James Grant

Many spectacular and sensational claims have been levelled at Oscar Pistorius in the media: that he has a bad temper, threated to break someone’s legs, is obsessed with firearms, fired a gun out of his sunroof, and another in a restaurant, was in illegal possession of ammunition and browsed pornographic websites on the night that he shot and killed Reeva Steenkamp. Presumably these claims – if proved – would tend to show that Oscar has a ‘bad character’. But there is a question that demands an answer: what does any of this matter? What would it matter if Pistorius can be shown to have a bad character? The answer is that it depends. During the plea explanation tendered by Pistorius, his defence team indicated that they would oppose the admissibility into evidence of ‘inadmissible’ character evidence as similar fact evidence. The admissibility and value of similar fact and character evidence is far from straightforward, but in light of the intense public interest in the Pistorius case, and the aspersions cast on his character, it seems necessary to attempt to deconstruct the law on this issue. Part of the complexity of trying to understand the place and admissibility of character and similar fact evidence in law is that it seems that the target is moving. When focused on similar fact evidence (concerning prior bad acts), it may seem that one is truly concerned with the character of the accused, and the opposite, when attempting to focus on character evidence, it may seem that one is truly concerned with similar fact evidence. These intuitions are, in my view, correct. That is because of the inextricable link between the two forms of evidence. Ordinarily, evidence of the disposition of a person is inadmissible because it is simply irrelevant. We know that good people do bad things, and bad people do good things. But there may be occasions on which, or people, in respect of which, sometimes, what they have done in the past, or the way they have done something, does seem to matter. In the classic case of R v Smith ((1915) 11 Cr App Rep 229) in which the accused was charged with murdering his wife – found dead in the bath. His defence was that she had suffered an epileptic seizure and drowned. There was evidence that he would benefit financially from his wife’s death. More damning though, was evidence that this had occurred on another two occasions: two other female partners had died in the bath and the accused stood to benefit from their deaths. This was all a terrible coincidence, or there was something about the accused that linked the incidents – that made it more likely that he had killed all three of them. This is indeed what the court found – the evidence of the other deaths made the claim of an accident in respect of the death of his wife implausible. It is at the point where we must say that there is something about the accused that makes it more likely that he is guilty that we are relying on evidence of disposition. I will argue also that, while ‘similar fact evidence’ is conventionally concerned with specific incidents from which inferences are sought to be drawn, and ‘character evidence’ is, perhaps not conventionally, but more realistically concerned with the disposition of an accused, the link between the two is that we can only draw valid inferences about the disposition of an accused from specific instances of prior conduct (similar fact evidence). When one is dealing with similar fact evidence – evidence of ‘prior bad acts’ – one is, in truth, dealing with whether this gives us some basis to draw inferences about the character – the disposition of the accused. For almost eternity (under Makin v Attorney General for New South Wales [1894] AC 57 (PC)), our law insisted that similar fact evidence was not admissible if it relied on aspersions on the character of the accused – if it operated upon the disposition of the accused. We seem now to be clear of this (DPP v Boardman [1975 AC 421 (HL); S v D 1991 (2) SACR 543 (A)) so that we no longer need to, nor should we (DT Zeffertt and AP Paizes The South African Law of Evidence 2nd ed (2009)), pretend that similar fact evidence is not, in truth, about disposition. Regarding character evidence, this is/was distinct on the basis that conventionally it referred exclusively to the general reputation of a witness – what the public generally thought of him or her. On this thinking, it was not evidence as to the disposition of the accused which may, in turn, be inferred from evidence of instances of improper conduct. This no longer appears to be the position (R v Malindi 1966 (4) SA 123 (PC) quoting with approval Stirland v Director of Public Prosecutions.[1944] AC 315). In essence then, both ‘bad character’ evidence and similar fact evidence seem to be concerned with instances of prior improper conduct of an accused from which inferences of an errant disposition can be inferred. The alleged misconduct of Pistorius regarding firearms and ammunition form the basis of several charges, in addition to the murder of Reeva Steenkamp which Pistorius is now facing. Importantly though, the question that I wish to address here is whether the evidence of his alleged misconduct regarding firearms and ammunition, and beyond that, of the other (character/dispositional) evidence against him, show that it is more likely true that he is guilty of the murder of Reeva Steenkamp. The general rule, in respect of both similar fact and character evidence, is that such evidence is inadmissible. The rationale is that it is simply irrelevant: its persuasive value is marginal at best and it is typically highly prejudicial – it triggers stereotypical thinking and wastes time and money, and causes the court to be drawn off point. It raises collateral issues which, in itself is problematic, but worse, an accused may be ill prepared to explain everything s/he has done in his/her life. However, there are several grounds on which evidence as to the ‘bad’ disposition of an accused can be exceptionally admitted. In terms of ‘character evidence’, at common law, the most prominent is where the accused leads evidence of his/her good character, the state may respond and present evidence of bad character – by cross-examining the accused or other witness or leading its own witnesses to testify as to the bad character of the accused. In addition, s 197 of the Criminal Procedure Act allows for the cross examination of accused person as to his bad character, (a) where s/he places his good character in issue (as under the common law), or impugns the character of a state witness; (b) gives evidence against a co-participant to the offence, (c) the charges relate to the possession of stolen property; or (d) proof of some prior bad act is admissible to show s/he is guilty on the current charge. It is worth repeating that this section relates only to the cross examination of the accused. The specific wording of the section is as follows: An accused who gives evidence at criminal proceedings shall not be asked or required to answer any question tending to show that he has committed or has been convicted of or has been charged with any offence other than the offence with which he is charged, or that he is of bad character, unless – (a) he or his legal representative asks any question of any witness with a view to establishing his own good character or he himself gives evidence of his own good character, or the nature or conduct of the defence is such as to involve imputation of the character of the complainant or any other witness for the prosecution; (b) he gives evidence against any other person charged with the same offence or an offence in respect of the same facts; (c) the proceedings against him are such as are described in section 240 or 241 [relating to possession of stolen property] and the notice under those sections has been given to him; or (d) the proof that he has committed or has been convicted of such other offence is admissible evidence to show that he is guilty of the offence with which he is charged.
It is worth dwelling on the meaning and import of the final exception (197(d)). As mentioned, this effectively permits cross-examination as to some ‘prior bad act/s’ where this would be admissible to prove the current charge. This is an indirect incarnation of the rules relating to similar fact evidence – here making cross examination contingent on whether the similar fact rules would be satisfied. What is similar fact evidence, and when is it admissible? The law on similar fact evidence was governed for a long time by the judgement from the case of Makin (Makin v Attorney-General for New South Wales 1894 AC 57 (PC)). In Makin, the skeletal remains of a baby were found in the garden of a residence rented by the Makins. Their defence was that, while they conceded they had improperly disposed of the body of the baby, the baby had died of natural causes: they had not murdered the child. What made the case interesting, is that the prosecution sought to introduce evidence that the skeletal remains of other babies were discovered in other former residences of the Makins. The court was careful that evidence which tended to show that the accused were of bad character was inadmissible, but such evidence could be admitted if it in some other way disproved the Makin’s defence: ‘It is undoubtedly not competent for the prosecution to adduce evidence tending to show that the accused has been guilty of criminal acts other than those covered by the indictment, for the purpose of leading to the conclusion that the accused is a person likely from his criminal conduct or character to have committed the offence for which he is being tried. On the other hand, the mere fact that the evidence adduced tends to show the commission of other crimes does not render it inadmissible if it is relevant to an issue before the jury, and it may be so relevant if it bears upon the question whether the acts alleged to constitute the crime charged in the indictment were designed or accidental, or to rebut a defence which would otherwise be open to the accused.’ Makin spawned a legacy of lawyers who thought about similar fact evidence in terms of categories of evidence, rather than degrees of relevance. Such evidence was admissible only if it qualified somehow, into a neat exceptional category, such as, did the evidence go to show that the conduct in question was designed and intentional, rather than a mistake, or did it somehow otherwise rebut a defence available to an accused. We have now developed, at least in theory, to the point at which our courts seem to appreciate that the true question is whether the evidence (of prior improper conduct) is relevant: whether it properly proves anything (DPP v Boardman [1975 AC 421 (HL); S v D 1991 (2) SACR 543 (A)) so that we no longer need to, nor should we (DT Zeffertt and AP Paizes The South African Law of Evidence 2nd ed (2009)) pretend that the disposition of the accused is not the real issue. While the position is that similar fact evidence is generally inadmissible – exceptions to this rule are also accepted. Our courts have previously insisted on some form of a link or nexus (R v Bond [1906] 2 KB 389 (CCR)) and currently, a ‘striking similarity’ between the prior improper conduct and the conduct in question (DPP v Boardman [1975 AC 421 (HL); S v D 1991 (2) SACR 543 (A)). Nevertheless, it seems that at issue is really only whether any sort of pattern can be discerned from the former conduct, that will sustain inferences being drawn to the conduct in question. It is not clear that our courts are yet comfortable with drawing inferences based on the disposition of the accused, but there appears to be no good reason to shy away from doing so where the disposition of the accused, by virtue of his prior conduct, is relevant (DT Zeffertt and AP Paizes The South African Law of Evidence 2nd ed (2009)). Thus, evidence as to specific prior improper conduct is admissible as ‘similar fact evidence’ where it would establish a pattern which reflects the disposition of the accused, from which the court could draw an inference as to his/her guilt. The topic of ‘character evidence’ does not carry this much further. Putting aside the specific statutory provisions relating to co-accused people and possession of stolen property, the rules regulating character evidence as character evidence allow for the leading of ‘bad character’ evidence or the cross-examination of the accused about his bad character if s/he pursues the line that s/he cannot be guilty because he is a good character. It also allows cross examination of the accused if s/he impugns the character of a state witness. Beyond that the rules governing character evidence as character evidence allow for the cross examination of the accused to establish what would be permitted by the rules regarding similar fact evidence. The bottom line is that if evidence of prior improper conduct by an accused would tend to properly (without, for instance, operating on stereotypes and without wasting time) show that the accused is more likely guilty, the evidence is relevant and admissible. How much weight it is given in the final analysis is another matter entirely. It is worth considering whether this is the case for Pistorius. Presuming that the state can prove allegations of prior improper conduct, would the evidence of a bad temper, a threat to break someone’s legs, having fired a gun out of his sunroof, and another in a restaurant, that he was in illegal possession of ammunition and viewed pornographic material make it more likely true that he intentionally and unlawfully shot and killed Reeva Steenkamp? This is not evidence that he shot and killed several previous girlfriends, or even evidence of repeated domestic violence. I don’t see a court accepting this evidence as establishing any sort of pattern on the basis of which it would be prepared to infer that Pistorius intended to kill Reeva Steenkamp. Of course, if the state were to produce evidence of domestic violence, then valid inferences as to his character (disposition) and ultimate guilt may be drawn. Our ability to discern patterns from complex information make us an intelligent species. But we must be careful that just because the identification of patterns help us understand our world, does not mean the world is full of patterns.  *I wish to thank Professor Andrew Paizes (co-author of Zeffertt and Paizes, The South African Law of Evidence) for his helpful comments on an earlier draft. I have amended this post in line with his suggestions. Nevertheless, any mistakes remain entirely mine.