Case CCT 71/13 International Cases [2014] Part 3 Case 9 [CCZA]



Gaston Savoi

- vs -

National Director of

Public Prosecutions











20 MARCH 2014


Madlanga J

(Moseneke ACJ, Skweyiya ADCJ, Cameron J, Dambuza AJ, Froneman J, Jafta J, Mhlantla AJ, Nkabinde J and Zondo J concurring)


  1. This application concerns the Prevention of Organised Crime Act[1] (POCA), about which Jafta J remarked as follows in Elran[2]:

    At the outset we must remind ourselves of the nature of the legislation we are concerned with. POCA was enacted in pursuit of legitimate and important government purposes of combating serious organised crime and preventing criminals from benefiting from the proceeds of their crimes.

  2. The KwaZulu-Natal High Court, Pietermaritzburg (High Court) declared section 2(1)(a)(ii), (b)(ii), (c)(ii), and (f)[3] of POCA constitutionally invalid only to the extent that each of the paragraphs of the subsection contains the words “ought reasonably to have known”. This is an application to confirm that declaration,[4] coupled with what is, in essence, an application for leave to appeal to the extent that the High Court refused to make other declarations of constitutional invalidity that the applicants[5] sought.[6] Stated broadly, the applicants’ challenge is that the impugned provisions[7] are void for vagueness; overbroad; retrospective; and violate the fair trial rights of an accused contained in section 35 of the Constitution. The respondents[8] oppose the confirmation of the declaration of constitutional invalidity and seek leave to cross-appeal the part of the High Court’s order that declared part of section 2(1) of POCA constitutionally invalid.


  3. The applicants are charged with various offences which include racketeering, fraud, corruption and money laundering before the KwaZulu-Natal High Court and in the Northern Cape. The charges relate to alleged unlawful conduct in connection with tenders for the procurement of water purification plants, gas generation systems and dialysis machines.

  4. The prosecution has been stayed pending the finalisation of the applicants’ constitutional challenge. In that challenge the applicants sought – before the High Court – an order declaring the definitions of “pattern of racketeering activity” and “enterprise” in section 1 and the entire Chapter 2 of POCA unconstitutional and invalid on the grounds that–

    1. the definition of “pattern of racketeering activity”[9] in section 1 is void for vagueness and thus unconstitutional;

    2. the definition of “enterprise”[10] in section 1 of POCA is overbroad and unconstitutional;[11]

    3. section 2(1)(a), (b), (c), (d), (e), (f), and (g) of POCA, which is predicated on the definitions of “pattern of racketeering activity” and “enterprise”, is consequently void for vagueness and constitutionally invalid;

    4. section 2(2) of POCA is unconstitutional and invalid because it violates the fair trial rights of an accused in section 35 of the Constitution;[12] and

    5. Chapter 2 of POCA is unconstitutional in its entirety because it operates retrospectively in violation of section 35(3)(l) of the Constitution and the rule of law.

  5. The High Court did not uphold a single one of these challenges. What it did instead was to declare paragraphs (a)(ii), (b)(ii), (c)(ii) and (f) of section 2(1) of POCA unconstitutional and invalid to the extent only that they contained the words “ought reasonably to have known” in each of the paragraphs. The effect of this declaration was that these words should be struck out. Let me pause and emphasise that the applicants had not asked for this. Then followed the confirmation proceedings and the applications for leave to appeal and cross-appeal.

  6. In their submissions in this Court[13] the respondents repeatedly bemoaned the fact that the applicants brought this challenge without indicating, by reference to the charges they are facing, how the impugned provisions are constitutionally invalid: once armed with the indictment the applicants simply rushed to the High Court. The respondents contend that this makes the application abstract.

  7. The issues that emerge from all this are–

    1. the abstract nature of the application;

    2. the definitional challenge;

    3. the procedural challenge;

    4. retrospectivity; and

    5. the High Court’s declaration of invalidity.

    Leave to appeal and cross-appeal

  8. Before dealing with the issues, let me dispose of the two applications for leave. Both relate to obvious constitutional issues of import. From what follows it will become clear that an interpretation of POCA is vital. The question whether leave to appeal should be granted depends upon whether doing so is in the interests of justice.[14] I take the view that it is in the interests of justice to grant both applications for leave. The basis for this view will be apparent as I deal with the issues, which I proceed to do.

    Abstract nature of the application

  9. The respondents characterise this application as an abstract challenge because the applicants elected not to place a particular set of facts before the Court.  Whatever the true nature of this application, there is no question that the applicants do have standing. In Ferreira[15] Chaskalson P for the majority said the following[16]:

    In the present case the applicants allege that section 417(2)(b) is inconsistent with section 25(3) of the Constitution. This is a matter which this Court has jurisdiction to enquire into, and it can do so in the present case if the applicants have standing to seek such an order from it.  Ordinarily a person whose rights are directly affected by an invalid law in a manner adverse to such person, has standing to challenge the validity of that law in the courts. There can be no question that the applicants have such an interest in the present case. Their right to refuse to answer questions that incriminate them is in issue and they seek to vindicate that right by challenging the only obstacle to their assertion of it. It was argued, however, that this does not apply to the present applicants because section 7(4) of the Constitution limits constitutional challenges to persons whose constitutional rights have been impaired or threatened. And, so the argument went, this could occur only if they are charged with a criminal offence and the evidence given by them at the enquiry is tendered against them at the criminal trial.

    [emphasis added and footnote omitted]

  10. He went on to say[17]:

    Whilst it is important that this Court should not be required to deal with abstract or hypothetical issues, and should devote its scarce resources to issues that are properly before it, I can see no good reason for adopting a narrow approach to the issue of standing in constitutional cases. On the contrary, it is my view that we should rather adopt a broad approach to standing. This would be consistent with the mandate given to this Court to uphold the Constitution and would serve to ensure that constitutional rights enjoy the full measure of the protection to which they are entitled.

  11. The same applies to this case; the applicants are entitled to challenge the constitutional validity of the Act under which they are currently charged. The impugned provisions are pertinent to the impending criminal proceedings. The applicants contend, amongst others, that the definitions of the very offences that they are charged with are so vague as to be unintelligible. Assuming for a moment that there is substance in that, it would be unfair to expect the applicants to plead to charges the inner and outer contours of which they have no idea.

  12. The broadly framed provisions on standing are an indication that the Constitution is consciously expansive so as to promote, rather than unduly restrict, the protection of constitutional rights. Unless an applicant’s claim to standing is truly unmeritorious, courts should be slow to shut the door on them. In my view, the present matter deserves to be entertained.

  13. So, the applicants plainly have standing to bring this challenge. This does not, however, make it irrelevant that this challenge is brought in the abstract.  Courts generally treat abstract challenges with disfavour. And rightly so.  Will hearsay, similar facts or evidence of previous convictions be led at the applicants’ trial?  At this stage we simply do not know. Abstract challenges ask courts to peer into the future, and in doing so they stretch the limits of judicial competence. For that reason, the applicants in this case bear a heavy burden – that of showing that the provisions they seek to impugn are constitutionally unsound merely on their face. The analysis that follows demonstrates just how heavy that burden is.

    Definitional challenge

  14. This challenge relates to the definitions of “pattern of racketeering activity”[18] and “enterprise”[19] in section 1 of POCA. By extension it also relates to section 2(1)(a) to (g) of POCA which, in creating certain offences, relies on those definitions. The challenge must be viewed in context. Part of that context is the purpose of POCA. In Mohamed NO[20] this Court stated the purpose thus [21]:

    The Act’s overall purpose can be gathered from its long title and preamble and summarised as follows: the rapid growth of organised crime, money laundering, criminal gang activities and racketeering threatens the rights of all in the Republic, presents a danger to public order, safety and stability, and threatens economic stability. This is also a serious international problem and has been identified as an international security threat. South African common and statutory law fail to deal adequately with this problem because of its rapid escalation and because it is often impossible to bring the leaders of organised crime to book, in view of the fact that they invariably ensure that they are far removed from the overt criminal activity involved. The law has also failed to keep pace with international measures aimed at dealing effectively with organised crime, money laundering and criminal gang activities. Hence the need for the measures embodied in the Act.

  15. POCA seeks to ensure that the criminal justice system reaches as far and wide as possible in order to deal with the scourge of organised crime in as many of its manifestations as possible. The question is whether it does so in a manner that is constitutionally impermissible. Insofar as it relates to “pattern of racketeering activity” the contention is that the definition is void for vagueness. The definition of “enterprise” is said to be overbroad and, therefore, unconstitutional. The applicants make other related contentions.

  16. In Affordable Medicines[22] Ngcobo J said [23]:

    The doctrine of vagueness is founded on the rule of law, which .... is a foundational value of our constitutional democracy. It requires that laws must be written in a clear and accessible manner. What is required is reasonable certainty and not perfect lucidity. The doctrine of vagueness does not require absolute certainty of laws. The law must indicate with reasonable certainty to those who are bound by it what is required of them so that they may regulate their conduct accordingly.

    [footnotes omitted]

  17. There is some intrinsic worth in laws being framed in general terms. On this the Supreme Court of Canada said that “laws that are framed in general terms may be better suited to the achievement of their objectives”.[24]

  18. The applicants argue that the factors that affect the requirements for establishing a pattern of racketeering activity and the underlying predicate offences are so numerous and varied that the entire concept of a “pattern of racketeering” is rendered vague. Let us look at those factors. Planning may be discrete as to the individual offences referred to in the definition. That is, each offence is planned as an individual offence. Likewise, there may be individualised, planned participation in more than one of the mentioned offences. But in the definition the focus of “planned” is not just in participation or involvement in the individual offences; it is in the fact that the participation and involvement must be ongoing, continuous or repeated. There must be interconnectedness between the offences with the result that they must form a sequence. They all have to be part of an elaborate plan. In S v Eyssen[25] the Supreme Court of Appeal held that “‘planned’ .... qualifies all three” (that is, “ongoing, continuous or repeated”). There must be forward planning that there will be participation or involvement in the offences referred to in Schedule 1.

  19. A useful treatment of “pattern” is to be found in Eyssen [26]:

    The relevant meaning of ‘pattern’ is given in the Oxford English Dictionary as ‘an order or form discernible in things, actions, ideas, situations, etc. Frequently with of as pattern of behaviour = behaviour pattern.’ .... In my view, neither unrelated instances of proscribed behaviour nor an accidental coincidence between them constitute a ‘pattern’ and the word ‘planned’ makes this clear.

  20. The participation must be in an offence specified in Schedule 1. Unless there is something the matter with what is contained in the Schedule, this too seems clear enough. The applicants refer to the fact that, amongst others, Schedule 1 lists “offences relating to coinage”[27] and “any offence relating to exchange control”,[28] and complain that these offences are so obscure as to render the Schedule vague. What the rule of law requires is reasonable certainty, not absolute or perfect lucidity.[29] In a different but, in my view, apt context[30] the Appellate Division in De Blom[31] quotes the following with approval from De Wet and Swanepoel:[32]

    I wish to agree with the view that knowledge of wrongfulness is an elementum essentiale of intention. A person only acts dolo malo when he acts unlawfully with full knowledge that he is doing so. This does not mean the wrongdoer must know that he is contravening section W of Act X of 19YZ, or that the wrongdoer must know that what he intends doing is punishable with this or that punishment, but only that he must be aware of the fact that what he intends doing is unlawful.  This does also not mean that the wrongdoer must know for sure that what he intends doing is unlawful, but only that he must have realised that what he intends doing could possibly be unlawful and that he has reconciled himself with this possibility.[33]

    [emphasis in original]

  21. Offences relating to coinage[34] and exchange control[35] do exist and I do not understand the applicants to suggest otherwise. That they may be hard to find, as the applicants suggest, is neither here nor there. The question is whether, in accordance with the proposition by De Wet and Swanepoel, the applicants must be aware of the existence of offences. It is interesting to note that in their affidavits the applicants do not touch on this issue.  Surely, at the very least, most reasonably informed South Africans are aware that they cannot stash R50 million into a suitcase and take it out of the country without official sanction and that they cannot mint their own current coins. Without substantiation on affidavit this complaint is stillborn. There is nothing so obscure with items 21 and 26 of Schedule 1 as to be vague.

  22. The applicants make the further point that the offences listed in Schedule 1 are numerous and varied. In particular – they add – item 33 refers to “any offence the punishment wherefore may be a period of imprisonment exceeding one year without the option of a fine”. They then argue that the list is so wide as to extend to “ordinary” or “garden variety” commercial crime which, by whatever stretch of imagination, cannot fall under what POCA is about, ie organised crime.[36] And on this the constitutional attack, raised in written and oral argument, is that POCA is overbroad and because of this, admits of prosecutorial abuse,[37] which gives rise to an unfair trial in violation of section 35 of the Constitution. In the affidavits[38] the definitional challenge, insofar as it relates to “pattern of racketeering activity”, is founded on the rule of law, ie the void for vagueness argument.

  23. The answer to the pleaded challenge is that there is simply no vagueness. There is reasonable certainty on what Schedule 1 is referring to. That the list may possibly incorporate offences which, when viewed individually, may not be expected to fall under the sort of notions that readily come to mind when one thinks of organised crime, does not necessarily translate to vagueness. As I seek to demonstrate immediately below, the applicants’ contention ignores the modus operandi (mode of operation) of organised crime. The argument must fail. Whether or not the expansive reach of the Schedule may found some other constitutionally cognisable cause of action is not before us.

  24. As the respondents argued, there is some inherent value in the concept of a “pattern of racketeering activity” and its utilisation in criminal prosecutions instead of framing indictments with respect to one or more of the offences listed in Schedule 1. That is so because it is a feature of organised crime that its organisational reach is wide and its tentacles stretch into many areas of commercial and governmental activity. What the concept of “pattern of racketeering activity” seeks to prohibit are the connections between conduct that might otherwise seem disparate but are in fact connected through the orchestrated activities of an organised criminal enterprise.

  25. The respondents add – correctly – that targeting specific offences for exclusion from the Schedule will fail to reach the true nature of criminal activity engaged in by criminal syndicates, both as to its scale and those who are ultimately responsible for it. Criminal syndicates work in a complex web-like manner. They operate in different areas of economic activity; utilise different agents and organisations; and thereby commit various offences − some relatively minor at face value − over time, in complex combination. It is the diversity of criminal activity, situated in complex organisational structures, occurring over time, where the lines of authority are deliberately obscured, that renders legislation in the nature of POCA a necessity. The concept of a “pattern of racketeering activity” is thus tailored to meet the multifarious ways in which organised crime manifests itself.

  26. To illustrate by reference to what, on the face of it, may be viewed as relatively minor individual offences: common assaults in the form of threats of violence[39] or actual application of force may be the order of the day in the organised criminality of a criminal syndicate. A ready example is where an organisation that deals in drugs on a large scale protects its turf and gains new turf – to use the colloquialism – to sustain and increase its sales by requiring its henchmen to force competitors into submission by means of threats of violence and actual violence amounting to no more than common assault. Quite conceivably, these offences might fall under the catch-all item 33 of Schedule 1. This would fit the definition of “pattern of racketeering activity” perfectly. I give this example to show that it is idle to attack the definition by isolating individual offences, forming an opinion on how relatively minor they are individually and concluding that they are, therefore, unsuited to the notion of organised crime and “pattern of racketeering activity”. That is shutting one’s eyes to how organised crime works.

  27. In short, what may appear to be “ordinary” or “garden variety” commercial criminality may, in fact, be very much part of organised crime. And that is a question of fact.

  28. In further support of the vagueness argument, the applicants placed strong reliance on the concurring judgment of Scalia J in the United States Supreme Court case of HJ Inc. v Northwestern Bell Telephone.[40]  In that matter he held that the majority’s definition of “pattern” in “pattern of racketeering activity” in RICO[41] is vague. The problem with this is that RICO’s definition of “pattern of racketeering activity” differs markedly from that of POCA.[42]  RICO says nothing about “planned, ongoing, continuous or repeated” participation in the offences concerned. To a significant extent, the differences of opinion between his judgment and the majority related exactly to that Court’s attempt to grapple with something RICO has not provided for, which POCA has. Therefore, this criticism of the RICO definition is unhelpful.

  29. Coming to the unfair trial component of the definitional challenge, it cannot succeed purely because it was not pleaded. It was raised for the first time in the applicants’ written submissions. It was not raised even in the affidavit filed in support of the application before this Court. The dismissal of the argument in this fashion is not technical. A fair-trial right challenge, being based on the Bill of Rights, would have involved a two-stage analysis,[43] and the respondents would have had to be alerted that this was the challenge. That is so because at the second stage – the justification stage – they are entitled to place before court evidence to justify the limitation at issue.[44] Because the challenge is founded on the rule of law, which does not entail this two-stage analysis, the respondents were effectively denied an opportunity to place necessary evidence before court.

  30. Relative to “enterprise”, the applicants complain that the definition is so all encompassing that it is overbroad. It is true that the definition covers a wide spectrum. On how broad the definition is, Cloete JA says [45]:

    It is difficult to envisage a wider definition. A single person is covered. So it seems is every other type of connection between persons known to the law or existing in fact; those which the Legislature has not included specifically would be incorporated by the introductory word ‘include’.

  31. Placing reliance on the judgment of Mokgoro J in Case,[46] the applicants then contend that the definition of “enterprise” and section 2(1), to the extent that some offences created by it are pegged on that definition, are unconstitutional. This argument is misconceived. In our constitutional jurisprudence overbreadth is not a self-standing ground of statutory constitutional invalidity. It comes into the equation in the justification analysis in terms of section 36(1) of the Constitution[47] once a law of general application has been found to limit a right in the Bill of Rights. It is interesting to note that the very paragraphs inCase that the applicants rely on are, in so many words, dealing with justification under section 33 of the interim Constitution, the equivalent of the present section 36.

  32. Whether some other constitutional attack based on the wide sweep of the definition – properly framed – is not feasible is something else altogether. And I should not be heard to suggest that it is feasible. For obvious reasons, I say nothing more on the subject.

  33. In sum, the entire definitional challenge must fail.

    Procedural challenge

  34. This attack targets section 2(2) of POCA, which provides:

    The court may hear evidence, including evidence with regard to hearsay, similar facts or previous convictions, relating to offences contemplated in subsection (1), notwithstanding that such evidence might otherwise be inadmissible, provided that such evidence would not render a trial unfair.

    [emphasis added]

  35. Ordinarily hearsay evidence, similar fact evidence and evidence of previous convictions is inadmissible. This is as a result of exclusionary rules that I deal with below. Besides these three types of evidence, there are several others that are also generally inadmissible. More on the others later. Suffice it to say for now, the relevance of these other categories is that the three mentioned specifically in section 2(2) do not constitute a closed list of the “otherwise inadmissible evidence” to which the section refers.[48]

  36. Apparently because of an assumption that the admission of any evidence that is subject to an exclusionary rule would of necessity render a trial unfair, the applicants contend that section 2(2) of POCA infringes the right to a fair trial contained in section 35(3) of the Constitution. For reasons that will become clear later, I must quote the sum total of this challenge as set out in the founding affidavit filed at the High Court.[49] If, in all instances, the admission of evidence that is subject to an exclusionary rule would always lead to an unfair trial, the premise on which the applicants proceed is on solid ground.[50] The question is: is that true in all instances?  If it is not, the entire substratum of the argument is gone.

  37. I answer the question by first dealing with the three categories of evidence mentioned in section 2(2). Although on a proper reading of the complaint it seems that the applicants take issue only with these categories,[51] for completeness I do deal with other types of inadmissible evidence later. I have alluded to the reasons for so doing: section 2(2) is about more than just the three mentioned categories.  To answer the question, it is necessary to look at the creation of the exclusionary rules, the rationale for each and why each has exceptions. In it I draw heavily from English law because it is the main source of our law of evidence.[52]


  38. The law’s aversion to the admission of hearsay is its general unreliability.[53] In Ndhlovu,[54] Cameron JA said [55]:

    Not only is hearsay evidence – that is, evidence of the statement by a person other than a witness which is relied on to prove what the statement asserts – not subject to the reliability checks applied to the first-hand testimony (which diminishes its substantive value), but its reception exposes the party opposing its proof to the procedural unfairness of not being able to counter effectively inferences that may be drawn from it.

    [emphasis added and footnotes omitted]

  39. Despite this rationale for the exclusion of hearsay, quite early on there was unhappiness with it. The hearsay rule “led to the exclusion of much reliable evidence”.[56] Some judicial critics even labelled the rule as technical and absurd.[57] It is not surprising that, on an ad hoc basis, exceptions to the rule were created.[58] From the historical account that Lord Reid gives,[59] it is apparent that this happened over time.

  40. This case by case creation of exceptions to the hearsay rule was finally pronounced to be at an end in Myers.[60] In South Africa the death knell was sounded by Vulcan Rubber Works.[61] If, as we have seen, hearsay was excluded because of its general unreliability and the unfairness it visited upon the party against whom it was tendered, the fact that the common law saw it fit to make possible the admissibility of certain types of hearsay evidence – through the exceptions – must mean that hearsay thus admissible was found reliable and not objectionable like that hit by the exclusionary rule.

  41. Why all this historical material?  The applicants submit that “[t]he right to a fair trial is immediately compromised by permitting the admission of otherwise inadmissible evidence” and that “[h]earsay evidence, similar fact evidence and evidence of previous convictions is ordinarily inadmissible for good reason: such evidence is inherently unreliable, prejudicial or unfair”. In Myers, Lord Reid did not say that the reason there could no longer be new exceptions to the hearsay rule was because the fountain whence the exceptions existing up to that point had been drawn had gone dry, with the result that no exceptions worthy of recognition could possibly ever come out of it again. What he did say was that he was averse to the piecemeal introduction of exceptions by the court as this would result in uncertainty, which is undesirable. Here are his words [62]:

    If we are to extend the law it must be by the development and application of fundamental principles. We cannot introduce arbitrary conditions or limitations: that must be left to legislation. And if we do in effect change the law, we ought in my opinion only to do that in cases where our decision will produce some finality or certainty. If we disregard technicalities in this case and seek to apply principle and common sense, there are a number of other parts of the existing law of hearsay susceptible of similar treatment, and we shall probably have a series of appeals in cases where the existing technical limitations produce an unjust result. If we are to give a wide interpretation to our judicial functions questions of policy cannot be wholly excluded, and it seems to me to be against public policy to produce uncertainty. The only satisfactory solution is by legislation following on a wide survey of the whole field, and I think that such a survey is overdue. A policy of make do and mend is no longer adequate. The most powerful argument of those who support the strict doctrine of precedent is that if it is relaxed judges will be tempted to encroach on the proper field of the legislature, and this case to my mind offers a strong temptation to do that which ought to be resisted.

  42. From what Lord Reid says, the very facts of Myers were clamant that the evidence be admitted. That much is clear from the last sentence of the quotation where, in essence, he says that the case offers a strong temptation to depart from precedent and admit otherwise inadmissible hearsay. He went further and implicitly accepted the reliability of the contested evidence [63]:

    At the end of their judgment the Court of Criminal Appeal [the court below] gave a different reason. ‘In our view the admission of such evidence does not infringe the hearsay rule because its probative value does not depend upon the credit of an unidentified person but rather on the circumstances in which the record is maintained and the inherent probability that it will be correct rather than incorrect.’  That, if I may say so, is undeniable as a matter of common sense.

    [emphasis added]

  43. The general unreliability of hearsay notwithstanding, Tapper[64] and Zeffertt and Paizes[65] explicitly make the point that hearsay falling outside of the exceptions to the hearsay rule may, in certain instances, be reliable. Zeffertt and Paizes say [66]:

    Hearsay could not be received if it did not fall within the corners of a recognised common-law or statutory exception. The fact that an item of hearsay evidence was highly relevant, or indeed reliable, did not alter this fact. Yet the primary reason for the exclusion of hearsay was its general unreliability – the fact that it rested for its evidential value on the untested memory, perception, sincerity and narrative capacity of a declarant or actor who was not subjected to the oath, cross-examination or any of the other procedural devices to which our adversary system of trial procedure subjects a witness giving original evidence.  Where in a specific case these objections were overcome the rationale for the exclusion disappeared.  In the absence of a recognised exception, however, the evidence had to be excluded.

    [footnote omitted and emphasis added]

  44. That it must be so – that is, there must still be categories of hearsay evidence not falling within the recognised exceptions, which are nevertheless reliable and thus deserving of admission – is quite understandable. I do not read the cases and other learning on the subject to say that when Myers finally drew the curtain, thus shutting out the possibility of new court-created exceptions, this was because no new, worthy exceptions could be found. The justification given aside, it cannot be gainsaid that the curtain was drawn at an arbitrary point: the creation of new exceptions could easily have been halted earlier or later or, as will be observed from the Canadian position below, not halted at all. This is illustrated by the words of Lord Blackburn in Sturla v Freccia [67]:

    [U]ndoubtedly the law is that, as a general rule, hearsay evidence is not admissible. But to that a great many exceptions have been introduced.  I do not say that if we were but beginning to make the law, we should be able to say exactly why so much should be admitted and no more.

    [emphasis added]

  45. Canadian courts refused to follow Lord Reid. They opted to continue finding new exceptions to the hearsay rule on a principled basis.[68] This buttresses the point that outside of the recognised exceptions to the hearsay rule there are other categories of hearsay the admission of which will not necessarily lead to unfair trials. In the Canadian case of R v Khan the Court held:[69]

    The hearsay rule has traditionally been regarded as an absolute rule, subject to various categories of exceptions .... While this approach has provided a degree of certainty to the law on hearsay, it has frequently proved unduly inflexible in dealing with new situations and new needs in the law. This has resulted in courts in recent years on occasion adopting a more flexible approach, rooted in the principle and the policy underlying the hearsay rule rather than the strictures of traditional exceptions.

  46. Adverting to the applicants’ complaints, it cannot be correct to say, without more, that all hearsay not admissible under the exceptions is of necessity unreliable and that its admission would result in unfairness. This is not to say hearsay evidence is without dangers. Quite the contrary.[70] Just as some hearsay evidence not falling within the recognised exceptions may not be objectionable, some may be. Admitting objectionable hearsay in criminal proceedings may well result in unfairness to an accused, resulting in a violation of the section 35(3) fair trial right. It must be as a result of this reality that a filter in the form of the proviso has been appended to section 2(2) of POCA.

  47. The section serves a twin purpose. First, it effectively does away with the hearsay rule on charges under section 2(1) of POCA and, indeed, other exclusionary rules. In this respect this section differs from section 3(1) of the Law of Evidence Amendment Act,[71] which retains the exclusionary rule[72] but permits the admission of hearsay evidence only under the circumstances set out in paragraphs (a) to (c) of the section.[73]

  48. Second, section 2(2) insists that hearsay should not be admitted if doing so would render the trial unfair. In the light of the copious discussion above, I see nothing unconstitutional with this. Any unconstitutionality that there might be would be a function of the improper admission of hearsay by the court concerned; that is, a failure to use the filter in a constitutionally compliant manner. As mentioned earlier, the applicants have not placed before the Court a specific set of facts, but have brought an abstract challenge. Therefore the question whether the section 2(2) proviso has been applied constitutionally does not arise.

  49. Needless to say, it would be ill-advised to attempt to anticipate instances where the admission of hearsay would be so unfair as to infringe an accused’s fair trial right. That is something best left to a trial court. There issues like, to mention but a few, the nature of the evidence, its reliability or lack of it, its probative value and prejudice to the accused[74] would have to be considered.

    Similar fact evidence

  50. In South Africa the admission of similar fact evidence is surrounded by some degree of confusion;[75] but perhaps less so in recent times. At the centre of this confusion is Makin v Attorney-General for New South Wales.[76] Similar fact evidence is inadmissible because it is inherently prejudicial.[77]

  51. Makin is one of three major English cases on similar fact evidence. The other two are DPP v Boardman[78] and DPP v P.[79] The legal principle that Makin is famous for goes [80]:

    It is undoubtedly not competent for the prosecution to adduce evidence tending to [show] that the accused has been guilty of criminal acts other than those covered by the indictment, for the purpose of leading to the conclusion that the accused is a person likely from his criminal conduct or character to have committed the offence for which he is being tried. On the other hand, the mere fact that the evidence adduced tends to [show] the commission of other crimes does not render it inadmissible if it be relevant to an issue before the jury, and it may be so relevant if it bears upon the question whether the acts alleged to constitute the crime charged in the indictment were designed or accidental, or to rebut a defence which would otherwise be open to the accused.

  52. What the first part of this formulation says is that, if all that evidence of similar facts shows is proclivity of a particular kind, it is not admissible regardless of the probative value of that disposition. It is not readily apparent why it was that propensity in and of itself might not, in a given set of circumstances, be sufficiently relevant to an issue before a trial court. Presumably because of its reference to design, accident and rebuttal of a defence, the Makin formulation came to be understood as having laid down rigid categories in which similar fact evidence would be relevant and admissible;[81] the converse being the inadmissibility of similar facts not falling within those categories.

  53. The House of Lords decision in Boardman[82] is credited with having relaxed the stereotypical approach to the admission of similar fact evidence.[83]  Lord Wilberforce said:[84]

    The basic principle must be that the admission of similar fact evidence (of the kind now in question) is exceptional and requires a strong degree of probative force. This probative force is derived, if at all, from the circumstance that the facts testified to by the several witnesses bear to each other such a striking similarity that they must, when judged by experience and common sense, either all be true, or have arisen from a cause common to the witnesses or from pure coincidence. The jury may, therefore, properly be asked to judge whether the right conclusion is that all are true, so that each story is supported by the other(s).

    I use the words ‘a cause common to the witnesses’ to include not only .... the possibility that the witnesses may have invented a story in concert but also the possibility that a similar story may have arisen by a process of infection from media of publicity or simply from fashion.[85]

  54. A major and – I would add – welcome relaxation of the English law on similar fact evidence came with DPP v P.[86] In a unanimous judgment the House of Lords – expressly overruling Boardman on this point – held that “it is not appropriate to single out ‘striking similarity’ as an essential element in every case in allowing evidence of an offence against one victim to be heard in connection with an allegation against another”.[87]

    I would deduce the essential feature of evidence which is to be admitted is that its probative force in support of the allegation that an accused person committed a crime is sufficiently great to make it just to admit the evidence, notwithstanding that it is prejudicial to the accused in tending to show that he was guilty of another crime. Such probative force may be derived from striking similarities in the evidence about the manner in which the crime was committed and the authorities provide illustrations of that of which Reg v Straffen [1952] 2 Q.B. 911 and Rex v Smith (1915) 11 C.r. App. R 229, provide notable examples. But restricting the circumstances in which there is sufficient probative force to overcome prejudice of evidence relating to another crime to cases in which there is some striking similarity between them is to restrict the operation of the principle in a way which gives too much effect to a particular manner of stating it, and is not justified in principle. Hume on Crimes, 3 ed. (1844), vol. II, p. 384, said long ago:

    The aptitude and coherence of the several circumstances often as fully confirm the truth of the story, as if all the witnesses were deponing to the same facts.

    Once the principle is recognised, that what has to be assessed is the probative force of the evidence in question, the infinite variety of circumstances in which the question arises, demonstrates that there is no single manner in which this can be achieved. Whether the evidence has sufficient probative value to outweigh its prejudicial effect must in each case be a question of degree.[88]

    This is a salutary proposition.

  55. What is the position in South Africa?  It is not necessary to track the historical development. I refer only to two decisions: S v D[89] and S v Nduna.[90] In S v D the Appellate Division quoted with approval the passage in Boardman[91] that says the probative force of similar fact evidence is derived from a “striking similarity” of the facts testified to by the several witnesses.[92] The insistence on striking similarity may lead to sophistry and technicality and raise more questions than provide answers.[93] The real question should be whether, when looked at in its totality, evidence of similar facts “has sufficient probative value to outweigh its prejudicial effects”;[94] and that is a matter of degree in each case.

  56. S v D, which was rejected in the Zimbabwean case of S v Banana,[95] appears to be still law in South Africa. And so is the more recent Supreme Court of Appeal judgment in Nduna.[96]  In Ndunathe Supreme Court of Appeal, although relying in the main on R v Katz[97] and R v Matthews,[98] followed the Makin formulation and the category based admission that seems to have come after Makin. The category in Nduna was the accused’s modus operandi in committing acts of robbery.

  57. Kruger says[99]:

    If in certain circumstances it appears that evidence of a tendency is highly relevant, the question is then not whether the evidence indicates a tendency but what the relative evidential value of that evidence is. In S v Banana 2000 (2) SACR 1 (Z) Gubbay CJ pointed out that the reformulation by Lord Mackay in R v P [1991] 3 All ER 337 (HL) of the test for admission of similar facts emphasises that the test concerns the evidential value of the evidence. Answering the test is thus a question of logic and sound common sense, not of legal philosophy. Whether the evidence has sufficient evidential value to overshadow its negative effects depends on the particular facts of each case and, of course, on the court’s balanced value judgement.

  58. I will make no attempt at suggesting what the ideal development of the law on similar fact evidence should ultimately be. That is not called for in this case. That said, in my view, the debate above adequately demonstrates that in South Africa there is still ample room for a less restrictive approach to the admission of similar facts.

  59. What emerges from this is that not all similar fact evidence that is inadmissible according to South African law would automatically render a trial unfair if admitted.

    Previous convictions

  60. In terms of section 211 of the Criminal Procedure Act[100] evidence of previous convictions is admissible only under certain specified circumstances and section 2(2) of POCA is not covered.[101]  The language of section 2(2) declares admissible evidence of previous convictions that would otherwise be inadmissible in terms of section 211. Is this constitutionally impermissible in terms of section 35(3) of the Constitution?  I think not.

  61. The rationale for the restriction on admissibility of evidence of previous convictions is that it is prejudicial to an accused to divulge her previous convictions.[102]

  62. Evidence of previous convictions might be used where it would serve as relevant similar facts. Thus the debate and conclusion on similar fact evidence finds application here as well. Kruger does make the connection with similar fact evidence [103]:

    It is logical that the doctrine of similar facts .... will encroach on [section 211] because the doctrine makes previous offences, irrespective of whether there was a conviction or not, admissible on the basis of their relevance to the facts in dispute.

    Other categories of inadmissible evidence

  63. To recapitulate, in addition to the three specified types of evidence, section 2(2) of POCA provides for the admissibility of other categories of otherwise inadmissible evidence.[104]

  64. From a treatment of the other three types of evidence, there has been a distillation of a principle. This is that it is misconceived to suggest that under all circumstances the admission of otherwise inadmissible evidence would automatically result in a trial being unfair in violation of section 35(3) of the Constitution. This principle seems applicable in respect of other categories of inadmissible evidence.

    Other issues on procedural challenge

  65. There are two more strings to the applicants’ bow. Firstly, they argue that, because there is no way of knowing in advance what criteria govern the admissibility of otherwise inadmissible hearsay, an accused cannot anticipate or guard against the inherent prejudice. That is because the section is silent on the criteria. Secondly, they contend that the admission of the evidence is dependent on the discretion of both the prosecutor and presiding judge with no guidelines on the exercise of that discretion.

  66. At first blush – but no further – the first argument appears sound. If the want of criteria were constitutionally objectionable, the Constitution would set criteria for instances where courts have to pass a value judgment on an issue. Well, an example of an instance where there are no criteria and yet courts are expected, should the need arise, to make a judgment call is section 35(5) of the Constitution. The section reads:

    Evidence obtained in a manner that violates any right in the Bill of Rights must be excluded if the admission of that evidence would render the trial unfair or otherwise be detrimental to the administration of justice.

  67. This section does not give an indication of circumstances where the admission of what I will loosely call unconstitutionally obtained evidence would render a trial unfair or be detrimental to the administration of justice. Also, we know on authority[105] that what is itemised in section 35(3)(a) to (o) does not constitute a closed list: the right to a fair trial is wider. If one uses what is itemised as a mirror to help guide one on what is unfair, beyond that mirror – that is, outside the list – one cannot anticipate what more would render a trial unfair. The judgment call in each of these instances will turn on what is before court. Is this objectionable?[106] I believe not. This leads me to the conclusion that there is simply no unfairness or prejudice to an accused arising from the lack of guidelines in section 35(5). Albeit sourced from an ordinary Act of Parliament, section 2(2) of POCA is analogous on the issue complained of. In principle I do not see why the position should be any different. Whether or not the admission of otherwise inadmissible evidence will render a trial unfair is something to be determined by the trial court based on the material before it.

  68. The public interest may also have to come into the equation when considering what is fair. In King the Supreme Court of Appeal held [107]:

    There is no such thing as perfect justice .... Fairness is not a one-way street conferring an unlimited right on an accused to demand the most favourable possible treatment but also requires fairness to the public as represented by the state. This does not mean that the accused’s right should be subordinated to the public’s interest in the protection and suppression of crime; however, the purpose of the fair trial provision is not to make it impracticable to conduct a prosecution. The fair trial right does not mean a predilection for technical niceties and ingenious legal stratagems, or to encourage preliminary litigation – a pervasive feature of white collar crime cases in this country.  To the contrary: courts should within the confines of fairness actively discourage preliminary litigation. Courts should further be aware that persons facing serious charges – and especially minimum sentences – have little inclination to co operate in a process that may lead to their conviction and ‘any new procedure can offer opportunities capable of exploitation to obstruct and delay.’  One can add the tendency of such accused, instead of confronting the charge, of attacking the prosecution.

    [footnotes omitted]

  69. The second argument referred to in [65] above is totally misconceived. Judicial officers are not administrative functionaries.[108]  The judgment call expected of them in terms of section 2(2) of POCA lies at the heartland of judicial function. Any judicial officer worth her salt should be perfectly placed to make the necessary value judgment. In this context, the reference to an exercise of discretion by a prosecutor is misplaced. What a prosecutor does is to decide on the arsenal to be deployed. In the line up there may be evidence in respect of which the prosecutor intends invoking section 2(2) of POCA. It is the trial court that will either admit or reject that evidence.

  70. The unfairness of a trial is a notion so wide that it would be an exercise in futility to try to contain it within four corners. It is a matter to be determined by the court when that issue arises. On that, the court may be proactive because it has a duty to protect an accused from being subjected to an unfair trial. Likewise, the accused may alert the court to possible unfairness. Axiomatically, it would be inadvisable to attempt to suggest instances where the admission of evidence in terms of section 2(2) of POCA might render a trial unfair. That is especially so in a case like the present, where – as shown in footnote 49 above – the challenge is scantily framed and does not relate to any specific evidence led or sought to be led at a trial.

  71. Since it is pre-eminently the function of the trial court to decide on the admissibility of evidence, including a decision whether the admission of evidence of a particular type would render a trial unfair,[109]that issue should only receive the attention of this Court if a constitutional issue pertaining to the performance of that function is raised.[110]

  72. In the end, I remain unswayed: section 2(2) of POCA does not limit the right to a fair trial.


  73. Section 35(3)(l) of the Constitution guarantees the right of an accused person “not to be convicted for an act or omission that was not an offence under either national or international law at the time it was committed”. Although expressed as a fair trial right, this right forms part of the principle of legality[111] which, in turn, is a sub-set of the rule of law.[112] In criminal law, long before the Constitution, this found articulation in the maxim nullum crimen, nulla poena sine lege (no crime, no punishment without law).[113]

  74. In National Director of Public Prosecutions v Basson[114] Nugent JA states that a “statute is said to operate retrospectively if it creates consequences for conduct only after that conduct has occurred”. The issue for determination arises from the definition of “pattern of racketeering activity” in section 1 of POCA.[115] The offences created by section 2(1)(a) to (g) of POCA are predicated on a “pattern of racketeering activity”. From the definition one or more of the offences that constitute the component parts of the “pattern of racketeering activity” may have occurred before the commencement of POCA. It is to this that the challenge relates.

  75. To understand the true content of the section 35(3)(l) right, it is necessary to consider its rationale. Central to the rule against retrospectivity is the need to forewarn people that conduct of a particular kind is proscribed and punishable criminally. With that forewarning, they are in a position so to order their lives that they do not fall foul of the proscription. This observation was made as far back as the 18th century by Blackstone [116]:

    There is still a more unreasonable method than this, which is called making of laws ex post facto; when after an action is committed, the legislator then for the first time declares it to have been a crime, and inflicts a punishment upon the person who has committed it; here it is impossible that the party could foresee that an action, innocent when it was done, should be afterwards converted to guilt by a subsequent law; he had therefore no cause to abstain from it; and all punishment for not abstaining must of consequence be cruel and unjust. All laws should be therefore made to commence in futuro, and be notified before their commencement; which is implied in the term ‘prescribed’.  

    [emphasis added]

  76. In similar vein, in National Director of Public Prosecutions v Carolus,[117] the Supreme Court of Appeal held that individuals should be afforded an opportunity to know what the law is in order to conduct themselves accordingly. If a statute creates a criminal offence retrospectively, there is, as Blackstone said centuries ago, an element of intrinsic cruelty in it,[118] which in my view, may not only implicate the section 35(3)(l) right, but possibly even the rights protected in section 12(1)(a) and (e) of the Constitution.

  77. So deeply rooted is the principle against retrospectivity that there has always been a presumption of ancient origin against the retrospective operation of laws. Some legal historians find evidence of it in Greek law.[119] In Roman law it manifested itself by way of several prohibitions in the Corpus Juris Civilis.[120]

  78. In pre-constitutional South Africa the notion of retrospectivity served no more than as a tool of interpretation: laws were presumed not to have been meant to operate retrospectively.[121] Nothing stood in the way of Parliament – in accordance with the principle of parliamentary supremacy, which we were subject to – to enact laws that operated retrospectively.[122] Converting a general principle of interpretation into a fundamental right signifies the intrinsic worth the framers of the Constitution saw in not having criminal laws that operate retrospectively.

  79. Chapter 2 of POCA creates various offences in section 2(1). Those are predicated, not on complete offences that predate POCA and for which an accused could now be convicted under that Act, but rather on a “pattern of racketeering activity”. A “pattern of racketeering activity” requires at least two offences listed in Schedule 1, one of which may predate POCA but the other not. The second offence must have taken place within 10 years (excluding any period of imprisonment) of the first. Thus there can be no conviction under POCA based only on offences that predate POCA. The question is: does this render Chapter 2 of POCA retrospective and thus unconstitutional?

  80. Keeping in mind what I see as central to the rule against retrospectivity, no criminal sanction flows, without more, from acts that predate POCA. For a conviction to attach there must be an act that postdates the coming into operation of POCA. When POCA came into effect, all were forewarned that if, after it came into operation, we commit an act that, in combination with one that predates POCA, completes the circle of racketeering activity, we will be guilty in terms of section 2(1) of POCA. A person that completes the circle does so with their eyes wide open: they must live with that choice. The inherent injustice, unfairness and cruelty, which make a criminal offence created retrospectively constitutionally impermissible, are simply not there.

  81. Section 2(1) in my view criminalises current conduct. The current conduct contemplated by those sections relates to the “pattern of racketeering activity”. That pattern must manifest itself post the coming into operation of POCA. That is what is criminalised in those sections; not individual offences committed pre-POCA or even a pattern of racketeering activity that predates POCA. That is in my view the clear meaning of the sections.

  82. Organised crime works like modern business and even organises its structures like those of business entities. Just as a particular business operates in a particular fashion, so does an entity involved in organised crime: it will have a particular mode of operation and that will be ongoing. Of course, like business, it may diversify or abandon the current category of activities and embark on something new altogether. As business becomes sophisticated, so do operations of organised crime. Cameron J in Elran said [123]:

    The second reason POCA-like legislation is indispensable is the intricacy and complexity of modern law-breaking. No longer is economic crime committed only through romantically imaginable methods like piracy, highway robbery and smuggled contraband. All that, if not past, is now of comparatively lesser importance. Most modern crime is committed through infinitely more sophisticated means – indirect and electronic. More importantly even, it is then concealed through those same means.

  83. I see nothing offensive to constitutionalism or criminal justice if the law were to peer over the shoulder and look at past conduct to determine whether, in conjunction with present conduct, it evinces a pattern of criminality; in particular, a pattern of racketeering activity. That will be a current – not past – pattern. There is nothing retrospective about that, at least not in the sense envisaged in section 35(3)(l) of the Constitution. It is to this current, planned, ongoing, continuous criminality that the section 2(1) offences relate and the criminal sanction attaches. To suggest otherwise would be but maudlin benevolence to criminality. And that approach would be deleterious to our relatively nascent democracy.[124]

  84. Yes, people facing criminal charges, not least the guilty, have a constitutionally guaranteed right to a fair trial: it behoves this Court to guard it jealously. But here there is simply no right under threat. Without doubt, the retrospectivity challenge must fail.

    High Court’s order of invalidity

  85. When declaring section 2(1)(a)(ii), (b)(ii), (c)(ii) and (f) of POCA unconstitutional only to the extent that it contains the words “ought reasonably to have known”, the High Court reasoned: first, the words expose an accused “to conviction for an offence he had not committed” and “the possibility of punishing an unintended, insensible or unconscious conduct”; and second, these words are vague and unintelligible. The High Court correctly identified the words to signify that the fault (mens rea) element is negligence.

  86. The general rule of our common law is that criminal liability does not attach if there is no fault or blameworthy state of mind.[125] This is expressed by the maxim: actus non facit reum nisi mens sit rea (an act is not unlawful unless there is a guilty mind).[126] The fault element may take the form of either intention or negligence.[127] This is true of both common law and statutory offences.[128] Thus a statute creating a criminal offence cannot be invalidated simply on the ground that it identifies negligence for the fault element. That is a choice that lies within the purview of the Legislature’s competence, and Parliament must be given the necessary leeway. Of course, that does not mean the Legislature is given free rein to choose the negligence standard as it pleases and under whatever circumstances: not in a constitutional democracy like ours. Pronouncing both on the need for flexibility and the constitutional curb, O’Regan J said in Coetzee [129]:

    [T]he appropriate form of culpability may well be affected by the nature of criminal prohibition as well as other factors. In addition, it should be borne in mind that significant leeway ought to be afforded to the Legislature to determine the appropriate level of culpability that should attach to any particular unlawful conduct to render it criminal. It is only when the Legislature has clearly abandoned any requirement of culpability, or when it has established a level of culpability manifestly inappropriate to the unlawful conduct or potential sentence in question, that a provision may be subject to successful constitutional challenge.

    It was not on any of the bases mentioned in the last sentence by O’Regan J that the High Court invalidated part of section 2(1).

  87. Nothing was placed before the High Court suggesting that the negligence standard was not suited to the offences concerned. Therefore, there are no grounds upon which we can conclude that the negligence standard provided for in section 2(1) is not constitutionally compliant. In fact, the High Court made the order of invalidity of its own motion. Needless to say, no evidence had been proffered by any of the parties on the issue. To put it bluntly, the order of constitutional invalidity was made on a case the respondents were never called upon to meet.

  88. What remains is whether the phrase “ought reasonably to have known” is unintelligible and vague as the High Court found. This language classically captures the idea of negligence. The measure for negligence is the so-called objective reasonableness standard. That explains the use of “reasonably” in the phrase. Negligence is determined by asking whether a reasonable person in the same circumstances would have acted in the same way.[130] To add, “[n]egligence can thus be said to be the failure to act as the reasonable man or woman would have acted”,[131] the so-called objective test.

  89. My view of the meaning of the contested phrase is reinforced by the provisions of section 1(3) of POCA which reads:

    For purposes of this Act a person ought reasonably to have known or suspected a fact, the conclusions that he or she ought to have reached are those which would have been reached by a reasonably diligent and vigilant person having both–


    the general knowledge, skill, training and experience that may reasonably be expected of a person in his or her position; and


    the general knowledge, skill, training and experience that he or she in fact has.

  90. [90] Paragraph (b) of this section does bring in an element of subjectivity. In my view this is more consonant with the interests of justice than the purely objective test for negligence. In its traditional formulation the objective test ignores the individual attributes of people: their level of education; background; personal beliefs – religious and otherwise; idiosyncrasies; fears; and so on. It has been argued that there is a potential for injustice when a completely objective criterion of negligence is applied.[132] On this the words of Hefer JA in S v Melk bear repetition:[133]

    The question is: by whom are [publications which are identifiable as prohibited] supposed to be identifiable?  If the so-called objective test of negligence is applied, as it generally is, the court hearing a case brought for an alleged contravention of section 56(1)(c) will be called upon to answer this question according to its own objective assessment of the reasonableness of the accused’s failure to identify the publication; and in doing so, it will not take the accused’s personal capabilities into account (S v Ngubane 1985 (3) SA 677 (A) at 687); the unsophisticated and uneducated shepherd will be treated no differently from the professor and no heed will be taken of the ‘widely differing standards of culture, education and social awareness of the various groups of persons to whom, as citizens of South Africa, this Act applies’.

  91. I am cognisant of the reality that in a setting more familiar to the shepherd, a higher standard of reasonableness may be expected of the shepherd than the professor if, for example, the professor is a complete stranger to the shepherd’s rustic lifestyle and that lifestyle is what is at issue. All that this is about is infusing an element of subjectivity to the pure objective test; a recognition of individual frailties and shortcomings; and, indeed, even positive attributes. That is what section 2(1)(a)(ii), (b)(ii), (c)(ii) and (f) seeks to achieve: the objective criterion of negligence tempered with a measure of subjectivity. That is quite plain and I fail to see the unintelligibility that the High Court saw. Nor is there anything constitutionally objectionable with this formulation.

  92. In the result, the respondents’ cross-appeal on this issue must succeed with the consequence that the High Court’s order of invalidity cannot be confirmed.


  93. In accordance with the general rule that an unsuccessful litigant in constitutional litigation should not be ordered to pay costs,[134] a costs order against the unsuccessful applicants is uncalled for in these proceedings. There has been no suggestion that this application “is frivolous or vexatious, or in any other way manifestly inappropriate” with the result that the applicants should not be “immunise[d] .... against an adverse costs award”.[135]  Moreover, the applicants’ underlying grievance relates to the criminal prosecution against them, making it generally inappropriate to mulct them in costs.


  94. The following order is made:

    1. Leave to appeal and cross-appeal is granted.

    2. The applicants’ appeal is dismissed.

    3. The applicants’ challenge to the constitutional validity of–

      1. the definitions of “pattern of racketeering activity” and “enterprise” in section 1 of the Prevention of Organised Crime Act 121 of 1998 (the Act);

      2. section 2(1)(a), (b), (c), (d), (e), (f) and (g) of the Act;

      3. section 2(2) of the Act; and

      4. Chapter 2 of the Act, fails.

    4. The respondents’ cross-appeal is upheld.

    5. The High Court’s order of constitutional invalidity is not confirmed.

    6. No order is made as to costs.

[1] 121 of 1998.

[2] National Director of Public Prosecutions v Elran [2013] ZACC 2; 2013 (1) SACR 429 (CC); 2013 (4) BCLR 379 (CC) (Elran) at para 22.

[3] The High Court’s order refers to section 2(1)(f)(ii). This is a patent error as section 2(f) does not have any subparagraph.

[4] In terms of section 167(5) of the Constitution: for an order of constitutional invalidity of an Act of Parliament, a provincial Act or conduct of the President to have force, it must first be confirmed by this Court.

[5] The first applicant is the chairman of various companies in the Intaka group of companies (Intaka Group). The second applicant is Intaka Holdings (Pty) Ltd. The third applicant is Fernando Praderi, an employee of the Intaka Group.

[6] The applicants assert that they are seeking a variation of the High Court’s order, to the extent that they did not succeed, and are asking for leave to appeal in the alternative. As I do not see the substantive difference, I will simply treat the application as one for leave to appeal.

[7] The definitions of “pattern of racketeering activity” and “enterprise” in section 1 of POCA; section 2(1)(a), (b), (c), (d), (e), (f) and (g) of POCA which is premised on these definitions; Chapter 2 of POCA in its entirety; and section 2(2) of POCA. The exact nature of the challenge in respect of each of these provisions is set out below.

[8] The first respondent is the National Director of Public Prosecutions, cited in his official capacity. The second respondent is the Minister of Justice and Constitutional Development. He too is cited in his official capacity and as the Minister responsible for POCA.

[9] In terms of section 1 of POCA a “pattern of racketeering activity” means–

the planned, ongoing, continuous or repeated participation or involvement in any offence referred to in Schedule 1 and includes at least two offences referred to in Schedule 1, of which one of the offences occurred after the commencement of this Act and the last offence occurred within 10 years (excluding any period of imprisonment) after the commission of such prior offence referred to in Schedule 1.

[10] Section 1 of POCA provides that “enterprise” includes–

any individual, partnership, corporation, association, or other juristic person or legal entity, and any union or group of individuals associated in fact, although not a juristic person or legal entity.

[11] The applicants refer to the challenges listed in (a) and (b) collectively as the “definitional challenge”.

[12] The applicants call this the “procedural challenge”.

[13] Written and oral.

[14] Everfresh Market Virginia (PtyLtd v Shoprite Checkers (Pty) Ltd [2011] ZACC 30; 2012 (1) SA 256 (CC); 2012 (3) BCLR 219 (CC) at para 17 and Albutt v Centre for the Study of Violence and Reconciliation [2010] ZACC 4; 2010 (3) SA 293 (CC); 2010 (5) BCLR 391 (CC) at para 20.

[15] Ferreira v Levin NO; Vryenhoek v Powell NO [1995] ZACC 13; 1996 (1) SA 984 (CC); 1996 (1) BCLR 1 (CC).

[16] Id at para 162.

[17] Id at para 165.

[18] Above n 9.

[19] Above n 10.

[20] National Director of Public Prosecutions v Mohamed NO [2002] ZACC 9; 2002 (4) SA 843 (CC); 2002 (9) BCLR 970 (CC) (Mohamed NO).

[21] Id at para 14.

[22] Affordable Medicines Trust v Minister of Health [2005] ZACC 3; 2006 (3) SA 247 (CC); 2005 (6) BCLR 529 (CC) (Affordable Medicines).

[23] Id at para 108.

[24] R v Nova Scotia Pharmaceutical Society (1992) 93 D.L.R (4th) 36; (S.C.C.) ((1992) 10 C.R.R (2d) 34) at 58 quoted with approval by Ngcobo J in Affordable Medicines above n 22 at para 108.

[25] [2008] ZASCA 97; 2009 (1) SACR 406 (SCA) at para 8.

[26] Id.

[27] Item 21.

[28] Item 26.

[29] Affordable Medicines above n 22 at para 108.

[30] There the context was mens rea (state of mind) where ignorance of the law was asserted.

[31] S v De Blom 1977 (3) SA 513 (A) at 530A-B.

[32] Strafreg 3 ed (LexisNexis Butterworths, Durban 1992) at 140.

[33] The translation comes from De Blom above n 31 at 530A. The original Afrikaans version reads:

Met die opvatting dat wederregtelikheidsbewussyn elementum essentiale van opset is, wil ek saamstem. ’n Mens handel slegs dan dolo malo wanneer hy hom willens en wetens in stryd met die regsorde stel. Dit wil nie sê dat die dader moet weet dat hy art. W van Wet X van 19YZ oortree nie, of dat die dader moet weet dat wat hy voornemens is om te doen met hierdie of daardie strawwe strafbaar is nie, maar slegs dat hy bewus daarvan moet wees dat wat hy wil doen regtens ongeoorloof is. Dit wil ook nie sê dat die dader seker moet wees dat wat hy wil doen wederregtelik is nie, maar slegs dat hy hom die voorstelling gemaak het dat wat hy wil doen moontlik regtens ongeoorloof kan wees, en hy hom met hierdie moontlikheid versoen.

[emphasis in original]

“Elementum essentiale” means “essential element”. Literally “dolo malo” means “with bad intention” but in context it means “with unlawful intention”.

[34] In terms of the South African Reserve Bank Act 90 of 1989 it is an offence to copy or counterfeit any South African banknote or coin. Section 2 of the Prevention of Counterfeiting of Currency Act 16 of 1965 lists a range of offences relating to current coins and banknotes.

[35] Regulation 22 of the Exchange Control Regulations as promulgated by Government Notice R1111 of 1 December 1961 reads:

Every person who contravenes or fails to comply with any provision of these regulations, or contravenes or fails to comply with the terms of any notice or order or direction issued or any permission or exemption granted under these regulations, or who obstructs any person in the execution of any power or function assigned to him by or under these regulations, or who makes any incorrect statement in any declaration made or return rendered for the purposes of these regulations (unless he proves that he did not know, and could not by the exercise of a reasonable degree of care have ascertained, that the statement was incorrect) or refuses or neglects to furnish any information which he is required to furnish under these regulations, shall be guilty of an offence.

[36] Offences itemised in Schedule 1 are: 1. murder; 2. rape; 3. kidnapping; 4. arson; 5. public violence; 6. robbery; 7. assault with intent to do grievous bodily harm; 8. indecent assault; 9. the statutory offence of (a) unlawful carnal intercourse with a girl under a specified age, (b) committing an immoral or indecent act with a girl or a boy under a specified age, (c) soliciting or enticing such girl or boy to the commission of an immoral or indecent act; 10. any offence under any legislation dealing with gambling, gaming or lotteries; 11. contravention of section 20(1) of the Sexual Offences Act 23 of 1957; 12. any offence contemplated in section 1(1) of the Corruption Act 94 of 1992; 13. extortion; 14. childstealing; 15. breaking or entering any premises whether under the common law or a statutory provision, with intent to commit an offence; 16. malicious injury to property; 17. theft, whether under the common law or a statutory provision; 18. any offence under section 36 or 37 of the General Law Amendment Act 62 of 1955; 19. fraud; 20. forgery or uttering a forged document knowing it to have been forged; 21. offences relating to coinage; 22. any offence referred to in section 13 of the Drugs and Drug Trafficking Act 140 of 1992; 23. any offence relating to the dealing in or smuggling of ammunition, firearms, explosives or armament and the unlawful possession of such firearms, explosives or armament; 24. any offence in contravention of section 36 of the Arms and Ammunition Act 75 of 1969; 25. dealing in, being in possession of or conveying endangered, scarce and protected game or plants or parts or remains thereof in contravention of a statute or provincial ordinance; 26. any offence relating to exchange control; 27. any offence under any law relating to the illicit dealing in or possession of precious metals or precious stones; 28. any offence contemplated in sections 1(1) and 1A(1) of the Intimidation Act 72 of 1982; 29. defeating or obstructing the course of justice; 30. perjury; 31. subornation of perjury; 32. any offence referred to in Chapter 3 or 4 of this Act; 33. any offence the punishment wherefore may be a period of imprisonment exceeding one year without the option of a fine; and 34. any conspiracy, incitement or attempt to commit any offence referred to in this Schedule.

[37] Part of the argument is articulated thus:

The crimes with which the applicants have been charged do not come close to fitting the model of organised crime which, on the respondents’ version, POCA was specifically designed to combat. None of the ‘special features’ of organised crime referred to by the respondents to justify the wide ambit of POCA is present in the indictments with which the applicants have been charged. This fact points squarely to the defects in POCA, which is overbroad such that it can cover both organised and non-organised criminal activity, and be used by the prosecuting authorities as a pressure tool in respect of one activity (alleged non organised criminal activity) that has nothing to do with the purpose of POCA (the fight against organised crime).

The effect of the aforegoing is that excessive enforcement discretion is delegated to the executive branch through the [National Director of Public Prosecutions], and through POCA an accused is at a material risk of having racketeering charges selectively and erroneously applied (or not applied) in a manner which is motivated by personal and/or political predilections of police and prosecutors. POCA, then, may simply be used as an arbitrary tool for negotiation by the prosecution, and as a penalty-enhancer in order to encourage the accused to plead guilty to the predicate offences in violation of the accused’s fair trial rights in section 35 of the Constitution. It thus allows, for example, a joinder of accused persons where no such joinder would probably be permissible. Such accused is then caught up in a lengthy, costly and massive criminal trial, the bulk of which is in relation to subject matter that has nothing to do with him.

[38] Founding, supplementary and replying affidavits.

[39] Snyman Criminal Law 5 ed (LexisNexis, Durban 2008) at 455:

Assault consists in any unlawful and intentional act or omission–


which results in another person’s bodily integrity being directly or indirectly impaired, or


which inspires a belief in another person that such impairment of her bodily integrity is immediately to take place.

[emphasis added]

[40] [1989] USSC 141; 492 U.S. 229 (1989).

[41] RICO, legislation from the United States of America, the Racketeer Influenced and Corrupt Organisations 18 U.S.C. §§ 1961-1968.

[42] RICO provides that “‘pattern of racketeering activity’ requires at least two acts of racketeering activity, one of which occurred after the effective date of this chapter and the last of which occurred within ten years (excluding any period of imprisonment) after the commission of a prior act of racketeering activity”. RICO then defines “racketeering activity” to mean a number of specified offences.

[43] Ferreira above n 15 at para 82.

[44] S v Makwanyane [1995] ZACC 3; 1995 (3) SA 391 (CC); 1995 (6) BCLR 665 (CC) at para 102.

[45] Eyssen above n 25 at para 6.

[46] Case v Minister of Safety and Security; Curtis v Minister of Safety and Security [1996] ZACC 7; 1996 (3) SA 617 (CC); 1996 (5) BCLR 609 (CC) (Case) at paras 49-50.

[47] Section 36(1) reads:

The rights in the Bill of Rights may be limited only in terms of law of general application to the extent that the limitation is reasonable and justifiable in an open and democratic society based on human dignity, equality and freedom, taking into account all relevant factors, including–


the nature of the right;


the importance of the purpose of the limitation;


the nature and extent of the limitation;


the relation between the limitation and its purpose; and


less restrictive means to achieve the purpose.

[48] The list is not exhaustive because of the use of the word “including” in the section. On the general rule on the use of “including” in a statute, see Minister of Health v New Clicks South Africa (Pty) Ltd [2005] ZACC 14; 2006 (8) BCLR 872 (CC); 2006 (2) SA 311 (CC) at para 455.

[49] The relevant paragraphs of the founding affidavit read:


Section 2(2) of POCA allows for hearsay, similar fact evidence or evidence relating to the previous convictions of an accused person to be admitted in a prosecution under POCA ‘notwithstanding that such evidence might otherwise be inadmissible, provided that such evidence would not render a trial unfair.


I am informed that the laws of evidence in criminal matters have developed in order to safeguard the fair trial rights of the accused, first under the common law and more recently under the Constitution.  Evidence that is excluded in criminal trials (be it hearsay, similar fact evidence or evidence relating to previous convictions) is excluded by the Court on the ground that the admission of such evidence would render the trial unfair. Whenever evidence is admitted under section 2(2) of POCA which is contrary to the law of evidence in criminal trials, it will a fortiori render the trial unfair, in violation of an accused’s rights under section 35 of the Constitution.

[emphasis in original]

[50] That, of course, would be subject to a justification analysis in terms of section 36 of the Constitution.

[51] Paragraphs 70 and 71 of the founding affidavit above n 49.

[52] Schwikkard and Van der Merwe Principles of Evidence 3 ed (Juta & Co Ltd, Cape Town 2009) at 24-7. Section 42 of the Civil Proceedings Evidence Act 25 of 1965 reads:

The law of evidence including the law relating to the competency, compellability, examination and cross-examination of witnesses which was in force in respect of civil proceedings on the thirtieth day of May, 1961, shall apply in any case not provided for by this Act or any other law.

In relation to criminal law, the Criminal Procedure Act 56 of 1955 generally retained the reference to the law of evidence as it stood on 30 May 1961 (Schwikkard and Van der Merwe above at 26-7). On what the law was as at the mentioned date, in S v Desai 1997 (1) SACR 38 (WLD) at 43G, Flemming DJP stated that our law of evidence “is essentially a part of the law which is tied to the law of England. (That, of course, does not exclude different developments at a date subsequent to the end of the statutory tying of our law of evidence to that of the law of England.)”  The “statutory tying of our law of evidence” can only be a reference to the law of evidence as at 30 May 1961 as provided for in the Acts referred to. Likewise, in Schwikkard and Van der Merwe at 26 the point is made that the reference to the law as at 30 May 1961 was a circuitous way of referring to the law of England; circuitous because of the perceived inappropriateness of the direct reference – in parliamentary Acts of the then new Republic – to the laws of another country. For completeness, I might mention that 30 May 1961 is the date South Africa ceased to be a colony of Britain and became a Republic.

[53] Zeffertt and Paizes The South African Law of Evidence 2 ed (LexisNexis, Durban 2009) at 386.

[54] S v Ndhlovu [2002] ZASCA 70; 2002 (2) SACR 325 (SCA).

[55] Id at para 13. See also S v Molimi [2008] ZACC 2; 2008 (3) SA 608 (CC); 2008 (5) BCLR 451 (CC) at fn 65.

[56] Tapper Cross & Tapper on Evidence 12 ed (Oxford University Press, New York 2010) at 551.

[57] See Myers v Director of Public Prosecutions [1965] AC 1001 at 1019G. In Jones v Metcalfe [1967] 3 All E.R. 205 at 208A-B Lord Diplock expressed himself in similar terms, going so far as to say that the hearsay rule “has little to do with common sense”. In Ndhlovu above n 54 at para 15, the following appears:

[The Law of Evidence Amendment Act 45 of 1988] was thus designed to create a general framework to regulate the admission of hearsay evidence that would supersede the excessive rigidity and inflexibility – and occasional absurdity – of the common law position.

[emphasis added]

[58] Tapper above n 56 at 551.

[59] Myers above n 57 at 1019G-1022B.

[60] Id at 1023D-E.

[61] Vulcan Rubber Works (Pty) Ltd v South African Railways and Harbours 1958 (3) SA 285 (A) at 296F-297B. See also Ndhlovu above n 54 at para 13.

[62] Myers above n 57 at 1021F-1022B.

[63] Id.

[64] Tapper above n 56 at 551 and 557.

[65] Zeffertt and Paizes above n 53 at 386.

[66] Id.

[67] (1880) 5 App Cas 623 HL at 640, quoted in Myers above n 57 at 1020G-1021A.

[68] See for example R v Khelawon 2006 CSC 57; [2006] S.C.R. 787; [2006] 2 R.C.S. 787; R v Mapara 2005 SCC 23; [2005] 1 S.C.R. 358; [2005] 1 R.C.S. 358; and R v Khan [1990] 2 S.C.R. 531; [1990] 2 R.C.S. 531.

[69] Above n 68 at 540f-i. See also Ares v Venner [1970] S.C.R. 608.

[70] Ndhlovu above n 54 at para 13.

[71] 45 of 1988.

[72] Ndhlovu above n 54 at para 14.

[73] Section 3(1) provides:

Subject to the provisions of any other law, hearsay evidence shall not be admitted as evidence at criminal or civil proceedings, unless–


each party against whom the evidence is to be adduced agrees to the admission thereof as evidence at such proceedings;


the person upon whose credibility the probative value of such evidence depends, himself testifies at such proceedings; or


the court, having regard to–


the nature of the proceedings;


the nature of the evidence;


the purpose for which the evidence is tendered;


the probative value of the evidence;


the reason why the evidence is not given by the person upon whose credibility the probative value of such evidence depends;


any prejudice to a party which the admission of such evidence might entail; and


any other factor which should in the opinion of the court be taken into account,

is of the opinion that such evidence should be admitted in the interests of justice.

[74] By “prejudice” I am not referring to the fact that the hearsay evidence in issue may tend to incriminate an accused. I am referring to prejudice that inheres in evidence of a hearsay nature.

[75] Writing more than three decades ago on similar fact evidence, in “Similar-fact Evidence in Criminal Proceedings” (1977) 94 SALJ 399 at 399, Zeffertt expressed himself thus:

To write on similar-fact evidence, were one untrammelled by authority, would seem less like blundering into a minefield or, to use the metaphor of Wigmore, into a ‘vast morass of authority’ where it is ‘hopeless to reconcile the precedents’. Wigmore said that at least 37 years ago. Since then the precedents have burgeoned, the morass has deepened, its paths have become more tortuous, slippery and false. To use another metaphor – from Lord Hailsham’s speech in DPP v Boardman – one is ‘constrained .... to traverse the pitted battlefield of “similar fact” evidence’.

[footnotes omitted]

[76] [1894] AC 57.

[77] Paizes “A Different Approach to Similar Facts” in Visser (ed) Essays in Honour of Ellison Kahn (Juta & Co Ltd, Cape Town 1989) 238 at 247 – of course, disregarding the references to prejudice attendant to jury trials.

[78] [1975] AC 421.

[79] [1991] 2 AC 447.

[80] Makin above n 76 at 65.

[81] Schwikkard and Van der Merwe above n 52 at 74 and Zeffertt and Paizes above n 53 at 277. Examples of the categories are “system”, “innocent association”, and “issue of identity”.

[82] Above n 78. On laudatory commentary about the case, see Hoffmann “Similar facts after Boardman” (1975) 91 LQR 193.

[83] Some of the Law Lords, in making their propositions, affirm the correctness of Lord Herschell’s words in Makin: Lord Morris at 438F-441D; Lord Hailsham at 450H-453E; and Lord Salmon at 461C-G. Thus these Law Lords imported into Boardman whatever may have been questionable in the Makin rule. In fact, what the exact ratio decidendi of Boardman is, is not without ambiguity. It is worth noting that two of the Law Lords – Lord Wilberforce and Lord Cross – do not make any reference to Makin.

[84] In the interests of brevity, I have deliberately refrained from referring to all the speeches of the other Law Lords. Suffice it to say that it is difficult to discern the ratio decidendi of the judgment. I refer to this one speech because of its reference to the concept of “striking similarity”, the significance of which will become apparent when I deal with the prevailing position in South Africa.

[85] Boardman above n 78 at 444D-F.

[86] Above n 79.

[87] Id at 460D.

[88] Id at 460E-461A.

[89] 1991 (2) SACR 543 (A).

[90] [2010] ZASCA 120; 2011 (1) SACR 115 (SCA).

[91] Above n 78 at 444D-E.

[92] S v D above n 89 at 546E-F.  S v D does not refer to DPP v P which overruled Boardman on this aspect. Perhaps the reason is that DPP v P had been decided merely three months to the day prior to the delivery of S v D.

[93] To illustrate how unhelpful the notion of striking similarity may be, let us take the facts of R v D 1958 (4) SA 364 (A). This case involved sexual intercourse across the colour line in contravention of section 1 of the Immorality Act 5 of 1927.  The common features of the testimony of the women were that in each case the woman testified that she had been employed as a domestic helper by the man’s wife; that the man had come to the kitchen at 05:30 still in his pyjamas and dressing-gown; that he had suggested intercourse; and that intercourse had taken place in the pantry while his wife was still asleep. Would the similarity not have been striking enough if, instead of always being in pyjamas, the accused was clad differently on each occasion?  What about if the women had testified that the sexual encounters had taken place at times ranging from the early morning to the afternoon and at different places in the house, and not just the pantry?  This immediately shows us that the idea of striking similarity confounds rather than illuminates the issue.

[94] DPP v P above n 79 at 461A.

[95] 2000 (2) SACR 1 (Z) at paras 8-10.

[96] Above n 90.

[97] 1946 AD 71.

[98] R v Matthews 1960 (1) SA 752 (A).

[99] Kruger Hiemstra’s Criminal Procedure Issue 1 (LexisNexis, Durban 2013) at 24–14.

[100] 51 of 1977.

[101] Section 211 reads:

Except where otherwise expressly provided by this Act or the Child Justice Act 2008, or except where the fact of a previous conviction is an element of any offence with which an accused is charged, evidence shall not be admissible at criminal proceedings in respect of any offence to prove that an accused at such proceedings had previously been convicted of any offence, whether in the Republic or elsewhere, and no accused, if called as a witness, shall be asked whether he or she has been so convicted.

[102] Hiemstra’s Criminal Procedure above n 99 at 24–19.

[103] Id.

[104] Examples of other categories of inadmissible evidence are previous self-consistent statements; opinion evidence; evidence of character; facts discovered by means of inadmissible confession; and confession against another.

[105] S v Zuma [1995] ZACC 1; 1995 (2) SA 642 (CC); 1995 (4) BCLR 401 (CC) at para 16 states:

The right to a fair trial conferred by that provision is broader than the list of specific rights set out in .... the sub-section.

[106] I consciously refrain from saying “constitutionally objectionable” because the absence of guidelines is in the Constitution itself.

[107] National Director of Public Prosecutions v King [2010] ZASCA 8; 2010 (7) BCLR 656 (SCA) at para 5, which this Court recently quoted in Estate Agency Affairs Board v Auction Alliance (Pty) Ltd [2014] ZACC 3 at para 71.

[108] Dawood v Minister of Home Affairs; Shalabi v Minister of Home Affairs; Thomas v Minister of Home Affairs[2000] ZACC 8; 2000 (3) SA 936 (CC); 2000 (8) BCLR 837 (CC) at para 46 reads:

There is, however, a difference between requiring a court or tribunal in exercising a discretion to interpret legislation in a manner that is consistent with the Constitution and conferring a broad discretion upon an official, who may be quite untrained in law and constitutional interpretation, and expecting that official, in the absence of direct guidance, to exercise the discretion in a manner consistent with the provisions of the Bill of Rights. Officials are often extremely busy and have to respond quickly and efficiently to many requests or applications. The nature of their work does not permit considered reflection on the scope of constitutional rights or the circumstances in which a limitation of such rights is justifiable. It is true that as employees of the state they bear a constitutional obligation to seek to promote the Bill of Rights as well. But it is important to interpret that obligation within the context of the role that administrative officials play in the framework of government, which is different from that played by judicial officers.

[footnote omitted]

[109] In De Vries v The State [2011] ZASCA 162; 2012 (1) SACR 186 (SCA) at para 52, the Court held that a trained judicial officer is able to restrict the effect of otherwise inadmissible evidence to charges in respect of which it is admissible and also to exclude it from consideration in respect of charges in which it is not.

[110] See Ferreira above n 15 at para 14.

[111] Burchell South African Criminal Law and Procedure Volume I: General Principles of Criminal Law 4 ed (Juta & Co Ltd, Cape Town 2011) at 34 says:

In its simplest form, the principle of legality proclaims that punishment may only be inflicted for contraventions of a clearly defined crime created by a law that was in force before the contravention.

See also Snyman above n 39 at 36.

[112] In Fedsure Life Assurance Ltd v Greater Johannesburg Transitional Metropolitan Council [1998] ZACC 17; 1999 (1) SA 374 (CC); 1998 (12) BCLR 1458 (CC) at para 56 Chaskalson P stated:

The rule of law − to the extent at least that it expresses this principle of legality − is generally understood to be a fundamental principle of constitutional law.

[113] Burchell above n 111 at 34-5; Snyman above n 39 at 36; Director of Public Prosecutions, Western Cape Minister of Justice v Prins [2012] ZASCA 106; 2012 (2) SACR 183 (SCA) at paras 7-8.

[114] [2001] ZASCA 111; 2002 (1) SA 419 (SCA) at para 1.

[115] See above n 9 for the definition.

[116] Quoted by Sampford in Retrospectivity and the Rule of Law (Oxford University Press, New York 2006) at 13, quoting Bailey Revolution to Reconstruction: A Biography of William Blackstone (1723–1780). The more unreasonable method alluded to by Blackstone at the beginning of the quotation is a ploy attributed to Caligula, a Roman Emperor. He is reported to have written laws “in a very small character and hung them above high pillars the more effectually to ensnare people.”  (See Sampford at 11.)

[117] [1999] ZASCA 101; 2000 (1) SA 1127 (SCA) at para 36.

[118] Sampford above n 116 at 11.

[119] Id at 10.

[120] See Smead “The Rule against Retroactive Legislation: A Basic Principle of Jurisprudence” (1935-36) 20 Minnesota Law Review 775.

[121] Genrec Mei (PtyLtd v Industrial Council for the Iron, Steel, Engineering, Metallurgical Industry [1994] ZASCA 143; 1995 (1) SA 563 (A) at 572E:

It is settled law that there is a strong presumption against retrospectivity of a statute, and that hence its operation should be construed as prospective only unless the Legislature clearly expressed a contrary intention.

See also S v Mhlungu [1995] ZACC 4; 1995 (3) SA 867 (CC); 1995 (7) BCLR 793 (CC) at para 37.

[122] See Executive Council of the Western Cape Legislature v President of the Republic of South Africa [1995] ZACC 8; 1995 (4) SA 877 (CC); 1995 (10) BCLR 1289(CC) at para 52.

[123] Elran above n 2 at para 67.

[124] In Mohamed NO above n 20 at para 15, Ackermann J noted that the problems POCA targets “make a severe impact on the young South African democracy”.

[125] S v Coetzee [1997] ZACC 2; 1997 (3) SA 527 (CC); 1997 (4) BCLR 437 (CC) (Coetzee) at paras 162 and 176.

[126] Burchell above n 111 at 54 and Snyman above n 39 at 152.

[127] Coetzee above n 125 at para 177.

[128] Burchell above n 111 at 55.

[129] Above n 125 at para 177.

[130] Burchell above n 111 at 423 and Snyman above n 39 at 208.

[131] Burchell above n 111 at 423.

[132] Id at 433.

[133] [1987] ZASCA 148; 1988 (4) SA 561 (A) at 578E-G.

[134] Biowatch Trust v Registrar, Genetic Resources, [2009] ZACC 14; 2009 (6) SA 232 (CC); 2009 (10) BCLR 1014 (CC) (Biowatch) at paras 21-3 and Affordable Medicines above n 22 at para 138.

[135] Biowatch above 134 at para 24.


G Marcus SC, K Kemp SC, M du Plessis and S Pudifin-Jones (instructed by Edward Nathan Sonnenbergs) for the applicant.

D Unterhalter SC, R Naidu, L Sisilana and C Sibiya (instructed by the State Attorney) for respondent.

all rights reserved