---

FREE STATE HIGH COURT, BLOEMFONTEIN

REPUBLIC OF SOUTH AFRICA

Appeal No. : A180/09

In the appeal between:-

STEPHEN PHINDA MBATHA ….............................................Appellant

and

THE STATE ….....................................................................Respondent

_____________________________________________________

CORAM: RAMPAI, AJP et NAIDOO, AJ

_____________________________________________________

HEARD ON: 12 MARCH 2012

_____________________________________________________

JUDGMENT BY: RAMPAI, J

_____________________________________________________

DELIVERED ON: 22 MARCH 2012

_____________________________________________________

[1] The matter came to us by way of an appeal. The appellant was convicted by the regional court on 19 October 2006 in respect of the first and the third charges, in other words, rape and robbery with aggravating circumstances respectively, but acquitted in respect of the second charge, in other words, the second count of rape.

[2] On the same day he was sentenced to 15 (fifteen) years imprisonment for rape and 4 (four) years imprisonment for robbery with aggravating circumstances.

[3] The appellant was aggrieved. On 3 April 2007 he unsuccessfully applied for leave to appeal against his conviction and sentence. He was again aggrieved. He challenged, by way of a petition, the regional court magistrate’s refusal to grant him leave to appeal. He now comes to us on appeal with the leave of this court granted on 4 June 2009 by Kruger J et C.J. Musi J.

[4] The respondent alleged in the charge sheet that the appellant raped and robbed the victim, Ms Selloane Loama Mokhatla at Zamdela, Sasolburg on Saturday, 11 September 2004. Notwithstanding his plea of not guilty, the appellant was found guilty and sentenced as I have earlier indicated.

[5] The version of the state was narrated by one witness only, the victim herself. In addition to her oral evidence, a medical report by Dr. M.S. Baloyi was, by agreement, handed in as documentary evidence in support of the victim’s version – vide exhibit ‘A’.

[6] The version of the defence was also narrated by one witness only, namely Mr. Steven Phinda Mbatha, the appellant himself. He was also known as Fifi.

[7] The strict cautionary rule relating to female victims of sexual offences was abolished in S v JACKSON 1998 (1) SACR 470 (SCA). However, the rule as modified still residually applies to cases where something more can be found in a case than the mere status of a victim, as a single witness. Where the rape complainant is a single witness and some unusual feature(s) emerges in the testimony which loudly cries out for its application to the peculiar facts of the specific case, then the residual invocation of the rule is permissible.

[8] Where a charge against an accused is one of a sexual nature, a trial court is not at liberty to indiscriminately or recklessly convict. In a rape case with an unusual feature(s), the criminal courts have to exercise extreme caution and critically evaluate such a complainant as an uncorroborated witness and thoroughly analyse such witness’ evidence. The severe gravity of the prescribed minimum sentence ordained for the crime of rape, makes it particularly imperative to the dispensers of justice to be constantly alert and alive to any deviation from the behavioural norm - S v VAN DER ROSS 2002 (2) SACR 362 (C).

[9] The rule must be meaningfully applied and not cosmetically cited. It should appear from the way the evidence was assessed that the trial court indeed pragmatically invoked and applied the rule - S v JONES 2004 (1) SACR 420 (C). Any misapplication of the rule will undoubtedly produce skewed outcome.

[10] In S v GENTLE 2005 (1) SACR 420 (SCA) the court held that the natural sympathy which a court may be inclined to have for the woman who averred she had been raped, should not be allowed to blind a trial court in determining whether or not the ultimate onus required in a criminal trial, had been satisfied. So much about the law pertaining to the victims in sexual offences.

[11] In R v SHEKELELE AND ANOTHER 1953 (1) SA 636 (T) at 638 f – g the court remarked that honest but mistaken identification frequently caused gross injustices. To avoid such injustices the court remarked that, in all cases that turn on identification of an alleged offender by a witness, the greatest care should be taken to test the evidence. A bald statement that the accused was the person who committed the crime, was not enough. Answers to relevant questions about the alleged culprit’s physique, complexion, peculiar features and wearing apparels, if not properly interrogated, just like untested and unexplored bold statement which has not been inquisitively investigated, can leave the door wide open for the reasonable possibility of a big mistake.

[12] In S v MEHLAPE 1963 (2) SA 29 (A) the court held that the witness must not only be honest, but that an identifying witness must also be trustworthy regarding the identification of the person identified as the culprit.

[13] In the often quoted decision of S v MTHETWA 1972 (3) SA 766 (A) at 768 A – C Holmes JA eloquently elucidated the cautionary rule of identification. He enumerated guidelines that could be used to test the reliability of a witness’ observation.

[14] The inherent pitfalls of a witness’ subjective identification were highlighted as follows in S v CHARZEN AND ANOTHER 2006 (2) SACR 143 (SCA) at 149 g – l par. [19] Cameron JA hit the nail on the head.

Where there is a measure of perceptible doubt about a single witness’ identification of a culprit, physical evidence may provide a measure of objective assurance against the dangerous pitfalls of subjective identification. For instance an article such as a ring stolen from a victim during robbery and subsequently recovered from a suspect shortly afterwards tends to connect such a suspect to the crime.

The case-law I have cited adequately encapsulates the general law relating to identification.

[15] In R v DIFFORD 1937 AD 370 on 373 the court quoted, with unanimous approval, the observation by the trial court, Greenburg J, that firstly no onus rested on the accused to convince the court about the truth of any explanation he gave. Secondly, that if he gave an explanation even if that explanation was improbable, the court was not entitled to convict unless it was satisfied, not only that the explanation was improbable but that beyond any reasonable doubt, it was false and thirdly, that if there was any reasonable possibility of his explanation being true, then he was entitled to his acquittal.

[16] It is sufficient for the accused to be acquitted if the trial court reckons that there is a reasonable possibility that the accused’s explanation may be substantially true – R v M 1946 AD 1023 at 1027 per Davies AJA.

[17] It is incumbent upon the prosecution to establish the guilt of the accused beyond a reasonable doubt. It is permissible to test the version of the accused against the inherent probabilities. However, his version can only be rejected on the strength of the inherent improbabilities if it can be said that that such version was so improbable that no reasonable possibility whatsoever existed, that it may be substantially true – S v SHACKELL 2001 (2) SACR 185 (SCA) per Brand AJA, as he then was.

[18] Where the prosecution case consists of the version of the victim only, the trial court is called upon to critically analyse the evidence of the victim as a single witness and to warn itself against the inherent danger of uncritical acceptance of such evidence - S v HESLOP 2007 (1) SACR 461 (SCA).

[19] In S v MAVININI 2009 (1) SACR 523 (SCA) the court held that a trial court must not only take moral responsibility on evidence and inference for convicting an accused, but that he or she must also vouch for the integrity of a system producing the ultimate conviction. The court stressed that the trial court’s own subjective satisfaction of the guilt of an accused, was not enough to secure proper conviction, but that subjective satisfaction has to be attained through proper application of the rules of the system.

[20] Enough has been said about the generalities of the law relating to evidence on onus of proof. I now turn to the specifics of this particular matter. The victim testified that shortly before midnight on Saturday, 11 September 2004, she was on her way from a stokvel to her place of residence at Zamdela. She was walking on foot with two male companions, namely Mojalefa and Lefa. They were walking on a tarred street. In the vicinity of a container, a group of boys emerged. The group split into two. Some confronted Mojalefa while others confronted Lefa. She was isolated and frightened. The thought of running back whence she came, crossed her mind, but she did not act on it.

[21] Suddenly the tide turned against her. Two members of the group rushed to her, sandwiched her, grabbed her and pulled her towards a pawnshop at Chris Hani. The appellant undressed her. Initially he ordered her to kneel down. He then knelt down behind her, rested on top of her back and vaginally penetrated her from the back. Later on he changed the position. He ordered her to lie down on her back. He then knelt down between her legs, rested on top of her and again vaginally penetrated her.

[22] The members of the group warned him, after a while, that his time was up. He ignored them and carried on. Then they physically removed him from her. As the second member of the group was preparing himself to rape her, a certain Morena appeared on the scene. The show was over. The appellant and his companions took to flight and vanished under cover of the night.

[23] The court a quo, however, accepted the testimony of the victim as a logical and clear version. The court a quo correctly recognised that the victim was a single witness; that her main complaint against the appellant was of a sexual nature; that her evidence had to be treated with caution and that she was also an identifying witness whose evidence similarly had to be treated with caution. It found that her evidence was not blemished by any contradictions, inconsistencies, evasive aspects or unreliable features. It then convicted the appellant.

[24] During her direct evidence the prosecutor asked the following question and the victim answered in the following manner:

“Do you know the accused before court? --- Yes, I know him.

How do you know the accused? --- I know the accused from the 12^th of the 9^th month 2004. I was from my place ... (intervenes)”

The victim’s answer was obviously evasive.

[25] During her indirect evidence the following exchange between Ms Leoto, attorney for the defence, and Ms Mokhatla, the complainant was recorded:

“The name Steven Mbatha where did you get it, at the time you want (sic) to press charges? --- The first time we came here to court they were looking for Steven Mbatha and we told them that he is not present because he is incarcerated.”

The answer was clearly evasive. The complainant evaded saying from whom she had learned that the appellant’s name was Steven Mbatha.

[26] Still during her indirect evidence, the following exchange between the lawyer and the complainant was also recorded:

“ME LEOTO: And didn’t you on the day of the incident tell Morena that you don’t know and you are not able to can identify the people who raped you? --- No, I did not tell Morena he could realise as we went to the hospital with a detective that is where he came to know that I was raped.”

The crux of the point pursued here was whether she could identify her eventual rapist on the day of the incident, Saturday 11 September 2009.

[27] There were only three reasonable possibilities where she would have had the opportunity to do so. The first was on the main street in the vicinity of the container where she was attacked. The second was in the vicinity of the pawnshop where she was raped. The third was somewhere between the container and the pawnshop where she was pulled. The complainant evaded the question in a manner which suggested that she did not really see how her assailant was dressed at the time of the incident.

[28] I have extracted only three passages from the complainant’s evidence to demonstrate just how evasive a witness she really was.

“The witness was not evasive.”

So found the regional magistrate. In my respectful view, the finding was not borne out by the evidence of the witness. Contrary to the finding of the court a quo it is my respectful view that the witness was very evasive.

[29] The court a quo found that the complainant had ample opportunity to observe the accused. It found that she had ten minutes to make a proper observation while the accused and his friend were pulling her away from the main street to the scene of the raping behind the shop. It was estimated that she was pulled over a distance of 45 metres. Her initial answer was that she could not estimate for how long the two members of the gang pulled her over such a distance. Her subsequent answer, which came out when she was pressured, was ten minutes. The court a quo accepted this as a reasonable estimate.

[30] In my view the answer was a wild guess. A distance of 45 metres was a very short distance, hardly half the length of a soccer field. The victim’s forced movement from the container to the shop was no stroll in the park. The assailants were undoubted in a hurry. Certainly they did not want to be seen. They could not afford to stroll for 10 long minutes to cover such a short distance. Because they wanted to accomplish their mission quickly, the pulling probably endured for a whole lot shorter time than the alleged ten minutes.

[31] The scene was mobile and not static. There were two assailants pulling her from a spot on the main street, which was reasonably illuminated to a secluded and dark sport behind a shop, which was not. Her attention was accordingly divided. She could not have exclusively focussed on only one of the two. Although she was very close to the two assailants while they were in transit, she apparently did not have ample opportunity to make proper observation. In saying so I am fortified by the victim’s own averment. During cross-examination she was asked as to how she knew that the person who penetrated her from behind was the accused, now the appellant. She answered:

“I saw him clearly as they were running away.”

[32] I have difficulties with that answer. At that stage the culprit’s back was turned towards the victim. He was facing away from her. The distance between the two was not narrowing but widening. The culprit was probably running away from and not towards the street lamp to take advantage of the poorly lit section of the neighbourhood. Common sense logically suggests that the time the group was in flight, would have been the most difficult stage for the witness to make a proper observation of the rapist.

[33] The aforegoing answer implicitly meant two important things: Firstly, that the complainant did not clearly see the rapist, despite the proximity between their faces, while she was lying on her back and while he was lying on top of her. That, however, did not surprise me because visibility behind the shop was rather comparatively poor. Secondly, her answer tacitly meant or suggested that during the course of the pulling, which preceded the rape, she did not clearly see those who were pulling her to the scene where one of them raped her. Again that did not come as a surprise to me. She was moderately intoxicated. Her mental faculties to make a good observation were slightly impaired. She was frightened. She could not properly concentrate on one assailant, because there were two. The scene was mobile. In the end the good lighting that there was in the vicinity of the container did not seemingly help her very much to make proper observation of those that were attacking and pulling her.

[34] During intense cross-examination pressure the complainant frankly admitted that she was able to identify the appellant as her rapist at the local police station the next day, Sunday 12 September 2004. Even such post ex facto identification was not done in accordance with the rules of fair play. It was unclean pointing out. Before the complainant went to the police station to lay charges against her robbers and rapists, she was taken to the appellant’s parental home in the same neighbourhood by three gentlemen.

[35] They found the appellant home. The complainant apparently had a glimpse of the appellant there. They demanded the goods of which she was robbed the previous night from the appellant but recovered none. The fact remained, though, that her visit was not fruitless. By going there she gained an unfair advantage concerning the peculiar identificative features of the appellant, on account of such unprocedural identification pointing out.

[36] The complainant, after a long struggle and persistent questioning by Ms Leoto, relented and admitted that she went to the police station to lay criminal charges against the appellant on the strength of the information she obtained from Morena, the man who surprised the members of the group on the scene. From her admission it became clear and obvious that her friend, Morena, knew the appellant; that he probably told the complainant that the appellant was among the members of the group; that he pointed out the appellant’s home and perhaps the appellant himself to the complainant before she went to the police station. All these strengthened the contention that, but for Morema, the complainant would not have been able to independently identify the appellant or to give the police any constructive information about the identity of her rapist.

[37] When it was put to her that she did not personally know who her rapist was and that Morena was the person who actually told her that the person who had raped her was the appellant, also known as Fifi, she denied the suggestion. But she was apparently so stunned by the suggestions that she became speechless. The court had to commandatively call upon her to speak up:

“Talk, we are listening.” So said the court.

Her response was that she independently knew her rapist was Steven Mbatha. Now, that was untrue and inconsistent, with her direct evidence. It will be recalled that her earlier evidence was that she did not know the appellant before the encounter.

[38] She wanted the court a quo to believe that she also identified the appellant, among other features, through a horizontal scar on the hairline of his forehead, yet she could not answer the question as to how she managed to see such a scar on the dark scene of the crime behind the shop. From the witness box she could not see the scar, but she was reluctant to admit that the scar was difficult to see even during daytime. She had to be afforded an opportunity by the court to approach the dock in order to take a closer look of the face of the appellant.

[39] She admitted that she had previously seen the appellant several times at court. It was therefore improbable that the scar was a distinctive feature through which she identified her rapist on the scene. She admitted that shortly before she gave evidence, she and the prosecutor discussed the appellant. Ms Leoto insinuated and Mr. Marabo, the prosecutor, did not object that the prosecutor assisted her to identify the appellant who was paraded as accused number 2 in the dock. Although the complainant denied the insinuation that she was so assisted, she admitted the discussion, but averred that she merely described to the prosecutor how the appellant was clad.

[40] If the complainant was certain about the identity of her rapist, it would not have been necessary at all for her to unorthodoxically give his description or to point him out to the prosecutor before the trial commenced. Soon after that discussion the prosecutor asked the complainant to identify her rapist. He leadingly asked her:

“Before you proceed any further madam, is the one that you say he undressed you present in this court today? --- Yes, he is.”

A fair question should have been: “Who undressed you?”

[41] The complainant, unfairly assisted by the prosecutor, then fingered the appellant. In my view the whole exercise was highly irregular. This case is a classic example of how dock identification can be manipulated. The complainant would not have needed anyone’s help if she was certain about the identity of the real rapist.

[42] The complainant’s evidence was at variance with her police statement as regards her description of the appellant. I deem it unnecessary to even go into the finer details of the dissimilarities. The finding of the court a quo that the complainant’s testimony was consistent with her previous statement, did not accord with the proven facts. The complainant herself frankly acknowledged that her two statements were inconsistent. The inconsistency concerned a very crucial aspect in the case, the identity of her rapist.

[43] There were unusual features in this case. The plaintiff did not tell Morena, the first person she met after the incident, that one member of the group had raped her. She hardly gave him any description of her rapist. Soon after the incident she called her brothers, Mojalefa and Lefa from her parental home, in connection with the incident, and discussed her robbery only but not her rape as well. The next morning a jacket of one of her brothers, stolen during the incident, was recovered from a man she regarded as a friend to the appellant. There was neither any indication that such a suspect was ever arrested and charged nor an explanation why that was the case.

[44] The complainant was taken to the appellant’s home before the rape was reported to the police. There was no physical evidence of any incriminating goods stolen from the victim but found in the appellant’s possession. Hours after the rape incident and the visit, the complainant again saw the appellant at the police station. She identified him then and there presumably to some police officer. However, it would seem that the appellant was not arrested then and there. Instead he was only arrested on 30 December 2004, approximately 15 weeks later. The appellant’s brother, whose name was never disclosed, tried to offer the complainant a reward if she dropped the charges. So alleged the complainant.

[45] In the light of all this it can be seen, therefore, that there were at least five possible witnesses who could have testified in support of the complainant, namely Lefa, Mojalefa, Morena, the appellant’s brother and the other suspect from whom the stolen jacket was recovered. Yet none of them was called either by the prosecutor or the trial magistrate. These then were some of the peculiar features of the case which justified that the principle in S v JACKSON be residually applied to the complainant in this case. The residual rule was cosmetically cited and not pragmatically applied – S v JONES, supra.

[46] What emerges from the aforegoing appellate critique is that the evidence of the victim was characterised by a number of unreliable features concerning the observation of the culprit. The victim as a single witness in a rape case was, with respect, not critically evaluated. Extreme caution was not exercised to scrutinise and to explore the weak and unreliable blind spots in the case – S v VAN DER ROSS, supra. The natural sympathy for a reasonably credible victim seemed to have clouded the open mind of the court a quo to such an extent that the highly unreliable features, coupled with some peculiarities and procedural irregularities, were overlooked – S v GENTLE, supra.

[47] The victim’s subjective identification was plagued by misdescription, peculiar aspects and unreliable features. Worst still, there was lack of physical evidence to provide a strong objective redeeming measure for the weak subjective identification. The pitfalls thereof still loomed large on the horizon when the court a quo reached its verdict on 19 October 2006 – S v CHARZEN, supra.

[48] In my view, the observation made by the victim failed the objective test of reliability. Unless we exercise appellate intervention in favour of the appellant we too shall have misapplied the law. The integrity of the rules of the system have to underpin the subjective human conviction of a trial court that the accepted evidence, as critically scrutinised and fairly explored, established the guilt of the accused beyond a reasonable doubt – S v MAVININI, supra. In the present matter gross injustice was done approximately five and a half years ago.

[49] In the circumstances I have reached the conclusion that the court a quo materially erred in concluding that the uncorroborated evidence of the victim, with all its peculiarities and irregularities was reliable to secure proper conviction of the appellant. However, I have painstakingly endeavoured to show, and I hope I succeeded, that, on the facts, such conclusion was unsustainable.

[50] I am persuaded by Mr. Pretorius’ submission that there exists a reasonable possibility that the version of the appellant, notwithstanding its questionable aspects, was substantially true – S v M, supra. Ms Bester admirably conceded that the guilt of the appellant was not established beyond a reasonable doubt. I would, therefore, interfere with both convictions.

[51] It follows, as a matter of logic, that if the convictions fall away, the sentences cannot remain standing in a vacuum.

[52] Accordingly I make the following order:

1. The appeal succeeds in toto.

2. The convictions in respect of the first and third charges are set aside.

3. The sentences in respect of the first and the third convictions are also set aside.

________________

M.H. RAMPAI, AJP

I concur and it is so ordered.

_____________

S. NAIDOO, AJ

On behalf of appellant: Adv. K. Pretorius Instructed by:

Bloemfontein Justice Centre

BLOEMFONTEIN

On behalf of respondent: Ms A. Bester

Instructed by:

Director Public Prosecutions

BLOEMFONTEIN

/sp

---