . could not be considered as “organised crime”, and that the Act was never intended to be applied in such situations’ (at [54]). It took ‘respectful note’, too, of this response to these submissions’ by the Supreme Court of Appeal (at 239f–g):

We cannot agree with this construction, which radically truncates the scope of the Act. It leaves out portions of the long title, as well as the ninth paragraph of the preamble. These show that the statute is designed to reach far beyond ‘organised crime, money laundering and criminal gang activities’. The Act clearly applies to cases of individual wrongdoing.

The court in York Timbers remained of the ‘respectful view’ that the words in the ninth paragraph could not be interpreted to mean that the object of the Act, what was said in the long title and what was said in the preamble ‘could be ignored for purposes of considering and adjudicating upon disputes arising from the provisions of the Act’. It considered, too, that ‘each case must be treated on its own merits to decide whether the ill complained of falls within the ambit of what the Act seeks to prevent and what the legislature had in mind when passing the legislation’. At the very least, it had to be proved on a balance of probabilities that the appellant ‘derived, received or retained, directly or indirectly’, ‘any property or any service, advantage, benefit or reward’ as a result of the unlawful activity. The respondent had not discharged this onus.

 (iv) Appeal and Review

Bail: Abolition of the automatic right of appeal to the Supreme Court of Appeal against refusal of bail by a High Court sitting as court of first instance

S v Banger [2015] ZASCA 79 (unreported, SCA case no 195/2015, 28 May 2015)

In S v Banger (supra) the Supreme Court of Appeal considered the effect of the Superior Courts Act 10 of 2013—which came into operation on 23 August 2013—on the so-called automatic right of appeal directly to the Supreme Court of Appeal against a refusal of bail by the High Court sitting as a court of first instance. The basis of this automatic right of appeal was set out in S v Botha en ’n ander 2002 (1) SACR 222 (SCA) where it was held that s 21(1) of the now-repealed Supreme Court Act 59 of 1959 was sufficiently wide to accommodate such a right, especially since neither the Criminal Procedure Act nor any other law provided for an appeal against the refusal of bail by the High Court sitting as a court of first instance. In Botha (at [14]) reliance was also placed on s 35(3)(o) of the Bill of Rights, where the right ‘of appeal to, or review by, a higher court’ is identified as an element of the right to a fair trial.

In Banger (supra) the refusal of bail was by the High Court sitting as a court of first instance, although two judges were involved (at [5] and [12]). Van der Merwe AJA (Cachalia and Mbha JJA concurring) pointed out that there could be no doubt that there must be a right to appeal against the refusal of bail, and the true issue was really ‘the procedure applicable to an appeal against the refusal of bail by the High Court sitting as a court of first instance . . .’ (at [5]).

With the repeal of the Supreme Court Act and in the absence of any other law or provision in the Criminal Procedure Act regulating the right concerned, the matter had to be addressed in terms of Chapter 5 of the Superior Courts Act and more specifically ss 16 and 17 (which replaced ss 20 and 21 of the Supreme Court Act).

At [11] and [12] it was noted that, in terms of s 16(1)(a), an appeal against any decision of the High Court as a court of first instance requires that leave to appeal should first be obtained from that court (see s 17(2)(a)) or, in the event of a refusal, from the Supreme Court of Appeal (see s 17(2)(b)). In Banger the appellant, believing that he had ‘an automatic right of appeal’ (at [3]), never applied to the High Court concerned for leave to appeal against its dismissal of his application for bail. The Supreme Court of Appeal therefore had no jurisdiction to entertain the matter (at [13]). The matter was struck from the roll (at [15]). It should be noted that the provisions of para (ii) in s 116(1)(a) could not come to the rescue of the appellant: the provision in para (ii) that an appeal against any decision of the High Court as a court of first instance lies to the Supreme Court of Appeal if the court consisted of more than one judge, remains subject to the provision ‘upon leave having been granted’ as stipulated in s 16(1)(a).

Van der Merwe AJA provided the following guidelines for purposes of future appeals against bail refusals by the High Court sitting as a court of first instance (at [14]):

Bail appeals are inherently urgent in nature. An accused person should not be deprived of his or her constitutional rights to freedom and to freedom of movement for longer than is reasonably necessary. The majority of appeals against the refusal of bail by the High Court as a court of first instance, will arise from a court that consists of a single judge and will not require the attention of this court. In these matters application for leave to appeal should generally be made immediately after the refusal of bail and, upon leave to appeal having been granted, a full court of that Division of the High Court should generally dispose of these appeals more expeditiously and cost-effectively than was the position before the advent of the Superior Courts Act.

The abolition of the so-called automatic right of appeal as explained and confirmed in Banger is probably welcomed by the Supreme Court of Appeal which has in the past expressed its discomfort with this automatic right, which really developed as a result of legislative oversight. See generally S v Masoanganye & another 2012 (1) SACR 292 (SCA) at [15]; S v Viljoen 2002 (2) SACR 550 (SCA) at [26] and S v Kock 2003 (2) SACR 5 (SCA) at 14g–h. The discussion of s 65A in Commentary, sv Right of accused to appeal to the Supreme Court of Appeal against decision by a superior court must now be read subject to Banger (supra).

 Table of Cases

Click here to see the full Table of Cases of the Criminal Justice Review No 1 of 2015


Click here for a printable .pdf version of this Review


Transformative Justice, A: Essays in Honour of Pius Langa
South African Sheriffs’ Guide: Practice and Procedure, The (Print)
Legal Terminology/Regsterminologie
Forensic Investigation
Child Offenders in South African Criminal Justice: Concepts and Process



> Click here to opt-in to continue to receive these newsletters.

> Click here to unsubscribe from this communication


> 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