Thebus and Another v The State (2003) AHRLR 230 (SACC 2003)
Abduraghman Thebus and Moegamat Adams v The State
Constitutional Court, 28 August 2003
Judges: Chaskalson, Langa, Ackermann, Goldstone, Madala, Mokgoro, Moseneke, Ngcobo, O'Regan and Yacoob
Extract: Moseneke and separate opinion of Goldstone and O'Regan
Previously reported: 2003 (6) SA 505 (CC)
The right to fair trial, constitutional supremacy and constitutional interpretation vis-à-vis the common law
Constitutional supremacy (24, 29, 39)
Interpretation (development of common law, 25, 28, 31; international standards, 92)
Dignity (common purpose doctrine, 36)
Personal liberty (common purpose doctrine, 40)
Fair trial (presumption of innocence, common purpose doctrine, 42, 43; right to silence, 52-59, 65-68, 79, 81-86, 84, 92, 93; cross-examination on why remained silent on alibi, 69, 70, 91; information on consequences of remaining silent, 87; effect of misdirection, 73, 74, 79; adversarial process, 83, 87)
[1.] This is an appeal against the judgment and orders of the Supreme Court of Appeal (the SCA) handed down on 30 August 2002, confirming the convictions of both appellants in the Cape High Court on 14 September 2000 on one count of murder and two counts of attempted murder.
[2.] On 14 November 1998, a group of protesting residents in Ocean View, Cape Town, gathered and approached the houses of several reputed drug dealers in the area, including the house of one Grant Cronje. They allegedly caused damage to the property of Cronje before moving on. The protestors drove through the area in a motorcade of about five to six vehicles. As the motorcade approached a road intersection Cronje opened fire on the group. In response, some members of the group alighted from their vehicles and returned fire. In the resulting crossfire, a seven-year-old girl, Crystal Abrahams, was fatally shot and two others, Riaan van Rooyen and Lester September, were wounded.
[3.] Thereafter, the two appellants were arrested on suspicion of having been part of the group involved in the shooting incident. After the arrest of the first appellant, Sergeant McDonald of the South African Police Services warned him that he was not obliged to make any statement and that if he did it may be used in evidence against him. In this regard Sergeant McDonald testified as follows:
Tydens die onderhoudsverklaring ... toe ek hom nou gewaarsku het van sy regte. Toe vra ek hom of hy vir my 'n verduideliking wil gee, toe sê hy ja. Hy het toe vir my sy weergawe gegee. Ek het dit, soos hy praat het ek dit genotuleer, maar hy wou nie hê dat, ek moes dit in 'n verklaringvorm sit nie. Dit wou hy nie gehad het nie.
The first appellant readily admitted that after his arrest he was informed of the charges of which he was suspected and warned that he need not make a statement. He, nonetheless, made an oral statement before Sergeant McDonald. In this regard, his evidence is as follows:
Ja. So met ander woorde mnr McDonald het vir u gesê daar is getuies wat sê u was betrokke, maar u het geweet dat u eintlik by u tweede vrou was daardie tyd. - Ja.
Het u dit vir mnr McDonald gesê? - Ek het gesê die familie was in Hanover Park gewees. Maar ek het nie gesê waar ek was nie.
Enige rede daarvoor? - Nee, ek het nie rede gehad nie.
So met ander woorde u het vir mnr McDonald gesê die familie was in Hanover Park, maar u het niks sê van uself nie. - Van myself nie.
En u sé daar was geen spesifieke rede daarvoor. - Nee.
[4.] Thereafter, the first appellant refused to make a written statement to the police. Nearly two years passed before the appellants were brought to trial. Neither of the appellants disclosed his alibi defence until the trial before the High Court.
[5.] At the trial, the State led evidence placing both appellants in the vicinity of the shooting. A witness for the State, Gregory Edward Kiel (Kiel), testified that he had seen the first appellant standing near a vehicle holding a pick-handle, while the second appellant was retrieving spent cartridges discharged from the firearms of other members of the group. He also testified that the second appellant held a firearm but that he had not seen him shooting. Mitchell AJ found Kiel to be an impressive and forthright witness, whose evidence concerning the first appellant was beyond reproach.
[6.] The first appellant testified in support of his alibi defence and called two witnesses. Both witnesses testified that on the date and at the time of the shooting, the first appellant was at a place other than the scene of the shooting. The trial Court rejected this alibi defence. It concluded that both appellants had been part of the protesting group and were present at the scene of the shooting. Applying what is commonly referred to as the doctrine of common purpose, Mitchell AJ found both appellants guilty of one count of murder and two counts of attempted murder.
[7.] The trial Court sentenced each of the two appellants to eight years' imprisonment, suspended for a period of five years on certain conditions. Both appellants were granted leave to appeal against their conviction and the State leave to appeal against the sentences.
[8.] In May 2002, the SCA heard both appeals. The majority of the SCA (per Lewis AJA and Olivier JA concurring) dismissed the appeal against the convictions and upheld the appeal of the State against the sentences. The SCA ordered that each of the sentences imposed by the High Court be replaced by a sentence of 15 years' imprisonment. In a separate judgment, Navsa JA concurred in some respects with and dissented in others from the majority judgment.
[9.] Thereafter, the appellants made an application in terms of Rule 20 for special leave to appeal to this Court against the judgment and order of the SCA. This Court granted leave to appeal and issued directions calling for argument on two constitutional issues. Firstly, in the case of both appellants, whether the SCA failed to comply with its duty in terms of section 39(2) 4 of the Constitution to develop and apply the common-law doctrine of common purpose so as to bring it in line with the constitutional rights to dignity, freedom and security of the person and the right to be presumed innocent. Secondly, whether the SCA erred in drawing a negative inference from the first appellant's failure to disclose an alibi defence prior to trial, in violation of his right to silence as contained in the Constitution.
The High Court
[10.] The trial Court was persuaded that the State had made out a proper case to warrant a conviction of both appellants based on the common-law doctrine of common purpose as laid down in S v Mgedezi and Others. 9 In that regard the trial Court held that:
(T)he events of that afternoon took place in a sequence which commenced with the gathering at the Raven's home. The evidence shows that some of those persons were armed and that there was no apparent attempt to conceal this from others in the group. The intent was to confront and intimidate persons alleged to be drug dealers. In these circumstances it can hardly be said that any member of the group did not appreciate the possibility that violence could erupt and persons could be killed by the use of the group's armaments. By participating in the further activities of the group, each member signified his acceptance of that possibility. Such possibility became a reality when the shooting took place. There is no doubt ... that the shots which killed Crystal and wounded Riaan and Mr September came from ... the group of which (the first appellant) and (the second appellant) were part.
Later in the judgment the trial Court observed that:
They were present on the scene; they were aware that the shooting was taking place; they were throughout making common cause with the group, including the gunman, and they acted in association with him - (the first appellant) by standing guard and [the second appellant] by collecting the cartridge cases ... they had the requisite intention, albeit by way of dolus eventualis, to commit murder ... .
[11.] The first appellant denied having been present at the scene of the shooting. In support of his alibi defence, the first appellant testified that at approximately 13h00 on the day of the shooting he travelled by taxi from Ocean View to Fish Hoek train station. The purpose of the trip was to visit his second wife, Ms Faranaaz Jacobs, in Parkwood Estate. In the taxi, the first appellant met with a fellow resident of Ocean View, Ms Brenda Van Rooy. He and Ms Van Rooy took the 15h10 train to Wynberg. On arrival in Wynberg, the first appellant went to the local mosque where he led the afternoon prayers. On his version, the first appellant spent the rest of the afternoon and evening with his second wife. He returned to Ocean View only on the following day. In their evidence, Ms Van Rooy and Ms Jacobs corroborated the version offered by the first appellant.
[12.] Mitchell AJ rejected as untrustworthy the alibi evidence put up by the first appellant and his two witnesses. The trial Court took into account that both witnesses had claimed that they had not discussed their evidence with each other or with anybody else; that Ms Van Rooy was informed one month and Ms Jacobs, one week before the trial that they had to testify about events which had occurred nearly two years earlier and that these witnesses remembered with remarkable detail and accuracy the occurrences of the day in question. Mitchell AJ concluded that the close correlation between the evidence of the two witnesses and of the first appellant had cast doubt on its credibility.
[13.] The trial Court found that the evidence of the State witness, Kiel, placing the first appellant on the scene of the shooting was satisfactory and adequate to secure a conviction against first appellant. It rejected the first appellant's claim that he chose to disclose his alibi defence only during his trial and not at any time after his arrest. The trial Court reasoned that the first appellant was a man of considerable stature within the Ocean View community. He was the assistant Imam at the local mosque. He was arrested one month after the shooting incident and spent nearly a week in custody before he was granted bail. According to his second wife, the community had known of his arrest and that it concerned the shooting incident in which a young child had been killed. To the first appellant and his second wife these unfounded accusations should have amounted to an obvious error. The trial Court rejected the alibi as false and in doing so it took into account, amongst other factors, the unlikelihood that the first appellant would have preferred to remain silent rather than gainsay the 'false accusations'. The trial Court took the view that, before the trial, the first appellant could easily have dispelled the baseless accusations against him by disclosing his whereabouts to the police on the day of the shooting. Moreover, worshippers at the Wynberg mosque, present on the afternoon in question, would have had no conceivable difficulty in confirming that the appellant had led the afternoon prayers.
The Supreme Court of Appeal
[14.] Lewis AJA, writing on behalf of the majority of the SCA, held that the reliability of Kiel's identification of the first appellant had to be weighed carefully against his alibi and the testimony of the two witnesses who supported his alibi. The SCA, as did the trial Court, held that the close correlation and the detailed precision of the evidence of the alibi witnesses, taken together with the evidence of the first appellant, attracted justified suspicion. The SCA found that the version put up by the first appellant and his two witnesses had been 'concocted' and 'carefully rehearsed'. The SCA reminded itself that such suspicion was not enough to dismiss the version as false beyond reasonable doubt. Following the reminder, Lewis AJA remarked that: 'What is more telling ... is that the version was raised only at the trial, some two years after the incident.' As a result, the majority concluded that:
The only inference that can be drawn from (the first appellant's) failure to advise the police, and from the other witnesses' failure to do so, is that the alibi had no truth in it at all.
[15.] The SCA held that the trial Court had properly rejected the alibi defence of the first appellant and that the appeal against his conviction had no merit. The majority of the SCA confirmed the convictions without reference to the basis of the conviction being common purpose. In his minority judgment, Navsa JA upheld the finding of the trial Court that the requirements of common purpose had been met. Navsa JA found that on the facts the members of the vigilante group who were at the scene were party to a common purpose that rendered them liable for the murder of the child and the attempted murder of two other persons. It was on this basis that Navsa JA confirmed the second appellant's conviction of murder and attempted murder. In this regard Navsa JA states, at 578d-f:
By coming to Ocean View armed and behaving in the manner described earlier in this judgment members of the vigilante group were demonstrating that they were intent on confrontation and violence. By stopping and standing in the middle of a populated area, firearms blazing away wild-west style, members of the group placed themselves and others in the community in danger. It is clear that members of the vigilante group acted in concert as they went about their business in Ocean View. No member of the group, whether in motor vehicles or in the street, dissociated himself from violent actions perpetrated by others in the group.
[16.] The majority judgment accepted the findings of Navsa JA on common purpose. Navsa JA parted ways with the majority by holding that Kiel's identification of the first appellant on the scene of the shooting was not sufficient to found his conviction. Olivier JA and Lewis AJA accepted the testimony of Kiel as a reliable and compelling identification of the first appellant as a participant in the crimes of murder and attempted murder. They placed reliance on the fact that the first appellant and Kiel had known each other since their childhood and that Kiel had called him by his nickname. In contrast, Navsa JA reasoned that Kiel was a single witness and that his testimony was not 'satisfactory in all material respects'. Moreover, Kiel's identification of two other accused, whom he had claimed were at the scene of the shooting, had been discarded by the trial Court as open to doubt and erroneous. But these accused were not known to Kiel. The trial Court accepted that Kiel had made an honest but mistaken identification. In the case of the first appellant, however, there was no room for such a mistake as Kiel and the first appellant had known each other since they were children. Thus Kiel's identification of the first appellant carried considerable weight. Navsa JA, as did the majority judgment, held that the alibi defence of the first appellant was fabricated. However, the learned Judge reasoned that the rejection of the first appellant's alibi as fabricated did not redeem Kiel's testimony. Navsa JA concluded that the appeal of the first appellant against his conviction should succeed as there was a reasonable possibility that he had not been present at the scene of the shooting.
[17.] Two substantive constitutional issues fall to be decided in this appeal. The first issue is whether, in regard to both appellants, the SCA failed to develop the common-law doctrine of common purpose in conformity with the Constitution, as required by section 39(2) and thereby failed to give effect to their rights to dignity, freedom of the person and a fair trial, which includes the right to be presumed innocent. The second issue is whether the negative inference drawn from the first appellant's failure to disclose his alibi defence before trial has infringed his right to silence.
[18.] The doctrine of common purpose is a set of rules of the common law that regulates the attribution of criminal liability to a person who undertakes jointly with another person or persons the commission of a crime. Burchell and Milton define the doctrine of common purpose in the following terms:
Where two or more people agree to commit a crime or actively associate in a joint unlawful enterprise, each will be responsible for specific criminal conduct committed by one of their number which falls within their common design. Liability arises from their ''common purpose'' to commit the crime.
Snyman 18 points out that 'the essence of the doctrine is that if two or more people, having a common purpose to commit a crime, act together in order to achieve that purpose, the conduct of each of them in the execution of that purpose is imputed to the others'. These requirements are often couched in terms which relate to consequence crimes such as murder.
[19.] The liability requirements of a joint criminal enterprise fall into two categories. 20 The first arises where there is a prior agreement, express or implied, to commit a common offence. In the second category, no such prior agreement exists or is proved. The liability arises from an active association and participation in a common criminal design with the requisite blameworthy state of mind. In the present matter, the evidence does not prove any such prior pact.
[20.] The trial Court found that the first appellant was a party to an unlawful common enterprise during which the child was murdered. In convicting the accused the Court relied on the decision of S v Mgedezi, in which it was held that for the doctrine to be invoked:
In the first place, he must have been present at the scene where the violence was being committed. Secondly, he must have been aware of the assault on the inmates of room 12. Thirdly, he must have intended to make common cause with those who were actually perpetrating the assault. Fourthly, he must have manifested his sharing of a common purpose with the perpetrators of the assault by himself performing some act of association with the conduct of the others. Fifthly, he must have had the requisite mens rea; so, in respect of the killing of the deceased, he must have intended them to be killed, or he must have foreseen the possibility of their being killed and performed his own act of association with recklessness as to whether or not death was to ensue.
The appellants contended that the principles enunciated in S v Mgedezi should have been developed in accordance with the requirements of section 39(2) of the Constitution, and if this had been done, they would have been entitled to be acquitted.
[21.] The rules which make up the doctrine of common purpose deal with a number of different situations in which an accused person might be held liable for a crime committed in the course of a common enterprise. Since S v Mgedezi, the application of these rules have been refined and developed by various decisions of the SCA. In the present case it is not necessary to consider all of these developments. We are concerned here with a case in which the accused were present at the scene of the crime. What needs to be decided is whether the principles applicable to such a case, as stated in S v Mgedezi and developed by the SCA in later cases, calls for further development. It is neither necessary nor desirable to consider other situations. This judgment therefore deals only with the existing law insofar as it is relevant to the facts of the present case.
[22.] After S v Mgedezi there remains no doubt that where the prosecution relies on common purpose as a basis for criminal liability in a consequence crime such as murder, a causal connection between the conduct of each participant in the crime and the unlawful consequence caused by one or more in the group, is not a requirement. 24 Rules of criminal liability similar or comparable to common purpose are found in many common-law jurisdictions, including England, 25 Canada, 26 Australia, 27 Scotland 28 and the USA. 29 In all these legal systems, a causal nexus is not a prerequisite for criminal liability. In civil legal systems, such as France and Germany there appear to be no rules, which, in substance, approximate our rule of common purpose.
Did the SCA fail to develop the doctrine of common purpose in accordance with section 39(2) of the Constitution?
[23.] The main thrust of the appellants' contention is that the pre-constitutional requirements of common purpose unjustifiably limit the appellants' rights to dignity, freedom and security of the person and a fair trial including the right to be presumed innocent. However, the appellants stopped short of asserting that the doctrine of common purpose is unconstitutional in its entirety. They submitted that the High Court and the SCA erred in failing to develop, apply and elucidate the following requirements that:
(a) there must be a causal connection between the actions of the appellants and the crime for which they were convicted;
(b) the appellants must have actively associated themselves with the unlawful conduct of those who actually committed the crime; and
(c) the appellants must have had the subjective foresight that others in the group would commit the crimes.
[24.] Since the advent of constitutional democracy, all law must conform to the command of the supreme law, the Constitution, from which all law derives its legitimacy, force and validity. Thus, any law which precedes the coming into force of the Constitution remains binding and valid only to the extent of its constitutional consistency. The Bill of Rights enshrines fundamental rights which are to be enjoyed by all people in our country. Subject to the limitations envisaged in section 36, the State must respect, protect, promote and fulfil the rights in the Bill of Rights. The protected rights therein apply to all law and bind all organs of State, including the Judiciary.
[25.] It is in this context that courts are enjoined to apply and, if necessary, to develop the common law in order to give effect to a protected right, provided that any limitation is in accordance with section 36. 35 Section 39(2) makes it plain that when a court embarks upon a course of developing the common law it is obliged to 'promote the spirit, purport and objects of the Bill of Rights'. In the Pharmaceutical Manufacturers case, Chaskalson P observes that:
The common law supplements the provisions of the written Constitution but derives its force from it. It must be developed to fulfil the purposes of the Constitution and the legal order that it proclaims - thus, the command that law be developed and interpreted by the courts to promote the 'spirit, purport and objects of the Bill of Rights''. This ensures that the common law will evolve within the framework of the Constitution consistently with the basic norms of the legal order that it establishes. There is, however, only one system of law and within that system the Constitution is the supreme law with which all other law must comply.
[26.] The appellants have urged this Court to develop the common-law doctrine of common purpose beyond the existing precedent. In Carmichele 39 this Court decided that, faced with such a task, a court is obliged to undertake a two-stage enquiry. The first enquiry is whether, given the objectives of section 39(2) of the Constitution, the existing common law should be developed beyond existing precedent. If it leads to a negative answer, that would be the end of the enquiry. If it leads to a positive answer, the next enquiry would be how the development should occur and whether this Court or the SCA should embark on that exercise.
[27.] Section 39(2) requires that 'when' every court develops the common law it must promote the spirit, purport and objects of the Bill of Rights. This section does not specify what triggers the need to develop the common law or in which circumstances the development of the common law is justified. In Carmichele 40 this Court recognised that there are notionally different ways to develop the common law under section 39(2), all of which might be consistent with these provisions. It was also held that the Constitution embodies an 'objective normative value system' and that the influence of the fundamental constitutional values on the common law is authorised by section 39(2). It is within the matrix of this objective normative value system that the common law must be developed. 41 Thus under section 39(2), concepts which are reflective of, or premised upon, a given value system 'might well have to be replaced, or supplemented and enriched by the appropriate norms of the objective value system embodied in the Constitution'.
[28.] It seems to me that the need to develop the common law under section 39(2) could arise in at least two instances. The first would be when a rule of the common law is inconsistent with a constitutional provision. Repugnancy of this kind would compel an adaptation of the common law to resolve the inconsistency. 43 The second possibility arises even when a rule of the common law is not inconsistent with a specific constitutional provision but may fall short of its spirit, purport and objects. Then, the common law must be adapted so that it grows in harmony with the 'objective normative value system' found in the Constitution.
[29.] When there is a constitutional challenge to legislation the test for its constitutional validity is in two parts. Kriegler J, in In re S v Walters 45 delineates the process thus:
First, there is the threshold enquiry aimed at determining whether or not the enactment in question constitutes a limitation on one or other guaranteed right. This entails examining (a) the content and scope of the relevant protected right(s) and (b) the meaning and effect of the impugned enactment to see whether there is any limitation of (a) by (b). Subsections (1) and (2) of section 39 of the Constitution give guidance as to the interpretation of both the rights and the enactment, essentially requiring them to be interpreted so as to promote the value system of an open and democratic society based on human dignity, equality and freedom. If upon such analysis no limitation is found, that is the end of the matter. The constitutional challenge is dismissed there and then.
If there is indeed a limitation, however, the second stage ensues. This is ordinarily called the limitations exercise. In essence this requires a weighing-up of the nature and importance of the right(s) that are limited together with the extent of the limitation as against the importance and purpose of the limiting enactment. Section 36(1) of the Constitution spells out these factors that have to be put into the scales in making a proportional evaluation of all the counterpoised rights and interests involved.
[30.] Thus, if the impugned legislation indeed limits a guaranteed right, the next question is whether the limitation is reasonable and justifiable, regard being had to the considerations stipulated in section 36. If the impugned legislation does not satisfy the justification standard and a remedial option, through reading in, notional or actual severance is not competent, it must be declared unconstitutional and invalid. In that event the responsibility and power to address the consequences of the declaration of invalidity resides, not with the Courts, but pre-eminently with the legislative authority.
[31.] A different approach is required when a Court deals with a constitutional challenge to a rule of the common law. The common law is its law. Superior Courts are protectors and expounders of the common law. The Superior Courts have always had an inherent power to refashion and develop the common law in order to reflect the changing social, moral and economic make-up of society. 49 That power is now constitutionally authorised 50 and must be exercised within the prescripts and ethos of the Constitution.
[32.] In a constitutional challenge of the first type, referred to in paragraph , to a common law rule, the Court is again required to do a threshold analysis, being whether the rule limits an entrenched right. If the limitation is not reasonable and justifiable, the Court itself is obliged to adapt, or develop the common law in order to harmonise it with the constitutional norm.
[33.] The appellants have criticised the doctrine of common purpose principally on the ground that it does not require a causal connection between their actions and the crimes of which they were convicted. During argument, the appellants correctly conceded that in a joint criminal activity, the action of an accused need not contribute to the criminal result in the sense that but for it the result would not have ensued. What was urged on us is to develop the common law by requiring that the action of the accused must be shown to facilitate the offence on some level. Such facilitation would occur when the act of the accused is a contributing element to the criminal result. This argument does not constitute a direct challenge to the principles set out in S v Mgedezi.
[34.] In our law, ordinarily, in a consequence crime, a causal nexus between the conduct of an accused and the criminal consequence is a prerequisite for criminal liability. The doctrine of common purpose dispenses with the causation requirement. Provided the accused actively associated with the conduct of the perpetrators in the group that caused the death and had the required intention in respect of the unlawful consequence, the accused would be guilty of the offence. The principal object of the doctrine of common purpose is to criminalise collective criminal conduct and thus to satisfy the social 'need to control crime committed in the course of joint enterprises'. The phenomenon of serious crimes committed by collective individuals, acting in concert, remains a significant societal scourge. In consequence crimes such as murder, robbery, malicious damage to property and arson, it is often difficult to prove that the act of each person or of a particular person in the group contributed causally to the criminal result. Such a causal prerequisite for liability would render nugatory and ineffectual the object of the criminal norm of common purpose and make prosecution of collaborative criminal enterprises intractable and ineffectual.
[35.] The appellants argue that the doctrine of common purpose undermines the fundamental dignity of each person convicted of the same crime with others because it de-individualises him or her. It de-humanises people by treating them 'in a general manner as nameless, faceless parts of a group'. On this contention, a crime like murder carries a stigma greater than a conviction on an alternative charge or competent verdict such as public violence, conspiracy, incitement, attempt and accomplice liability. The appellants claim that the doctrine of common purpose violates their right not to be deprived of freedom arbitrarily, because this mode of criminal liability countenances the most tenuous link between individual conduct and the resultant liability. The appellants further argue that the doctrine of common purpose violates the presumption of innocence by dispensing with or lowering the threshold of proof for certain elements of a crime. That, the appellants contend, is at odds with the rule that the State must prove all the elements of a crime beyond a reasonable doubt. In the last instance, the appellants submit that the violation of any of their constitutionally protected rights is not justifiable as the primary rationale for the doctrine of common purpose is convenience of proof in favour of the prosecution.
[36.] I am unable to agree that the doctrine of common purpose trenches upon the rights to dignity and freedom. It is fallacious to argue that the prosecution and conviction of a person de-humanises him or her and thus invades the claimed rights. The entire scheme of sections 35 and 12(1) of the Bill of Rights authorises and anticipates prosecution, conviction and punishment of individuals, provided it occurs within the context of a procedurally and substantively fair trial and a permissible level of criminal culpability. 57 The essence of the complaint must be against the criminal norm in issue. The doctrine of common purpose sets a standard of criminal culpability. It defines the minimum elements necessary for a conviction in a joint criminal enterprise. The standard must be constitutionally permissible. It may not unjustifiably invade rights or principles of the Constitution. Put differently, the norm may only 'impose a form of culpability sufficient to justify the deprivation of freedom without giving rise to a constitutional complaint'. 58 However, once the culpability norm passes constitutional muster, an appropriate deprivation of freedom is permissible.
[37.] The definitional elements or 'the minimum requirements necessary to constitute a meaningful norm' for a common-law crime are unique to that crime and are useful to distinguish and categorise crimes. Common minimum requirements of common-law crimes are proof of unlawful conduct, criminal capacity and fault, all of which must be present at the time the crime is committed. Notably, the requirement of causal nexus is not a definitional element of every crime.
[38.] Thus, under the common law, the mere exclusion of causation as a requirement of liability is not fatal to the criminal norm. There are no pre-ordained characteristics of criminal conduct, outcome or condition. Conduct constitutes a crime because the law declares it so. Some crimes have a common-law and others a legislative origin. In a constitutional democracy, such as ours, a duly authorised legislative authority may create a new, or repeal an existing, criminal proscription. Ordinarily, making conduct criminal is intended to protect a societal or public interest by criminal sanction. It follows that criminal norms vary from society to society and within a society from time to time, relative to community convictions of what is harmful and worthy of punishment in the context of its social, economic, ethical, religious and political influences.
[39.] In our constitutional setting, any crime, whether common-law or legislative in origin, must be constitutionally compliant. It may not unjustifiably limit any of the protected rights or offend constitutional principles. Thus, the criminal norm may not deprive a person of his or her freedom arbitrarily or without just cause. The 'just cause' points to substantive protection against being deprived of freedom arbitrarily or without an adequate or acceptable reason and to the procedural right to a fair trial. The meaning of 'just cause must be grounded upon and (be) consonant with the values expressed in section 1 of the Constitution and gathered from the provisions of the Constitution'.
[40.] Common purpose does not amount to an arbitrary deprivation of freedom. The doctrine is rationally connected to the legitimate objective of limiting and controlling joint criminal enterprise. 62 It serves vital purposes in our criminal justice system. Absent the rule of common purpose, all but actual perpetrators of a crime and their accomplices will be beyond the reach of our criminal justice system, despite their unlawful and intentional participation in the commission of the crime. Such an outcome would not accord with the considerable societal distaste for crimes by common design. Group, organised or collaborative misdeeds strike more harshly at the fabric of society and the rights of victims than crimes perpetrated by individuals. Effective prosecution of crime is a legitimate, 'pressing social need'. The need for 'a strong deterrent to violent crime' is well acknowledged because 'widespread violent crime is deeply destructive of the fabric of our society'. There is a real and pressing social concern about the high levels of crime. 66 In practice, joint criminal conduct often poses peculiar difficulties of proof of the result of the conduct of each accused, a problem which hardly arises in the case of an individual accused person. Thus there is no objection to this norm of culpability even though it bypasses the requirement of causation.
[41.] At a substantive level, the conduct of the appellants, as found by the trial Court, answers beyond a reasonable doubt to the prerequisites of the criminal liability norm set by the rule. Moreover, their complaint is not against the procedural fairness of their trial but against the substantive constitutional compatibility of the rule. It may be added that a person who knowingly, and bearing the requisite intention, participates in the achievement of a criminal outcome cannot, upon conviction in a fair trial, validly claim that his or her rights to dignity and freedom have been invaded.
Presumption of innocence
[42.] I now turn to the appellants' claim that their conviction under the doctrine of common purpose denied them the right to be presumed innocent. Section 35(3)(h) accords to every accused person the right to a fair trial, which includes the right to be presumed innocent. In S v Bhulwana; S v Gwadiso, 67 O'Regan J, speaking for the Court, held that:
(T)he presumption of innocence is an established principle of South African law which places the burden of proof squarely on the prosecution. The entrenchment of the presumption of innocence in section 25(3)(c) must be interpreted in this context. It requires that the prosecution bear the burden of proving all the elements of a criminal charge. A presumption which relieves the prosecution of part of that burden could result in the conviction of an accused person despite the existence of a reasonable doubt as to his or her guilt. Such a presumption is in breach of the presumption of innocence and therefore offends section 25(3)(c). 68
[43.] Of course, the doctrine of common purpose does not relate to a reverse onus or presumption which relieves the prosecution of any part of the burden. The appellants argued that the substantive effect of the doctrine of common purpose is to dispense with the requirement of a causal nexus between the conduct of the accused and the criminal result. As found earlier, the doctrine of common purposes sets a norm that passes constitutional scrutiny. The doctrine neither places an onus upon the accused, nor does it presume his or her guilt. The State is required to prove beyond a reasonable doubt all the elements of the crime charged under common purpose. In my view, when the doctrine of common purpose is properly applied, there is no reasonable possibility that an accused person could be convicted despite the existence of a reasonable doubt as to his or her guilt. In my view, the common purpose doctrine does not trench the right to be presumed innocent.
[44.] Some text writers have raised two principal criticisms against the doctrine of common purpose. The first is that, in some cases, the requirement of active association has been cast too widely or misapplied. The second criticism is that there are less invasive forms of criminal liability short of convicting a participant in common purpose as a principal. The appellants echoed these complaints.
[45.] In my view, these criticisms do not render unconstitutional the liability requirement of active association. If anything, they bring home the duty of every trial court, when applying the doctrine of common purpose, to exercise the utmost circumspection in evaluating the evidence against each accused person. A collective approach to determining the actual conduct or active association of an individual accused has many evidentiary pitfalls. The trial court must seek to determine, in respect of each accused person, the location, timing, sequence, duration, frequency and nature of the conduct alleged to constitute sufficient participation or active association and its relationship, if any, to the criminal result and to all other prerequisites of guilt. Whether or not active association has been appropriately established will depend upon the factual context of each case.
[46.] It was submitted that the findings of the trial Court and the SCA were, on the facts, wrong. The appellants did not associate themselves actively with the crimes for which they were convicted. Both Courts, it was argued, ought to have found that the appellants took no action to support the members of the group who actually fired their weapons and that the first appellant was merely a bystander and the second appellant was not even present at the time of the fatal shooting. To the extent that these submissions deal only with the factual findings of the SCA, they have no merit. Where there is no other constitutional issue involved, a challenge to a decision of the SCA on the basis only that it is wrong on the facts is not a constitutional matter.
[47.] The appellants also submitted that the SCA misapplied the liability requirement of active association as formulated in S v Mgedezi 71 and applied in subsequent case law. 72 On this argument, both Courts adopted too wide a concept of active association and failed to satisfy themselves that the first appellant was a party to the common purpose prior to the infliction of the fatal shot. There is no merit in this criticism. The trial Court and the SCA held that throughout the shooting both appellants were present on the scene and made common cause with the group, including the gunman. The appellants also complained that the legal requirements of active association were misapplied. The application of a rule by the SCA may constitute a constitutional matter if it is at variance with some constitutional right or precept. 73 No such case has been made out. There is no constitutional ground in the present case to justify interference by this Court with the credibility findings or application of the requirement of active association by the trial Court or the SCA.
[48.] The argument on the relative degree of the invasiveness of common purpose in comparison to other forms of liability such as accomplice liability and competent verdicts is, in essence, a proportionality argument. It rests on the assumption that common purpose invades a constitutionally protected right to a degree disproportionate to the need and objective of crime control. In the light of the finding in this judgment that the doctrine of common purpose does not limit any of the rights asserted by the appellants, this contention need not detain us.
[49.] The appellants contend that the trial Court and the SCA omitted to apply the existing requirement that the State must prove that the appellants had the subjective foresight that others in the group would commit the crimes of which they were convicted. This complaint rests on the assertion that the evidence does not even prove that they were present and that neither Court made any attempt to determine the individual intention of the two appellants. I can find no merit in any of these submissions. This criticism of the factual findings of the trial Court and of the SCA is not borne out by the record. Moreover, the appellants have not advanced any need, nor could I find any, to adapt or elucidate the existing requirement of subjective fault. The common-law precedent is, in this regard, clear and consistent with the Constitution. It appears that that was the approach adopted by both the trial Court and the SCA. If the prosecution relies on common purpose, it must prove beyond a reasonable doubt that each accused had the requisite mens rea concerning the unlawful outcome at the time the offence was committed. That means that he or she must have intended that criminal result or must have foreseen the possibility of the criminal result ensuing and nonetheless actively associated himself or herself, reckless as to whether the result was to ensue.
[50.] Despite the evocative history of the application of the doctrine of common purpose in political and other group prosecutions, I am of the view that the common-law doctrine of common purpose in murder, as set out in S v Mgedezi and cases considered in this judgment, 75 does pass constitutional muster and does not, in the context of this case, require to be developed as commanded by section 39(2).
Right to silence
[51.] In the present matter, the first appellant disclosed his alibi defence for the first time at trial. He now contends that the trial Court and the SCA drew an adverse inference from his failure to disclose his alibi defence until his trial and that such an inference constitutes an infringement of his right to silence as contained in section 35(1)(a) 76 of the Constitution.
[52.] The central issue raised by this appeal is whether an adverse inference may be drawn from a failure to disclose an alibi prior to trial. In this regard three questions arise, being whether it is permissible to: (a) draw an adverse inference of guilt from the pre-trial silence of an accused, (b) draw an inference on the credibility of the accused from pre-trial silence and (c) cross-examine the accused on the failure to disclose an alibi timeously, thus taking into account his or her responses.
Scope and objects of the right
[53.] The pre-trial right to silence under section 35(1)(a) must be distinguished from the right to silence during trial protected by section 35(3)(h). This Court has authoritatively pronounced on constitutional claims premised on the right to silence during trial. From the various dicta it appears that the objective of the right is to secure a fair trial. Thus, though procedural, this protection is an integral part of the substantive right to a fair trial. The protection of pre-trial silence is buttressed by the constitutional requirement under section 35(1)(b) to inform an arrested person promptly of the right to remain silent and the consequences of not remaining silent.
[54.] The rights to remain silent before and during trial and to be presumed innocent are important interrelated rights aimed ultimately at protecting the fundamental freedom and dignity of an accused person. This protection is important in an open and democratic society which cherishes human dignity, freedom and equality.
[55.] The protection of the right to pre-trial silence seeks to oust any compulsion to speak. Thus, between suspicion and indictment, the guarantee of a right to silence effectively conveys the absence of a legal obligation to speak. This 'distaste of self-incrimination,' as Ackermann J puts it, is a response to the oppressive and often barbaric methods of the Star Chamber 78 and indeed to our own dim past of torture and intimidation during police custody. It is therefore vital that an accused person is protected from self-incrimination during detention and police interrogation which may readily lend itself to intimidation and manipulation of the accused.
[56.] In S v Manamela this Court affirmed that: '(T)he right to silence, like the presumption of innocence, is firmly rooted in both our common law and statute', and 'is inextricably linked to the right against self-incrimination and the principle of non-compellability of an accused person as a witness at his or her trial'.
In S v Boesak, Langa DP, speaking for the Court, pointed out that the right to remain silent has different applications at different stages of a criminal prosecution. On arrest a person cannot be compelled to make any confession or admission that may be used against her or him; later at trial there is no obligation to testify. The fact that she or he is not obliged to testify does not mean that no consequences arise as a result. If there is evidence that requires a response and if no response is forthcoming, that is, if the accused chooses to exercise her or his right to remain silent in the face of such evidence, the Court may, in the circumstances, be justified in concluding that the evidence is sufficient, in the absence of an explanation, to prove the guilt of the accused. This will, of course, depend on the quality of the evidence and the weight given to that evidence by the Court. 82 In Osman 83 Madala J held that
the fact that an accused has to make such an election is not a breach of the right to silence. If the right to silence were to be so interpreted, it would destroy the fundamental nature of our adversarial system of criminal justice.'
Inference of guilt
[57.] In our constitutional setting, pre-trial silence of an accused person can never warrant the drawing of an inference of guilt. This rule is of common-law origin. In R v Mashelele and Another Tindall JA, relying on the English decision of R v Leckey 85 formulated the rule thus:
(I)f the silence of the accused could be used as tending to prove his guilt, it is obvious that innocent persons might be in great peril; for an innocent person might well, either from excessive caution or for some other reason, decline to say anything when cautioned. And I may add that an accused person is often advised by his legal advisers to reserve his defence at the preparatory examination. It would, also, in my opinion, have been a misdirection to say that the silence of the accused was a factor which tended to show that their explanation at the trial was concocted.
[58.] It is well established that it is impermissible for a court to draw any inference of guilt from the pre-trial silence of an accused person. Such an inference would undermine the rights to remain silent and to be presumed innocent. 87 Thus, an obligation on an accused to break his or her silence or to disclose a defence before trial would be invasive of the constitutional right to silence. An inference of guilt from silence is no more plausible than innocence. The majority of the US Supreme Court in Doyle v Ohio reminds us that 'every post arrest silence is insolubly ambiguous'. 88 To hold otherwise, the mandatory warning under section 35(1)(b) will become a trap instead of a means for finding out the truth in the interests of justice.
Inference of credibility and an alibi defence
[59.] A distinction may properly be made between an inference of guilt from silence and a credibility finding connected with the election of an accused person to remain silent. In the dissenting judgment in Doyle v Ohio 90 a comparable distinction is drawn between the 'permissibility of drawing an inference on the credibility of the accused from silence and the impermissibility of drawing a direct inference of guilt'. In the latter, the presumption of innocence is implicated. In the former, a court would have regard to the factual matrix within which the right to silence was exercised.
[60.] An alibi defence has often generated judicial debate on whether it is an exception to the right to silence. In R v Cleghorn the peculiarity of an alibi defence is explained as follows:
(T)here is good reason to look at alibi evidence with care. It is a defence entirely divorced from the main factual issue surrounding the corpus delicti, as it rests upon extraneous facts, not arising from the res gestae. The essential facts of the alleged crime may well be to a large extent incontrovertible, leaving but limited room for manoeuvre whether the defendant be innocent or guilty. Alibi evidence, by its very nature, takes the focus right away from the area of the main facts, and gives the defence a fresh and untrammelled start. It is easy to prepare perjured evidence to support it in advance.
The minority in this case held that the requirement to disclose an alibi was an exception to the right to silence.
[61.] More recently, the South African Law Commission 94 has recommended that legislation should be introduced to permit a court to draw an inference from the pre-trial silence of an accused person in certain circumstances. The draft legislation proposes that a court should be authorised to make an inference appropriate to that case from the failure of an accused person to disclose an alibi during or before plea proceedings. The approach to disclosure of an alibi defence in the proposed enactment is not dissimilar to the one adopted by the majority in Cleghorn.
[62.] Canadian courts treat a failure to disclose an alibi timeously as being a factor which can properly be taken into account in the evaluation of the evidence as a whole:
(T)he consequence of a failure to disclose properly an alibi is that the trier of fact may draw an adverse inference when weighing the alibi evidence heard at trial. 96
[63.] That a failure to disclose an alibi timeously has consequences in the evaluation of the evidence as a whole is consistent with the views expressed by Tindall JA in R v Mashelele. After stating that an adverse inference of guilt cannot be drawn from the failure to disclose an alibi timeously, Tindall JA goes on to say:
But where the presiding Judge merely tells the jury that, as the accused did notdisclose his explanation or the alibi at the preparatory examination, the prosecution has not had an opportunity of testing its truth and that therefore it may fairly be said that the defence relied on has not the same weight or the same persuasive force as it would have had if it had been disclosed before and had not been met by evidence specially directed towards destroying the particular defence, this does not constitute a misdirection.
[64.] As pointed out earlier, an arrested person has the right to remain silent. This, indeed, is part of the warning given to the person, including that if he or she chooses to say anything it may be used in evidence against him or her. Drawing an inference on credibility in these circumstances has the effect of compelling the arrested person to break his or her silence, contrary to the right to remain silent guaranteed by section 35(1)(a) of the Constitution. To this extent, drawing an adverse inference on credibility limits the right to remain silent.
[65.] The rule of evidence that the late disclosure of an alibi affects the weight to be placed on the evidence supporting the alibi is one which is well recognised in our common law. As such, it is a law of general application. However, like all law, common law must be consistent with the Constitution. Where it limits any of the rights guaranteed in the Constitution, such limitation must be justifiable under section 36(1). Whether this rule is justifiable in terms of section 36(1) is a question to which I now turn.
[66.] I have already alluded to the importance of the right to remain silent. What is also important is that the accused receives no prior warning that his or her failure to disclose an alibi to the police might be used against him or her in evaluating the alibi defence. On the contrary, the accused is warned of his or her right to remain silent and that anything that he or she says might be used against him or her. The absence of a warning that his or her constitutional right to remain silent might be limited is a relevant consideration in the justification analysis. However, what weighs heavily with me is the extent of the limitation.
[67.] Firstly, the late disclosure of an alibi is one of the factors to be taken into account in evaluating the evidence of the alibi. Standing alone it does not justify an inference of guilt. Secondly, it is a factor which is only taken into consideration in determining the weight to be placed on the evidence of the alibi. The absence of a prior warning is, in my view, a matter which goes to the weight to be placed upon the late disclosure of an alibi. Where a prior warning that the late disclosure of an alibi may be taken into consideration is given, this may well justify greater weight being placed on the alibi than would be the case where there was no prior warning. In all the circumstances, and in particular, having regard to the limited use to which the late disclosure of the alibi is put, I am satisfied that the rule is justifiable under section 36(1).
[68.] The failure to disclose an alibi timeously is therefore not a neutral factor. It may have consequences and can legitimately be taken into account in evaluating the evidence as a whole. In deciding what, if any, those consequences are, it is relevant to have regard to the evidence of the accused, taken together with any explanation offered by her or him for failing to disclose the alibi timeously within the factual context of the evidence as a whole.
[69.] An election to disclose one's defence only when one appears on trial is not only legitimate but also protected by the Constitution. However, a related issue is whether it is permissible to cross-examine an accused on why she or he opted to remain silent on an alibi or indeed on any other defence. Such a line of enquiry is, in my view, permissible. It is quite proper, and often necessary, to probe, in cross-examination, the preference to remain silent. This goes to credit and would not unjustifiably limit the content of the right to remain silent. It may advance 'the truth-finding function of the criminal trial' and test the veracity of a belatedly disclosed or fabricated defence.
[70.] However, there are limits to such cross-examination. An explanation that the accused chose to remain silent as of right may in a particular context be an adequate answer. Thus such cross-examination must be exercised always with due regard to fairness towards both the accused and the prosecution and without unduly encroaching upon the right to remain silent or limiting a proper enquiry for the delayed disclosure of a defence.
[71.] It seems to me that there is no reason why this Court should not have regard to the failure by the first appellant to mention the alibi when he responded to questions put to him by Sergeant McDonald. Had this been a trial before a jury, there may have been a level of concern about that line of cross-examination. Where a jury is concerned it may be difficult for its members to evaluate the nuances involved in credibility findings, if matters which may be prejudicial but capable of explanation are put before them. Moreover a jury is not obliged to deliver an open and reasoned judgment on its factual findings. But in a trial before a Judge, in my view, it is quite permissible to ask questions on why the alibi was not mentioned earlier and to take the response thereto into reckoning when evaluating the evidence as a whole. Ultimately it is a matter of what is fair and just in the light of the requirements of a fair trial.
Submissions of the first appellant
[72.] The foundational submission of the first appellant is that the majority finding of the SCA rests entirely and precariously on an inference drawn from his silence regarding his alibi. This assertion is not without merit. The majority judgment of the SCA appears to have been premised on the reasoning that the mere suspicion about the version of the first appellant was not in itself enough; what justified his guilt was that 'the version was raised only at the trial, some two years after the incident'. The learned Judges of appeal then concluded that the appellant's failure to advise the police justified an inference that 'the alibi had no truth at all'. An inference of guilt from the disclosure of an alibi defence only at trial unjustifiably limited the appellant's right to pre-trial silence. Such an approach has, in effect, imputed guilt from pre-trial silence and thus trenched his constitutional guarantee to remain silent before his trial.
[73.] The resultant issue is whether this impermissible approach adopted by the SCA adversely prejudiced or undermined the substantive fairness of the trial. The full record of proceedings before the trial Court and the SCA is before us. This Court has had the benefit of full argument and is consequently in no different position from the trial Court or the SCA to consider facts which are connected or relevant to the proper adjudication of a constitutional issue. Such evidence, in my view, would itself be an issue connected to a decision on a constitutional matter. 100 Any further remission of this already protracted case would not serve the interests of justice. Moreover, both counsel were agreed that the matter should be brought to finality by this Court. It is thus competent and in the interests of justice for this Court to decide the matter.
[74.] In my view, the misdirection of the SCA would be relevant only if it would be an issue which materially alters the outcome of the trial 101 or compromises its substantive fairness, to which the appellant is entitled under section 35(3) of the Constitution. Put otherwise, the applicable test is whether, 'on the evidence, unaffected by the defect or irregularity, there is proof of guilt beyond reasonable doubt'. 102 If this Court were to find that such proof has been established, it must follow that the conviction must stand.
[75.] The credibility findings of the trial Court pose an insurmountable obstacle to the first appellant's case. The trial Court made it clear that the alibi evidence was not credible. Both the trial Court and the majority of the SCA correctly held that there was no reasonable possibility that Kiel's identification could be mistaken. The majority of the SCA held that Kiel's identification of the appellant was beyond reproach and that his evidence was reliable and compelling. Both Courts, inclusive of the minority judgment of Navsa JA, rejected the alibi evidence as false.
[76.] After his arrest, the first appellant was confronted by the police with the allegation that he had been present at the scene of the shooting. After having been warned of his rights he was asked by the police, prior to his arrest, what he had to say about these allegations. He chose to proffer an explanation, albeit a truncated one. His response that the family was in Hanover Park is hardly consistent with the alibi subsequently asserted. The only explanation he could give was that he was referring to his family and not to himself. This disingenuous explanation for the failure to disclose the alibi when confronted with the evidence against him can legitimately be taken into account in the evaluation of the evidence. Having regard to the fact that a late disclosure of an alibi carries less weight than one disclosed timeously, the cogency of Kiel's evidence and the unsatisfactory nature of the first appellant's evidence, the trial Court was entitled to reject the evidence of the alibi, and to convict the first appellant.
[77.] The trial Court properly convicted the first appellant on a consideration of the totality of the evidence. The appellant's explanation of why he chose to remain silent, the lateness of the disclosure of his alibi defence, the unacceptable evidence which was tendered by two of his witnesses and the cogency of the evidence tendered by Kiel taken together, entitled the trial Court to return a verdict of guilt against the first appellant.
[78.] Such is the adversarial nature of our criminal process. Once the prosecution had produced sufficient evidence which established a prima facie case, the first appellant had no duty to testify. However, once he had chosen to testify it was quite proper to ask him questions about his alibi defence, including his explanation on his election to remain silent. When his evidence was found not to be reasonably possibly true, as did the trial Court, he ran the risk of a conviction. Thus, absent a credible version from the first appellant, the version advanced by the prosecution, if found credible, was likely to be accepted. In S v Dlamini and Others, 103 Kriegler J emphasised the importance of freedom of choice in a democracy. However, liberty to make choices brings with it a corresponding responsibility and 'often such choices are hard'.
The appeals of the first and second appellant are dismissed.
Chaskalson CJ and Madala J concurred in the judgment of Moseneke J.
Goldstone J et O'Regan J
[79.] We agree with the order made by Moseneke J, and with his reasons for rejecting the appellants' arguments in relation to the doctrine of common purpose. However, in our view, the Supreme Court of Appeal, in drawing an adverse inference from the first appellant's failure to disclose his alibi, breached his constitutional right to silence. Given that a judgment of the Supreme Court of Appeal is binding on all courts other than this, we think it important that the correct constitutional approach to the question of the drawing of adverse inferences from the silence of an accused be explored in this judgment even though, after careful consideration of the record, we consider that this breach makes no difference to the outcome of the appeal. On a conspectus of all the evidence, 105 but without drawing any adverse inference from his failure to disclose his alibi prior to the trial, we are satisfied that the first appellant was proved to have been guilty beyond a reasonable doubt of all three charges.
[80.] The right to silence is entrenched in section 35(1)(a) and (b) and section 35(3)(h) of the Constitution as follows:
35(1) Everyone who is arrested for allegedly committing an offence has the right - (a) to remain silent; (b) to be informed promptly - (i) of the right to remain silent; and (ii) of the consequences of not remaining silent; ...
35(3) Every accused person has a right to a fair trial, which includes the right - ... (h) to be presumed innocent, to remain silent, and not to testify during the proceedings.
It is important to note that sections 35(1)(a) and (b) entrench not only the right to silence, but also the right to be informed of the consequences of not remaining silent.
[81.] This Court has acknowledged that the right to silence is firmly rooted in our common law. 106 The precise scope of the phrase, however, both in our law and that in other jurisdictions has remained uncertain. As Lord Mustill noted in R v Director of Serious Fraud Office, Ex parte Smith, the right to silence is best understood not as denoting a single right, but a disparate group of immunities. Lord Mustill identified six: an immunity from being compelled on pain of punishment to answer questions posed by anyone; an immunity from being compelled on pain of punishment to provide answers to questions when the answers may be self-incriminatory; a specific immunity from being compelled to answer, on pain of punishment, questions put by police officers when under suspicion of having committed an offence; the specific immunity of those accused from being compelled to give evidence in their trial; the specific immunity of those arrested from having questions put to them by police officers; and a specific immunity possessed by accused persons from having adverse comment made on their failure to answer questions before trial or to give evidence at trial. In addition, Lord Mustill noted that there are different underlying reasons for the different aspects or immunities contained within the right to silence.
[82.] In each case in which a court considers a constitutional challenge based on the right to silence, it will need to consider which aspect of the right to silence is in issue and whether it falls within the right protected in our Constitution. We disagree therefore with Yacoob J (at paragraph ) when he says there is only one right to silence, and that there is no difference between pre-trial silence and trial silence. In each case concerned with the right to silence, a court must identify the underlying purpose of the relevant aspect of the right to silence and consider whether it has been infringed in the case before it. In this case, we are concerned with the last immunity described by Lord Mustill - the specific immunity of an accused from having an adverse inference drawn from his or her silence. We must decide whether it is constitutionally acceptable to draw an adverse inference from the failure of an accused to disclose an alibi to the police or to the court in the period before the trial commences in circumstances where the accused was advised of his right to remain silent.
[83.] Various reasons are given for the principle that adverse inferences should not be drawn from an accused person's silence. One identified by Lord Mustill is the following:
(T)he instinct that it is contrary to fair play to put the accused in a position where he is exposed to punishment whatever he does. If he answers, he may condemn himself out of his own mouth; if he refuses he may be punished for his refusal.
In our view, this does not provide a valid foundation for the principle under our constitutional order. This Court has held that an adversarial system of criminal procedure necessarily forces hard choices on an accused, not by the operation of an unfair rule of law, but by the fundamental nature of the adversarial process itself. This Court has held that such choices which flow from the character of the adversarial system do not constitute an infringement of the right to silence. 111 Once the prosecution has produced evidence sufficient to establish a prima facie case against the accused, the accused faces the choice of staying silent, in which event he or she may be convicted, or seeking to lead evidence which may or may not be incriminatory. This hard choice faced by the accused is the consequence not of an unfair rule of law, but of the operation of the adversarial system coupled with the absence of a valid defence. In an adversarial system there can be no immunity from facing such choices and having to make such a choice cannot offend the right to silence as entrenched in our Constitution.
[84.] Another explanation commonly given for the rule against adverse inferences is the principle that the State bears the onus of proving every element of an offence without the assistance of the accused. It is clear from our Constitution that the presumption of innocence implies that an accused person may only be convicted if it is established beyond a reasonable doubt that he or she is guilty of the offence. That, in turn, requires the proof of each element of the offence. However, our Constitution does not stipulate that only the State's evidence may be used in determining whether the accused person has been proved guilty. Indeed, our law has always recognised that the question of whether the accused has been proven guilty or not is one to be determined on a conspectus of all the admissible evidence, whatever its provenance. This principle, too, cannot therefore found a valid objection to the drawing of adverse inferences.
[85.] A third reason given for the rule against the drawing of adverse inferences is the importance of protecting arrested persons from improper questioning and procedures by the police. Unfortunately, in the past people arrested were coerced by improper police methods to confess (not infrequently, falsely) to crimes. Such practices need to be put firmly behind us. In our view, the need to reduce unconstitutional policing practices is of such importance in the light of our history, that the right to silence should protect an accused person from having an adverse inference drawn from pre-trial silence in the face of questioning from the police. This concern provides an important reason for not drawing adverse inferences from the silence of an arrested person in the face of police questioning. It is of no relevance to the silence of an accused in court.
[86.] A different but equally cogent reason for the rule against the drawing of adverse inferences from the silence of an arrested person relates to the warning given to people when they are arrested. Section 35(1)(b) requires the police to warn people when they are arrested that they have the right to remain silent and of the consequences of not remaining silent and thus a failure to give the warning will infringe section 35(1)(b). In our view, it is constitutionally impermissible to draw an adverse inference from an arrested person's silence once he or she has been informed of the right to remain silent. That warning, as currently formulated, clearly implies that the arrested person will not be penalised for silence. For the person arrested to be told that he or she may remain silent without more, and for that very silence thereafter to be used to discredit the person, in our view is unfair. We are not persuaded therefore by Yacoob J's reliance on section 35(5) of the Constitution. Nor are we persuaded that it can ever be fair to warn a person arrested and give him or her the impression that there is a right to remain silent without qualification, and then to draw an adverse inference from that silence.
[87.] The adversarial process imposes many hard choices upon the accused. This is inevitable and appropriate. What is neither inevitable nor appropriate, is that the accused should be misinformed of the implications of the course of action he or she adopts. As this Court stated in S v Dlamini; S v Dladla and Others; S v Joubert; S v Schietekat:
Each and every one of those choices (relating to what the accused should do in bail proceedings) can have decisive consequences and therefore poses difficult decisions. As was pointed out in Osman's case '(t)he choice remains that of the accused. The important point is that the choice cannot be forced upon him or her.' It goes without saying that an election cannot be a choice unless it is made with proper appreciation of what it entails. It is particularly important in this country to remember that an uninformed choice is indeed no choice.'
An accused person needs to understand the consequences of remaining silent. If the warning does not inform the accused that remaining silent may have adverse consequences for the accused, the right of silence as understood in our Constitution will be breached.
[88.] Moreover, in many cases, the fact of the warning itself will render the silence by the accused ambiguous. It will not be clear whether the accused remained silent because he or she is relying on the right to remain silent, or for another reason, whether legitimate or not. To the extent that the silence is ambiguous, of course, it will have little value in the process of inferential reasoning, especially where guilt must be proved beyond a reasonable doubt.
[89.] In this case, the first appellant was warned of his right to remain silent. Thereafter he made a brief statement to the police stating that at the time of the offence his family was at Hanover Park, but when asked if he wished to have that statement reduced to writing, he demurred, as he was entitled to do. Thereafter he said nothing prior to trial. In our view, to use this silence against the first appellant either as confirmation of his guilt, as the majority judgment in the Supreme Court of Appeal did, or to discredit him as a witness, is unfair to him and constitutes a breach of his right to silence and his right to a fair trial. We do, however, consider it acceptable to use the statement that he made to the police after being warned concerning his family's presence at Hanover Park. Using such statement to evaluate his evidence does not constitute a breach of his right to silence. Indeed the first appellant was duly warned that any statement he might make may be used against him in his trial.
[90.] One further point needs to be made. It should be clear from what we have said that we do not see that a valid distinction can be drawn in this context between adverse inferences going to guilt, and adverse inferences going to credit. There is of course a conceptual difference between inferences going to credit and inferences going to guilt. But in the context of an alibi, the practical effect of the adverse inference to be drawn for the purposes of credit, namely, that the alibi evidence is not to be believed, will often be no different to the effect of the inference to be drawn with respect to guilt, 114 namely that the late tender of the alibi suggests that it is manufactured and that the accused is guilty. 115 We disagree therefore with the distinction drawn by Moseneke J between an adverse inference to credit on the one hand and an adverse inference to guilt. Whether an adverse inference is drawn going to guilt or credit, in our view, the accused has been treated unfairly in the light of the warning given.
[91.] Moseneke J comes to the related conclusion that it is permissible for an accused person to be cross-examined 'on why she or he opted to remain silent on an alibi or indeed any other defence ...' We do not agree. In the first place, we are of the opinion that no accused person should have to account for the exercise of a right entrenched in the Constitution. This is especially so where that account may be used against the accused. Secondly, it would be unfair to allow such cross-examination in the light of the accused person having been informed of the right to silence without at the same time being informed that she or he might be requested to account for the positive exercise of the right at the trial.
We must emphasise that we are concerned only with cross-examination relating to the pre-trial silence of the accused. Nothing we have said should be understood as precluding other lines of cross-examination designed to test the veracity of the alibi.
[92.] The foregoing should make it plain that the constitutional position would be different were there to be a law of general application permitting the drawing of an adverse inference in circumstances where the accused has been properly informed of the consequences of a failure to raise an alibi timeously. No such rule presently exists at common law in South Africa. In our view, such a rule if properly tailored and, in particular, if accompanied by an appropriate revision to the warning issued to arrested persons would still limit the right to silence, but would pass constitutional muster under section 36 of the Constitution. In this case, were the first appellant to have been duly warned that his failure to disclose an alibi timeously could result in an adverse inference being drawn, the common law could have been developed to permit the drawing of an adverse inference by the Supreme Court of Appeal and such development would have been a justifiable limitation of his right to silence and to a fair trial. It should be noted that a rule requiring timeous disclosure of an alibi defence has existed at common law in Canada for many years and according to a majority of the Supreme Court of Canada it 'has been adapted to conform to Charter norms'. Limits on the right to silence have also recently been adopted in the United Kingdom. The European Court of Human Rights has also held that an adverse inference from silence is not necessarily incompatible with art 6 of the European Convention on Human Rights.
It appears that rules of this nature are proposed by the SA Law Reform Commission.
[93.] We conclude, however, that the right to silence was breached in this case, because an adverse inference was drawn from the failure of the first appellant to disclose an alibi after being informed of his right to remain silent. Nevertheless we are persuaded that the appeal of the first appellant should be dismissed for the record establishes his guilt beyond a reasonable doubt without reliance upon any adverse inference from his silence. The High Court found Kiel's evidence cogent and persuasive, while rejecting that of the two alibi witnesses as false. There is no basis for rejecting these findings. Moreover, the first appellant, when initially questioned by the police, said that his family had been at Hanover Park at the time of the offence, which is inconsistent with the alibi he subsequently raised. At best for the accused, his statement that 'the family was at Hanover Park' is ambiguous and evasive. It is not consistent with the alibi tendered later to the effect that he was with his second wife at Parkwood Estate, which is nowhere near Hanover Park. In the light of the rejection of the evidence of the two defence witnesses and the prior inconsistent statement made by the first appellant, the alibi evidence does not in the context of all the evidence in the case (particularly the strong evidence of Kiel) raise a reasonable doubt as to the innocence of the first appellant.
Ackermann J and Mokgoro J concurred in the judgment of Goldstone J and O'Regan J.