There is some suggestion that charges of crimen injuria for the racist insults of Penny Sparrow (and others) cannot attract liability in criminal law in SA because crimen injuria requires the relevant insult to be directed at an individual rather than a group.
Crimen injuria is defined as the intentional unlawful impairment of the dignity or privacy of another (Burchell, 2013, 4ed, 632).
The essential requirements of the offence are as follows:
1. Intention;
2. Unlawfulness;
3. Impairment of the dignity of another.
Some definitions include a qualification that the impairment must be serious (Milton, 1996, 492). However, it is clear that this qualification is accounted for under the requirement of unlawfulness – since the law does not concern itself with trivialities (de minimis non curat lex).
The suggestion under discussion here – that the the relevant insult must be directed at an individual rather than a group – would seem to be addressed to the intention requirement. That is, one may expect that this is a suggestion that the accused must intend to infringe the dignity of a particular specified victim. However, there appears no suggestion that the general principles relating to intention, according to which the identity of the actual victim is irrelevant, S v Pistorius 2015 SCA) do not apply to crimen injuria.
Our law recognises that one intends to kill whoever is killed if one throws a bomb into a crowd – even though one does not know who is in the crowd. This is known as dolus indeterminatus – general intention – in our law. In the same way, there seems not basis to suspect that this “bomb” could not take the form of an offensive social media post.
The application of general principles of intention to crimen injuria is supported in the leading text on common law offences in SA, South Africa, JRL Milton, South African Criminal Law and Procedure (vol II, 1996, p 516 n274) – although Milton was endorsing the view that general principles make dolus eventualis a valid form of intention for crimen injuria.
The authority for the proposition that the relevant insult must be directed at an individual rather than a group appears to arise out of Snyman (6th ed, p463). Snyman states:
“An attack, not against Y himself, but against some group to which he is affiliated (eg his language group, his religion, race or nationality) will normally not constitute a violation of his dignitas, unless there are special circumstances from which an attack on his self-respect can be deduced.”
Snyman cites the case of S v Tanteli 1975 2 SA 772 (T) as authority for this proposition.
It is a tragic irony, that this cases appears to be a possible obstacle – albeit a small one – to successful prosecution of statements such as “all black people are monkeys”. The alleged insult in Tanteli was that the language, Afrikaans, was only suitable to be spoken to or with black people [described, in that case, by the accused using the pejorative k* word].
The insult or sting arising out of the Penny Sparrow social media post – in case this has escaped some – that “black people are monkeys” is that it suggests that black people are, in themselves, inferior as of a lesser species.
There are several points to be observed here.
The first is that this case is a Transvaal Provincial Division case – now the Gauteng Provincial Division (situated in Pretoria). If it contains any precedent (and I will below argue it does not), that is, anything binding on cases that follow, it would in any event be restricted to Gauteng. Even in Gauteng, since the Tanteli case was a single Judge judgement, it does not bind courts sitting with more than one Judge (“higher” courts). It may, on the rules of precedent, be persuasive in “higher” courts in Gauteng or in other jurisdictions (provinces), but it is certainly not binding in “higher” courts or other jurisdictions.
Also, while the judgement is mentioned and relied upon in Snyman’s text on criminal law in South Africa, it is not mentioned in the other leading general text on criminal law – written by Burchell. Perhaps more importantly, it is mentioned in what is arguably the leading text on common law offences in SA (Milton, South African Criminal Law and Procedure (vol II, 1996) only as a case to be distinguished from the general principle. This may itself be read as an indictment on the case.
The second is that Snyman’s rendition of the case authority is in one respect not a true reflection of what the court held. The third is that, in another respect, Snyman’s rendition is true, which requires in turn some analysis of the judgement in the Tanteli case.
To begin, Snyman’s rendition is not correct in that Nicholas J in Tanteli did not say that only in exceptional circumstances can an insult against a group with which one associates provide a basis for a conviction of crimen injuria. He found that, on the facts in that case, the dignity of the victim was, in fact, not injured. He said as follows: “There may, of course, be cases in which an insult to a person’s language, or race, or religious persuasion or national group may, in the circumstances, constitute also an impairment of his dignitas, but this is not such a case.” (p. 775). There is not mention of a requirement of special circumstances.
Snyman’s rendition is correct however insofar as he reflects that the judgement did seem to distinguish between an attack on a person as opposed to an attack on an attribute of a group, with which s/he associates him/herself. It is worth considering if this is a valid distinction, or, at least, one that could possibly prevent a conviction of crimen injuria in scenarios where a group, as a whole, is insulted, where the victim is part of that group.
Already a difficulty becomes apparent in trying to distinguish between an insult against:
1. an attribute of a group with which the victim associates; and
2. a group to which one belongs.
The vague distinction between these two seems to turn on the distinction between associating with an attribute of a group as opposed to being part of a group.
The question being begged here is what it means to be part of a group. It is at least possible to regard belonging to a group as associating with the attributes of that group. Arguably, under this description, to “belong to a group” may require that it’s members associate with more than one attribute of a group – but this does not seem to be necessarily true. A group may, conceivably, be defined by a single attribute which it’s members have in common. That being so, the distinction drawn in Tanteli is unsound.
Apparently the crux of the judgement is not in its attempt to distinguish between individual and group insults, but rather between insults that can be considered to strike at a superficial level as opposed to an insult which strikes at one’s core. As dear as our languages may be to us, it must in a different league to suggest that my language is inferior, as opposed to that I am inferior.
This is revealed or reinforced if one imagines what the outcome would have been if, instead of an insult to the language of the complainant to the effect that his language was inferior, the insult was to the effect that he was inferior. There can be little question that a conviction for crimen injuria would have, and probably should have, followed.
The example above also illustrates the point made above – that there can be no valid distinction based on an insult to a group attribute versus the individual. It would scarcely have helped the accused to have argued that she had only said that say, Afrikaners are inferior, as opposed to that Afrikaans is inferior. The proposed distinction between an insult directed at a group as opposed to the individual breaks down because it is clear that the true distinction is between an insult that may be hurtful, but superficial, compared with one that goes to the core of the person.
In conclusion, the only authority for the proposition that an insult must be directed at a specified person as opposed to a group appears to come from the case of Tanteli. This case is limited in its binding effect – to single Judge courts in Gauteng. It is apparently identified by the leading text on common law crimes in South Africa as an anomaly. Finally, the distinction described in the case between an insult to group to which the complainant associates him/herself as opposed to one directed at the individual appears to break down on analysis. The true distinction appears to be whether insult is superficial as opposed to one that goes to the core of a person’s sense of self worth. The Tanteli case, and the suggestion that an insult must be directed at a specified person, cannot possibly stand in the way of a conviction for crimen injuria in cases where an insult was directed against a group, where the insult offends the very core of the members of the group. In such a case, far from providing a defence that no individual was insulted, the insult is aggravated in that it offends an entire group of individuals.

Intoxication is recognised in South African criminal law as the basis for a defence. This remains the case despite legislation which has attempted to alter the position. However, even though intoxication is recognised as the basis of a defence, the principles of our criminal law are sufficiently robust so that, even if one escapes liability for “conduct” once drunk, one may nevertheless trigger liability for for one’s conduct before one became drunk or for one’s conduct in getting drunk.
Two points should be observed first. By intoxication I am referring to intoxication by any means: alcohol or drugs.
Secondly, I am not dealing here with offences in respect of which being intoxicated is a requirement of the offence, such as, being in control of a motor vehicle while intoxicated. Instead I am dealing with the question of whether liability for other offences, such as assault, murder or culpable homicide can be imposed, where, because of intoxication, a requirement of liability is absent.
Intoxication was recognised as a defence, or rather, accepted as a basis for a defence in 1981 in S v Chretien (AD). Until the case of Chretien, our law applied what was known as the “specific intent” approach, which allowed for the conviction of a person for a lesser crime even if, because of intoxication, an element or requirement of criminal liability of the lesser crime was absent. For instance, in 1969 in S v Johnson (AD) the accused was convicted of culpable homicide even though the court accepted that he was so drunk that his conduct was involuntary. The point was that the courts took issue with the reason for the defence raised – that a drunk person should not be less liable to the criminal law than a sober person. The problem was that this approach was utterly contrary to the principles of criminal law – that if someone is to be convicted of an offence, all the requirements must be met – it should not matter why a requirement is not met.
Remember also, before considering Chretien, that murder, or attempted murder, requires intention. Intention in our law is an entirely subjective enquiry. It is concerned with what the accused actually thought at the relevant time. In South African criminal law it’s range extends to dolus eventualis: where an accused foresaw the risk of harm, reconciled, and proceeded reckless as to whether the consequence eventuated or not (Ngubane 1985 AD; Humphreys 2012 SCA; Pistorius 2015 SCA).
Thus, even if one proceeds under the hope that the harm will not eventuate, if one nevertheless foresees that it may occur and, accepting that risk, one proceeds, one has intention in our law in the form of dolus eventualis. Also, if one does not care whether the harm eventuates, if one nevertheless foresees the harm may eventuate, and one accepts this risk and proceeds, one has intention in our law in the form of dolus eventualis. In scenarios where an accused hoped or did not care, an almost inescapable inference arises that the accused must have forseen the risk – precisely because s/he hoped it wouldn’t eventuate, or didn’t care if it did.
In S v Chretien the accused was at a party at a house where he got very drunk. He left in his car, and had to navigate the road in front of the house which was still crowded with party goers. He drove into the crowd, killing one and injuring five.
He was charged with culpable homicide of the one he had killed, and attempted murder of the five he had injured.
Culpable homicide is the negligent unlawful killing of another person. Murder is the intentional unlawful killing of another human being. Attempted murder also requires intention. You must intend to murder, even though, for some reason, you fail.
The trial court convicted the accused in Chretien on the culpable homicide charge. This was not controversial or at odds with the law of the time. What was controversial and at odds with the prevailing law (of specific intent) was the courts findings on the attempted murder charges.
On the specific intent approach, an accused charged with attempted murder, which requires intention, could be convicted of common assault, even though this offence (common assault) requires intention and even though the accused, because of intoxication, had no intention to assault.
In response to these charges, the accused argued that, because he was intoxicated, he truly believed that the people in the road would move out of the road. This was a claim of a lack of intention to kill or assault. It is crucial to note that the court accepted this – that the accused did expect the people in the road to move and did not have any intention to kill or assault.
The scene was set for a conviction of common assault – following the specific intent approach. However, the trial court found that it could not convict the accused of common assault because it could not convict someone of a crime which required intention when the accused did not harbour the required intention. It acquitted him of the five charges of attempted murder and of common assault. This decision was confirmed by the Appellate Division (now the Supreme Court of Appeal). The Appellate Division agreed that principle must prevail. If a crime requires intention, an accused cannot be convicted of the offence in the absence of intention.
It is crucial to recognise that Chretien did not recognise intoxication as a defence, but only that, if an accused lacks a requirement for criminal liability because of intoxication, this reason (intoxication) will be irrelevant.
This is crucial to recognise because the ordinary principles continue to apply. If one, despite being drunk, does foresee the risk of harm, accepts the risk and proceeds, one has dolus eventualis. The point is that being drunk, in itself, is not a defence.
The decision in Chretien was an endpoint in the development of SA criminal law in accordance with principles rather than considerations of policy and expedience, of arbitrary rules and equally arbitrary exceptions.
It led in turn to further development in the areas of involuntariness and incapacity and is the basis for the recognition in our law of the defence of non-pathological criminal incapacity (aka temporary insanity).
The Chretien case was not received well by the legislature and in 1988 it intervened – enacting the Criminal Law Amendment Act (1 of 1988) by creating a crime (in s 1(1)) of: being not guilty of any offence. More specifically, it created the offence of being not guilty of an offence because of intoxication. It was however crippled by its ill-considered structure and the bad choice of wording. It was so badly worded that it placed a criminal onus on the prosecution to prove that, as alluded to above, an accused was not guilty of any offence (see Andrew Paizes and Roger Whiting’s devastating analyses).
Despite being a response to the Chretien case in which the accused escaped because he lacked intention, the Act targets an accused who escapes liability because of a lack of capacity. In other words, the Act does not even address itself to the problem it was created to solve.
Even worse, it places the prosecution in the invidious position that, while on the main charge of, say, assault, it has to prove, beyond reasonable doubt that the accused had capacity, if the court was not convinced and acquitted the accused, the prosecution had to, under s 1(1) of the Criminal Law Amendment Act, return to court and contradict everything it had argued up to that point regarding capacity. It has to argue that the accused lacked capacity beyond all reasonable doubt. Thus it has to go from arguing that the accused had capacity beyond all reasonable doubt, to arguing that the accused actually lacked capacity beyond all reasonable doubt.
Understandably then, while there have been convictions under this Act, there have been few, if any, convictions that have withstood the scrutiny of a review or appeal. It was bound to fail in any event given that it’s premise is to impose liability on someone who, on fundamental principles of criminal law, is not guilty of any crime.
This does not mean that one may rely on intoxication without consequence. This is because if all the requirements for liability line up and are present at the time that you start drinking, you may be convicted of committing the offence at that moment in time. This is known, originally as the doctrine of actio libera in causa (to liberate oneself in a cause), and now, as antecedent liability. There is no magic in this although it is often misunderstood. It is often misunderstood to be an exception to the rule that all requirements must be met for liability to be imposed. It is not an exception, instead it is actually an application of the principles that all requirements must all be present – but that whenever they are, you incur criminal liability. Let’s assume I decide to get very drunk, park my car on a hill above a busy road that I know will be packed with pedestrians when I leave, I leave the car unlocked because I know that when I return I will be too drunk to unlock the door and I go and get drunk. If, then, on my return, I manage to stumble into the car and release the handbrake, after which I pass out, I cannot escape liability for, at least culpable homicide, even though, at the moment my car knocks into and kills pedestrians, I was in a state of involuntariness. Our law will look backward in time for antecedent liability. Whether there was a moment in time, when I was acting voluntarily, possibly when I returned to the car, or even when I left the car, which is causally linked to the death of the pedestrians and at which time I was negligent – for culpable homicide. Arguably, there are two such moments: when I returned to the car and when I left it. It is even arguable that I may have forseen the possibility of causing the death of pedestrians and thereof be guilty of murder – again, not because of anything I did after I passed out – but because of what I did and thought before that.

Thus, in theory, intoxication in our law is a valid defence, although subject to the virtually crippled Criminal Law Amendment Act. However, this is no licence to drink and misbehave, because doing so naturally attracts attention to what you thought and did before you became too drunk to be accountable to the criminal law. If you expect that you will misbehave or even if you ought to know that you may, you would make yourself a candidate for antecedent liability.

By James Grant

Tomorrow’s Supreme Court of Appeal judgement in the Pistorius appeal is crucially important for South African law of evidence and criminal law.
It will decide whether the State can appeal on a question of law even if the trial court convicted the accused on a competent verdict (such as culpable homicide when the main charge was/is murder). The decision in the 1982 case of S v Seekoei (AD) precludes such an appeal and before the SCA addresses any issue of any substance, it will have to decide whether Seekoei was wrong. Only if it decides that Seekoei was indeed wrong, can it engage with any of the substantive questions before it.
Assuming it decides Seekoei was wrong, we should expect clarification on the following issues:
  1. How circumstantial evidence must be dealt with – whether a piece of circumstantial evidence can ever be ignored and, if so, when.
  2. The distinction between questions of fact as opposed to questions of law. This distinction is not easy particularly when it relates to rules that govern what a court may do – that is, in respect of rules evidence and procedure. It may be helpful (following decisions such as Basson CC) to regard questions of fact as an attempt to identify a state of reality – for instance, whether someone did, or thought something. Questions of law pertain to the standard of conduct that one must observe – what it is that a person must not do or must do. This seems to apply equally to procedural issues, such as what a court must and must not permit in proceedings before it.
  3. It can also be expected that the court will define and explain the difference between a defence of involuntariness as opposed to one of putative (mistaken) private/self defence. If the court follows the decision in De Oliviera, it is likely to distinguish the two on the basis of whether, for putative private defence, the accused was thinking, but mistaken in his thinking; whereas, for involuntariness, the accused was not thinking at all, or, at least, his thoughts did not direct his conduct.
  4. It will also clarify the concept of dolus eventualis, in particular, as applied to scenarios of error in objecto and scenarios of putative private/self defence. In other words, the SCA will clarify the requirements for an accused to escape a conviction of murder (or any crime requiring intention), in a scenario in which s/he was mistaken both as to the identity of the actual victim and as to whether s/he was actually entitled to kill that person.
  5. We will hear whether the court confirms that, a defence of putative private defence must fail, when dolus eventualis is sufficient for a finding of intention (as it is for murder), if the accused had foresight of the possibility, at the critical time, that any of the requirements of private/self defence may not be have been satisfied. The significance is that the true defence (that is, not the “putative” mistaken defence), requires, for the defence to succeed, that:
    1. The accused was under an unlawful attack;
    2. That the attack had commenced or was imminent;
    3. That the attack was directed at an interest of the accused which he was entitled to defend;
    4. That force was necessary to repel the attack;
    5. That no more force than was necessary to avert the attack was used, and that in any event, such force was reasonable given the circumstances; and
    6. The force used was directed at the attacker.
A defence of putative private defence is that, while there was, in reality, no valid claim of private defence, the accused was mistaken so that s/he believed that all requirements were satisfied.
We should expect the SCA to confirm that the defence of putative private defence must fail if there is dolus eventualis (forsight of the possibility and reckless persistence) relating to any of these requirements. In other words, that the defence must fail if the accused foresaw the possibility (and persisted nevertheless), that any of the requirements may not have been met. Thus, the defence must fail if the accused foresaw the possibility of all or of any of the following:
  1. That s/he was not under attack;
  2. That no attack had commenced or was imminent;
  3. That there was no attack on a legally protected interest that s/he was entitled to defend;
  4. That no force was necessary;
  5. That the force used was excessive in the circumstances; or
  6. The force used was not directed at the attacker.

If the court confirms this, the claim of putative private defence must fail.

There is alot at stake in the Pistorius appeal decision. Anyone interested in criminal law and procedure and the law of evidence will do well to pay close attention to this judgement.

The recommendation of the Farlam commission that some police members can only be charged with attempted murder because it could not be proved which officers had shot which miner is, on the law as it stands, wrong.

It is important to extract the facts that this recommendation appears to concede: that these officers unlawfully fired at the miners with the intention to unlawfully kill them. This places in issue only the link between the conduct of each officer and the death of any miner.

In circumstances such as these, our law resorts to the doctrine of common purpose. It is a controversial doctrine and was certainly abused under apartheid. For what it’s worth, I am opposed to its use and believe that there are adequate alternatives in our law (see James Grant Common Purpose: Thebus, Marikana and Unnecessary Evil 2014 1 SAJHR 1). Nevertheless, common purpose is relied upon in our criminal courts on a daily basis and it has been sanctioned by the Constitutional Court (in the case of Thebus v S 2003 (6) SA 505 (CC)). To my knowledge, the doctrine has never been invoked against the police, but there is nothing in principle why it cannot be used.

The doctrine is designed for exactly the scenario that Farlam says prevents a murder conviction – when one cannot be certain, which one, of a group of attackers, actually killed a victim.

The doctrine holds that each person acting together with others in an unlawful attack, is to be regarded as having done whatever everyone did. So, if A, B and C, agree to attack and kill Y, and that A will stand as lookout, while B will strike Y with a stick, and C will shoot Y, it does not matter that C, in shooting Y, causes his immediate death, all will be liable to a murder conviction for the death of Y. It is irrelevant that B’s conduct (with his stick) only inflicted minor and insignificant wounds, nor even that A is causally irrelevant to the death of Y. Indeed, both A and B are causally irrelevant to the death of Y. Nevertheless, on our law, as it stands, all three may be convicted of murder under the doctrine of common purpose. It is equally irrelevant that it cannot be established which police officer caused the death of which miner.

It is helpful to understand the work that common purpose actually does. It allows our courts to pretend that whatever anyone in a common purpose does, is what everyone does. Thus, on the example above, A does, on our law, what he does, but also whatever B and C does. And so with B and C. B does what he does, but also what A and C does. And so on.

If a police officer can be convicted of attempted murder for his/her part of the attack on the miners, they are almost invariably liable to a conviction of murder – on an application of common purpose. Every police officer who triggers the operation of common purpose, did whatever every other police officer did in that common purpose. Thus, an officer who fired but perhaps missed, also did what his fellow officers in the common purpose did – some of whom fired, struck, and killed the miners – but it doesn’t matter who.

It is true that common purpose has its own requirements and that there are two forms of it. In this case, recognition of these different forms and the requirements of each reveal that it is not only those officers present on the scene who may be convicted of murder, but that liability for murder may stretch up, to anyone who was a party to the plan to resort to the ‘tactical option’, and who foresaw that anyone may be unlawfully killed in the execution of that plan.

The two forms of common purpose are known as prior agreement and active association. Common purpose by prior agreement for murder requires that the parties must, at some prior meeting, foresee the possibility that people may be unlawfully killed upon the execution of their plan (S v Madlala 1969 (2) SA 637 (A); S v Nzo 1990 (3) SA 1 (A)). It is certainly possible that at all and any meetings that discussed the ‘tactical’ option, the unlawful death of miners was foreseen by some. Whoever in those meetings had this foresight – on this form of common purpose – are regarded as having done what they foresaw may happen – even if they were nowhere near the scene at the time of the killing. This has serious consequences, because it allows liability for murder to creep up the chain of command, restrained only by what was foreseen as possible.

The second form of common purpose, active association, is resorted to when the first, prior agreement, doesn’t apply. It is possible that many of the officers who fired their weapons on the day, were at a previous meeting at which they foresaw the possibility of unlawful killing, and thus already satisfied the requirements for common purpose to operate. However, if not, they may have formed a common purpose (by which they can be held liable for an unlawful killing by another officer), by active association, if (Mgedezi 1989 (1) SA 687 (A); Thebus above): 1) the officer was present on the scene; 2) s/he were aware of the attack; 3) s/he intended to be part of the attack; and 4) s/he did something to manifest his/her sharing of the common purpose.

Before we consider how this may apply to the scenario at the scene we must remember what Farlam concedes: that some officers unlawfully fired at the miners and intended to unlawfully kill them. Assuming this, the requirements of common purpose by active association may well be satisfied, as follows: 1) The officers who fired at the miners were present on the scene; 2) They must have known of the unlawful attack on the miners; 3) they must have intended to be part of the unlawful attack on the miners; 4) they did manifest their sharing by virtue of firing their weapons. Thus, even if they are not liable by prior agreement, they may well be liable to a conviction of murder by operation of common purpose by active association.

This is not a suggestion that all officers on the scene are guilty of murder, nor even that those who discharged their weapons are necessarily guilty of murder. It is the argument that if, as the Farlam recommendation appears to concede, some officers unlawfully fired at the miners with the intention to unlawfully kill them, the absence of a link between the conduct of each officer and the death of any miner can be cured by an application of common purpose. It means that if any officer could be convicted of attempted murder, s/he could be convicted of murder. Of course, the opposite is true, but is worth noting: if an officer cannot be convicted of, at least, attempted murder, there will be no basis on which to convict him/her of murder, by an application of common purpose.

Thus, by an application of the doctrine of common purpose, any officer who could be convicted of attempted murder, may be liable to a conviction of murder. Furthermore, liability based on common purpose for any murder may easily creep up the chain of command to anyone in any planning meeting who foresaw that their plan may result in unlawful killings.

– James Grant

Euthanasia or assisted suicide is arguably, now, both legal and illegal in South Africa. Here’s an attempt to explain. Three preliminary points are required.

Firstly, it’s important to understand that what is at issue is whether killing in euthanasia is unlawful (just like for self/private defence). It is a question of unlawfulness because the issue is whether the consent given to be killed is valid and recognised in our law.

Secondly, only decisions of our courts that are consistent with the Constitution actually reflect what the law is. Any decision inconsistent with the Constitution is invalid and not law. The importance of the Constitutional Court is that it has the last say on what it is that the Constitution permits or requires. This has implication for what the law is, even before the Constitutional Court decides what it is. The law is whatever the Constitution permits or requires ever since the coming into effect of the Constitution. This is crucial and bears repeating. Since the coming into effect of the Constitution, our law has been whatever the Constitution permits or requires, no matter what any legislation or any Court (other than the Constitutional Court) says it is.

Thirdly, the question raised in cases of euthanasia or assisted suicide is judged by the “legal convictions of the community”, now, as informed by the values in the Constitution (Carmichele [2001] ZACC 22).

Currently our law is in a state of conflict. There are several old Appellate Division (AD), now known as the Supreme Court of Appeal (SCA), decisions that euthanasia is unlawful (Robinson 1968; Grotjohn 1970).

There is also the case – progressive for its time – of Clarke v Hurst (1992 D) and now, that of Stransham-Ford in the Gauteng North High Court (of 29 April 2015), holding that euthanasia is lawful. Both are Provincial Division decisions.

To complete the picture, it must be mentioned that the Law Commission has recommended (in 1999) that passive euthanasia/assisted suicide be legalised – in terms of which it will be lawful to withhold care or medication. However, the Law Commission took no position on the issue of active euthanasia/assisted suicide – where one actively does something, such as administer a lethal dose of a drug, to hasten death. The status of this commission report is that it has gone nowhere and certainly doesn’t have the status of law. It does however give us a glimpse that the law commission is leaning in favour of the recognition of euthanasia/assisted suicide.

Returning to the court decisions we have the Provincial Division decisions which are in conflict with the AD/SCA decisions. Although AD/SCA decisions ordinarily override Provincial decisions, as indicated above, only decisions that are consistent with the Constitution actually reflect the law.
This leads to the obvious question here: which decisions reflect the Constitution? For this reason we need, at the very least, an appeal to the SCA to resolve the conflict between its own old decisions and the Provincial division decisions, but ultimately we need the Constitutional Court to tell us what the Constitution says. An appeal against the decision in Stransham-Ford’s case is therefore to be welcomed. It will drive the issue towards the Constitutional Court.

Until then, and despite the recent decision in the case of Stransham-Ford, no one can claim to know whether euthanasia or assisted suicide is lawful or not. It leaves those who are terminally ill, who are suffering and who want to die, their families and the doctors who want to help, in a state of uncertainty that can only cause more suffering.

In the meantime the inevitable question persists – what is the position in our law? I can’t answer that. This is for the Constitutional Court to say. I can only guess that it will give prominence to an individual’s right to dignity and to die with dignity.

What I can say in the meantime is what this state of uncertainty does for criminal liability. It is worth considering the possibilities given the importance of this question. There are, of course, two possible outcomes: that euthanasia or assisted suicide is either: 1) lawful; or 2) unlawful.
If the Constitutional Court finds that euthanasia/assisted suicide is lawful, it will mean that it was always lawful, since the inception of the Constitution. In this scenario, there can be no liability – no murder or culpable homicide conviction, even if the conduct is/was committed prior to the Constitutional Court decision saying so. There is an interesting scenario, still within the parameters of it being lawful, in which a person proceeds on the mistaken though firmly held belief that it is unlawful, although it is not. This is what is known as an attempt at the legally impossible, and also attracts no liability.

The second alternative, in which the Constitutional Court declares that euthanasia/assisted suicide is unlawful, will mean that it remained unlawful despite the inception of the Constitution.
I expect some may regard this as a silly proposition given that “ignorance of the law is no excuse”. However, in South Africa, since 1977 this is not our law (S v De Blom AD). A mistake of law is indeed a defence. It would undermine the requirement of intention for murder, and, probably, also the requirement of negligence for culpable homicide. To be clear, the effect of a mistake is to undermine the fault requirement of murder or culpable homicide.

If someone proceeded to kill another in circumstances that one can describe as euthanasia/assisted suicide, but knew/believed that it was unlawful, s/he could be convicted of murder.
However, if someone were to believe – mistakenly on this scenario – that euthanasia/assisted suicide was/is lawful, and was to proceed on this basis, s/he cannot be convicted of murder. This follows from simple and unquestioned principles of mistake in our law. The more difficult question arises though whether this person could be convicted of culpable homicide. This is the question whether making such a mistake is reasonable – could the reasonable person make this mistake? Remember, unlike intention, which is a subjective enquiry (what was the accused actually thinking), negligence is an objective enquiry. Negligence is concerned, not with what the accused did think, but with what s/he should have thought and done. The bottom line is that an accused is regarded as knowing only as much as would be reasonable for him/her to know.

In cases where the issue has been an error of law our Courts have distinguished between complete lay people and those who engage in a particular sphere of activity.  Those who engage in a sphere of activity are expected to know the law pertaining to that sphere of activity (De Blom 1977 AD; Du Toit 1981 C). Thus, a builder is expected to know the law and regulations relating to building, a motorist is expected to know the law and regulations relating to his or her mode of transport). This follows from the notion that the reasonable person would know the law pertaining to the sphere of activity in which s/he is engaged. It would probably be considered reasonable for a lay person to think that, given the right to dignity in the Constitution, or following the decision in Stransham-Ford, euthanasia/assisted suicide is now legal in South Africa.

The standard required of Doctors is going to be higher. They will be expected to know the law pertaining to their “sphere of activity”. In the circumstances what can the reasonable Doctor be expected to know? The law on the issue is unclear and we are waiting for the SCA and ultimately the Constitutional Court to interpret the Constitution and declare the law (and what it has been since the inception of the Constitution). We may guess that the Constitutional Court will probably recognise the right to die with dignity, but we have no certainty until they decide. What should a Doctor do? Ultimately the question is going to be this: would a reasonable Doctor foresee the significant possibility that the Constitutional Court may find that euthanasia/assisted suicide is unlawful? Is there a significant possibility that the Constitutional Court may find that it is unlawful? Regrettably, as much as I think the Constitutional Court should and will find that euthanasia/assisted suicide is lawful, there remains a chance that it will not – that it may find that it is unlawful. I say regrettably because I have to acknowledge a degree of uncertainty here that is unacceptable. If this is so, that there is a possibility that the Constitutional Court would find that euthanasia /assisted suicide is unlawful, it will not be a stretch for a Court to find that a reasonable doctor would have foreseen this prospect and would not have assisted. Regrettably again, given the uncertainty, I expect this is a basis upon which Doctors may pause in anticipation of the desperately needed clarity on this issue. However, much as I must acknowledge the possibility that the Constitutional Court may find euthanasia/assisted suicide to be unlawful, I doubt it could amount to a significant risk. The only case (that of Stransham-Ford) that has considered this question in light of the Constitution has concluded that the Constitution permits euthanasia/assisted suicide as lawful. Indeed, even pre-constitutionally, but as we approached our new Constitutional dispensation in 1992, the court in Clarke v Hurst found that the “legal convictions of the community”, the all-important test for unlawfulness, require that euthanasia/assisted suicide is lawful. Given these two judgements, that the AD/SCA decisions are old (1968 & 1970) and certainly pre-constitutional, and that the only other indication we have is from the law commission apparently in favour of euthanasia, it would seem entirely reasonable for a Doctor to believe that when this does finally reach the Constitutional Court, that Court will find that it is lawful. What would a reasonable Doctor make of the uncertainty? I expect a reasonable Doctor would acknowledge the uncertainty, would decry the uncertainty, but would, in my view, regard the risk that the Constitutional Court could find that euthanasia is unlawful as too remote and ultimately insignificant. Unfortunately, this is only, and can only be, a best guess.

In summary then, despite the conflicting court decisions in our law, whether euthanasia/assited suicide is unlawful or not is to be determined by reference to the Constitution. If anyone (lay person or Doctor) were to assist another (who is terminally ill and suffering) to die, no liability can attach to that conduct were the Constitutional Court to declare that such conduct is lawful, and always was (since the inception of the Constitution). If the Constitutional Court were to find however that such conduct remained unlawful despite the inception of the constitution, the fault requirement of the crimes of murder and culpable homicide may offer a defence to both lay persons and Doctors. As indicated, Doctors bear a higher burden than lay people in respect of the law and regulations applicable to the practice of medicine. But ultimately it must be recognised that right now, no one can claim to know what the law is and all we have to go on are the indications in the law commission proposal and the two provincial division decisions in favour of euthanasia being lawful.

Ultimately, and assuming whoever assists another to commit suicide genuinely believes that it is the right thing to do and that this belief will be vindicated by the Constitutional Court, s/he cannot incur liability for murder, no matter what the Constitutional Court decides. Assuming it is reasonable to hold this view, as I suspect it is, no one can be convicted of culpable homicide for doing so, again, no matter what the Constitutional Court decides.

These conclusions are by no means an attempt to reduce the importance of a decision on this from the Constitutional Court. It recognises that that decision is desperately needed and ultimately all important. In the meantime though, for those who are suffering, with no prospect of recovery, for their families and loved ones, and for the Doctors who want to help, I hope this answers some questions.

– By James Grant (@JamesGrantZA)

___________________________

* In honour of Robin Stransham-Ford and my mother. I am grateful to Professor Marius Pieterse of Wits University for his guidance on the constitutional law discussed here. All errors remain, naturally, all mine.

** This discussion is an attempt to answer some abstract questions that arise given the apparent state of legal conflict. It’s purpose is to provide a framework within which questions relating to the specific circumstances of any particular person may be considered. It is not encouragement or a licence to kill anyone. It cannot be regarded as legal advice which must be sought in every individual scenario in order for the specific circumstances to be considered and advised upon.

My apologies that I have made no contribution to this site for some time. I am currently doing my Advocate’s pupillage at the Johannesburg Bar and undergoing intense training that permits me little to no time to add to this site. I do intend to continue with the online textbook once I am through pupillage and hope, in the meantime, to post the odd note if I can find the time. In the meantime, I apologise for the break in transmission.

– James

The Seekoie Hurdle

Posted: October 29, 2014 in Uncategorized

Translated from the Beeld (29 Oct 2014), available at: Seekoei.Beeld

Story by Marida Fitzpatrick, translated (in a flash) by Darron West.

*** This is a revised and substantially expanded version of the post “Unsuccessful attempts to Justify Judge Masipa’s Errors”. It is organised around themes rather than individual commentators. ***

Introduction

There is something deeply disturbing about many of the attempts that have been made to justify Judge Masipa’s judgment in the Pistorius case. They seem to take the line that her conclusion of culpable homicide was conceivable on the law as it stands, and therefore, although her reasons were questionable, her conclusion was correct. It’s the argument that the end justifies the means. This type of justification is as problematic in law as it is in most other areas of life.

In law it is a licence to introduce uncertainty in our law, or worse, distort or destroy the proper functioning of the principles. Perhaps even worse than that, it seems to reconcile with the notion that the principles don’t matter, as long as the right conclusion was reached. This is the worst possible implication because it reconciles with a lawless society. We must not accept this. It is not even the thin edge of the wedge – it is the wedge, the whole of the wedge.

Another common theme can be observed – related to the point made above. It’s the notion that there was nothing wrong with Masipa’s reasons because they are in line with a defence raised even if she did not actually apply the defence or principles underlying the defence. This is all very well, except that if the Judge did not actually rely on that defence or the principles underlying the defence, we cannot pretend that she did. This is law by wishful thinking. I’m not sure it’s as bad as lawlessness. But if lawlessness was bad, this is madness.

At this point it doesn’t serve any purpose to attempt to respond to the arguments raised in justification of Masipa’s judgment on an individual basis – that is, organised by individual attempt. It seems more sensible to extract the common themes from comments made (to date) and respond to those.

It is necessary to explain what dolus eventualis, dolus indeterminatus, error and objecto and putative private defence are, before I move on to explain what they are not. All of this is easier to understand when illustrated by the crime of murder – which is the intentional unlawful killing of another human being.

To confirm – we only have one single crime of murder in our law. Premeditation relates to bail or sentence. Also, no matter what form of intention you have, if you have any form, and if you unlawfully kill another human being, you are guilty of murder.

For authority, other than case law, I will refer to and quote from the textbooks of, arguably, the two leading authors on South African criminal law: Burchell (Principles of Criminal Law, 2013, 4th ed) and Snyman (Criminal Law, 2008, 5th ed). Both have also commented on the judgment in the Pistorius case, which I will discuss below.

Dolus Eventualis

Dolus eventualis is the widest form of intention known in our law. It exists (on the law as it stands: Ngubane 1985 AD; Humphreys 2013 SCA) where the accused, at the time of the killing: 1) foresaw the possibility of killing a human being; 2) accepted this risk; and 3) proceeded nevertheless.

Quite correctly it has been observed that, originally, Molemo Maarohanye (known as Jub Jub) was convicted of murder because he was found to have foreseen the risk of killing his victims, and to have accepted/reconciled to that risk. This conviction has now, on appeal, been reduced to one of culpable homicide and a sentence of 8 years imprisonment imposed. Humphreys (2013 SCA) was convicted only of culpable homicide. The facts of Humphreys are, briefly, that Humphreys drove a taxi and was transporting children. He approached a railway crossing where the boom had come down to stop traffic – in anticipation of a train that was about to pass. Humphreys switched to the right hand lane, overtook the waiting cars, and the boom, and begun to cross the railway line – directly in front of the oncoming train. The train struck the taxi killing 11 of the children being transported. Humphreys was charged with the murder of the 11 killed, but convicted of culpable homicide because the court found that, although he foresaw the risk that the children in his taxi could be killed, he did not accept the risk. This is an exceedingly controversial decision. The court’s reasoning seemed to be that he could not have accepted the risk, because, since he was driving the taxi, he too would be killed, and there was no evidence that he was suicidal. The apparently strange reasoning arose out of an application of the law of dolus eventualis.

The definition of dolus eventualis has been controversial for years – since the decision of Jansen JA in S v Ngubane (1985 AD). The definition of dolus eventualis adopted in Humphreys was set out in Ngubane. The definition was subject to devastating academic criticism. Nevertheless, this criticism was not even mentioned in Humphreys, when the court uncritically adopted the Ngubane definition of dolus eventualis.

However woeful our law is in respect of the requirements of dolus eventualis, Masipa’s judgement did not turn on a finding – that was crucial in the Humphrey’s case – that, although the accused did foresee the possibility of killing whoever was behind the door, he did not accept that risk.[1] Again, that would have been a valid basis (on our law as it stands) to acquit the accused of murder. However, from the wording of her judgement, this was not the reasons she gave for acquitting the accused of murder.

It is also notable that the defence never argued (as some commentators are now suggesting) that the accused had not intended to kill – but only, perhaps, at most, intended to injure. This would have been a valid defence against murder, but it was not a defence claimed, and certainly not one – from the wording of her judgement – relied on by Masipa as a basis on which to acquit the accused of murder. I don’t see how it could possibly make Masipa’s judgement right to refer to a basis on which she could have relied, but did not.

Identity: Dolus Indeterminatus and Error in Objecto

Our law distinguishes one’s identity in the sense of one’s body as opposed to someone else’s body (which I will refer to as object/body identity) from one’s identity in the sense of your name and other characteristics, such as your personality (which I will refer to as nominal/name identity). Nominal/name identity is irrelevant in our law – just as the colour of a victim’s clothing is irrelevant on a charge of murder. You cannot raise the defence that you thought the victim was wearing a yellow shirt whereas it turns out to be blue. In the same way, you cannot raise the defence that your victim’s name happens to be Peter, and not John.  If you have intention to unlawfully kill a human body, and you succeed in unlawfully killing that human body, you have committed murder in SA law.

This distinction and the irrelevance of nominal/name identity is borne out by our law in recognising that intention may be general (dolus indeterminatus), and in the attitude our law has taken to scenarios of error as to object (error in objecto).

Dolus Indeterminatus

Dolus indeterminatus is at play in scenarios where the accused throws a bomb into a crowded room. The accused may have no idea who is in the room (at the very least in the sense of nominal/name identity),[2] but intends to kill whoever is in the room. The accused cannot raise the defence that s/he did not intend to kill because he had no specific (nominal/named) victim in mind. (Burchell (4th ed, 2013) explains dolus indeterminatus as follows:

“Where a person throws a bomb into a crowd or derails a train, the fact that he or she has no particular intention to kill a particular individual in the crowd or upon the train does not mean that the person lacks intention. Since he or she knows or foresees that someone will die, this person has what is called ‘general’ intention or dolus indeterminatus. Dolus directus, indirectus or eventualis may be general or indeterminate intention.”[3]

The point here is that it is no defence to say you had no particular victim in mind. On this form of intention, the accused need not have known who was in the room at all, but will be held to have intention to kill nevertheless.

This principle extends liability. It excludes and denies a defence of mistake as to identity, at least in the sense of nominal/name identity. There is no basis for any suggestion that it somehow permits for exceptions. The rule is that you will not be able to validly argue, when you killed B, C and D, that you did not know that B, C, and D were in the room. There appears to be no basis on which it may be argued that somehow this rule would or should yield if one had some reason for thinking that, say, D was not in the room.[4] It will not be a defence for an accused to argue that s/he only thought that B and C were in the room, or that B, C and E were in the room. The point is that it doesn’t matter who was in the room. It certainly cannot matter that an accused thought that B, C and E were in the room, because the accused thought that D was not in the room.[5]

It follows necessarily that no matter what reason Pistorius might have had for thinking – if indeed he did – that it was someone else behind the door, and not the deceased, this cannot be a defence and it is no ground upon which to find that he lacked intention to kill.

This is confirmed by our law’s approach to problems that are far more like that of the facts of the Pistorius case. In our law, if one aims a shot at a particular human body (identified by a particular object/body), fire at and kill that particular human body, it is no defence if it transpires that that human body has a different name to what one expected. This is the essence of the law on error in objecto (error as to object).

Error in Objecto

Burchell (4th ed, 2013) distinguishes aberatio ictus (going astray of the blow – missing one’s target) scenarios, from error in objecto scenarios as follows:

“Aberratio ictus must be distinguished from error in objecto, since difficulty arises if the two are confused.  Error in objecto occurs where A, intending to kill B, shoots and kills C whom he mistakenly believes to be B. In these circumstances A is clearly guilty of the murder of C. A’s intention is directed at a specific predetermined individual, although he is in error as to the exact identity of that individual. In other words, he intends to kill the individual regardless of whether the name of the individual is B or C. There is thus in the case of error in objecto, so to speak, an undeflected mens rea which falls upon the person it was intended to affect. The error as to identity is thus … irrelevant to the question of mens rea. On the other hand, in the aberratio ictus situation A intends to kill B, but misses him and kills C. A’s intention is ‘directed at one whom he knows and recognises to be [B].  It is through unforeseen and unintended factors that the blow falls upon [C]’. It follows that A has intention in respect of C only if he foresaw the possibility of C’s death or, for culpable homicide, if C’s death was reasonably foreseeable.”[6]

Snyman (5th ed 2008) gives a similar explanation:

“What is the position if X intended to shoot Z but it subsequently transpires that he mistook his victim’s identity and in fact shot Y? Here his mistake did not relate to whether it was a human being he was killing, but to the identity of the human being. Murder is committed any time a person unlawfully and intentionally kills a human being, and not merely if a person kills that particular human being who, according to his conception of the facts, corresponds to the person he wanted to be the victim. For this reason X in this case is guilty of murder.”[7]

There can be no question that the facts of the Pistorius case disclose an error in objecto scenario – assuming that Pistorius did not know it was Steenkamp behind the door. He aimed at whoever (whatever object/body) was behind the door, fired, and struck and killed that object/body. He did not miss – which would have brought into play the rules relating to aberratio ictus.

It is worth noting that neither author (Burchell or Snyman) make any mention of the reason why A thinks that C is B. This is in line with our settled law, that takes no account of the reason why A makes his/her mistake. If (nominal/name) identity is irrelevant, the reason for any error as to identity cannot be relevant. The identity of the actual victim is irrelevant as is, necessarily, any reason for a mistake as to identity.

Both Masipa and Burchell[8] correctly observe that the facts of the Pistorius case disclose a scenario of error in objecto – in which the nominal/name identity of the victim is irrelevant.[9]

After noting that, in error in objecto scenarios, the identity of the actual victim is irrelevant, Masipa asks what she says is required by the test of dolus eventualis (legal intention): Did the accused foresee the possibility of killing the deceased. This is the wrong question for two reasons. As discussed below, the accused’s defence was that he did not intend to unlawfully kill anyone. This focus – the very basis of the defence of putative private defence – is lost. But, putting that aside, secondly, it is wrong because it deviates from, even contradicts, the recognition that the facts disclose an error in objecto scenario and that the identity of the victim is irrelevant. The question then, correcting also for the defence of putative private defence, should have been: did the accused foresee the possibility of unlawfully killing whoever was behind the door.

Again, it does not help to say this cannot be true because he thought the deceased was in the bedroom. This is the argument that Masipa seems to have finally settled on – and it is fatally flawed in logic. She says: the accused could not have foreseen the possibility of killing the deceased, or anyone else for that matter, because he thought the deceased was in the bedroom. The fact that the accused thought that the deceased was in the bedroom says nothing about what he thought about the presence of someone else in the toilet. Indeed, it is his defence that he believed someone else was in the toilet.

Burchell perpetuates this error in logic. Referring to his scenario in which he notes that it is irrelevant that A intends to kill B, but kills C, he argues that there can be no intention to kill C if A had excluded in his mind the possibility that the person he thinks is B, could be C. But this is the definition of an error in objecto scenario – let’s take this slowly: A intends to kill B. He does not intend to kill C, but B. He thinks that B (the human body) is, well, B. In thinking that B is B, he does not think that B is actually C. So thinking that B is B and not C, he has excluded in his mind, the possibility that B is C. This is the definition of an error in objecto scenario and there is no dispute that the identity of the actual victim is irrelevant. That is, it cannot help the accused that he thought that B (the human body) was B – where he did not think that B was C. Therefore, it cannot assist an accused who has excluded from his mind the possibility that B is C.

Importantly, the reason for any error as to identity is also irrelevant. It cannot help an accused who thinks that B is B because B is not C. Our law is clear, in error in objecto scenarios, (nominal/name) identity is irrelevant, and it remains irrelevant no matter what an accused’s reasons are for making a mistake as to the identity of his victim.

Masipa’s error was similar. In her view the accused could not be convicted of murder because the accused did not foresee that he could kill C (or B), because he thought C was in the bedroom. In that case, it leaves him thinking that B was in the toilet. As we know from the rules of error in objecto, it is no defence to say that he thought that B was B, and not C.

Some have argued that the charge of murder was somehow defective and restricted the prosecution to prove that Pistorius knew (or foresaw) that it was specifically Steenkamp behind the door. Phelps has argued that: “ … the problem started off with the way that the indictment was drafted. It focused closely on him shooting at Reeva Steenkamp and not on whoever was behind the door”.[10] Yet the indictment read as follows, and could not conceivably have said anything else:

COUNT 1 – MURDER …IN THAT … the accused did unlawfully and intentionally kill a person, to wit, REEVA STEENKAMP, a 29 year old female.

Given that the crime of murder is the unlawful intentional killing of another human being, it is difficult to understand how this restricted the prosecution to having to prove that the accused knew it was Steenkamp behind the door. They name the human being that he did unlawfully kill. On the settled law of error in objecto – where nominal/name identity is irrelevant and only object/body identity matters – all they had to argue was that Pistorius unlawfully and intentionally killed the body of Reeva Steenkamp. Reading in the technicalities, the indictment reads as follows: the accused unlawfully and intentionally killed the human body of Reeva Steenkamp, a 29 year old female. It is not necessary however to state the charge in this peculiar fashion, because the settled law of error in objecto requires that it be understood that way.

Putative Private Defence

It is true that the accused’s (original) defence was/is putative private defence.

Private defence excludes the unlawfulness of one’s conduct – only when one is objectively actually (in reality) under and unlawful attack that has commenced or is imminent. If so, one is entitled to use necessary and reasonable force against the attacker.

If, in hindsight, one was not actually under attack, one may not rely on private defence to render your conduct lawful – however, our law recognises that if one was mistaken in thinking one was under attack and entitled to use the extent of force one does resort to, one may raise the defence of putative private defence. If the accused makes this mistake, whether he is mistaken about the facts, or (as Burchell[11] correctly adds) the law (De Blom 1977 AD), the accused cannot be convicted of murder. This is because an accused who is mistaken in this way has no intention to act unlawfully. Putting it positively, the accused believes s/he is acting lawfully. The enquiry is entirely subjective – what did the accused actually think or foresee? It is, as indicated, a complete defence to murder.

Pistorius was not actually under attack, and therefore he could not rely on the defence of private defence. He could however, and did, rely on a defence of putative private defence. This defence was the original defence raised by Pistorius and persisted in by his defence team in the alternative. This defence required that Masipa asked whether the accused foresaw the possibility that his conduct could unlawfully kill.

The problem is that, after specifically stating that she was turning her attention to this defence (of putative private defence), Masipa erroneously states that the question is whether the accused foresaw that his conduct could kill [full stop]. As discussed, this is not the right question.

It does not make her decision right to say, as many commentators have,[12] that if Masipa had asked the right question, she would have (possibly) arrived at the same or right conclusion. We cannot know what conclusion she would have come to if she had asked a different question.

It also does not follow that – as Snyman is reported to argue[13] – a defence of putative private defence somehow excludes a scenario from also being an error in objecto. Putative private defence describes the accused mental state – that s/he was mistaken in believing that whoever X (a human body) happens to be, X is about to attack. An error in objecto is an error as to who X actually is – B or C. These mistakes may obviously be related: an accused may have mistakenly believed s/he was under attack because s/he believed that X was B (who posed a threat), whereas it transpires, X was actually C (who posed no threat). This is indeed what Pistorius is claiming – and it invokes the principles relating to putative private defence and error in objecto.

Conclusion

In conclusion, we have seen several attempts to justify Masipa’s judgment, all of which seem to fail. It does not make Masipa’s judgment right to pretend that her reasons were reasons that would make her judgement right: that she decided that the accused had not accepted the risk of killing the deceased, that she recognised putative private defence, or that she found that the accused only intended to injure the intended target. These were not her reasons and, however valid they may be, it doesn’t help to pretend that they were her reasons. It also doesn’t help to attempt to justify the distortion of our law on dolus indeterminatus and dolus eventualis applied to error in objecto and the logical error made by Masipa. The law on dolus indeterminatus and error in objecto is clear. Nominal/name identity is irrelevant – for that very reason it cannot help the accused if he thought that it was B behind the door; nor can it help an accused who argues that he thought that it was not C. The identity of whoever was behind the door remains irrelevant – and the indictment did not need to remind the judge that it is not relevant. Finally, there is no reason to ignore this clear law (that nominal/name identity is irrelevant) just because the scenario triggers the rules relating to both error in objecto and putative private defence. There appears to be nothing to save us from the inevitable conclusion that Masipa made errors of law and errors of logic.

 

***

 

[1] See B Cubane, “JUDGE MASIPA GOT IT RIGHT: Oscar Pistorius and the intention to kill” http://shar.es/1aWktC via @thelawthinker.

[2] Possibly even in the sense of object/body identity – although this may be controversial and is unnecessary for present purposes.

[3] Footnotes omitted; p 348.

[4] As Kelly Phelps is reported to argue (Martin Williams, quoting K Phelps “Why Oscar Judge was right”, The Citizen, (Tuesday 16 September 2014).

[5] This is very similar to the argument made by Burchell discussed below (under the heading “Error in Objecto”) – that someone (C) may be excluded from an error in objecto if A thinks that B is not C.

[6] Footnotes omitted; p 396.

[7] p 193.

[8] Such as J Burchell “Masip’a decision to acquit Oscar of murder justified” (http://m.bdlive.co.za/opinion/?articleId=339191).

[9] Snyman disagrees on the basis that the defence was one of putative private defence. As discussed below (under the heading “Putative Private Defence”), this does not exclude the scenario from being one of error in objecto.

[10] http://citizen.co.za/241517/legal-experts-mixed-reaction-pistorius-judgment.

[11] Such as J Burchell “Masip’a decision to acquit Oscar of murder justified” (http://m.bdlive.co.za/opinion/?articleId=339191).

[12] Such as J Burchell “Masip’a decision to acquit Oscar of murder justified” (http://m.bdlive.co.za/opinion/?articleId=339191); N Taitz (“Judge Masipa was right on Dolus and murder” http://bit.ly/1wGgNdm&#8221); Martin Williams, quoting K Phelps “Why Oscar Judge was right”, The Citizen, (Tuesday 16 September 2014); http://m.bdlive.co.za/opinion/?articleId=339191; Sonja Carstens, quoting CR Snyman “Legal question that divides” http://www.beeld.com/stemme/2014-09-19-regsvraag-wat-verdeel?vc=e9eccde605d8837f80b9fa6ccc527a34.

[13] Sonja Carstens, quoting CR Snyman “Legal question that divides” http://www.beeld.com/stemme/2014-09-19-regsvraag-wat-verdeel?vc=e9eccde605d8837f80b9fa6ccc527a34: “Die situasie staan in die reg bekend as error in objecto . But it is not in Oscar’s case happened. Maar dit is nie wat in Oscar se saak gebeur het nie. It was not error in objecto not. Dit was nie error in objecto nie. Jack in the case I would have realized I act unlawfully on. In die Koos-geval sou ek besef het ek tree wederregtelik op. “Pistorius did not realize he was acting unlawfully stops.“Pistorius het nie besef hy tree wederregtelik op nie. He thought he was acting in self-defense. Hy het gedink hy tree op in noodweer. The storms were not objectively present. Die noodweer was nie objektief aanwesig nie. This just in his mind, his subjective representation exists, “says Snyman. Dit het net in sy gedagtes, in sy subjektiewe voorstelling bestaan,” verduidelik Snyman.”

Several attempts have been made to justify the decision of Judge Masipa to acquit Oscar Pistorius of the murder of Reeva Steenkamp. Regrettably, in my view, none of them are successful.

Burchell* falls into the same wishful thinking errors of Taitz and Cibane, and logical error as Masipa.

Burchell (and Taitz**) observe, correctly, that the (original) Pistorius defence was putative private defence. This is the defence that one mistakenly believes that one is entitled to act in private defence. If the accused makes this mistake, whether he is mistaken about the facts, or (as Burchell correctly adds) the law (De Blom 1977 AD), the accused cannot be convicted of murder. This is because an accused who is mistaken in this way has no intention to act unlawfully. Putting it positively, the accused believes s/he is acting lawfully. The enquiry is entirely subjective – what did the accused actually think or forsee. It is, as indicated, a complete defence to murder. This defence required that Masipa asked whether the accused foresaw the possibility that his conduct could unlawfully kill.

The problem is that, after specifically stating that she was turning her attention to this defence (of putative private defence), she erroneously states that the question is whether the accused foresaw that his conduct could kill [fullstop]. As discussed, this is not the right question.

It does not make her decision right to say, as Taitz and Burchell argue, that if Masipa had asked the right question, she would have (possibly) arrived at the same or right conclusion. We cannot know what conclusion she would have come to if she had asked another question.

Then there’s the problem of error in objecto and dolus eventualis. Error in objecto describes a scenario in which, as Burchell explains: “In terms of this rule, if A intends to kill B and shoots and kills C, whom he mistakenly believes to be B, then A’s mistake as to the precise identity of his victim is irrelevant and he could be liable for the murder of C.”

It is worth noting that there is no mention of the reason why A thinks that C is B. This is in line with our settled law, that takes no account of the reason why A makes his/her mistake. If identity is irrelevant, the reason for any error as to identity cannot be relevant. The identity of the actual victim is irrelevant as is, necessarily, any reason for a mistake as to identity.

Both Masipa and Burchell correctly observe that the facts of the Pistorius case disclose a scenario of error in objecto – in which the identity of the victim is irrelevant.

It is notable also that the defence never argued that the accused had not intended to kill – but only, perhaps, at most, intended to injure. This too would have been a valid defence against murder, but it was not a defence claimed, and certainly not one – from the wording of her judgement – relied on by Masipa as a basis on which to acquit the accused of murder. I don’t see how it could possibly make Masipa’s judgement right to refer to a basis on which she could have relied, but did not.

It is possibly also worth mentioning that her judgement did not turn on a finding – that was crucial in the Humphrey’s case (2013 SCA) – that, although the accused did forsee the possibility of killing whoever was behind the door, he did not accept that risk.*** Again, that would have been valid basis (on our law as it stands) to acquit the accused of murder. However, from the wording of her judgement, this was not the reasons she gave for acquitting the accused of murder.

Returning to what Masipa actually said, and how she actually reasoned. After noting that, in error in objecto scenarios, the identity of the actual victim is irrelevant, she asks what she says is required by the test of dolus eventualis (legal intention): Did the accused forsee the possibility of killing the deceased. This is the wrong question for two reasons. As discussed above, the accused’s defence was that he did not intend to unlawfully kill anyone. This focus – the very basis of the defence of putative private defence – is lost. But, putting that aside, secondly, it is wrong because it deviates from, even contradicts, the recognition that the facts disclose an error in objecto scenario and that the identity of the victim is irrelevant. The question then, correcting also for the defence of putative private defence, should have been: did the accused foresee the possibility of unlawfully killing whoever was behind the door.

Again, it does not help to say this cannot be true because he thought the deceased was in the bedroom. This is the argument that Masipa seems to have finally settled on – and it is fatally flawed in logic. She says: the accused could not have forseen the possibility of killing the deceased, or anyone else for that matter, becuase he thought the deceased was in the bedroom. The fact that the accused thought that the deceased was in the bedroom says nothing about what he thought about the presence of someone else in the toilet. Indeed, it is his defence that he believed someone else was in the toilet.

Burchell perpetuates this error in logic. Referring to his scenario in which he notes that it is irrelevant that A intends to kill B, but kills C, he argues that there can be no intention to kill C if A had excluded in his mind the possibility that the person he thinks is B, could be C. But this is the definition of an error in objecto scenario – let’s take this slowly:
A intends to kill B.
He does not intend to kill C, but B.
He thinks that B (the human body) is, well, B.
In thinking that B is B, he does not think that B is actually C.
So thinking that B is B and not C, he has excluded in his mind, the possibility that B is C.
This is the definition of an error in objecto scenario and there is no dispute that the identity of the actual victim is irrelevant. That is, it cannot help the accused that he thought that B (the human body) was B – where he did not think that B was C. Therefore, it cannot assist an accused who has excluded from his mind the possibility that B is C.

Importantly, the reason for any errror as to identity is also irrelevant. It cannot help an accused who thinks that B is B because B is not C. Our law is clear, in error in objecto secanrios, identity is irrelevant, and it remains irrelevant no matter what an accused’s reasons are for making a mistake as to the identity of his victim.

Masipa’s error was similar. In her view the accused could not be convicted of murder because the accused did not forsee that he could kill C (or B), because he thought C was in the bedroom. In that case, it leaves him thinking that B was in the toilet. As we know from the rules of error in objecto, it is no defence to say that he thought that B was B, and not C.

In conclusion, we have seen several attempts to justify Masipa’s judgment, all of which seem to fail. It does not make Masipa’s judgment right to pretend that her reasons were reasons that would make her judgement right: that she decided that the accused had not accepted the risk of killing the deceased, that she recognised putative private defence, or that she found that the accused only intended to injure the intended target. These were not her reasons and, however valid they may be, it doesn’t help to pretend that they were her reasons. It also doesn’t help to attempt to justify the distortion of our law on dolus eventualis applied to error in objecto and the logical error made by Masipa. The law on error in objecto is clear. Identity is irrelevant – for that very reason it cannot help the accused if he thought that it was B behind the door; nor can it help an accused who argues that he thought that it was not C. The identity of whoever was behind the door remains irrelevant. There appears to be nothing to save us from the inevitable conclusion that Masipa made errors of law and logic.

* http://m.bdlive.co.za/opinion/?articleId=339191.
** “Judge Masipa was right on Dolus and murder http://bit.ly/1wGgNdm”.
*** See Cubans B. , JUDGE MASIPA GOT IT RIGHT: Oscar Pistorius and the intention to kill http://shar.es/1aWktC via @thelawthinker.

While Masipa J has delivered judgement in the Pistorius case, there appear to be errors of law in her judgement. These errors of law may allow the prosecution to appeal.
Many commentators are saying that Masipa’s mistake was to misconceive the form of intention in our law known as dolus eventualis. A careful analysis reveals that the apparent error relates only partly to dolus eventualis. Instead, the apparent error related to how Masipa conceived of dolus eventualis as related to a far more complex issue in our law: the difference between the scenarios of error in objecto and aberratio ictus. It is only by understanding these scenarios, and the settled law on these scenarios, that one can understand where Masipa appeared to go wrong.
After dismissing the prospect of premeditated murder, Masipa turned her attention to whether Pistorius was nevertheless guilty of murder.
She indicated that this required that she deal with the defence argument that a conviction in the circumstances would require that the court revive an old doctrine rejected from our law: transfered intent. The spectre of this is daunting – because to understand that this doctrine is not in question, one must understand what it is, what it is not, and why it doesn’t apply. Correctly Masipa identified that this required an analysis of two scenarios in criminal law: error in objecto and aberratio ictus. Anyone who tells you these are easy to understand, probably hasn’t understood them. At its essence, scenarios of error in objecto (or where one is dealing with a person, error in persona) relate to circumstances in which one strikes/shoots at a particular object/person and one hit one’s target. If you have aimed at a human being (a particular human body), and you hit and kill that human being, you cannot argue that you thought that the person you killed was John, whereas, in fact, it was Peter. This (error in objecto/persona) is irrelevant.
Aberatio ictus on the other hand means “going astray of the blow” – and ultimately describes scenarios in which one misses one’s target. This is crucial because it is the essence of the distinction between scenarios of error in objecto and aberratio ictus. It is an aberratio ictus scenario where one aims and shoots at someone (John), but misses that person and strikes and kills another (Peter). The question arises whether this mistake is material and can form the basis of a defence.
In these scenarios, our law used to refuse to recognise that this mistake could possibly be a defence. It used to simply regard your intention to kill John as “transferred” onto Peter. Hence, “transferred intent”. Our law has shifted and rejected this transferred intent approach. It is now possible on our law to rely on this mistake as a defence against a murder charge in respect of Peter – but only if there was no other form of intention that actually fell on Peter. That is, our law will allow you to be convicted of the murder of Peter, if, while you intended to kill John, you also had intention in respect of Peter – such as perhaps dolus eventualis – you foresaw the risk of missing John and killing Peter, accepted the risk and proceeded. (For further discussion of the meaning of these two scenarios and the difference between them, please see http://wp.me/p3efQr-2)
Lets consider into which scenario the facts of Pistorius fits. Did Pistorius miss his target? No, he did not. He aimed at a particular human being and shot and killed that human being – whoever was behind the door. There can be no question that this is a scenario of error in objecto and that the identity of the person behind the door was irrelevant.
Masipa discussed these scenarios, correctly stated the law, and correctly identified that we (on the facts of the Pistorius case) are dealing with a scenario of error in objecto – in which (in the relevant sense) the identity of the victim was irrelevant. That is, it doesn’t matter who was behind the door. The significance of all of this is that it has direct implications for the way in which one enquires whether an accused had intention or not.
Then the first sign of trouble appeared. After turning her attention to the accused’s defence of putative private defence, Masipa stated that the question was: “whether the accused intended to kill.” Immediately it becomes apparent that there was a misconception regarding the nature of the defence of putative private defence. It is not the question of whether the accused intended to kill,  but whether he intended to unlawfully kill.
The question of whether the accused intended to unlawfully kill is the question of whether the accused believed he was under attack and entitled to resort to force in defence. This question was not engaged with. The question the judge pursued was whether he intended to kill.
At that point it seemed Pistorius was bound to be convicted of murder – given that there seemed little question that he did intend to kill whoever was in the toilet – despite his defence of putative private defence. But the judgement took another strange turn.
“I now deal with dolus eventualis or legal intent. The question is:
1. Did the accused subjectively forsee that it could be the deceased behind the toilet door;
2. Notwithstanding the foresight, did he then fire the shots, thereby reconciling himself to the possibility that it could be the deceased in the toilet.
The evidence before this court does not support the states contention that this could be a case of dolus eventualis. On the contrary, the evidence shows that, from the onset, the accused believed that at the time he fired the shots into the toilet door, the deceased was in the bedroom, while the intruders were in the toilet.”
She repeats this again twice: did he forsee the possibility of killing the deceased – although, on the third occasion coming closer to what ought to have been addressed: whether the accused foresaw the possibility of killing whoever was behind the door. She says, on the third occasion, that the accused did not forsee killing “the person behind the door, let alone the deceased, as he thought she was in the bedroom at the time”. But accepting that the accused thought that the deceased was in the bedroom does not exclude the possibility of there being someone else behind the door. Indeed, ironically, that is his own version: that he thought there was someone else behind the door. Applied to the undisputed law on error in objecto where one mistakes one person for another – which is immaterial – the question ought to have been: “did the accused forsee the possibility of killing whoever was behind the door”. This is an entirely different question which,  in turn,  begs the question whether the accused must have, and by inference did, forsee that he would kill whoever was in his toilet by firing four shots through the door.
It is true that, on day two of her judgement, she referred to several authorities on how one may reason to a finding of dolus (intention). These authorities make the valid point that one must be careful not to conclude that just because a reasonable person would forsee something (death of someone for our purposes), that the accused did. This is the usual and well founded caution against a logical error of thinking that just because something should be true, doesn’t make it true. Just because someone should have realised something doesn’t mean s/he did realise it.
She also summarised her findings and, in reference to dolus eventualis, said that “this court has already found that the accused cannot be guilty of murder dolus eventualis on the basis that, from his belief and conduct, it could not be said that he forsaw that either the deceased, or anyone else for that matter, might be killed when he fired the shots at the toilet door.” Regretably this takes things no further because it is a bare conclusion without the all important reasons for this statement. We are left having to rely on the reason she provided previously in her judgement – that he did not forsee killing “the deceased or anyone else for that matter” because, as she said previously, the accused thought that the deceased was in the bedroom. As discussed above, a belief that the deceased was in the bedroom does not exclude someone else being in the toilet and this is exactly what he believed, on his own version.
What is also revealing is that, on a defence of putative private defence, even if a court accepts that the accused acted in putative private defence (mistakenly believed he was under attack and was entitled to resort to force in defence), the problem of how much force in defence arises. It is not the question of whether the extent of force actually used was allowed, because, given that there was no attack, no force at all would be allowed. The question – a very necessary question – becomes, did the accused foresee that he was not allowed to resort to that extent of force. Masipa ought to have asked, if she engaged properly with a defence of putative private defence, whether, not only was the accused mistaken, but was he so mistaken that he could have believed he was entitled to fire four shots through a door at an intruder. The court could have gone either way on this, but that is not the point. The point is that this is another reason to think the Court did not properly engage with the defence of putative private defence.
On the charge of unlawful possession of ammunition Masipa seemed to conflate the mental requirement for possession (knowledge of possession), with the mental requirement (known as fault) for the crime – that is, can one only be guilty of this crime if one intends to unlawfully possess ammunition or even if one only negligently unlawfully possesses ammunition. The statute that creates this offence is silent on the issue, which requires, in turn, that a court must decide what form of fault, if any is required. This analysis is conspicuously missing from her judgement. The significance is that an analysis could have led her to the conclusion that only negligence was required. If that were so, his defence that he did not know he was not entitled to be in possession of the ammunition would have to stand up to the appropriate test of negligence: would the reasonable firearm owner know that this is prohibited?
In the final analysis, Masipa appears to have erroneously conceived of the defence of putative private defence, and to have misconceived the test of intention when dealing with a problem of error in objecto. Also, she appears to have conflated a requirement for possession for the fault required for the unlawful possession of ammunition. These are all, arguably, errors of law. As errors of law, the state may appeal. The effect is that, if the state does appeal, and one may well expect that it will, Pistorius continues to face the prospect of a murder conviction.

DISCLAIMER

This is translated and elaborated from an article, inspired by Marida Fitzpatrick, and published in the Beeld newspaper today (11 Sept 2014, see OSCAR: VAS OF VRY? http://www.beeld.com/nuus/2014-09-11-oscar-vas-of-vry via @Beeld_Nuus). It appears here subject to the disclaimer under which it was published.  To be clear,  Ulrich Roux and I wrote our respective pieces as an academic excercise to show what a guilty and not guilty verdict could look like; to help people to see both sides. We did not choose which verdict (guilty or not guilty) we would write; the views expressed are not necessarily our own views, and we don’t expect the court to necessarily adopt any of the views expressed.

NOT GUILTY

The criminal onus requires that the state must prove its case beyond a reasonable doubt and that if there is a reasonable doubt as to the accused’s guilt, he must be acquitted. A doubt means nothing more than a possibility. Is there a reasonable possibility that the accused is not guilty? In a world in which, in a relevant sense, everything is possible, the question for us becomes whether any of these possibilities in the accused’s favour, are reasonable.
The evidence that there were moments of discontent in the relationship, that the accused may have threatened to break someone’s legs, and other evidence to the effect that the accused may be of “bad character” might, in some respects be technically admissible because the defence led evidence of the accused’s good character – as provided by s 197 of the Criminal Procedure Act or at common law. However, in line with the general reason why this sort of evidence is ordinarily inadmissible, especially that the accused is not on trial for being of “bad character”, we regard it as of such little weight that we have excluded it from our considerations.
The testimony of the “ear-witnesses” is clearly in a different league. While it is correct that is “circumstantial evidence” in that it requires the court to draw inferences, most evidence, even so called direct evidence still requires that the court must draw inferences. Furthermore, we know that circumstantial evidence is not necessarily weaker than direct evidence. It is trite that in our law the drawing of inferences from circumstantial evidence is governed by the “cardinal rules of logic” from S v Blom 1939 AD to the effect that: the inference sought to be draw must be consistent with the proved facts, and it must be the only reasonable inference. The evidence of Dr Lin was to the effect that it is entirely possible that a person would mistake a male scream for that of a female. He did concede that it is possible that person could accurately distinguish, but this concession does not assist in finding that it was a woman who screamed – as the only reasonable inference. This is just another way of saying that there appears to be a reasonable doubt as to whether the screaming that was heard, was that of the deceased.
We disagree with the prosecution that these two defences (involuntariness and putative private defence) are incompatible. It is not the case that involuntariness is a claim to not have been thinking, while putative private defence is a defence that the accused was thinking, albeit mistakenly. This is to misconceive involuntariness – it is not a claim that one was not thinking, only that, if one was thinking, whatever one was thinking did not control one’s conduct.In that case, the two defences proposed by the defence in the alternative are compatible and amount to the claim that the accused was under a mistaken belief that there was an intruder in his toilet and 1) in his heightened state of anxiety he got a fright and involuntary discharged his firearm; 2) he voluntarily discharged his firearm in putative private defence (the mistaken belief that he was entitled to kill whoever was in the toilet).
The defence raised by the accused that he discharged his firearm as a result of a “fright” is, in our view, the defence if involuntariness.A state of involuntariness prevails if the accused’s conduct was not under the control of his conscious mind (Burchell, Principles of Criminal Law 2013; S v Johnson 1969 (1) SA 201 AD; S v Chretien 1981 (1) 1097 AD). This court has stated on numerous occasions that a court is permitted to draw the natural inference that, ordinarily, conduct is voluntary and that if an accused wishes to claim involuntariness, he must provide some basis for this claim. (S v Henry (1) 1999 13 SCA). We are of the view that this basis was indeed provided in the evidence of Professors Derman and  Vorster. In this, taken together with the evidence to the effect that the accused is an overly anxious person and likely to startle easily, we find a sufficient basis to place a reasonable doubt on the voluntariness of the accused’s conduct at the time of pulling the trigger.
In the absence of reliable evidence (particularly to the effect that the deceased was screaming that night) to support the version of the state as the only reasonable inference, we cannot accept the version proposed by the state. However, the question arises, on the accused’s own version whether we accept that there is a reasonable possibility that upon hearing a sound emanating from his bathroom, the accused immediately assumed an intruder or intruders had gained entrance to his home, that, throughout his response, until he fired the fatal shots, he continued to misinterpret further sounds and the situation generally as one of an attack, and that he did not foresee the possibility that he could be wrong and that it could be the deceased in the bathroom and toilet. Again, the question we must ask ourselves is not whether this could possibly be true – that is not enough for the accused. On the other hand, the question is also not whether we think this is true, nor whether this is what most likely or probably happened. The question is whether this could reasonably possibly have happened. Given that we have accepted that the accused is an overly anxious person, and that, ironically, because the State has convinced us that the accused is reckless with firearms, we are able to conceive of the reasonable possibility that he could have laboured under this mistake and fired his gun at the door in the mistaken belief that he was entitled to do so. This mistake excludes the intention to unlawfully kill and he cannot be convicted of murder.

***

See the article in the Beeld for Ulrich Roux’s guilty “verdict”. See also my article in the Telegraph to the effect that Oscar Pistorius is, in my view, in grave jeopardy: Legal View: Oscar Pistorius ‘in grave jeopardy’ of being convicted of Reeva Steenkamp’s murder | via @Telegraph http://fw.to/GnI3Kq

What is most important for parents, is that there is no basis recognised in law, upon which a parent may possess or distribute an image or video of his/her child that amounts to “child pornography”, simply by virtue of being a parent. If the images or videos you have of your own child amount to “child pornography”, your possession, as a parent, is unlawful – as will any distribution of these images and videos.

That said, after a careful analysis of both the Films and Publications Act, and the Sexual Offences Act, the following may be discerned: No matter whether you are a child’s parent, it is an offence to intentionally (or possibly negligently) unlawfully create, possess, or distribute an image or video of a child which is sexually explicit – where the primary objective (judged objectively and in context) is to illicit sexual arousal – showing a child:

(a) engaged in an act that constitutes a sexual offence;

(b) engaged in an act of sexual penetration;

(c) engaged in an act of sexual violation;

(d) engaged in an act of masturbation; or

(e) displaying the genital organs of such person in a state of arousal or stimulation;

(f) unduly displaying the genital organs or anus of such person;

(g) displaying any form of stimulation of a sexual nature of such person’s breasts;

(h) engaged in sexually suggestive or lewd acts;

(i) engaged in or as the subject of sadistic or masochistic acts of a sexual nature;

(j) engaged in any conduct or activity characteristically associated with sexual intercourse;

(k) showing or describing such person— (i) participating in, or assisting or facilitating another person to participate; or (ii) being in the presence of another person who commits or in any other manner being involved in, any act contemplated in paragraphs (a) to (j); or

(l) showing or describing the body, or parts of the body, of such person in a manner or in circumstances which, within the context, violate or offend the sexual integrity or dignity of that person or any category of persons under 18.

For the details see and a discussion of the actual offences, see Possession and Distribution of Child Pornography  Please read the full document before sending any questions – your questions may well be answered in the full discussion.

Does James Grant think he is an alien? 1/
Pistorius needs only to raise a reasonable doubt that he intended to UNLAWFULLY kill whoever was behind the door. 2/
Intention is judged subjectively – what was he actually thinking; A doubt = a possibility. Thus a reasonable doubt = a reasonable possibility. 3/
There must always be doubt about anything in the universe. It’s possible that Grant is an alien… 4/
But is it reasonably possible? We may dismiss this possibility as unreasonable, but can we dismiss the possibility that Grant THINKS he is an alien as unreasonable? Probably (I hope)5/
Similarly, for Pistorius, he has conceded that there is no reasonable possibility that he was, in fact, entitled to kill in private defence 6/
But, can we dismiss the possibility that Pistorius THOUGHT he was entitled to kill in private defence as unreasonable? 7/7

The finding of the panel which observed Oscar Pistorius, reported by Prosecutor Gerrie Nel, is both surprising, and unsurprising – it is both right and wrong. This is because nobody can say whether Oscar Pistorius suffered, at the relevant time (of the alleged offence), from a legally significant “mental illness or defect”. The concept is undefined in our law. The effect is that no one can know whether a “mental illness or defect” was present. The first problem is that a “mental illness or defect” is a legal concept, distinguished in law, from “mental disorders” that are the concern of psychology and psychiatry. The law insists that a “mental disorder” does not necessarily amount to a “mental illness or defect”. Courts must ultimately decide – and yet there exists no criteria upon which to base this decision.

I attach a working draft of a paper that sets out this argument at length – constructive comments welcome. It argues that the requirement of “mental illness or defect” in our law, as a threshold requirement for a defence of “pathological incapacity” (the insanity defence) is almost entirely meaningless, or that such meaning as it has is unfounded, ill considered, or a distraction, and that it is, is in any event, apparently redundant. Nobody can say whether a “mental illness or defect” was present or not. I argue that we must face this problem, and that the best solution is to abolish the requirement from our law.

ABSTRACT:
The defence of pathological criminal incapacity requires, as a threshold criterion, that an accused suffered from a ‘mental illness or defect’ at the relevant time. Analysis of what this threshold criterion means reveals that it is almost entirely meaningless, or that such meaning as it has is unfounded, ill considered, or a distraction, and is in any event, apparently redundant. It seems instead that the criterion is giving expression to surreptitious fears regarding the dangerousness of persons who may succeed with this defence. I recommend that we drop the criterion as intellectually dishonest and that we no longer distinguish pathological from non-pathological incapacity and that we ask what appears to be the real question: about dangerousness. We can ask this question in the correct context, that of the civil law which is already designed to address this question. This enquiry could follow automatically upon a successful incapacity defence. Finally I argue that there are several associated benefits to doing this, including that this would allow for all persons who claim incapacity to be treated equally, that the dubious ‘reverse burden’ placed on accused persons who attribute their incapacity to a mental illness or defect, which is not placed on those who claim incapacity but do not attribute it to a mental illness or defect, is withdrawn. Also it will introduce further equality in that the dangerousness of all accused who succeed with an incapacity defence will be considered. Currently dangerousness is surreptitiously presumed in respect of those who attribute their incapacity to a mental illness or defect, but it is not a question which is currently considered in respect of accused who do not attribute their incapacity to a mental illness or defect – and who currently receive, if successful, an unqualified acquittal.

DOWNLOAD HERE: THE MEANING OF THE MENTAL ILLNESS OR DEFECT REQUIREMENT FOR AN INSANITY DEFENCE.WORKING

Criminal Law in 4D

Posted: May 28, 2014 in Uncategorized

This is a working paper in which I attempt to reveal an aspect of criminal law that leads to great confusion. It is directed at law students, but anyone interested should be able to follow – although strong coffee will help. Inevitably, this is going to hurt a little. But a failure to understand it is one reason why the (original) Pistorius defence is prone to be misunderstood.

Criminal law can be confusing because conduct can be both reasonable and unreasonable, right and wrong, and things can exist or not, all at the same time. This might appear to be an indictment of criminal law as inherently contradictory, but it is not. It is an attempt to illuminate how all these things can be true at the same time – without contradiction.
The key is to recognise that the answers depend on the particular perspective or context from which the questions must be approached. It isn’t even as simply as a reminder that there is (presumably) a world out there, of reality, and that we have no direct access to this reality – we can only perceive it. This is only part of what must be recognised. In the final analysis I will identify four different dimensions or perspectives from which questions in criminal law must be approached.
It is first necessary to sketch out the requirement of criminal liability in brief, before discussing the different perspectives/dimensions applicable to these requirements.

The various requirements for liability (set out in more detail in the “Overview” (on this site)), are as follows:
Conduct: The accused must have done something or failed to do something in circumstances where our law requires the victim to act (that is, in the face of a legal duty to act). This conduct must be voluntary: subject to the control of the accused’s mind. In the case of crimes which are defined by the causing of a particular consequence (such as murder), the prosecution must also prove that you caused the prohibited consequence: that the accused caused the death of the victim.
Unlawfulness: conduct must be unlawful to attract criminal liability. This is the requirement that allows a court to take account of specific circumstances in which the conduct takes place and to judge whether the conduct (which is generally unlawful), was, in the specific circumstances, lawful and, therefore, without liability. This is ultimately tested by reference to the “legal convictions of the community” as informed by the values in the Constitution (Carmichele v Minister of Safety and Security 2001 CC). These exceptional circumstances are regarded as grounds of justification. Self/private defence is a ground of justification. If someone acts in self/private defence, his/her conduct is lawful – and cannot attract liability. Other grounds of justification include consent and necessity (also known as duress).
It is worth noting that in the context of omissions, which are only punishable if there was a legal duty to act on the accused, the question of whether such a legal duty existed is a question of unlawfulness. While our law presumes that positive conduct is unlawful, it presumes that omissions are ordinarily lawful. The question of whether any particular omission was actually unlawful is determined by the overarching ultimate tests of unlawfulness. Common scenarios in which our law regards an omission as unlawful is where the accused was in control of a dangerous animal or thing, in a protective relationship with the victim (such as a father or mother in respect of his/her child).
Capacity: the ability to appreciate the wrongfulness of his/her conduct and to act in accordance with this appreciation. This is where the defence formerly known as “insanity” (now known as pathological incapacity) is located. Our law also regards youth as a basis on which an accused may lack capacity.  More recently, our law has recognised intoxication (Chretien 1981 AD) and severe emotional stress (Wiid 1990 AD; Eadie 2002 SCA) as conditions which may deprive an accused of capacity.
Fault: Crimes usually require some form of fault: intention or negligence. Our law accepts, exceptionally, crimes that require no fault – known as strict liability crimes. Most serious crimes require intention, such as murder, theft, fraud, robbery, and assault. Culpable homicide requires only negligence. Crimes that require no fault (strict liability offences) are exceptional and are ordinarily relatively minor – such as some traffic offences.

The different possible perspective are as follows – starting with what is possibly the most neglected:
A. Objective Normative: Throughout our law, objective normative judgments are required. These are value judgments where the Court must exercise its discretion in determining what conduct was permitted in the specific circumstances – based on considerations of fairness and justice, ultimately guided by the values in the Constitution. Several areas exist where courts must make these value judgments, such as in considering whether an accused should be regarded as having killed the victim – causation is ultimately determined by reference to considerations of justice and fairness. Also, in considering whether conduct was justified (under the unlawfulness requirement) and in the context of negligence in deciding what the reasonable person would do. These normative or value judgments are questions of law and refine the law to apply to the particular circumstances – setting out what was permitted or prohibited in the particular circumstances. In the case of private/self-defence the value judgments (made by Courts) to date allow us to set out the requirements of the defence as follows:
1. You can only resort to force in response to an attack against you that is unlawful itself (you can’t resort to force against someone who is acting lawfully, such as in the execution of a warrant of arrest against you);
2. It must have commenced or be imminent (It is possibly worth noting the ground breaking decision of Satchwell J in (Engelbrecht 2005 WLD) in which she accepted that it is enough if the attack was ‘inevitable’.);
3. It must be an attack against a legally protected interest of yours or of another person (life, limb, or property of substantial value);
4. It must be necessary to resort to force;
5. That force must be directed at the attacker (not someone else); and
6. The extent of force must be necessary and reasonable – for instance, you cannot shoot someone for threatening you with a light assault. 
These requirements are objective normative requirements. They are statements of law. When a case comes before court and a claim is made of private defence, the court will consider whether the objective reality (discussed under B) fits with the objective normative requirements in deciding whether to accept the defence.
It is worth noting that these normative judgements are not fixed. In particular, in the context of grounds of justification, which are judged by the “legal convictions of the community” as informed by the values in the Constitution, our courts could develop the requirements and, ultimately change what is permitted or prohibited, in a particular circumstance.
B. Objective Reality – appropriate to judge questions of whether, on the facts, the accused’s conduct satisfies the conduct requirement and whether that conduct is, in the specific circumstances, unlawful. In the context of negligence, it is the question of whether the accused’s (actual) conduct fell short of the normative standard of the reasonable person.  Under unlawfulness, once it is determined what is normatively permitted in the circumstances, the test for whether one’s conduct was unlawful (that is, conformed with what is permitted) in those circumstances, is judged with regard to reality. Based on the value judgments to date, which require, for private/self defence (as set out above under A), that one must be under an unlawful attack, the question becomes whether the accused was, in reality, under attack or not. Fundamentally, this question takes no account of what an accused was subjectively thinking (see below under D), nor what s/he ought to have been thinking – as we may expect of the reasonable person, in the circumstances (objectively constrained – see above under C).
Finally, a reality based enquiry also determines the circumstances into which we must place the reasonable person for the purposes of an enquiry into negligence. One does not locate the reasonable person in circumstances that the accused thought s/he was in, but rather the actual circumstances of the accused .
C. Objective Constrained: The reasonable person in the circumstances of the accused. This enquiry is the basis for the test of negligence (under fault). It is, at its essence, a comparison between the normative judgement (discussed under A) and the actual conduct of the accused (discussed under B). However, it is important to recognise that the normative judgment (under A) is constrained for the purposes of the test of negligence, by the circumstances of the accused. The reasonable person in the circumstances on the accused can only know what the circumstances permit, and only what is reasonable to know in the circumstances. Fundamentally, this standard permits for the reasonable person to make mistakes. For example, a reasonable person in circumstances in which someone starts shouting, then reaches into his/her pocket, produces a shiny object, and advances on him/her, may well believe that s/he is under attack. This belief may be, in reality, quite untrue: the person may have been calling to someone else, have produced his/her car keys, and be walking to his/her car, behind the accused. Nevertheless, the belief of the accused may be entirely reasonable. Thus, in the context of negligence, there is the ever present possibility of a reasonable mistake.
D. Subjective: Question may be judged subjectively – in the sense of what the accused was actually thinking. Intention, voluntariness and capacity are judged subjectively. In this context the accused’s particular mental characteristics and vulnerabilities are relevant. These considerations could lead a court to conclude that the accused did not actually intend to say, kill, or to unlawfully kill, or that the accused lacked capacity. In this context reality and the reasonableness of the accused’s beliefs are irrelevant.

Our courts sometimes refer to the standard of the reasonable person to assist them in determining what is permissible as justified (lawful). This is a dangerous practice because the test of the reasonable person is conventionally utilised in the context of negligence – in which the question is: what would the reasonable person do in the circumstances – in which the appropriate perspective is one which is constrained by the accused’s circumstances (discussed under C). The reasonable person (in the circumstances of the accused) may be mistaken and so a mistake would be reasonable. On the other hand, in the context of unlawfulness, the question of whether an accused’s conduct meets the normative requirements (discussed under A) are judged against reality (discussed under B). In this context, there is no room for error on the part of the reasonable person. If one adopts a reasonable person to answer questions of unlawfulness, it is possible to commit an error if this distinction regarding perspective is not observed: to judge conduct as justified based on a mistaken reasonable person. If one must resort to the use of the reasonable person in this context (of unlawfulness), as our courts sometimes do, it must be a reasonable person who is not mistaken and who, essentially, knows everything. This is difficult to conceive and seems prone to error.

In application to a charge of murder (defined as the intentional unlawful killing of another human being), self/private defence excludes the unlawfulness requirement. One cannot be convicted of murder for the intentional lawful killing of another human being.
As discussed, justifications are judged objectively (normative and on the (actual) facts) and an accused’s subjective perceptions are irrelevant to any justification defence, including one of self/private defence.
However, an accused perceptions and beliefs are relevant to determining whether the accused had intention. If, as in the Pistorius case, he was not actually under an imminent attack (for whatever reason), the accused cannot claim to be acting in self/private defence. It is worth a pause here to point out that even if you are under an imminent attack, the other requirements (set out above under A) must be satisfied for a valid claim to self/private defence. For Pistorius, none of these were true (in reality), and so he has no claim to self/private defence.
But if he believed he was under imminent attack (and that all other requirements were met), although he acted unlawfully, his intention would have been to act lawfully. This is a defence of mistake, known as “putative self/private defence” – that is, supposed/mistaken self/private defence. It important to recognise that the effect of this defence (of mistake) is that it does not exclude the unlawfulness requirement. Just because you are mistaken in thinking you were under attack (and acting lawfully) does not make it true that you were acting lawfully. Your conduct remains unlawful. But, the effect of a mistaken belief that you were under attack and acting lawfully will exclude your intention to act unlawfully. The defence of mistake therefore excludes intention.
If that mistake is reasonable – one that a reasonable person in the (actual) circumstances of the accused may make – this reasonable mistake will exclude negligence with the effect that the accused cannot be convicted of culpable homicide.

Until now (and as I have indicated in other posts), our courts have refused to take account of disability of any kind in constructing the reasonable person against whom to compare the accused for the purposes of a negligence enquiry. Instead, if you do something that requires special skill and knowledge (for example, surgery and presumably owning a gun), you will be judged by the standard of the reasonable person who possesses the required skill and knowledge. I expect that our courts will compare Pistorius’s conduct with that of a reasonable gun owner.
Thus, if Pistorius’s defence is that he was mistaken in acting in self/private defence (that is, putative self/private defence), he must be acquitted on the murder charge if the court accepts that this is reasonably possibly true. But the court will then consider whether this mistake was one that a reasonable gun owner may make.

In conclusion, one must identify the appropriate perspective from which to approach any enquiry in criminal law: Is it a subjective or objective enquiry. If objective, one must go further. It is not enough to correctly identify that an objective perspective is required – because there remain three options within objective enquiries: normative, constrained, and reality. Without observing and respecting these different perspectives, one is bound to become disorientated and confused.

 

In the conduct of Pistorius’s defence, through the testimony of Dr Vorster, the criminal capacity of Pistorius has been placed in issue. Criminal capacity, together with the requirement that your conduct must be voluntary, are the bases upon which our law enquires whether you are a responsible person – whether it makes any sense to hold you criminally liable in law and to punish you. We take it that it would not be sensible to punish, for instance, rocks and trees. It wouldn’t make much sense, if the limb of a tree fell on you, to charge the tree with assault. In our law capacity requires that a person must possess two abilities: 1) to appreciate the wrongfulness of his or her conduct (referred to as “insight”); 2) to act in accordance with that appreciation (referred to as “self-control”).
In our law, one may lack capacity because of mental illness (this was previously called the “insanity” defence – now called pathological incapacity) and youth (very young children below 10 years are regarded as lacking capacity). More recently (although controversially) we have recognised that severe intoxication (S v Chretien 1981 AD), and severe emotional stress (S v Wiid 1990 AD; Eadie 2002 SCA) can deprive a person of capacity and responsibility. Cases of non-responsibility due to severe emotional stress have come to be called non-pathological criminal incapacity (“non-path”). In the case of Eadie the Supreme Court of Appeal confirmed that non-path is a valid and complete defence in South African Law – in theory. The judgment went further though and raised the bar for claims of this defence. The effect of this judgement (Eadie) has been to virtually shut down the defence in practice.

An astute observer may ask whether there is any relationship between a claim to lack capacity and the second apparent defence raised by Pistorius – in his own testimony – that he acted involuntarily (see “Pistorius’s Second Defence: Involuntariness” on this site). The answer is that there is. In the leading case of S v Eadie (referred to above – which effectively shut down the defence of non-pathological criminal incapacity in practice) the SCA indicated that the requirement of capacity to act in accordance with an appreciation of wrongfulness (the second leg of capacity) and voluntariness were equivalent. This is the reason why non-pathological incapacity became such a difficult defence – because involuntariness is a difficult defence to succeed with. The court in Eadie indicated that a defence of incapacity to conduct oneself in accordance with an appreciation of wrongfulness must be established on the same basis as a defence if involuntariness. A defence of involuntariness is exceedingly difficult to prove. The essence of involuntariness is that the accused’s mind did not control his/her conduct. Well known examples are instances of epilepsy and sleepwalking. Also, the prosecution is aided by a form of presumption (“a natural inference”) that the court will rely on: that conduct of an accused is ordinary voluntary and that if the accused wishes to disturb this natural inference, the accused will have to lay a sound basis for this. There seems to be no sound basis for a claim to incapacity for self control or involuntariness – unless the testimony of Dr Vorster is treated as such, or perhaps, the enquiry that her evidence triggered may provide. Her evidence, apparently (if media reports are correct) included statements that Pistorius would respond “differently” (presumably to “normal” people) and also that his mental condition may have affected his ability to control his conduct (act in accordance with an appreciation of wrongfulness).

If a person’s capacity is placed in issue, a court may refer the accused for an enquiry into his/her mental condition (s 78(2)). At this point, the provision is permissive – the Court may refer the accused. If, however, a mental illness is implicated as the reason why the accused’s capacity for self-control may have been absent, and the prospect of a pathological incapacity (“insanity”) defence is raised, the Court must refer the accused – the court has no discretion. Vorster diagnosed the accused with a psychiatric disorder (generalised anxiety disorder) and insisted that this did not constitute a mental illness for the purpose of a defence of pathological incapacity. The issue then turned on whether the condition that Vorster diagnosed did constitute a mental illness. A mental condition or disorder must constitute a mental illness (or defect) to form the basis of a pathological incapacity defence. The obvious question then is what is a mental illness or defect?

The defence of pathological incapacity (“insanity”) appears in s 78 (1) of the Criminal Procedure Act:
“A person who commits an act or makes an omission which constitutes an offence and who at the time of such commission or omission suffers from a mental illness or mental defect which makes him or her incapable—
(a) of appreciating the wrongfulness of his or her act or omission; or
(b) of acting in accordance with an appreciation of the wrongfulness of his or her act or omission, shall not be criminally responsible for such act or omission.”

Notably there is no definition in the Act or the law at all for a mental illness or defect. The closest the law has ever come to a definition, and this seems to be treated as the unofficial definition, was stated in S v Stellmacher (1983 SWA): mental illness or defect must be: 1) a pathology and 2) of endogenous (internal) origin. Regrettably, this is unhelpful. A pathology is simply a synonym for illness. There is also some suggestion that the disorder should be somehow “morbid” (S v Mabena 2007 SCA), but regrettably, this only means, once again, that the condition must be a disease condition, which is another synonym for illness.
On the (2nd) requirement that the disorder must be of “internal” origin, this is a virtually impossible criteria to apply and one that attracts deep controversy in the social sciences as to whether a valid distinction can be drawn on the basis of the source of a disorder. It is so contested that the Diagnostic and Statistical Manual of Mental Disorders (DSM, which lists all currently recognised mental disorders) only rarely comments on the source. The one clear exception is in respect of post traumatic stress disorder in respect of which the DSM stipulates that it is caused by an extraordinarily stressful event. The problem is that it seems well recognised that most disorders arise out of a combination of “internal” and “external” factors. While an internal factor, such as a person’s genes, may predispose the person to developing a particular disorder, the “trigger” for the disorder may well come from the person’s environment (external to the person).
In the circumstances, there is truly no reliable definition. At this point one may think that it is just as well that a matter is referred to an expert panel of mental health practitioners – surely they can tell the court. However, one must wonder how a mental health practitioner can possible tell the court whether a disorder is a mental illness or not because these concepts are “legal concepts” which remain undefined in law.
One may wonder how our law proceeds in the circumstances – you would do well to wonder. It is a source of deep mystery.

In the Pistorius case the court was confronted with testimony of a well respected mental health expert to the effect that Pistorius suffered from a mental disorder that would (indirectly) make him dangerous and took the view that this may reasonably possibly (S v Malatji 2013 GNP) constitute a mental illness. Once that was recognised and coupled with Voster’s testimony that Pistorius’s capacity for self-control at the time of the conduct in question may have been affected, the court had no choice. It had to refer him (s78(2)).

What remains of some interest is whether Pistorius will be committed for 30 days, as is customary, or be observed as an outpatient. It would unquestionably be exceptional for a referral to be observed as an outpatient, but that is perhaps not enough for anyone to question why Judge Masipa would even contemplate this. We should, of course, wait for her actual order and her reasons for the order she makes. We owe her at least that, and more.

However, while we will have to wait and see, it is worthwhile observing that an enquiry ordered in terms of s 79(2)(a) does not require commitment for 30 days – it permits it – the word is “may”. The wording of the sections is as follows:
“(2) (a) The court may for the purposes of the relevant enquiry commit the accused to a psychiatric hospital or to any other place designated by the court, for such periods, not exceeding thirty days at a time, …” (Authors emphasis)

Beyond that, perhaps there is already something in what Judge Masipa has said that we should celebrate. We should celebrate a judge who shows preparedness to do something unconventional – so long as it serves its purpose. Just because the convention is to commit accused people for 30 days doesn’t make it right, and doesn’t require that we continue this practice if there is a better alternative.
In my view an order that Pistorius is to be observed as an outpatient will not achieve the required purpose. A mental health enquiry is ordinarily done on an in-patient basis to permit for constant observation: 24/7. This allows psychiatrists, psychiatric nurses, social workers and psychologists to observe the patient in all contexts, in interaction with other patients, at night, to observe how much and how well an accused sleeps and eats, and allows the psychiatrists to take complete control of an accused’s medication. I doubt this extent of control and observation can be dispensed with. Nevertheless, it is always refreshing and encouraging to hear a judge question the status quo.

It is worth remembering – because this issue continues to be misunderstood: one may intentionally kill another human being – so long as you are justified (such as in self/private defence) and you know/believe you are justified.
If you are not justified (such as you are not under attack), you are only liable to a murder conviction if you know/believe you were not justified – if you know that you were not under attack. If you know you were not justified, you have intention in South African law. Intention is an entirely subjective enquiry. It turns on what the accused was actually thinking.
If you are not justified, and you did not know that you were not justified (that is, you were mistaken) you cannot be convicted of murder. In this scenario (of mistake) you can still be convicted of culpable homicide, if the mistake you made was not one that the reasonable person, in the circumstances, would make. The enquiry into negligence is objective and essentially a comparison between the conduct of the accused, and what the reasonable person would have done in the circumstances that the accused was in. 
In this enquiry, everything depends on what one regards as the “circumstances of the accused” and one’s construction of the reasonable person. The more characteristics of the accused that are attributed to the reasonable person, the more like the accused the reasonable person becomes, and the less likely a court will be to find that the accused was negligent. The significance of what constitutes the “circumstances of the accused” is whether this includes anything “internal” to an accused or anything beyond the immediate circumstances of an accused. The answer to this is that our law has, until now, only been prepared to take into account the immediate external
circumstances of the accused, and has not been prepared to attribute to the reasonable person any disorders, disabilities, or deficiencies on the part of an accused person (S v Ngubane 1985 AD). Quite the opposite, our law has instead attributed to the reasonable person the skill and knowledge that is required to perform the task or activity that the accused was engaged in. Thus, the conduct of an accused who performs a surgical operation from which the victim dies, will be compared to the conduct of a reasonable surgeon. The conduct of an accused who uses a firearm will be compared to that of a reasonable firearm owner.

All crimes/offences have an “unlawfulness” requirement built in. You are only liable to criminal punishment if your conduct is “unlawful”. Unlawfulness is judged by the “legal convictions of the community” but ultimately, by the values in the Constitution. Thus the unlawfulness requirement in every offence is really a requirement that one’s conduct must be unconstitutional. It is therefore not possible to be convicted of any offence, including any under the Protection of State Information Bill (once enacted), if one’s conduct is Constitutional. That is, if you would be permitted by the values in the Constitution (such as the rights to freedom of the press and to information and expression) to possess or disclose classified information, doing so would be lawful, and one cannot be convicted of any offence for doing so.
Although this defence which I will call the “Consitutional-lawfulness defence” must be recognised by our Courts, I appreciate that it will come as little consolation to anyone contemplating possessing or disclosing classified information – until the courts have recognised the defence and set out its parameters and the relevant factors for consideration. This raises a problem, as things stand, of uncertainty.
Certainty of law (which arises out of the rule of law – also entrenched in the Constitution) requires, at the very least, that people who are subject to the law must be able to determine with reasonable accuracy, what they are permitted and not permitted to do.
Although the overriding Constitutional-lawfulness defence must be recognised, I do think it would be preferable, even demanded by the rule of law, for the parameters and relevant factors of this defence to be spelt out in the Bill, in as much detail as possible, so that anyone contemplating ostensibly contravening the Protection of State Information Bill (once enacted), will have as much certainty as possible.
Although I am open to persuasion, it is not clear to me that the latest amendments to the Bill (in Bill B6D) achieve what is required. In the circumstances, and assuming that the Bill in its current form is made law and enforced, the Consitutional-lawfulness defence may need to be relied on.

* Thanks to David O Sullivan for taking an interest in this argument and for challenging me to summarise it as (I hope) I have done here.
** This argument was originally published in very brief form on Politicsweb in Dec 2011 (available at http://www.politicsweb.co.za/politicsweb/view/politicsweb/en/page71619?oid=271256&sn=Detail).
*** The full piece (including the argument that despite the existence of a defence in the “unlawfulness” requirement, the demands of legal certainty require that the defence is set out) was published in the South African Journal of Human Rights, Vol 28, Issue 2, 2012. Ironically, although I wrote it, I am not permitted to post it on the net. It is possible to access it, by subscribing to the journal, or otherwise purchasing a copy through Sabinet (for a whopping R200 – none of which I will ever see). If it were my choice, I would make it available freely.

Murder is the intentional unlawful killing of another human being. Culpable homicide is the negligent unlawful killing of another human being.
You are entitled (justified) in law to intentionally lawfully (such as, in self-defence) kill another human being. There is no criminal liability for doing so. Whether you are under attack and entitled to act in self defence is judged objectively, on the facts. Pistorius is not making this claim. He accepts he acted unlawfully – he was not entitled to shoot at anyone that night.
In addition, you will be excused on a murder charge for unintentionally unlawfully killing another human being. This is a mental state defence – it denies the required guilty mental state.
For murder, you must intend to unlawfully kill. If you are mistaken, and genuinely believe you are acting lawfully (such as in private-defence (the technical name for the defence under which self-defence is located)), whereas you are not acting lawfully, you cannot be convicted of murder because you don’t intend to act unlawfully. To escape a conviction of culpable homicide this mistake must be reasonable – one which the reasonable person may make. But on a murder charge, it is enough, for an acquittal, if the accused was subjectively mistaken.
This defence (that you mistakenly thought you were entitled to act in private-defence) is known as “putative private-defence”. Until his testimony, this was Pistorius’s defence. I say “until his testimony”, because during his testimony, he seems to be claiming that he fired at the toilet door by accident. This is vastly different – a claim of “accident” amounts in law to a claim of involuntariness. The defence of involuntariness is well recognised – examples include movement during an epileptic seizure and sleepwalking. The essence of the defence is that your mind did not direct or control your conduct. His testimony seems to be raising this defence. In Tasha’s, the gun in his hand simply went off by itself. His claim is he did not pull the trigger. This is not, at least, at odds with what he has said before on this firearms charge because his testimony is, to my knowledge, the first indication of his defence on this charge.
However, on the murder charge, his defence, until his testimony, has been that he mistakenly thought he was entitled to act in self defence. This, as I’ve said, is a valid defence. Yet in his testimony, he seems to be changing his defence. He seems to be claiming that the discharge of his firearm was an accident or at the very least, that his conduct was not under the control of his mind. This is again a claim that the gun had gone off in his hands, but he had nothing to do with it. This seems to keep happening to Oscar.
A claim to involuntariness is a difficult one because our courts assume that ordinary conduct is voluntary. If you have done something, you need to lay a basis for a claim to have done so involuntarily – because the courts presume voluntariness (S v Henry 1999 SCA). There appears to be no basis for this claim – at least nothing in the evidence that I have seen so far. To be fair, the interpretation of “accident” as a claim to involuntariness is a technical one. It is understandable that an accused who is unfamiliar with the law and the legal implications of what he is saying may make this mistake. This is not, in my view, the problem. The problem is that he seems unclear as to what his defence is. Until his testimony his defence has been one of putative private defence. It is consistent with such a defence to say: I intended to kill – although I thought I was doing so lawfully; I intended to kill the person behind the door who I thought was an intruder and that I had to use lethal force. On a charge of murder, there is nothing inconsistent with innocence to intentionally kill someone – so long as you think you are doing so lawfully (such as in private defence). But if this is your defence, it makes no sense to deny having intended to kill anyone. An accused who does so would appear to be unclear about his/her defence.
For me this gives rise to a crucial question: Why would an accused be so unclear about his defence that it seems to change as he testifies on the stand?

Adv Roux is right that s 197 of the Criminal Procedure Act only allows the prosecution to cross-examine the accused as to his bad character if the accused testifies as to his good character. This is the statutory position – but it is not exhaustive of the law on the issue.

At common law (the law made by court judgements), it seems that the prosecution enjoys the right to produce what may be regarded as new evidence of bad character, in response to an accused who testifies as to his/her good character. (Schwikkard, Principles of Evidence, p 61).

Zeffertt & Paizes (South African Law of Evidence) state that: “Although it is most unusual, the prosecution may counter evidence of good character by calling a witness to swear to the accused’s bad character” (p 249). Zeffertt and Paizes cite
R v Rowton ((1865) Le & CA 520, 169, ER 1497) on this point, and, to be fair, the concept of bad character, in this case, was understood in what appears now to be an outdated mode: of character as restricted to general reputation. Nevertheless, and however it is understood, it would seem that since the prosecution may produce new evidence as to bad character from a witness, it may, in principle, produce new evidence in response to an accused who puts his own character in issue.