THE SUPREME COURT OF APPEAL
OF SOUTH AFRICA
Case number : 500/03
REPORTABLE
In the matter between :
ANGELO HAMMOND APPELLANT
and
THE STATE RESPONDENT
CORAM : BRAND, CLOETE JJA, COMRIE AJA
HEARD : 23 AUGUST 2004
DELIVERED : 3 SEPTEMBER 2004
Summary: Evidence ─ sexual misconduct cases ─ the purpose for which evidence
of a first report in a sexual misconduct case may be used, analysed and defined.
_________________________________________________________
JUDGMENT
CLOETE JA/
CLOETE JA:
[1] Much confusion has been caused by the common law rule of evidence which permits the fact and contents of a complaint in a sexual misconduct case to be put before a court. The requirements for admissibility of such evidence are not in issue in the present appeal. Its permissible use is.
[2] The appellant was convicted of rape by a regional magistrate and sentenced to ten years imprisonment. An appeal to the Cape High Court was dismissed on the basis that there was no misdirection by the magistrate. In fact the magistrate misdirected himself in several fundamental respects. The court a quo refused leave to appeal further but granted bail to the appellant pending an application to this court for such leave. The application was made and granted.
[3] The evidence was to the following effect. The appellant, with his friends, came across the complainant and her friends at the beach at about noon on the day in question, which was a Sunday. The complainant and her friends were drinking beer. The appellant and his friends were drinking wine, which they shared with the complainant and her friends. At about four pm the complainant left with the appellant and his friends in the appellant’s motor car. He had promised to give her a lift home to Mitchells Plain. She had to collect her children, who were with what she termed her ‘parents-in-law’ (presumably the parents of her divorced husband) because the children had to attend school the following day. Instead, the appellant drove to Crossroads where beers were purchased and thereafter, to Westridge where brandy was purchased. The complainant admitted drinking four glasses of brandy mixed with Coca Cola as well as several beers during these detours. When they left Westridge, instead of taking her home, the appellant drove back to the beach ─ according to the complainant, despite her protestations and according to the appellant, with her express consent.
[4] The central dispute revolves around what happened next. It is common cause that the appellant had sexual intercourse with the complainant at a sandy place about fifteen metres from the vehicle. According to the complainant, the appellant dragged her by her hair during the course of which she stumbled and gashed her leg, and despite her protestations (she says she was hysterical) and the resistance she put up, he raped her. According to the appellant, he made amorous advances towards the complainant whilst they were in his motor car, she reciprocated and they had consensual sexual intercourse. The appellant’s version was that the complainant had gashed her leg earlier, when she stumbled and fell. The appellant also said that the complainant suggested that they return to her house so that they could continue their liaison. The appellant’s evidence that the complainant was a willing party was corroborated by a defence witness, Mr Mitchell. Mitchell, a close friend of the appellant, was with the appellant and the complainant when they returned to the beach. He said that he sat in the back seat of the vehicle and that after the appellant and the complainant had sat together in the front seat for a while, the two of them walked off together. He specifically denied the complainant’s version that she had been dragged away by her hair. It is common cause that whilst the two of them were having sexual intercourse, Mitchell walked over to them and asked the appellant how much longer they were going to be.
[5] It is also common cause that after the appellant and the complainant had had sexual intercourse, the appellant’s vehicle became stuck in the beach sand. A Land Rover arrived with four fishermen in it. Two ─ Messrs Steyn and English ─ gave evidence on behalf of the State. Steyn, somewhat hesitantly, estimated the time of their arrival to have been around nine to ten pm. Both said that the complainant approached them, alleged that she had been raped, attempted to climb into their vehicle and asked them to call the police. According to Steyn:
‘She looked ─ she looked very wild and haggard… She looked like she had been partying the whole afternoon… [S]he was a bit teary… She was teary. She was crying… It is just her hair looked very wild like she had just woken up… She looked like she was partying.’
According to English:
‘She was upset and her hair was all wild and things… she was upset, she ─ I think she was crying, ja. And anyway we proceeded to help these guys to get their car out of the sand. And in that time she wanted to get into the van and she was mumbling on about, you know, that she has been raped and things… She was upset. To my mind she was upset, but I know ─ I am not a person with a breathalyser or anything but to my mind she looked like she had been drinking, she had [had] alcohol of some sort.’
[6] The fishermen pulled the appellant’s vehicle out of the sand. Both Steyn and English said that the complainant did not want to leave with the appellant. Steyn said that he did not believe her story that she had been raped and the complainant confirmed that this had indeed been her impression. When the appellant’s vehicle got stuck in the sand again, one of the fishermen called the police ─ apparently because they wanted to get rid of the complainant.
[7] The police, including Inspector McNabb, arrived at the scene. McNabb said that he was in the charge office when the call to render assistance was received at about 11 pm and that he was at the scene about 10 minutes thereafter. McNabb said further that the complainant was considerably upset, that he had talked to her after he had put her in the front seat of the patrol van and that during this conversation she began crying.
[8] The appellant and his three companions were arrested. They and the complainant were taken to the police station. The complainant was later taken by Inspector Hendricks from the police station to the district surgeon, who examined her at 1:15 the following morning. The doctor testified that such injuries as he found to the complainant’s private parts could have been sustained during consensual intercourse. He also confirmed that the gash on her leg was consistent with both the State and the defence versions as to how it had occurred.
[9] The magistrate found that the complainant was a very good witness. He went on, however, to point out unsatisfactory aspects of her evidence but concluded that he could not find that her evidence was untruthful. The magistrate also found that despite the state of the complainant’s sobriety, he could safely rely on her evidence, particularly in view of the finding by the district surgeon ─ who, the magistrate said, had examined the complainant shortly after the incident ─ that she was ‘not drunk’. But Inspector McNabb, who had served in the police force for 9 years, said bluntly and without qualification, both in his evidence-in-chief and under cross-examination, that the complainant was drunk; Inspector Hendricks said that when he went to collect the complainant at about 1 am to take her to the district surgeon, she was asleep in the charge office; it was in fact at least two and possibly more than four hours after the incident when the district surgeon examined the complainant; and the district surgeon readily conceded in cross-examination that she could have sobered up in the meantime. Inspector McNabb’s evidence also impacts adversely on the complainant’s credibility. She claimed that she was sober and that her faculties had not been impaired even although she had had what she termed a little drink (‘’n drinkie’) ─ which she defined as six beers or a bottle of brandy. In view of the quantity and variety of alcohol which she had imbibed and the evidence of Inspector McNabb, her evidence as to her sobriety falls to be rejected.
[10] The magistrate found that the appellant had not contradicted himself; said that he had given his evidence very clearly; and commented that he had sketched a very good picture for the court of what had happened. He was obviously a good witness. The magistrate further found that there were no material contradictions between the evidence of the appellant and the defence witness, Mitchell. But the magistrate said that Mitchell did not impress him at all. This view was based partly on the fact that Mitchell, as a co-accused, had been present in court until he was discharged at the end of the State case and also the fact that he had constantly looked towards the appellant and his attorney before answering questions in cross-examination. The first point of criticism cannot carry much weight as Mitchell was presumably not present when the appellant gave his evidence-in-chief or was cross-examined, otherwise this would no doubt have been brought out by the prosecutor and would have been cause for comment by the magistrate. Of course Mitchell was present when the appellant’s version was put to the complainant; but that was also his version and he was cross-examined on it. So far as the second point of criticism is concerned, the magistrate obviously considered that Mitchell’s demeanour left much to be desired but there is no suggestion that he was being prompted by the appellant or his legal representative and he was not in any way tripped up in cross-examination. Despite his shortcomings, his evidence does provide corroboration of the appellant’s version that he, the appellant, had not dragged the complainant out of the vehicle by her hair and that she was a willing party; and Mitchell was peculiarly well placed to give this evidence as he was sitting immediately behind them in the appellant’s motor vehicle before they had sexual intercourse, he saw them leave and he went over to them whilst they were in the act.
[11] The magistrate gave essentially two reasons for convicting the appellant. The one which requires some detailed consideration appears from the following passages in the judgment (which I have translated):
‘The court must weigh the totality of the evidence. The court cannot adopt a compartmentalized approach to each witness’s evidence. The court must, as I have mentioned, have regard to the evidence as a whole, the complainant right through the whole spectrum up to the end of the evidence of the defence witness. And if the court [has regard to] that evidence, the court must look at the probabilities in this case that a woman, who was quite willing to go and have further intercourse, would suddenly say, well now, I was raped… The court must ask itself, is this conduct of the complainant consistent with the probabilities in this case. The court has considered the evidence, and weighed it. There is absolutely no reason why this woman, shortly after she had had consensual sexual intercourse, would suddenly complain or would shout that she had been raped.’
This approach by the magistrate constitutes a fundamental misdirection as to the purpose for which the evidence of a complaint in a case of sexual misconduct such as the present may be received.
[12] It is often said that the fact that a complainant in a sexual misconduct case made a complaint soon after the alleged offence, and the terms of that complaint, are admissible for two purposes, namely, to show the consistency of the complainant’s evidence, and to negative consent: See eg R v M 1959 1 SA 352 (A) 355G-H.
[13] In the seminal English case of R v Lillyman [1896] 2 QB 167 Hawkins J, giving the judgment of the court (the other members being Lord Russell of Killowen CJ, Pollock B, Cave and Willis JJ) said at 170:
‘It is necessary, in the first place, to have a clear understanding as to the principles upon which evidence of such a complaint, not on oath, nor made in the presence of the prisoner, nor forming part of the res gestae, can be admitted. It is clearly not admissible as evidence of the facts complained of: those facts must therefore be established, if at all, upon oath by the prosecutrix or other credible witness, and, strictly speaking, evidence of them ought to be given before evidence of the complaint is admitted. The complaint can only be used as evidence of the consistency of the conduct of the prosecutrix with the story told by her in the witness-box, and as being inconsistent with her consent to that of which she complains.’
The learned judge continued at 177:
‘The evidence is admissible only upon the ground that it was a complaint of that which is charged against the prisoner, and can be legitimately used only for the purpose of enabling the jury to judge for themselves whether the conduct of the woman was consistent with her testimony on oath given in the witness-box negativing her consent, and affirming that the acts complained of were against her will, and in accordance with the conduct they would expect in a truthful woman under the circumstances detailed by her.’
[14] Lillyman was followed in R v Osborne [1905] 1 KB 551 where Ridley J, delivering the judgment of the court (the other members being Lord Alverstone CJ, Kennedy, Channell and Phillimore JJ) said at 557-558:
‘By the judgment in Reg. v. Lillyman it was decided that the complaint was admissible, not as evidence of the facts complained of, nor as being a part of the res gestae (which it was not), but as evidence of the consistency of the conduct of the prosecutrix with the story told by her in the witness-box, and as being inconsistent with her consent to that of which she complains. Mr Marchant argued upon this that the reasons so given were one only, and that the consistency of the complaint with the story given by the prosecutrix was material only so far as the latter alleged non-consent. If, however, that argument were sound, the words in question might have been omitted from the sentence, and it would have been sufficient to say that the complaint was admissible only and solely because it negatived consent. We think, however, if it were a question of the meaning of words, that the better construction of the judgment is that while the Court dealt with the charge in question as involving in fact, though not in law, the question of consent on the part of the prosecutrix, yet the reasons given for admitting the complaint were two─first, that it was consistent with her story in the witness-box; and, secondly, that it was inconsistent with consent… [I]t appears to us that, in accordance with principle, such complaints are admissible, not merely as negativing consent, but because they are consistent with the story of the prosecutrix.’
[15] In Kilby v R [1973] 1 ALR 283 (High Court of Australia) Barwick CJ (in whose judgment McTiernan, Steven and Mason JJ concurred) said at 287 lines 27-46:
‘[E]vidence of a complaint at the earliest reasonable opportunity is exceptionally admitted only as evidence of consistency in the account given by the woman claiming to have been raped: that is to say, it is admitted as matter going to her credit (see R v Lillyman [1896] 2 QB 167, per Hawkins J at 170; [1895-9] All ER Rep 586; Sparks v R [1964] AC 964, at 979; [1964] 1 All ER 727). Because the account with which the complaint is said to show consistency is an account of intercourse without consent, it has often been said that the evidence of the complaint is evidence negating consent. In my opinion, this manner of expressing the function of the evidence of proximate complaint is not correct: though, as it shows consistency in her account of rape, the fact of the complaint buttresses her evidence of no consent or, as it was said in R v Lillyman, supra, is inconsistent with consent. At times also it is said with technical inaccuracy that the evidence of such a complaint is corroborative of the woman’s evidence of the rape. It is quite clearly not so corroborative (see R v Christie [1914] AC 545; Eade v R (1924) 34 CLR 154; 30 ALR 257), though it is so spoken of in American literature (see Wigmore on Evidence, 3rd ed, vol IV, p 219, para 1134 and p 227, para 1137; vol VI, p 173, para 1761).’
The learned Chief Justice then embarked on a careful analysis of Lillyman, Osborne and several English textbooks, in the course of which he said the following (289 lines 35-49; 290 lines 12-29; and 292 lines 1-8):
‘In my opinion, nothing in this judgment [ie Lillyman] lends any support to the proposition that evidence of the making of the complaint is evidence of any fact other than the fact of the making of the complaint itself and of the terms in which it is claimed to have been made. When Hawkins J in the first of the two passages which I have quoted from Lillyman’s Case [that at 170 quoted in para [13] above] spoke of the evidence of a complaint as being inconsistent with consent he was not, in my opinion, intending to place its admissibility upon a second and different ground from that of its tendency to show consistency in the conduct of the prosecutrix. He was merely indicating the extent of its effect on the credit of the prosecutrix.
In my opinion, the error which has been made by text writers and in subsequent decisions is in treating this remark of Hawkins J as if it did set up a second and independent ground of admissibility. In my respectful opinion, it did not.
…
In any case, to say that Lillyman’s Case recognizes that the evidence of a proximate complaint may be used to negative consent is to make an ambiguous statement. If it means that in so far as a complaint tends to buttress the evidence of the prosecutrix that what occurred did occur without her consent and in so far as belief in truth of her statement would negative consent, it may be an acceptable statement, though, I think, prone to be, as it has proved to be, misleading. If, of course, it means that the evidence of a complaint is direct evidence negativing consent, I am of opinion that the statement is completely unwarranted, both in point of precedent so far as Lillyman’s Case is concerned and in point of logic. It is true that Ridley J in R v Osborne [1905] 1 KB 551; [1904-7] All ER Rep 54, treated the evidence of proximate complaint as admissible on two grounds, founding himself on Lillyman’s Case. He did not intend to depart from the decision or to enlarge its reasoning. But as I have indicated, Lillyman’s Case does not really warrant the conclusion that there are two distinct grounds of admissibility of evidence of proximate complaint. Always the basic authority for the contrary proposition in the texts and in the decisions has been Lillyman’s Case.
…
The admission of a recent complaint in cases of sexual offences is exceptional in the law of evidence. Whatever the historical reason for an exception, the admissibility of that evidence in modern times can only be placed, in my opinion, upon the consistency of statement or conduct which it tends to show, the evidence having itself no probative value as to any fact in contest but, merely and exceptionally constituting a buttress to the credit of the woman who has given evidence of having been subject to the sexual offence.’
[16] The authors of one of the leading English textbooks, Cross and Tapper on Evidence (eighth ed.), support the conclusion reached in Kilby and say at 300:
‘Both in Lillyman and in Osborne reference was made to the complaint being used to show consistency with the victim’s testimony, and being inconsistent with consent. In Kilby v R, it is submitted rightly, the High Court of Australia emphasised that this could not be taken to mean that it amounted to evidence of the absence of consent, nor its absence to evidence of consent. That would be to make the very hearsay use of the complaint warned against by Hawkins J, in Lillyman. The correct view is that the victim’s testimony is evidence of lack of consent, and the complaint does no more than support the credibility of the victim in so testifying.’
I respectfully agree. The remarks made by this court in M’s case to which I referred in para [12] above are not a bar to the conclusion I have reached in as much as that case was concerned with the consistency of the complaint and not with whether a complaint can negative consent, as Schreiner ACJ made clear at 355G-H; and the question whether the latter purpose is separate from and independent of the former, did not arise for decision.
[17] I return to the reasoning of the magistrate in the present matter set out in para [11] above. It would have been correct for the magistrate to have had regard to the fact of the complaint and its terms as establishing consistency in the complainant’s evidence and therefore supporting her credibility. But that is not what the magistrate did in the passages quoted. The magistrate, in weighing up the totality of the evidence, had regard to the complaint and its terms as constituting a probability in favour of the State case which tended to disprove consent, which was an issue ─ indeed, the only issue ─ in the case. That is not permissible.
[18] The magistrate’s second reason for convicting the appellant may be dealt with comparatively briefly. The magistrate reasoned as follows: Both of the fishermen Steyn and English said that the complainant’s hair was ‘wild’; the complainant said that she had been pulled by her hair from the vehicle to the place where she was raped; her version was the only explanation before the court for the condition of her hair; and the appellant’s evidence that he did not notice her hair was an attempt by him to conceal the truth from the court. I find this reasoning entirely unconvincing. A woman who has been drinking steadily since noon, who is inebriated and had just had sexual intercourse on the beach, is likely to look more than a little unkempt. There was furthermore no particular reason for the appellant ─ as opposed to the fishermen, who were suddenly confronted by the complainant ─ to have particular regard to the complainant’s hair at any stage, and his denial that he had done so is accordingly an unsafe basis for a credibility finding against him. This is particularly so in view of the specific denial by the defence witness Mitchell that the appellant had dragged the complainant by her hair.
[19] The magistrate also found corroboration for the complainant’s version in the evidence of the district surgeon that her leg had been gashed whilst she was being dragged from the car to the place where the appellant had sexual intercourse with her. That finding ignored the evidence of the district surgeon in cross-examination that this injury was equally consistent with the appellant’s version that the complainant had stumbled and fallen earlier. Bearing in mind the inebriated state of the complainant, the appellant’s version is not improbable. The injury to the complainant’s leg provides no corroboration of her version and the magistrate’s finding to the contrary was a misdirection.
[20] The magistrate did not emphasize the emotional state of the complainant, coupled with the facts that she wanted to climb into their vehicle and was not willing to leave with the appellant, as testified to by Steyn and English, or the evidence of Inspector McNabb that she was considerably upset. Such evidence (unlike evidence of a complaint and its terms) is admissible to prove absence of consent, as is clearly established by two decisions of this court. In S v S 1990 (1) SACR 5 (A) this court held at 11a-c and 12a-c:
‘Na die gebeure op die plaas het die appellant die klaagster teruggeneem na haar koshuis. Daar het sy vir ‘n wyle voor die deur bly staan, en toe na ‘n vriendin, Louise Nel, in ‘n ander koshuis gegaan en aan haar vertel dat sy verkrag is. Louise Nel het ‘n besondere grafiese beskrywing van heirdie besoek gegee. Sy sê dat die klaagster by haar kamer ingebars het sonder om te klop, en dat sy in ‘n erge geskokte toestand voorgekom het ─ asof sy die dood self aanskou het. Die klaagster het nie die appellant se naam aan Louise Nel genoem nie daar sy gesê het dat die man verloof was, en sy nie sy verloofde in die verleentheid wou stel nie. Sy het egter vir haar vertel hoedat sy deur dié man na die plaas genooi is om na “tapes” te gaan luister, en hoe hy haar daar verkrag het. Hierdie getuienis waarin die uiters geskokte toestand waarin die klaagster verkeer het steeds beklemtoon word, is nie wesenlik in kruisverhoor aangeval nie, en dit is ook nie voor ons betoog dat die landdros verkeerd was om dit te aanvaar nie. Dit bied ongetwyfeld sterk stawing vir die klaagster se getuienis dat sy verkrag is.
…
Hierbenewens word sy [the complainant] gestaaf deur die getuienis van Louise Nel, en, tot ‘n mindere mate, deur die getuienis van haar geneesheer en van die verteenwoordiger van “Rape Crisis” met wie sy gesels het. Die uiters geskokte toestand waarin Louise Nel haar gevind het kort nadat die appellant haar by haar koshuis afgelaai het, is sekerlik nie op die getuienis van die appellant verklaarbaar nie. Dit strook volkome met die getuienis van die klaagster dat sy verkrag was, en dit blyk trouens dat dit op geen ander redelike veronderstelling verklaarbaar is nie.’
In S v Jackson 1998 (1) SACR 470 (SCA) this court held at 477g-h:
‘Furthermore, on both versions the complainant fled from the car, leaving her plimsoles there. This is incompatible with the accused’s version of consensual and non-violent love-making. When the complainant reached her sister and friends, she was hysterical and immediately complained of having been raped. The district-surgeon also reported that when he examined her, she was in a state of shock. This is incompatible with the accused’s version.’
It must be emphasized that in neither case did this court say that the fact or contents of the complaint corroborated the complainant’s evidence or created a probability in favour of its acceptance. In each case the court had regarded the complainant’s emotional state and her conduct as creating that probability. The references to the complaint in the passages quoted from the S case were part of the narrative of the sequence of events. The reference to the complaint in Jackson must be interpreted as showing consistency on the part of the complainant and no more.
[21] Caution must be exercised when the emotional state of a complainant is taken into account. The English cases on this point are collected and discussed in Ramesh Chauhan (1981) 73 CAR 232. In one of those cases, Redpath (1962) 46 CAR 319 at 321-2 Lord Parker CJ said:
‘It seems to this court that the distressed condition of a complainant is quite clearly capable of amounting to corroboration. Of course, the circumstances will vary enormously, and in some circumstances quite clearly no weight, or little weight, could be attached to such evidence as corroboration. Thus, if a girl goes in a distressed condition to her mother and makes a complaint, while the mother’s evidence as to the girl’s condition may in law be capable of amounting to corroboration, quite clearly the jury should be told that they should attach little, if any, weight to that evidence because it is all part and parcel of the complaint. The girl making the complaint might well put on an act and simulate distress. But in the present case the circumstances are entirely different.’
The circumstances in Redpath were that the distressed condition of the little girl who had been the subject of the assault was observed by someone whom the little girl did not know to be there.
[22] I should perhaps say for the sake of completeness that evidence of the distressed state of the complainant is also admissible to show that sexual contact took place, where this is denied. The facts in Ramesh Chauhan provide a good example. In that matter the appellant accompanied his sister to premises where she had applied for employment, and while she was being interviewed he waited in another room where a female employee was working alone. They entered into conversation, whereupon it was alleged by the victim that the appellant touched her breast and tried to kiss her. She extricated herself and ran upstairs to the ladies’ lavatory crying. A fellow employee heard her cries and followed her. The victim explained to her fellow employee what had happened. The police later interviewed the appellant who admitted being alone with the victim but denied that any incident had taken place in which he had touched her. He maintained that when she left the room she had been behaving normally. The appellant was charged with indecent assault. At the end of the prosecution case counsel for the appellant submitted that there was insufficient corroboration for the issue to be left to the jury. The recorder ruled that the jury were entitled to regard the victim’s distressed condition described by the fellow employee as corroboration if they thought it right to do so. The appellant was convicted. The Court of Criminal Appeal held that the evidence had sufficient weight to be left with the jury, who were carefully and properly directed on the need for corroboration and the possibility of regarding the evidence of distress as corroboration. The appeal was dismissed.
[23] In the present matter, the evidence as to the complainant’s emotional state is of little ─ if any ─ assistance, in as much as it may have been due to other factors. Her friends knew that she had left the beach with complete strangers at about 4 pm. It had became late. She was intoxicated. She would have had some explaining to do to her ‘parents-in-law’ as to why she had failed to collect her children during the afternoon so that they could attend school the following day. And her fiancé, whom she said she would have seen that night, would no doubt have asked questions as to her whereabouts ─ particularly if he had seen her in the condition testified to by the two fishermen, Steyn and English. Or she may have been overcome by remorse, perhaps induced by the quantity of alcohol she had consumed. All of these possibilities have a factual foundation in the evidence led, as required by Jackson at 477c-d.
[24] On the face of it, the complainant’s conduct would, on the appellant’s version, appear improbable (and this is what led the magistrate and the court below into error): One moment she was happy to have sexual intercourse and even to continue their involvement at her home; the next moment she was upset and looking to the fishermen for assistance. Such a change in attitude would indeed be improbable in a person who was behaving rationally. But the complainant was not. She left the beach with the appellant because he was going to take her home so that she could fetch her children. Yet when there had already been a detour to Crossroads, she was quite happy to continue the party: She did not walk home from Westridge, as she admitted in cross-examination she could have. She further admitted that at Crossroads she called for a pair of pliers to cut off her engagement ring, because, according to her, it was pinching her. And according to the appellant, whilst they were at Westridge, the complainant climbed out of the vehicle and urinated on the ground in front of a queue of men who were waiting to be served at the shebeen. The complainant denied this conduct but, when asked in cross-examination why she had not mentioned the detour to Westridge in her evidence-in-chief, she said that she had forgotten about it. Her evidence on this point was unsatisfactory. In all the circumstances it would be unsafe to find that the change in the complainant’s attitude which must have taken place on the appellant’s version, renders that version improbable ─ much less false beyond a reasonable doubt.
[25] To sum up: The complainant did make a report to the fishermen that she had been raped, as one would have expected her to do. That is a factor which supports the consistency of her evidence and therefore supports her credibility. She was also upset and unkempt shortly after the incident, when the fishermen arrived. She was still upset when the police arrived. As against these facts, she was drunk and had behaved irrationally earlier that afternoon, and several reasons appearing from the evidence suggest themselves as to why she may have been in the emotional condition she was and why she may have behaved as she did. She was furthermore a single witness. Aspects of her evidence were unsatisfactory and she lied as to the state of her sobriety. A cautionary approach to her evidence was required for that very reason: Jackson at 476f and 476i-477a. By way of contrast, the appellant’s evidence was beyond reproach. There is furthermore no gainsaying the fact that his version was corroborated by Mitchell, even if one approaches the latter’s evidence with reservations (because he was a good friend of the appellant) and with caution (because of his demeanour). The magistrate’s two reasons for convicting the appellant, namely, the state of her hair and the probability constituted by the complaint she made to the fishermen, cannot be sustained. In all the circumstances, it cannot be said that the appellant’s guilt was proved beyond reasonable doubt.
[26] The appeal succeeds. The order of the court a quo is set aside and the following order substituted:
‘The appellant’s conviction and the sentence imposed are set aside.’
________________
T D CLOETE
JUDGE OF APPEAL
Concur: Brand JA
Comrie AJA