The same rules of evidence apply to both defence evidence and state evidence.
In our system of law all persons accused of crimes are presumed innocent until their guilt has been proved beyond reasonable doubt. The rule in our system of law is that the burden or onus of proof is on the State to prove the guilt of X. X does not have to prove his or her innocence.
The same rule, namely that the onus is upon the State to prove the guilt of X, applies in respect of statutory crimes. However, sometimes in statutes the onus is placed upon X to prove certain things. The most glaring examples of such shifting of the onus of proof to X are to be found in some of the security offences contained in the Criminal Law (Codification and Reform) Act.. Normally, however, the shifting of onus is done as a matter of convenience and the onus is only placed on X to prove something peculiarly within his or her knowledge on a balance of probabilities. See, for example, s 40 as read with s 61(7) of Road Traffic Act, s 78(1) of Road Traffic Act and s 385(5) CPEA.
In our system, the State has to prove the guilt of X beyond reasonable doubt. Proof beyond reasonable doubt cannot be subject to exact measurement. For judges and magistrates it becomes a matter of experience and intuition rather than analysis. It is a matter of degree. Proof beyond reasonable doubt does not mean proof to an absolute degree of certainty. It means that there should be such proof as leaves no reasonable doubt in the mind of an ordinary man capable of sound judgment and of appreciating human motivations. It means a high degree of probability, not proof beyond a shadow of a doubt. The State does not have to close every avenue of escape and fanciful or remote possibilities can be discounted as these do not lead to reasonable doubt. See Isolano 1985 (1) ZLR 62 (S) at 64-65. To be a reasonable doubt, the doubt must not be based on pure speculation but must be based upon a reasonable and solid foundation created either from the positive evidence or gathered from reasonable inferences not in conflict with or outweighed by the proved facts. (It is sometimes said that X should not be convicted unless there is moral certainty as to his or her guilt.) However, it is not necessary for the State to prove every single individual fact in a criminal case beyond reasonable doubt although the State must prove beyond reasonable doubt a fact which is particularly vital and upon which the whole State case hinges. The question which needs to be asked is: do all the facts taken together prove guilt beyond reasonable doubt? Even a number of lines of inference, none of which would be decisive, may in their total effect lead to there being proof beyond reasonable doubt.
If X gives some explanation, he or she must be acquitted even if the court is not satisfied that his or her explanation is true if, nonetheless, the explanation might reasonably be true. The onus is not on X to prove that his or her story is true. Even if he or she gives an explanation which is improbable, X cannot be convicted unless the court is satisfied beyond reasonable doubt that it is false. Again X must be acquitted. Even if his or her story is not believed in all of its details, X must be acquitted if there is a reasonable possibility that his or her story is substantially true.
In Mupatsi 2010 (1) ZLR 529 (H) the court reiterated that there is no onus resting on X to convince the court of the truth of any explanation which he gives. It he gives and explanation, even if that explanation is improbable, the court is not entitled to convict unless it is satisfied not only that the explanation is improbable, but beyond and reasonable doubt it is false. If there is any reasonable possibility of his explanation being true, he is entitled to his acquittal.
Drawing of inferences
In S v Mtetwa HH-63-15 the court pointed out that even in the most straightforward of cases, a court must ultimately draw inferences. Some evidence requires fewer inferences – direct evidence – whereas other evidence – circumstantial evidence – will require more inferences. The court is never free of drawing inferences and therefore the rules that govern the drawing of inferences govern the court in its ultimate evaluation of all evidence. The question ultimately becomes: how is a court to evaluate the evidence? The law draws no distinction between circumstantial evidence and direct evidence in terms of weight or importance. Either type of evidence may be enough to establish guilt beyond a reasonable doubt, depending on the facts of the case. Because circumstantial evidence requires the drawing of inferences, it is incumbent for the court to restate the process involved in analysing that evidence and what a court must do before returning a verdict of guilty based solely on circumstantial evidence. Initially, the court must decide, on the basis of all of the evidence, what facts, if any, have been proved. Any facts upon which an inference of guilt can be drawn must be proved beyond a reasonable doubt. After the court has determined what facts, if any, have been proved beyond a reasonable doubt, then it must decide what inferences, if any, can be drawn from those facts. Before the court may draw an inference of guilt, however, that inference must be the only one that can fairly and reasonably be drawn from the facts; it must be consistent with the proven facts; and it must flow naturally, reasonably and logically from them. The evidence must also exclude, beyond a reasonable doubt, every reasonable hypothesis of innocence. If there is a reasonable hypothesis from the proven facts consistent with the accused's innocence, then the court must find the accused not guilty. If the only reasonable inference the court finds is that the accused is guilty of the crime charged, and that inference is established beyond reasonable doubt, then the court must find the accused guilty of that crime. In the drawing of inferences the court must take into account of the totality of the evidence, and must not consider the evidence on a piecemeal basis.
With all defences except insanity, once the evidential onus of introducing sufficient evidence of the defence to put the defence in issue is discharged (or to put it another way as soon as a foundation has been laid for the defence), the substantive onus to disprove the defence rests squarely on the prosecution.
The case of Mapfumo & Ors 1983 (1) ZLR 250 (S) at 253 (ZS), clarified further the matter of onus in relation to defences. It pointed out that there was no onus as such upon X to establish some defence. Once there is some material, whether adduced by the defence or emerging from the prosecution case, suggesting that such certain defence may be available the Court must consider the defence. Evidence sufficient to raise a defence does not have to be evidence sufficient to establish the factual basis on a balance of probabilities; all that is required is that there be sufficient material evidence to make it a realistic issue and this evidence could emerge from the State case, from X’s confession, from the cross-examination of State witnesses or from evidence from the defence. Thus once there is a sufficient foundation for the defence from whatever quarter this evidence comes, the onus rested squarely on the prosecution to prove that the defence does not apply. In Machakaire S-30-92, the court held that the magistrate had wrongly cast the onus on X to prove his or her defence.
Insanity is the one exception to the rule that the onus rests upon the State throughout to disprove X’s defence. In respect of the defence of insanity, it is laid down that the defence must prove on a balance of probabilities that X was insane at the time he or she committed the crime. In practice this means that the defence lawyer must call psychiatric evidence to prove this defence on a balance of probabilities. However, State and the court may decide to request psychiatric examination of X if there seems to be doubt as to the mental stability of X when he or she committed the crime or when he or she is about to stand trial.
Reid-Rowland 21-5 – 21-9
Corroboration means evidence, other than that of the complainant, which is consistent with the complainant’s version of the facts and which tends to show the guilt of X. To be of evidential weight, the fact or facts corroborated must be material ones.
In the situations dealt with below, what is important is for there to be “implicatory corroboration”. This means evidence that implicates X in the commission of the offence.
The corroboration can come from evidence adduced by another witness or from the evidence of X. The confession of an accused can be used as evidence corroborating other evidence. Even the failure of X to tell the truth can sometimes be a corroborative factor: Katerere S-55-91.
In certain situations dealt with below, there is a cautionary rule which applies. In these situations, the court must be alive to the dangers which arise from accepting certain types of evidence, especially if that evidence is uncorroborated. It is not enough that the court warn itself on a token basis of the dangers of accepting these types of evidence. This warning must be put into practice exercising great caution before accepting this evidence.
It should be noted that a witness cannot corroborate himself.
Reid-Rowland 18-25; 21-4
Where there is only one single witness to the crime certain special evidential rules apply.
A single witness may or may not also be a “suspect witness”. If the single witness is also a “suspect witness” then the court must apply the special rules relating to single suspect witnesses. These rules are dealt with later. This present section deals with single witnesses who are not also suspect witnesses.
An accused can be convicted on the basis of the uncorroborated testimony of a single competent and credible State witness. However, the credibility and reliability of this witness must be very carefully assessed to see whether it is safe to convict on the basis of his or her testimony alone.
There is obviously a risk which attaches to convicting X on the basis of the uncorroborated testimony of a single witness. There is a scarcity of evidence in the case and the testimony of the witness is the sole proof of X’s guilt. In this situation the court must be particularly alive to the dangers of poor observation, faulty recollection, reconstruction of evidence after the event, bias and any other risk that the circumstances of the case suggest. The quality of evidence must make up for the lack of quantity.
The credibility and reliability of this witness must be very carefully assessed to see whether it is safe to convict on the basis of his testimony alone. In Nduna & Anor HB-48-03 it was held that where a conviction relies on the evidence of a single witness, discrepancies in the witness’s evidence are not necessarily fatal. The discrepancies must be of such magnitude and value that it goes to the root of the matter to such an extent that their presence would no doubt give a different complexion of the matter altogether. Discrepancies whose presence do not usher in that change should be regarded as immaterial and as such of no value in the determination of the truth or otherwise of the matter at hand.
In Zimbowora S-7-92, the appellant had been convicted of three counts of contravening the Labour Relations Act. The State case had rested entirely on the evidence of the complainant. On appeal, the Supreme Court said that although the trial court was entitled to convict the appellant on the single evidence of the complainant, it was necessary for such evidence to be clear and satisfactory in every material respect. As the complainant was a witness with an interest to serve, the trial court was not only required to approach her evidence with caution but should also have sought corroboration of her evidence. The conviction was set aside by the Supreme Court as the complainant’s evidence was not satisfactory in all material respects and no evidence was led to corroborate her assertions.
In Mandebvu HH-96-11 X was convicted of two counts of having illegal sexual intercourse with a minor. He appealed against conviction and sentence. The complainant and X were related and lived in the same house together. The complainant was a quiet and reserved person. Almost a year after the incidents in question, she reported the abuse to her former school teacher but did not disclose the perpetrator’s identity. The teacher in turn reported the matter to Police leading to the appellant’s arrest. The complainant was the only witness to the incidents complained of. X argued that the complainant was not a credible witness primarily because of the delay in making a report against X. He also cited inconsistencies in the number of times the complainant claimed to have been raped. The court held that it is permissible in terms of s 269 CPEA for a court to convict a person on the single evidence of a competent and credible witness. The trial judge must weigh his evidence, will consider its merits and demerits and decide whether it is trustworthy and whether, despite the fact that there are shortcomings or defects or contradictions in the testimony, he is satisfied that the truth has been told .The exercise of caution must not be allowed to displace the exercise of common sense. Of course, such evidence must be approached with caution and the merits thereof weighed against any factors which militate against its credibility. In essence, a common sense approach must be applied. If the court is convinced beyond a reasonable doubt that the sole witness has spoken the truth, it must convict, notwithstanding that he was in some respects unsatisfactory. Where the evidence of the single witness is corroborated in any way which tends to indicate that the whole story was not concocted, the caution enjoined may be overcome and acceptance facilitated. But corroboration is not essential. Any other feature which increases the confidence of the court in the reliability of the single witness may also overcome the caution. The court confirmed the conviction.
In Nduna & Anor HB-48-03 it was held that where a conviction relies on the evidence of a single witness, discrepancies in the witness’s evidence are not necessarily fatal. The discrepancies must be of such magnitude and value that it goes to the root of the matter to such an extent that their presence would no doubt give a different complexion of the matter altogether. Discrepancies that do not affect the change should be regarded as immaterial and as such of no value in the determination of the truth or otherwise of the matter at hand.
In Chingurume HH-454-14 the court pointed out that there is a need to exercise extreme caution when one has to rely on the evidence of a single witness in order to guard against possible deception in the whole process. The right to convict on the evidence of a single credible witness, stated without qualifying words in s 269 of CPEA, should not be regarded as putting the evidence of one witness on the same footing in regard to the cogency as the evidence of more than one. Although the evidence of one witness may in any particular case be more convincing than of a number, it remains true that, given the same apparent quality in the witnesses, the more there are, the more reason there is to accept their story. It is not a mere rule of thumb: if there are two or more witnesses to the same facts their version can be checked against each other to see if they have given honest and accurate evidence. Elements of corroboration may of course appear from the circumstances; the fact that an accused person has given no evidence may be an element. The apparent reluctance to easily accept the evidence of a single witness is demonstrated by the proviso to the s 269, which renders it incompetent for the court to rely on such evidence in respect of certain offences specified therein. Even in other offences like assault, our courts have espoused the need to exercise caution when dealing with the evidence of a single witness. The courts should avoid the “boxing match” approach: the tendency, especially in assault cases, to throw the two protagonists into the ring with the magistrate as referee. At the end of the bout the magistrate awards points for demeanour and probability, and names the winner, who is usually the complainant.
In Mandebvu HH-96-11 X was convicted of two counts of having illegal sexual intercourse with a minor. He appealed against conviction and sentence. The complainant and X were related and lived in the same house together. The complainant was a quiet and reserved person. Almost a year after the incidents in question, she reported the abuse to her former school teacher but did not disclose the perpetrator’s identity. The teacher in turn reported the matter to Police leading to the appellant’s arrest. The complainant was the only witness to the incidents complained of. X argued that the complainant was not a credible witness primarily because of the delay in making a report against X. He also cited inconsistencies in the number of times the complainant claimed to have been raped. The court held that it is permissible in terms of s 269 CPEA for a court to convict a person on the single evidence of a competent and credible witness. The trial judge must weigh his evidence, will consider its merits and demerits and decide whether it is trustworthy and whether, despite the fact that there are shortcomings or defects or contradictions in the testimony, he is satisfied that the truth has been told .The exercise of caution must not be allowed to displace the exercise of common sense. Of course, such evidence must be approached with caution and the merits thereof weighed against any factors which militate against its credibility. In essence, a commonsense approach must be applied. If the court is convinced beyond a reasonable doubt that the sole witness has spoken the truth, it must convict, notwithstanding that he was in some respects unsatisfactory. Where the evidence of the single witness is corroborated in any way which tends to indicate that the whole story was not concocted, the caution enjoined may be overcome and acceptance facilitated. But corroboration is not essential. Any other feature which increases the confidence of the court in the reliability of the single witness may also overcome the caution. The court confirmed the conviction.
The fact that the single witness is himself guilty of some unlawful conduct does not make him an accomplice in the crime which is charged. Where the accused, who were policemen, arrested and robbed a person who was crossing the border illegally, that person was not an accomplice.
In S v Chingurume HH-454-14 the court pointed out that there is need to exercise extreme caution when one has to rely on the evidence of a single witness in order to guard against possible deception in the whole process. The right to convict on the evidence of a single credible witness, stated without qualifying words in s 269 CPEA, should not be regarded as putting the evidence of one witness on the same footing in regard to the cogency as the evidence of more than one. Although the evidence of one witness may in any particular case be more convincing than of a number, it remains true that, given the same apparent quality in the witnesses, the more there are, the more reason there is to accept their story. It is not a mere rule of thumb: if there are two or more witnesses to the same facts their version can be checked against each other to see if they have given honest and accurate evidence. Elements of corroboration may of course appear from the circumstances; the fact that an accused person has given no evidence may be an element. The apparent reluctance to easily accept the evidence of a single witness is demonstrated by the proviso to the s 269, which renders it incompetent for the court to rely on such evidence in respect of certain offences specified therein. Even in other offences like assault, our courts have espoused the need to exercise caution when dealing with the evidence of a single witness. The courts should avoid the “boxing match” approach: the tendency, especially in assault cases, to throw the two protagonists into the ring with the magistrate as referee. At the end of the bout the magistrate awards points for demeanour and probability, and names the winner, who is usually the complainant.
In S v Mupfumburi HH-64-15 the court pointed out that with crimes other than perjury and treason, the court is entitled to convict an accused on the basis of the uncorroborated evidence of a single competent and credible witness. There is obviously a risk which attaches to convicting the accused on the basis of the uncorroborated testimony of a single witness. There is a paucity of evidence in the case and the testimony of the witness is the sole proof of the accused's guilt. In this situation the danger arises of poor observation, faulty recollection, and reconstruction of evidence after the event, bias and any other risk that the circumstances of the case suggest. Before the court relies on such evidence it must be satisfied that the quality of evidence must make up for the lack of quantity. The uncorroborated evidence of a single witness should only be relied upon if the witness’s evidence is clear and satisfactory in every material respect. Slight imperfections would not rule out reliance on that evidence, but material imperfections would. Single witness evidence should not be relied upon where, for example, the witness has an interest adverse to the accused, has made a previous inconsistent statement, has given contradictory evidence or had no proper opportunity for observation. There is no rule of thumb to be applied when deciding upon the credibility of single witness testimony. The court must simply weigh his evidence and consider its merits and demerits. It must then decide whether it is satisfied that it is truthful, despite any shortcomings, defects or contradictions in that testimony. The court must have rational grounds to conclude that the evidence of the single witness is reliable and trustworthy and is a safe basis for convicting the accused.
Corroboration is regarded by many as a cornerstone of the criminal justice system. It is perceived to be an important check which helps to ensure, so far as practicable, that miscarriages of justice are kept to a minimum. Corroboration is biblical in origin, its roots being found in references in both Old and New Testaments to a fact needing to be established by two or more witnesses. The purpose of the requirement is to protect an accused from being convicted on the basis of a single witness, who may be either fallible or dishonest. Where there is a single witness to the crime itself, corroboration may be by facts and circumstances proved by other evidence than that of a single witness who is to be corroborated. There is sufficient corroboration if the facts and circumstances proved are not only consistent with the evidence of the single witness, but more consistent with it than with any competing account of the events spoken to by him. Accordingly, if the facts and circumstances proved by other witnesses fit in to his narrative so as to make it the most probable account of the events, the requirements of legal proof are satisfied.
Proper investigation of criminal cases will usually uncover corroborating evidence and it is seldom necessary to rest the entire State case upon the uncorroborated testimony of a single witness. Police officers and prosecutors should not be content with the production of evidence from a single witness. However, where it appears to a court that there are other witnesses who may be called, it has the power to call these witnesses itself in appropriate cases.
In the South African case of Mokoena 1956 (3) SA 81 (A) at 85-86, it was laid down that the uncorroborated evidence of a single witness should only be relied upon if the evidence was clear and satisfactory in every material respect. Slight imperfections would not rule out reliance on that evidence but material imperfections would. The court stated that single witness evidence should not be relied upon where, for example, the witness had an interest adverse to X, has made a previous inconsistent statement, has given contradictory evidence or had no proper opportunity for observation. However, in the later case of Sauls & Ors 1981 (3) SA 172 (A) the South African Appellate Division stated that there was no rule of thumb to be applied when deciding upon the credibility of single witness testimony. The court must simply weigh the evidence and consider its merits and demerits. It must then decide whether it is satisfied that the testimony is truthful, despite any shortcomings, defects or contradictions in it. The approach adopted in the Sauls case was followed in the case of Nyabvure S-23-88. See also Worswick S-27-88; Mukonda HH-15-87; Nemachera S-89-86.
Judge Beck in his article in Prosecutors Bulletin Vol. 1 No 1 at p 18, advises that in assessing the quality of the single witness’ evidence in order to decide whether X should be convicted on the basis of this evidence, the court should take the most attentive note of the witness. It should take particular note of his apparent character, his intelligence, his capacity for observation, his powers of recall, his objectivity and things like that. The evidence should be carefully weighed against the objective probabilities of the case, and against all the other evidence which is at variance with it. The court must have rational grounds to conclude that the evidence of the single witness is reliable and trustworthy and is a safe basis for convicting X. In Worswick S-27-88, it was said that the evidence of an accomplice must be examined critically and with caution.
Judge Beck points out that proper investigation of criminal cases will usually uncover corroborating evidence and that it is seldom necessary to rest the entire State case upon single uncorroborated testimony. The police and prosecutors should be discouraged from being content with the production of only such evidence. In appropriate cases the court itself should use its powers to call other witnesses if it appears there are such witnesses.
In Muzonza & Ors S-217-88 the Supreme Court stated that as a general rule it is undesirable to rely solely and entirely on the evidence of the complainant, particularly in assault cases and more particularly where there are counter allegations of provocation, self-defence or justification in one form or another. The complainant in such cases has a clear bias and a reason to place himself in a favourable light and X in an unfavourable light.
In Tamba S-81-91, the Supreme Court stated that in assault cases where there are other witnesses to the incident in addition to the complainant, these witnesses should be called and the case against X should not be left to rest only upon the testimony of the complainant alone. It is wrong to deal with such cases as if they were a “boxing match” between the complainant and X. These two protagonists should not, as it were, be thrown into the ring with the magistrate as referee who, at the end of the bout, having awarded points for demeanour and probability, would name the winner who would usually be the complainant. It is even worse if the magistrate is, as often seems to be the case, a biased referee who works on the unspoken assumption that the police would not have charged X if he or she was not the guilty one. This approach, said the Supreme Court, was dangerous, especially in assault cases where almost invariably the parties give conflicting versions of what was the cause of the fight. Without evidence from bystanders, it was almost impossible to determine which version of the facts was the true one although often both versions are partially untrue or exaggerated.
Crimes of perjury or treason
With crimes perjury and treason, the court may not convict an accused on the basis of the uncorroborated evidence of a single competent and credible State witness: s 269 CPEA.
In M B Ziko (Pvt) Ltd & Anor v Cestaron Invstms (Pvt) Ltd & Anor S-68-07 it was held that an appellate court may still disagree with the finding of the trial court if on examination of the evidence and considering all the circumstances (such as inferences from unquestioned facts and probabilities) of the case, it comes to the conclusion that the trial court's findings on the credibility of witnesses cannot be supported. Whilst demeanour is an important factor to be taken into account in the assessment of a witness's credibility the weight to be placed on it in determining the question whether the evidence given is reliable and probative of the facts in issue must depend on all the circumstances of the case.
In rape and other cases of a sexual nature, such as aggravated indecent assault, indecent assault and sexual relations with a person under the age of 16, the courts used to adopt the approach that because of the danger of false incrimination in such cases, a cautionary rule applies. Essentially this cautionary rule meant that in sexual cases the court had not only to believe the complainant, but in addition it had to be satisfied, by an application of the cautionary rule, whether it might still not have been deceived by a plausible witness. It therefore must seek corroboration or evidence tending to exclude the danger of false incrimination. This was laid down in a series of cases: Mupfudza 1982 (1) ZLR 271 (S); Chitiyo 1989 (2) ZLR 144 (S); Chigova 1992 (2) ZLR 206 (S); Makanyanga 1996 (2) ZLR 231 (H); Zaranyika 1997 (1) ZLR 539 (H).
However, in the case of Banana 2000 (1) ZLR 607 (S) the Supreme Court ruled that the cautionary rule in sexual cases is based on an irrational and outdated perception, and has outlived its usefulness. It unjustly stereotypes complainants in sexual assault cases (overwhelmingly women) as particularly unreliable. There were no convincing reasons for its continued application. It exemplified a rule of practice that placed an additional burden on victims in sexual cases which could lead to grave injustice to the victims involved. It is no longer warranted to rely on the cautionary rule of practice in sexual cases. Despite the abandonment of the cautionary rule, however, the courts must still consider carefully the nature and circumstances of alleged sexual offences.
In the context of sexual cases, usually the strongest evidence which the State will be able to lead will be medical evidence. If the accused has admitted that he or she had sexual relations with the complainant but maintains that the complainant was a consenting party, medical evidence of injuries consistent with forced sexual relations will be cogent evidence of the complainant's allegation of rape. Where, on the other hand, the accused denies sexual relations, medical evidence indicating that the complainant was raped or at least that the complainant has had sexual relations with someone does not prove the identification of the accused as the culprit. Corroboration often takes the form of testimony from the complainant's mother or sister. As regards the corroborative evidence, its cumulative effect must be considered. V v A 1984 (2) ZLR 139 (S).
In Mandeyi HH-34-13 the court observed it is permissible for a court to convict, in a sexual case, even if there is no corroboration of the complainant, but only where the merits of the complainant and the demerits of the accused are without question. On the other hand, corroboration will not secure a conviction unless the court is in any event satisfied that the complainant is credible. In the case of young children, the degree of corroboration or other factors required to reduce the danger of relying on the child’s evidence will vary with the age of the child and other circumstances of the case. The evidence of a single witness must be approached with caution and its merits weighed against any factors that militate against its credibility. A common sense approach must be adopted. Where the evidence of a single witness is corroborated in any way that tends to indicate that the whole story was not concocted, the caution may be overcome, as it may be by any other feature that increases the confidence of the court in the reliability of the single witness. Corroboration is not, however, essential.
In S v Musumhiri HH-404-14 Tsanga J pointed out that in assessing the prospects of success on appeal in cases involving rape, it is necessary that such cases are looked at, not just from the perspective of the person who has been convicted of rape, but also from the lens of the complainant who has experienced the rape. This is even more so in cases where the alleged rape has taken place between parties who are known to each other, as it is precisely in such cases that the administration of justice can be hampered. In such situations, applicants for bail more often than not, when convicted, seek to take advantage of the fact that the two were known to each other: the conduct of their victims may generally fall short of the standard that society has so relentlessly crafted in terms of the expected behaviour of its ideal rape victim. She must scream – very loudly. She must show evidence of physical resistance. She must be battered and bruised if she is a genuine victim. If she knows her assailant she instantly loses credibility and the understanding is that she was not raped. It is the duty of the court to assess an application for bail pending appeal in rape cases unfettered by such dangerous myths which can clearly threaten the quest for substantive justice.
Research on cultural inhibitors to reporting gender-based violence and sexual assault indicates that silence cannot be equated with acquiescence. Fear of lack of support from the family, fear of the consequences that might befall the complainant, which may include being totally blamed for the event, being thrown out of the home, or being forced to marry the rapist are some of what keeps many women from not reporting. With women often held culturally as custodians of what is deemed to be appropriate sexual conduct, and with the responsibility for sexual restraint being placed on a woman’s shoulder, regardless of her age or power imbalances, it is understandable that a complainant may fail to report even when she was now free of the sexual assault. The requirements to be met by a rape complainant should therefore not be divorced from the cultural context that might contribute heavily to swift action not being pursued. A young girl who has been raped may not make a voluntary report because her cultural context makes it difficult for her to do so without being re-victimised. She may fail to report without delay as expected by the law, because in her lived reality she has no idea if she will receive support or condemnation, if not eternal damnation. She may not report to the first person she could reasonably be expected to report for fear of being reduced to a liar and a tease. It is these realities that must therefore, with equal measure, inform the scrutiny of the likely prospects of an appeal in a rape case.
Children often have vivid imaginations and have a tendency to fantasize. They may believe their fantasies and relate them as reality because they believe them. Immature children are also susceptible to suggestions made by others. In cases of sexual molestation of children, parents may jump to wrong conclusions about the culprits and may prompt or intimidate their children in the direction of implicating innocent persons. Or the persons doing the prompting of children may want maliciously to get other people into trouble.
In Sibanda S-55-94, the court set out the dangers inherent in testimony from children as follows: The court accepted that the 6 main objections to a child’s evidence are:
a) Children’s memories are unreliable — particularly for detail;
b) Children are egocentric and not likely to consider the effect of their statements on others — particularly pre-school children;
c) Children are highly suggestible;
d) Children have difficulty distinguishing fact from fantasy;
e) Children make false allegations, particularly of sexual assault; and
f) Children do not understand the duty to tell the truth.
Judge Beck advises these steps to reduce the risks arising out of evidence from children:
1. The judge and prosecutor must put the child at ease in the court. Efforts must be made to make the child feel relaxed rather than feeling that the atmosphere is threatening and strange.
2. The judge must use his or her authority to control the examination and especially the cross-examination of child witnesses. There must be no unfair questioning aimed at overbearing, overpowering or confusing the child or trying to prompt the child unduly. The child should be allowed to respond naturally and spontaneously.
3. The child’s mental development and maturity must be assessed very carefully. One has to be careful in applying the normal tests of credibility such as demeanour, consistency and probabilities to child witnesses. A seven-year-old cannot be expected to behave in the same rational way as an adult. When assessing the probabilities the judge should take into account the child witness’ age and maturity. The child should not be expected to behave like a mature, adult witness.
In Musasa 2002 (1) ZLR 280 (H) Hlatshwayo J held that while the evidence of child witnesses must be approached with caution, such caution must be creative or positive caution, where a judicial officer uses knowledge of psychology or other relevant disciplines in order to maximise the value of such testimony. Psychological research has established that young children do not fantasize about being raped and other unusual, horrific occurrences but that their fantasies and play are characterized by their daily experiences.
In Ncube 2014 (2) ZLR 297 (H) the court decided that while the evidence of child witnesses must accordingly be approached with caution, such caution must be a creative or positive caution, where a judicial officer uses knowledge of psychology or other relevant disciplines in order to maximize the value of such testimony. Psychological research has established that young children do not fantasize about being raped and other unusual horrific occurrences, but that their fantasies and play are characterized by their ordinary daily experiences. It is highly unlikely for very young complainants to make serious allegations without any basis at all. There is certainly no psychological research or medical case study material which suggests that children are in the habit of fantasizing about the sort of incidents that might result in court proceedings; for example, observing road accidents or being indecently assaulted. Unusual fantasies are seen by psychiatrists as highly suspicious: the cognitive and imaginative capacities of three-year-olds do not enable them to describe anal intercourse and spitting out ejaculate, for instance. Such detailed descriptions from small children, in the absence of other factors, should be seen as stemming from the reality of the past abuse rather than from the imagination.
To overcome the dangers which are inherent in testimony from children such as the danger arising out of their tendency to fantasize, the court should see whether from the evidence the events related by the child really did happen.
Wherever possible, corroboration of the child’s evidence implicating X should be looked for. The existence of corroborative evidence is the safest assurance against wrong conviction. There is, however, no rigid requirement that a child’s evidence must be corroborated. The court can convict on the basis of the uncorroborated testimony of the child witness, provided it is satisfied that the dangers inherent in founding a conviction on the child’s uncorroborated evidence have been eliminated. Ponder 1989 (1) ZLR 235 (S); J 1958 (3) SA 689 (SR); Sikurlite 1964 (3) SA 151 (SR).
In Elsworth S-55-95 a number of women who had been caught by a farm guard stealing firewood lodged a complaint with the police that the farmer had stripped them of their clothes. The appeal court said that the trial court should have approached their evidence with caution.
The initial allegation on remand was that they were released by the farmer to go home stark naked; the charge alleged that they were ordered to remain in a state of nudity or semi-nudity; the women’s evidence was merely allegations of partial undress. This was coupled with the fact that no two witnesses gave the same account of the brief order the farmer allegedly issued, to be weighed against the consistency and credibility of his own account and the support of an impartial witness. “Where a wrongdoer, caught red handed, turns around to make a counter allegation of a more serious offence against the person whom he or she has wronged, his or her testimony against that other must be received with the utmost caution.”
The trial court took the variety of allegations made by the complainants and stuck them together with the mortar provided by a biased and discredited witness — a proven liar to whose evidence should have been accorded no weight. Conviction and sentence were set aside.
Reid-Rowland 21-3 – 21-5
An accomplice is a person who has participated or assisted in the commission of a crime together with others.
In the case of Mamoche HH-80-15 the court said that in the basic sense an accomplice witness means a witness to a crime who, either as principal, accomplice, or accessory, was connected with the crime by unlawful act or omission on his or her part, transpiring either before, at time of, or after commission of the offence, and whether or not he or she was present and participated in the crime. The word ‘accomplice’ has not been defined by the CPEA. However, a perusal of the case law appears to suggest that in Zimbabwe, an accomplice is one of the guilty associates or partners in the commission of a crime or who in some way or the other is connected with the commission of crime or who admits that he has a conscious hand in the commission of crime. It can also be said that an accomplice is one concerned with another or others in the commission of a crime or one who knowingly or voluntarily cooperates with and helps others in the commission of crime. An accomplice, in this sense, is a competent witness provided he is not a co-accused under trial in the same case. But such competency which has been conferred on him by a process of law does not divest him of the character of an accused. An accomplice by accepting a pardon under s 267 (2) CPEA becomes a competent witness and may, as any other witnesses, be examined on oath; the prosecution must be withdrawn and the accused formally discharged under s 267(2) before he can become a competent witness. Even if there is an omission to record such discharge an accused becomes a competent witness on withdrawal of prosecution.
The accomplice who is testifying against others may or may not be on trial himself. A person who is jointly charged with others may deny that he or she was involved in the crime at all and testify that he or she witnessed one or more of those jointly charged commit the crime or he or she may admit that he or she was involved but claim that his or her involvement was minor and that the major role was played by his or her fellow criminals. The other type of case is where an accomplice testifies after the authorities have dropped charges against him or her in return for his or her giving evidence against his or her fellow criminals.
There are a number of cogent reasons for treating accomplice evidence with considerable caution.
- An accomplice is a person who is himself guilty of criminal conduct and might easily be a person of bad character who does not have a high regard for the truth.
- The accomplice may tell lies about another person in the hope that, by testifying against another person, he or she will secure an indemnity from prosecution or he or she will receive a lighter sentence or, if he or she has already been sentenced, that he or she will receive clemency. He or she may have received promises from the police that they will go easy on him or her provided that he or she testifies against another person.
- The accomplice may wish to implicate another falsely to shield someone else. He or she may do this, for instance, because he or she is afraid of the real culprit or he or she may greatly exaggerate the role played by a fellow criminal in order to minimise his or her own role in the crime.
- The inside knowledge of the accomplice of how, when and where and by whom the offence was committed gives him or her a golden opportunity to engage in convincing deception when giving evidence. The only thing he or she has to change in what is otherwise an entirely true version of the facts is the identity of his or her accomplice. By virtue of his or her inside knowledge, he or she is peculiarly equipped to convince the unwary that his or her lies are true: Ngara 1987 (1) ZLR 91 (S).
However, it should be noted that there are varying types of accomplices. Some are more culpable than others and thus some are more dangerous to believe than others. In Moyo S-170-90, the Supreme Court said that the accomplice was simply a person caught up in an evil system and was not trying to shift the blame from his shoulders and therefore there was no danger in relying upon his testimony.
Because of the danger of false incrimination which exists with accomplice evidence, this evidence must be approached with extreme caution. The court must be satisfied that the inherent dangers of relying on this testimony have been eliminated and that the evidence can safely be relied on. There are special rules which apply in respect of single accomplice evidence. These are dealt with later.
It is obligatory for magistrates to warn accomplices who are testifying in conformity with the direction given in Simakonda 1956 R&N 463 (SR) at 465B-C. See also Ncube & Anor 1975 (2) RLR 150 (A) at 151H-152A and Ngara 1987 (1) ZLR 91 (S) at 96G. The court should warn the accomplice that what is expected of him or her is to tell the truth.
The warning that X must answer questions “to the satisfaction of the court” can be misleading when translated into the vernacular as it may give the impression to the accomplice witness that the evidence required of him or her to satisfy the court is evidence which incriminates X, even if this evidence is not the truth. The judge, in warning the accomplice, must emphasise that the court is interested only in the truth, whether it incriminates or exculpates X.
As regards the accomplice who has already been convicted and who is presently serving his or her sentence, he or she should be advised that exaggerating the part allegedly played by X or minimising his or her own role will not affect the sentence in any way.
As regards the unconvicted accomplice in line with s 267 CPEA, he or she must be advised that
- he or she is not obliged to give evidence;
- if he or she testifies, questions may be put to him or her which might incriminate him or her in regard to the specified offence;
- he or she will be obliged to answer any question that may be put to him or her despite the fact that the answer might incriminate him or her in respect of the specified offence (or some other offence for which he or she could be found guilty on the basis of the charge relating to the specified offence);
- if he or she answers all questions put to him or her frankly and honestly he or she will be discharged from prosecution in respect of the specified offence (or from any other offence for which he or she could be found guilty on the basis of that charge).
See Hoffmann and Zeffertt South African Law of Evidence 3rd Ed pp 196-197.
Where the case against X rests on the evidence of one single accomplice s 270 CPEA (not s 269) applies. This says that a court may convict an accused on the basis of the evidence of a single accomplice, provided there is competent evidence other than the single and unconfirmed evidence of the accomplice which proves to the satisfaction of the court that the crime was actually committed.
The courts have interpreted this provision to mean that even where there is no proof aliunde (from another source than the evidence of the accomplice) of the commission of the offence, X can still be convicted if there is corroboration in a material respect of the evidence of the accomplice. In Mubaiwa 1980 ZLR 477 (A) at 479H-480A this is stated as follows:
“the purpose of this section is that the court must be satisfied that the crime to which the accomplice testifies has, in fact, been committed. If not, there can be no conviction at all. Even where there is no proof aliunde that the crime has been committed, the statutory requirement can still be satisfied if there is corroboration in a material respect which convinces the court that the accomplice can safely be relied on when he or she says the crime was committed, though it need not directly implicate X. In such a case, the requirement is satisfied because, despite the lack of proof aliunde of the commission of the offence, the accomplice is no longer ‘single and unconfirmed’.”
In Lawrence & Anor 1989 (1) ZLR 29 (S) the Supreme Court laid down that with single accomplice testimony there should be a two-pronged inquiry. The court must first satisfy itself that the offence with which X is charged has been committed before convicting. Secondly, the court must look for corroboration, for if there is no evidence aliunde proving the commission of the offence then there can still be a conviction if the court is satisfied that there is corroboration of the evidence of the accomplice sufficient to satisfy the court that the witness is to be believed. See also Moyo 1989 (3) ZLR 250 (S).
Thus if the evidence of the accomplice is single and unconfirmed there must be proof aliunde of the commission of the offence. If, on the other hand, there is material corroboration of the testimony of the accomplice, the evidence is no longer single and unconfirmed and there need not be proof aliunde of the commission of the offence.
Where there are imperfections in an accomplice’s evidence and there is no corroboration of his or her evidence implicating X, the question remains whether there are other features which reduce the dangers of false incrimination and, if there are, whether they reduce it to the point where there is no reasonable possibility that X has been falsely incriminated: Juwaki & Anor 1964 RLR 604 (A).
In Lawrence & Anor 1989 (1) ZLR 29 (S) despite the imperfections in the accomplice’s evidence there was sufficient corroboration to eliminate the danger of false incrimination.
In Vengesayi S-26-93 the court said the warning administered to an accomplice witness is not for the benefit of X but is a promise to the witness. Failure to give it is an irregularity but does not make the witness’s evidence wholly inadmissible. Even if there is ample corroborative evidence of an accomplice’s testimony, the court must warn itself of the danger inherent in basing a conviction solely on the evidence of a suspect witness; but the conviction will only be quashed if there has been a substantial miscarriage of justice.
In Chitongo HH-53-95, the court found that although there were some unsatisfactory aspects in the accomplices’ evidence, they were credible and their evidence was corroborated. On the other hand, X’s testimony was wildly improbable and his testimony was not credible. Thus although there was a need for caution in respect of the evidence of accomplices, everything pointed to the overall truth of their testimony.
Where two or more persons are jointly charged with an offence and each gives evidence blaming the other for the offence, the evidence of each is admissible against the other, but the court must approach the evidence with care since there is a risk that either or both may be seeking to protect himself by telling lies: Sambo S-22-90.
Note that the statement of X in reply to police questions is only evidence against the maker of the statement and is not evidence against any other person: s 259 CPEA. This is because there is no opportunity for cross-examination of the person who made this statement when he or she makes his or her statement. But if the maker goes into the witness box and repeats on oath what he or she said in the statement, he or she renders himself liable to cross-examination by an accused who is jointly charged with him or her and thus such evidence on oath is admissible against the co-accused.
In Nyathi & Ors S-52-95, first appellant’s fingerprint was found at the scene of a housebreaking/murder. He had an absolute need to shift the blame, and named the two other appellants. Strong corroboration was essential. The involvement of one was corroborated by his possession of a lot of the stolen property and by his own evidence. But the accomplice evidence implicating the other was only supported by the latter’s possession of a cheap disposable lighter and his explanation about how he came to be in possession of it. The second accomplice did not mention him. This was not sufficiently strong corroboration to prove his guilt beyond reasonable doubt; his conviction was set aside.
What should a judicial officer do where co-accused have not been warned? Where a prosecutor fails to inform the court that a witness is to be treated as an unconvicted accomplice or where a witness is not warned he or she would be discharged from liability if he or she gave his or her evidence satisfactorily and that he or she could decline to answer questions which might tend to incriminate him, it was held that it would be contrary to interests of justice for him or her not to be discharged. The witness had admitted to further crimes with which he was subsequently charged and the effect was held to be to deprive witness of the right to fair hearing contrary to s 18(2) of Constitution: Sivako v A-G 1999 (2) ZLR 271 (S).
The safest way to eliminate the risk of false incrimination of another by an accomplice is to look for corroborative evidence implicating X. In Ngara 1987 (1) ZLR 91 (S) the Supreme Court said, however, that the corroborative evidence need not necessarily show that X is the offender. It said that the corroboration looked for was corroboration in some material respect which shows, or tends to show, that the accomplice is reliable, though it does not necessarily have to show that X is the offender.
The evidence of one accomplice can corroborate the evidence of another. The court, however, must be satisfied that the testimony of both accomplices is credible and that there has not been an opportunity for the accomplices to conspire together before testifying in order to concoct a false story to implicate X.
In Zata S-64-91, a visitor to Zimbabwe alleged that he had paid a bribe on demand to a junior customs officer who had handed it over to the appellant. The junior officer confirmed his story. As the visitor was a stranger to the customs official and they did not have the same interests to serve, their evidence could be given credence and could be used to corroborate each other’s testimony.
Where there are several counts, each testified to by a single witness, the evidence on one count may be taken to corroborate the evidence on others, provided that the acts spoken of by each accomplice bear a striking similarity to one another: Ngara 1987 (1) ZLR 91 (S).
It is usually dangerous to convict without corroboration of the accomplice’s evidence. Thus in Machakata S-106-89, there was no corroboration and the court quashed the conviction. The appellant had been found guilty of stock theft. It was alleged that he had instructed two of his employees to go and steal cattle for him. One of these employees, P, gave evidence for the State and the entire State case rested on this testimony which was not corroborated. The appellant denied that he had given such an instruction to the two employees and another of his employees, E, corroborated his testimony. The Appeal Court found that the trial court had only paid lip service to the cautionary rule. P’s evidence had not been rigorously examined to ascertain whether or not he may have falsely implicated the appellant. P’s uncorroborated evidence was open to question.
However, sometimes even without corroborative evidence, the court can convict a person on the basis of the evidence of even a single accomplice if the circumstances are such that the court can properly be quite satisfied that the accomplice is telling the truth. For example, if the accomplice gives convincing evidence against another and X adamantly refuses to give evidence and maintains his or her right to silence, there is a reduced risk of relying on the evidence of the accomplice in convicting X because if X is innocent one would have expected him or her to have vigorously denied the false testimony against him. So too, there is a reduced risk of reliance on testimony by an accomplice against a person with whom he or she has a very close relationship and with whom he or she has been on good terms previously because here it would be unlikely that the accomplice would implicate this person falsely.
In Lawrence & Anor 1989 (1) ZLR 29 (S), the Supreme Court indicated that it is desirable that the court be informed of any inducement or promise made to an accomplice when the accomplice is called upon to testify because the danger of the false incrimination is greater when an accomplice has been promised a pardon or remission. It further pointed out that it is the court, not the Attorney-General, which should decide whether or not the accomplice has given satisfactory evidence justifying fulfilment of the undertaking to recommend remission.
Before the accomplice gives evidence the State should advise the court that the witness who is being called is either a convicted or unconvicted accomplice.
Accomplice evidence must be approached with extreme caution because of the dangers of false incrimination by the accomplice. Although it is not essential that the accomplice evidence be corroborated, the presence of corroborative evidence is usually the best safeguard against false incrimination. Slight imperfections in the accomplice’s evidence do not necessarily discredit it, especially if the material portions of that evidence are corroborated. The evidence of one accomplice can corroborate the evidence of another, provided that the court is satisfied that the accomplices did not conspire together to give an agreed false story against X.
Accomplices must be given a warning when they are testifying.
With single accomplice evidence the judge should ask himself:
1. Are you satisfied that the crime charged has been committed?
If no, acquit;
If yes, proceed to next stage.
2. Is there material corroboration for the witness’ testimony?
If yes, may convict even if no proof aliunde of commission of crime;
If no, may only convict if proof aliunde that crime committed.
For a lie told by X in or out of court to be capable of amounting to corroboration of the testimony of a State witness, the following criteria apply:
- the lie must be deliberate;
- it must relate to a material issue;
- the motive for the lie must be a realisation of guilt and a fear of the truth;
- the statement must be clearly shown to be a lie by evidence other than that of the witness who is to be corroborated.
Caution should be adopted in this regard. Too much weight should not be attached to lies told by X. The court must guard against drawing an inference of X’s guilt solely on the basis of lies told by an accused person: Nyoni S-118-90.
In Masawi & Anor HH-111-94, two persons were charged with the kidnap-murder of a woman. They told lies about their involvement with her, the whereabouts of their car and themselves over the crucial period and other matters. The court found that their lies, transversing all the crucial parts of the case, could only be indicative of guilt. They could not be explained by fear or other reasons. Both were convicted of murder.
The courts are obliged to approach accomplice evidence with extreme caution because of the dangers of false incrimination by the accomplice. Although it is not essential that the accomplice evidence be corroborated, the presence of corroborative evidence is usually the best safeguard against false incrimination. Slight imperfections in the accomplice's evidence do not necessarily discredit it, especially if the material portions of that evidence are corroborated. The evidence of one accomplice can corroborate the evidence of another, provided that the court is satisfied that the accomplices did not conspire together to give an agreed false story against the accused.
Where the evidence of a single accomplice is relied on, the judicial officer is obliged to consider whether there is material corroboration for the witness' testimony. If there is such corroboration, the court may convict even if there is no proof aliunde of commission of crime. If there is not, it may convict the accused only if there is evidence aliunde that the crime was committed.
It is provided in s 273 CPEA that a court may convict X on the basis of a confession proved to have been made by him, although the confession is not confirmed by any other evidence, provided that the offence has been proved to have been committed by competent evidence other than such confession.
The court may thus convict on the basis of a confession either:
- where there is proof that the crime was committed, although there is no evidence other than the confession to connect X with the crime; or
- where there is direct evidence to confirm X’s confession, even though there is no direct proof of the commission of the crime:
Tsorayi 1985 (1) ZLR 138 (H).
In Cloete 1994 (1) SACR 420 (A), the court stated that the rule in Valacia 1945 AD 826, that a judicial officer must take into consideration everything contained in an extra-curial statement made by X, including exculpatory portions, applies equally to statements made by X in explanation of his or her plea under s 163(4) CPEA.
Similar fact evidence is evidence of similar acts done previously by the accused. Similar fact evidence is not admissible if its only relevance is to show that the accused is of bad character and is therefore likely to have committed the offence. It is, however, admissible if it is relevant and is of sufficient probative force to warrant its reception despite its apparently prejudicial nature.
Previously the approach of the courts was that the similar facts had to bear a striking resemblance to the case in hand. See Mutsinziri 1997 (1) ZLR 6 (H); Ngara 1987 (1) ZLR 91 (S). However, in Banana 2000 (1) ZLR 607 (S) the Supreme Court said that the test for the admissibility of similar fact evidence used to be whether the similar facts were of such a striking similarity that it would be an affront to common sense to assume that the similarity was explicable on the basis of coincidence. However, the courts have moved away from this test. Striking similarity is not a pre-requisite to admissibility. What has to be assessed is the probative force of the evidence in question; there is no single manner in which this can be achieved. Like corroboration, this is a matter of logic and common sense.
This is in contrast to the position in South Africa. In the South African case of M & Ors 1995 (1) SACR 667 (BA) contains an exhaustive review of the authorities on the admissibility of similar fact evidence. In the course of its review, the court agreed with authorities which held that for such evidence to be admissible, the similar facts must bear a striking similarity to the evidence adduced in relation to the offence charged: “The use of the word ‘striking’ — strengthens the concept that the admission of similar fact evidence requires a ‘strong degree of probative force’, bearing in mind the basic principle that its admission is out of the ordinary and unusual.
In Mutsinziri 1997 (1) ZLR 6 (H) the court pointed out that where there are multiple counts, the fact that each one must be looked at separately does not prevent material which could be admissible under the rules relating to similar fact evidence from being received. Even evidence on one count which ultimately leads to an acquittal may be used but, for such evidence to be receivable and acted upon by a court of law, those discreditable acts of the accused must share with the discreditable conduct in issue, features of such an unusual nature and striking similarity that it would be an offence to common sense to assert that the similarity was explicable on the basis of coincidence. Similar fact evidence may be admitted on one count in order to bolster evidence on another count, where there is an issue as to identity.
In Mupah 1989 (1) ZLR 279 (S) the court said that evidence of a previous offence is admissible to rebut a defence of accident or innocent intent and to show a systematic course of conduct by X. This is so whether or not X has been convicted of that offence. There must not only be similarity between the previous acts and those in issue, but such a concurrence of common features that the various acts are naturally to be explained as caused by a general plan of which they are the individual manifestations. The discreditable acts of X must share with the discreditable conduct in issue, features of such an unusual nature and striking similarity that it would be an affront to common sense to assert that the similarity was explicable on the basis of coincidence. Where X has been acquitted in one case, it would be wrong, in order to obtain a conviction in a later case, to seek to show that X was guilty in the first case. This does not mean that evidence relating to the first case may not be called to show what X’s intent was in the second; it means that it is impermissible, in the second case, to rely on X’s guilt in the first if he or she has been acquitted in the first case.
In the case Mupah 1989 (1) ZLR 299 (S) the Supreme Court set out when similar fact evidence is admissible. It stated that:
- Evidence as to previous conduct is admissible to prove that the acts alleged to constitute the crime charged were intentional or part of a systematic course of conduct or to rebut a defence of accident or innocent intent;
- The conduct must demonstrate such an underlying unity or such a concurrence of common features that it would be an affront to common sense to assert that the similarity was explicable on the basis of coincidence;
- X person need not have been convicted of an offence in relation to the previous conduct before such conduct becomes admissible as similar fact evidence;
- On the other hand, if X has been acquitted of an offence in relation to the previous conduct, evidence as to that conduct is not admissible as similar fact evidence as its admission would require the court to assume that X was guilty of the offence of which he or she was acquitted; such evidence may be admissible, however, to show, for example, what X’s intent was in relation to his or her subsequent conduct.
In Tirivanhu S-170-91, the Supreme Court stated that where an accused faces several charges, the evidence on one count cannot be taken as evidence on another unless it would have either been relevant to that other charge if it were the only charge against him, and justify an inference about it, or it shows a certain peculiar course of conduct.
In this case, the appellant was charged with the theft of five pumps, all of which were found in his possession, three shortly after the theft, and he gave a false explanation for each. The Supreme Court said that it would be an affront to common sense to accept that the possession of all these stolen pumps by him was explicable on the basis of coincidence.
In order for a person to give expert evidence, his or her special expertise must first be established. The professional qualifications and the professional experience of the person concerned must be established: Makuni HH-75-84.
In Machona HH-450-15 the court pointed out that the function of an expert is to assist the court to reach a conclusion on a matter on which the court itself does not have the necessary knowledge to decide. It is not the mere opinion of the witness which is decisive but his ability to satisfy the court that, because of his special skill, training or experience, the reasons for the opinions he expresses are acceptable. Any expert opinion which is expressed on an issue which the court can decide without receiving expert opinion is in principle inadmissible because of its irrelevance.
In SNdzombane 2014 (2) ZLR 197 (S) the court dealing with psychiatric evidence dealt with the value to be attached to expert evidence. It pointed out that expert opinion evidence is admitted to assist the court to reach a just decision by guiding the court and clarifying issues not within the court’s general knowledge. It is not the mere opinion of the expert witness which is decisive but the expert’s ability to satisfy the court that, because of the special skill, training and experience, the reasons for the opinion expressed are acceptable. However, in the final analysis, the court itself must draw its own conclusions from the expert opinion and must not be overawed by the proffered opinion, and simply adopt it without questioning or testing it against known parameters. The expertise of a professional witness should not be elevated to such heights that sight is lost of the court’s own capabilities and responsibilities in drawing inferences from the evidence. The court can only do this well if it requires the expert witness to give oral evidence in the clarification and elucidation of an affidavit that is otherwise technically dense and incomprehensible, contradictory or inadequate in all respects except the conclusion. A court errs when it merely adopts the conclusions of an expert report without exercising its mind on it by, for example, calling for oral testimony or drawing the necessary inferences from the evidence.
Human observation is very fallible and experience has shown that genuine errors can easily be made by witnesses who have identified culprits. In a number of cases in Britain it emerged that X were wrongly convicted on the basis of mistaken identification evidence. In S v Mthetwa 1972 (3) SA 766 at 768 A-C the Judge stated:
“Because of the fallibility of human observation, evidence of identification is approached by the courts with some caution. It is not enough for the identifying witness to be honest; the reliability of his observation must also be tested. This depends on various factors such as lighting, visibility and eyesight; the proximity of the witness; his opportunity for observation, both as to time and situation; the extent of his prior knowledge of the accused; the mobility of the scene; corroboration; suggestibility; the accused’s face, voice, build, gait and dress; the result of identification parades, if any; and of course the evidence by or on behalf of the accused. The list is not exhaustive. These factors, or such of them as are applicable in a particular case, are not individually decisive, but must be weighed against the other, in the light of the totality of the evidence and the probabilities.”
Whenever the State case depends wholly or mainly on evidence of visual identification special caution must be exercised, especially if there is only a single witness who has made the visual identification. Even if the court decides that the witness is entirely honest and truthful in his or her testimony and he or she has asserted that he or she is completely certain that he or she has identified the correct person as the culprit, it must nonetheless ask itself whether there is a danger that the witness was mistaken: Mutters & Anor S-66-89; Makoni & Ors S-67-89.
It should also be borne in mind that ordinary members of the public are not trained in accurate observation as are policemen. The ability of different people to observe carefully and recollect later varies widely. Also memories fade over time and accurate identification long after the event is difficult.
The quality and reliability of the identification evidence must be most carefully assessed. The court must look at the circumstances in which the identification was made and whether the witness has the capacity for careful observation. Factors which should be taken into account include the following:
- the amount of time that the witness had X under observation;
- the distance between the witness and X at the time of observation;
- the lighting conditions at the time;
- whether there were any objects in the way which would have prevented or obscured observation;
- whether the witness has good or poor eyesight if the witness wears glasses, whether he had them on at the time;
- whether the witness was able to see clearly the face of X as well as the rest of his or her body;
- whether or not the witness had known X previously and, if he or she had, how well had he or she known him or her;
- whether there is any special reason why the witness would have been likely to have remembered X, such as the fact that X has distinct facial scars or a pronounced limp or is extraordinarily tall or short;
- whether the person identified belonged to a different ethnic group to the witness because if he or she did there may be doubt that the witness was able to distinguish accurately between different persons in that other ethnic group.
See Mutters & Anor S-66-89; Nkomo & Anor 1989 (3) ZLR 117 (S) and Ncube & Anor HB-55-13.
The identification will be unreliable if, for instance, the witness caught only a fleeting glimpse of a person from a considerable distance in poor light. The identification in these circumstances would be dubious, whether or not the person was a person he or she had known previously. On the other hand, close-range observation for a reasonable period of time in good lighting conditions where the witness clearly saw and carefully studied the person’s facial features will be far more reliable, particularly if the person was well known to the witness or had some very distinctive features which made him or her easy to identify: Nkomo & Anor 1989 (3) ZLR 117 (S).
In Musakwa S-1-95, the complainant was conned out of money by two men. One chatted to her for 10 minutes and she saw X at the same spot shortly afterwards and identified him. He denied any involvement and said that he had been elsewhere. The police did not check his alibi. The trial court believed the complainant and accepted her identification. The appeal court held that mistakes happen; the question is not simply whether there was the opportunity for reliable recognition: the State must disprove the alibi.
The bald assertion by the witness that he or she is certain that he or she has identified the right person should not be accepted at face value. The objective basis of his or her identification must be carefully probed.
Witnesses should be asked by what features they made their identification. The witness should be questioned as to the height, build, complexion and apparel of the person observed. Where X is undefended, the court should carefully examine the circumstances of the identification and test its reliability.
The dangers of wrong identification are reduced if several witnesses independently identify X.
In Madziwa S-191-90, it was pointed out that weak evidence of identification is not made any more reliable by the mere fact that appellant was in the vicinity at the time and lied about this fact, as even an innocent person can lie out of a sense of panic.
Identification evidence can be very unreliable. In order to ensure that it is safe to rely upon the identification evidence the circumstances of the identification must be carefully probed and the powers of accurate observation and recall of the witness properly tested.
If the police decide to ask the witness to identify the culprit by looking at photographs, this exercise must be fairly conducted if any identification based thereon is to have reliable evidential value. For the identification process to be fair:
- the witness should be asked to look at a reasonable number of photographs and should not be shown only one photograph and asked whether this is the culprit;
- the names of the persons photographed should not be on the photographs and the photograph of a person whom the police already suspect should not be ringed or specially marked and that photograph should not be of an entirely different size from all the others;
- the police should allow the witness to make his or her own independent selection and should not prompt the witness in the direction of selecting one particular photograph;
- only photographs of reasonable quality should be shown because there is an increased danger of wrong identification from poor quality photographs;
- the identification process will not, however, be vitiated simply because the witness sees a label on the cover of the album indicating that the photographs in the album are of persons convicted of crimes similar to those for which the culprit is presently being sought:
Nkomo & Anor 1989 (3) ZLR 117 (S).
A witness to a crime is sometimes asked to identify the culprit from photographs so that the police can then make an arrest. In cases where police investigations have hitherto failed to lead to the arrest of a suspect, the police may ask a witness to a crime to look through an album containing photographs of persons with previous convictions to see if he or she can identify the culprit. If the witness is able to make an identification in this way, the police can investigate the case and possibly discover tangible evidence which can be produced in court. If this happens, the case against X will not have to be based solely on identification evidence.
However, if they cannot uncover such further evidence, and it is intended to proceed on the basis of identification evidence, it is best to hold an identification parade to see whether the witness will pick out the same person originally identified from photographs: Nkomo & Anor 1989 (3) ZLR 117 (S).
If, however, the police already have a suspect, the best way to test whether the person is the one the witness saw is to hold an identification parade, rather than asking the witness to look at photographs. The effect of showing the witness photographs upon his or her subsequent identification of X at a parade or in court needs to be considered. A witness who is first shown photographs may be predisposed towards identifying a person at a parade or in court on the basis of the photographic image he or she has selected rather than on the basis of his or her original observation at the time of the crime. This danger is reduced if the witness has made his or her selection from a large number of photographs. But even here there is a danger of false identification. The witness may have only caught a fleeting glimpse of the criminal committing the crime. He or she may then be asked by the police a long time later to pick out the culprit from photographs in an album. If, with only a hazy recollection of the culprit, he or she selects one of the photographs as being that of the culprit, he or she will obviously have looked carefully at the photograph. He or she will thus see in close-up all the person’s facial features. That image will then stick in his or her mind and when he or she is called to identify the culprit at a parade or in court, he or she will be making his or her selection mostly on the basis of the photograph. This is not always so. Where the witness had a good opportunity to observe the face of the culprit at the scene of the crime he or she may be able to make a positive and reliable identification from both a photograph and at a later identification parade.
The State should always disclose if the witness previously identified the culprit from police photographs because this will have a bearing on the reliability of the witness’ subsequent identification of the culprit at a parade or in court: Ndlovu S-3-88.
There is a difficulty in the State producing in evidence the photograph album which the witness examined to make an identification. The State may want to produce this to show that the identification process was a fair one and that the witness was asked to go through a large number of photographs. Because of the way in which the album is labelled its production to the court may have the effect of revealing that X has at least one previous conviction for the crime. If this is the case it cannot be produced because this would be prejudicial to X. If, however, X raises the matter of his or her identification from photographs and alleges that the process was unfairly conducted, then the State is at liberty to produce the album to prove the fairness of the process. See 1978 Criminal Law Review 343.
In examining the reliability of evidence of identification from photographs, the judge must ascertain whether the identification exercise was fairly conducted. The selection should have been made from a substantial number of photographs and there must have been no special markings on the photograph of X. The witness must be allowed to make his or her own independent selection without any prompting. After identifying a suspect from photographs, the witness should normally have been asked to confirm his or her identification at an identification parade.
In Mutsinziri 1997 (1) ZLR 6 (H) the court said that, where identity is in issue and X is identified by means of a photograph of him or her in the possession of the police, it is not inevitable in all cases that the witness would thereby conclude that X is known to the police and has a known or suspected criminal record. The police may come by photographs of the suspect in perfectly innocent circumstances and it is not necessarily harmful to X, where a policeman does no more than say that he or she was able to identify the accused from a photograph or to say that he or she tried to locate X by distributing photographs.
The reliability of evidence of identification from photographs can be attacked either on the basis that the witness’ identification is unreliable or on the basis that the photographic identification exercise was unfairly conducted.
In Mutsinziri 1997 (1) ZLR 6 (H) the court pointed out that a “dock identification”, where a witness is asked whether the person in the dock is the offender, suffers from considerable disadvantages. Everything about the atmosphere of the court proceedings points to the accused, and to him or her alone, as the person who is to be identified by the witness. These circumstances are inevitable unless one insists that any dock identification take the form of an identity parade. The manner in which a dock identification is elicited from witnesses by the prosecutor can be done the right way or the wrong way. The wrong way is one which makes it virtually impossible for the witness to say anything other than that the accused is the culprit. This way constitutes an irregularity. The better way is to get the witness to recount all the events without reference to the accused in the dock, and only when the witness has said all he or she has to say about the events should he or she be asked whether any person in the court is recognised. This form of identification still carries the defective feature of a dock identification, that the accused is obviously the person who is suspected of committing the offence, but it avoids leading questions and putting the identification into the witness’s mouth.
In the South African case of Maradu 1994 (2) SACR 410 (W) the court held that the danger of a dock identification is the same as that created by a leading question in examination-in-chief: it suggests the answer desired. As the latter type of question is inadmissible, there is no reason why a dock identification should also not be inadmissible, save in special circumstances. (The court found that the witness’s dock identification of the appellant was unreliable for a number of reasons.)
If the State case is likely to hinge on identification evidence identification parades should be held wherever possible. This should be done in order to overcome the obvious dangers of the witness being left simply to identify the person in the dock as the culprit. In identifying the person in the dock, the witness will be likely to assume that the police have got hold of the right person and, if there is any similarity between this person and the person he or she saw commit the crime, he or she will be inclined to assume that the person in the dock is the person he or she saw commit the crime.
To produce reliable evidence an identification parade must be carried out fairly. It must be carefully carried out to obviate errors. These are the basic requirements for a fair parade:
- It must be conducted by an officer who has had nothing to do with the investigations into the case;
- The officer conducting the parade should not call the witness to the parade;
- A sufficient number of persons, say ten, should form the parade;
- The persons on the parade must be approximately the same build, height and complexion and they should all wear similar clothing and, preferably, clothing similar to that worn when the crime was committed;
- The witness must be kept somewhere where he or she cannot see the prisoner being guarded by police officers or the parade being assembled;
- The witness should not be told that the suspect is on the parade but should only be asked if the person he or she saw commit that crime is on the parade;
- The witness must be left to pick out the person he or she saw commit the crime, if he or she can, without out any form of assistance or prompting;
- The police should not attempt to point out or suggest someone either before or during the parade;
- It is also important that the identification parade be held as soon after the commission of the crime as possible to avoid memory impairment occurring with the passage of time. Mavunga 1982 (1) ZLR 63 (S).
In Gomo HH-21-93 the court stated that evidence must be led showing the proper conduct of an identification parade in accordance with the established police procedures, with all its safeguards. Unless these essential requirements are complied with, any evidence of identification at a parade must not be relied on.
In Masawi & Anor HH-111-94, a woman saw her daughter being kidnapped at gunpoint at 7 am. She made no immediate statement to anyone present or to the police that she had recognised the assailant, who was 30 paces off and wore dark glasses and a hat pulled down over his face. Despite her claim in evidence that she recognised the man immediately as someone she had met two weeks before, it was obviously an ex post facto identification after she had had time to think about it. This was not inherently objectionable, but the particular circumstances made her identification unreliable, so it could only be accepted as evidence that the assailant was someone who looked very like X. Other circumstantial evidence then convinced the court that it was in fact X. On legal advice, X declined to participate in an identification parade. The police then organised an ‘informal parade’ in which the witness was invited to look around the police station and identify the suspect if she saw him. She did so. The police and the court accepted that, because of the handcuffs and his dishevelled state, this procedure did not enhance the identification but simply excluded the possibility that she would say it was not him.
In determining the reliability of the evidence derived from identification of a suspect at an identification parade, the judge must explore whether the parade was fairly conducted. A sufficient number of persons of similar build and appearance and similarly dressed should have formed the parade and the witness must have made his or her identification independently and without any prompting.
In Nkomo HB-160-2010 the applicant, who was seeking bail pending trial, was identified as the culprit by the complainant. The circumstances in which the identification took place were that as the applicant was brought to the charge office barefooted and was asked to put on his shoes in the presence of the complainant, a procedure which was unusual as it exposed him to the complainant who no doubt regarded him as the culprit and therefore proceeded to identify him as such.
The court held that the identification parade was conducted in this manner was a mockery of justice. An identification parade is a procedure where a complainant or witness should independently identify the suspect or the wrongdoer without being given a clue which is designed to expedite police investigations. Identification parades should be conducted in a manner that excludes the possibility of any suspicion of bias or unfairness.
Where an identification of a suspect has been made easier by a police officer’s conduct, conscious or otherwise, the courts should be ready to condemn such proceedings without more ado. What occurred in this case could not pass the test of an identification parade. There should be fairness in the process. The probative value of personal identification at a parade conducted in a manner which does not guarantee fairness carries less weight as it would have been calculated to prejudice the accused.
In Ncube & Anor HB-55-13 the identification parade was faulty for a variety of reasons such as that he was told that the suspects were present in the parade and that accused one was not dressed in similar clothes with others.
Where the State case rests exclusively or substantially on evidence that the prints found at the scene of the crime matched those taken from X, an expert must testify as to the basis upon which he or she reached his or her conclusion that the prints belonged to one and the same person.
In Mutsinziri 1997 (1) ZLR 6 (H) the court said that where fingerprint evidence is given by an expert, the court ought not insist on its own ability to make a fingerprint identification by study of a comparison chart between the latent print (that found at the scene) and the inked print (that recorded from the suspect). Nevertheless, the court is still faced with a decision as to whether or not to accept the expert’s evidence when he or she purports to find sufficient points of identity between the latent and the inked print. The court must take into account the witness’s experience and the apparent weight and reliability of his or her opinion. The court may, if it considers it necessary, insist on a study of the comparison chart; and where the court is in a position to make its own examination of the comparison, it may, to the extent which it considers proper, attempt of its own accord to confirm the validity of the expert’s opinion. In an appropriate case it may depart from the expert’s opinion, if it is unable to find on the chart the points of similarity which so impressed the expert.
Fingerprint or handprint evidence is led in order to show that X was present at the place where the crime was committed. It is damning evidence if, for instance, it is established that X’s fingerprints were found inside a house which had been burglarised and into which X had no right of entry. On the other hand, there is nothing incriminating for the fingerprints of X to be found inside the same house if he or she is a domestic worker who enters the house on a daily basis to carry out his or her work. The same would apply to the finding of the fingerprints of an accused who is a messenger, in offices into which he or she has free access in connection with his or her work.
When dealing with the issue of identification, it is permissible for the court to rely on evidence relating to bare footprints. However, a number of precautions have to be observed before such evidence can be accepted. It is certainly not enough for a witness to make a bald assertion that the footprints were those of X, even if he or she says he or she had lived together with X in the same area for some time and he or she knew those prints well. The witness must be asked by what characteristics or peculiarities, marks or indications he or she recognised the footprints as being those of X. The ability to give a precise and detailed description and to point to features of unique distinction will point in the direction of reliable identification.
In Menzou HH-90-93, the court said there should be details of the length of the trail, its clarity and the expertise of the witnesses who followed it. The identification of a footprint as belonging to an accused must be precise and set out enough unusual features to convince the court beyond reasonable doubt.
In cases in which the identity of the footprints of the suspect forms a vital part of the evidence upon which the State relies for conviction, the police should, whenever possible, take a cast or other impression of the footprint at the scene of the crime and a comparative footprint from X. These should then be produced in court as exhibits so that comparisons can be made. See Menzou HH-90-93.
In respect of bare footprints an expert in handprints may also have the expertise to conduct an expert comparison between the two sets of footprints.
Reliance on shoeprints is obviously fraught with danger, especially where the shoeprint is from a type of footwear which is in widespread use: Mavunga 1982 (1) ZLR 63 (S).
Casts and photographic evidence should be produced and a scientific expert must testify as to the common features.
There is obviously substantial risk of error if identification is made on the basis of witness testimony that the voice of the suspect is the same as that of the culprit. This sort of evidence would need to be probed extremely carefully. The court should approach this sort of evidence in the same way as it does visual identification evidence. Questions such as the following must be put to the witness:
- What was there about the voice heard which made him or her sure that it was a particular person’s voice?
- Did the voice have a timbre or quality that set it apart from the voice of others and, if so, what was this quality?
- Were the words spoken when the crime was committed spoken in a soft or loud voice and for how long did the culprit speak?
- How good is the witness’s hearing?
- Were there other background noises at the time which would have made it difficult to hear the voice properly?
- Did the witness know X previously and was he or she familiar with his or her voice?
See Denhere & Ors GS-235-80; Chitate 1966 RLR 251 (A), 1966 (2) SA 690 (RA).
Expert medical and scientific evidence is obviously required.
When assessing the reliability of a person’s identification of property such as a stolen motor vehicle, the court does not look at each feature or point of identification in isolation but has regard to the cumulative effect of the various features by which the person has identified the property: Nyamaro 1987 (2) ZLR 222 (S).
As regards the identification of dead bodies see s 278(6) CPEA and as regards evidence on the physical condition of the deceased; see s 278(5) CPEA.
There is provision for handwriting evidence to be given in terms of s 261 CPEA. An expert may be called to testify as to whether the handwriting in a document produced in court is that of X. The person usually called to give evidence is the Questioned Documents Examiner of the Police Forensic Science Laboratory. The expert will testify as to the results of the comparison between the handwriting on the questioned document with the handwriting of X.
It is desirable that the handwriting expert should produce photographs supporting his or her evidence to show points of similarity between X’s handwriting and the handwriting on the questioned document. The expert should point out the similarities and the significance of these.
The vital question is whether the similarities are so strong as to exclude any reasonable possibility that the handwriting is that of any one other than X.
It is the duty of the court to satisfy itself that the handwritings are those of the same person, X. In doing this, the court is entitled to take account of its own observations regarding the similarities and dissimilarities between the handwritings.
See Chidota 1966 RLR 178 (A); Chibi v Minister of Internal Affairs 1970 (1) RLR 88 (GD); Mayahle 1968 (1) RLR 133 (A); Sibanda (2) 1963 R&N 601 (SR); Dedza v Ekblad S-196-91. See also Questioned Documents Examiners Evidence in Court by E. Dzvairo in (1988) Legal Forum Vol. 1, No. 1, p 16.
In (1987) Prosecutors Bulletin Vol. 1, No. 3 p 12, there is a useful article by Chief Inspector Haley of the Police Forensic Firearms Identification Office entitled “Ballistics Evidence in Court”. This article deals with the procedure for obtaining a ballistics report and also with the problems encountered when presenting ballistics evidence in a criminal case.
Presumption in relation to crimes only committed by those possessing particular qualification or acting in particular capacity
In terms of s 315 CPEA, where a crime can only be committed by a person possessing a particular qualification or vested with a particular authority or acting in a particular capacity it will be presumed that X had the qualification or authority or was acting in that capacity until the contrary is proven.
Where the conviction of an accused depends upon circumstantial evidence and the drawing of inferences from all the established facts, then the inference sought to be drawn must be consistent with all the proved facts and the facts should be such that they exclude every reasonable inference from them, save the one sought to be drawn: Blom 1939 AD 288; Edwards 1949 SR 30; Marange & Ors 1991 (1) ZLR 244 (S).
In Attorney-General v Benett 2011 (1) ZLR 396 (S) the court dealt with the probative value of circumstantial evidence. It pointed out that in assessing the probative value of such evidence the court must not take each circumstance separately and give the accused the benefit of any reasonable doubt as to the inference to be drawn from each one so taken. It must carefully weigh the cumulative effect of all the circumstances together. Only after it has done so is the accused entitled to the benefit of any reasonable doubt which the court may have as to whether the inference of guilt is the only reasonable inference which can reasonably be drawn. In the present case each of the circumstances relied on by the State had very little, if any, probative value. Taken together, the various circumstances did not make a case for the accused to answer.
In Mtetwa 2014 (2) ZLR 533 (H) the court set out how to deal with circumstantial evidence:
Even in the most straightforward of cases, a court must ultimately draw inferences. Some evidence, such as direct evidence, requires fewer inferences whereas other evidence, like circumstantial evidence, will require more evidence. The court is never free of drawing infeences and therefore the rules that govern the drawing of inferences govern the court in its ultimate evaluation of the evidence. The question ultimately becomes: how is the court to evaluate the evidence? The law draws no distinction between circumstantial and direct evidence in terms of weight or importance. Either type of evidence may be enough to establish guilt beyond a reasonable doubt, depending on the facts of the case. When deciding upon guilt on the basis solely of circumstantial evidence the court must decide, on the basis of all the evidence, what facts, if any, have been proved. Any facts upon which an inference of guilt can be drawn must be proved beyond reasonable doubt. After the court has determined what facts, if any, have been proved beyond reasonable doubt, it must then decide what inferences, if any, can be drawn from those facts. Before a court may draw an inference of guilt, however, that inference must be the only one that can fairly and reasonably be drawn from the facts; it must be consistent with the proven facts; it must flow naturally, reasonably and logically from them. The evidence must also exclude, beyond reasonable doubt, every reasonable hypothesis of innocence. If there is a reasonable hypothesis from the proven facts consistent with the accused’s innocence, then the court must find the accused not guilty. If the only reasonable inference the court finds is that the accused is guilty of the crime charged, and that inference is established beyond reasonable doubt, then the court must find the accused guilty of that crime. In the drawing of inferences the court must take into account the totality of the evidence, and must not consider the evidence on a piecemeal basis.
A person can be convicted of murder even if no body is found, on the basis of circumstantial evidence if that evidence is consistent with no other reasonable inference than that the victim is dead and was murdered by the accused. See Shonhiwa 1987 (1) ZLR 215 (S) and Masawi & Anor 1996 (2) ZLR 472 (S).
In the Commonwealth Magistrates Handbook this advice is to be found on circumstantial evidence:
“Means, motive and opportunity are all examples of what is called circumstantial evidence. Where direct evidence of a particular act or state of affairs is not available, one may, and indeed must, have resort to indirect means of establishing the facts … Since the direct evidence of a witness is open to all the weaknesses of observation and recollection, … evidence of a circumstantial kind may be less contestable and more easily relied on. To show that a defendant had the means, a motive and the opportunity may go some way towards convincing us of his or her guilt. It may raise a prima facie case against him or her which he or she is called upon to answer.”
Hearsay evidence is testimony not of what the witness himself saw, heard or otherwise observed, but what he or she heard others tell him or her or say about the matter under investigation. The general rule is that hearsay evidence is not admissible. The reason for this is that it is not the best evidence in that the actual observer is not giving the evidence and therefore the credibility of his or her evidence cannot be tested by cross-examination. There is also the risk that a second-hand report of what the actual observer said may be garbled or inaccurate.
There are many exceptions to the rule against hearsay evidence and the relevant textbooks should be consulted if there is a dispute as to whether the case falls within a particular exception. The main exceptions are statements made in the course of duty and dying declarations. As regards statements made in the course of duty this is provided for in s 253 CPEA. The provision applies where the person who made the statement is dead or unfit to give evidence due to bodily injury or mental condition or he or she cannot, with reasonable diligence, be identified or found or brought to court and the person made the statement in the ordinary course of duty, contemporaneously with the facts stated and without motive to misrepresent. See “Hearsay Evidence in Outline” by W.A. Hope in 1961 Rhodesia and Nyasaland Law Journal 130.
In Mvambo 1995 (1) SACR 180 (W) the court said that where an accused or his or her legal representative, in the course of cross-examining a State witness, elicits evidence of a confession made by X, the evidence is admissible, provided that —
- the witness’s answer constitutes a direct and fair answer to the cross-examination; and
- in the case of an unrepresented accused, the court is satisfied that X is fully aware of the risk attached to the question.
Considerations of fairness to X require that the courts should be vigilant to enforce these provisos. Furthermore the presumption underlying the second proviso, that legal practitioners know what they are about and are aware of the risks attaching to the questions, is not always true — particularly with inexperienced pro deo counsel. In such cases, the court should be vigilant, when counsel puts a dangerous question, or seems about to do so, to warn of the dangers and to permit its withdrawal before it is answered.
In Gquma & Ors (2) 1994 (2) SACR 182 (C) the court said it is a settled general principle that the issue of the admissibility of a statement or confession must be kept clearly distinct from the issue of guilt, which object is achieved by insulating the inquiry into voluntariness (the trial within the trial) in a compartment separate from the main trial. But once the statement has been ruled admissible X and his or her witnesses, if they give evidence in the main trial, may be cross-examined on the evidence given by them in the trial within a trial.
In Shezi 1994 (1) SACR 575 (A) the SA court stated that an accused person has the right to have the question of the admissibility of his or her statement tried as a separate and distinct issue. Hence the evidence at a trial within a trial to determine the admissibility of a statement cannot be relied upon in reference to the ultimate verdict. A trial within a trial and the main trial are separate in substance as well as form, and the former is restricted to evidence relating to the admissibility of X’s statement.
Reid-Rowland 16-37; 20-8
In terms of the CPEA the court can draw certain adverse inferences from X’s silence at different stages. The court may draw such adverse inferences from silence at these different stages as appear proper. Also, on the basis of such inferences, the silence may be treated as evidence corroborating any other evidence given against X.
Adverse inferences can be draw at the trial if, at proceedings to confirm a statement allegedly made by X to the police, X remains silent and does not mention any fact which, in the circumstances, he or she could reasonably have been expected to have mentioned. If then at his or her trial he or she challenges the statement on the basis that he or she did not in fact make it or he or she did not make it freely and voluntarily, the court may draw adverse inferences from his or her earlier failure to mention the facts: s 115 CPEA.
Adverse inferences can be drawn if X pleads not guilty or the judge enters a plea of not guilty because X refuses to plead to the charge, and on being called upon to give his or her defence outline he or she fails to mention any fact relevant to his or her defence which, in the circumstances existing at the time, he or she could reasonably have been expected to have mentioned. Such adverse inferences can be drawn by the court from his or her earlier failure to mention these facts when it determines his or her guilt for the offence charged or any other crime which he or she may be convicted of on that charge: s 189(2) CPEA.
This provision must not, however, be taken too far in relation to X who is not legally represented. In Pandehuni 1982 (2) ZLR 133 (S), the Supreme Court stated that even where X has been properly warned of the fact that adverse inferences may be drawn from his or her failure to mention relevant facts in his or her defence outline, it must still be borne in mind that an accused who is required to give an ex tempore summary of the features of his or her defence may very easily fail to marshal his or her thoughts so as to include in the outline everything that should be included. The outline is after all only that; it is not a detailed and comprehensive exposition such as would be expected in the course of evidence in chief. X must therefore be allowed to add facts to his or her evidence in chief which are not in his or her initial outline. This is why it is wrong to use the shortcut device of prompting X simply to adhere to his or her initial outline when giving evidence in chief before cross-examination commences. This applies even when he or she is asked if he or she wishes to add anything to that outline. By circumscribing X in this way he or she is not able to give a full exposition of his or her case in his or her evidence in chief and this means that he or she does not have a proper opportunity to give full evidence covering in detail the additional facts not in his or her initial outline.
If X is circumscribed in this way, the scepticism which arises from the fact that he or she has mentioned salient facts for the first time only when under cross-examination may be entirely unjustified.
If X refuses to answer any questions put to him or her when he or she gives evidence in his or her defence, and his or her refusal is not for a just cause (on the grounds of privilege), adverse inferences may be drawn by the court from the failure to answer the questions: s 199 CPEA.
In Masawi & Anor HH-111-94, the facts were that at 7 am on 15 December 1992 a woman was kidnapped; in late December a second suspect was arrested and after being warned and cautioned simply said he denied the charge. The court drew an adverse inference from his failure to give any indication that his defence was an alibi or any details of what he was doing during the crucial period. Although the court accepted that he had been assaulted by police at the time of his arrest and then denied both a timeous remand and access by a lawyer for some days, it was found to be an insufficient justification for giving no facts at all.
Even if X has declined to give evidence in his or her defence he or she can nonetheless be questioned by the prosecutor or the court. If, without just cause, he or she refuses to answer such questions adverse inferences may be drawn by the court from the failure to answer the questions: s 199 CPEA.
If X’s statement to the police is admissible, the statement needs to be carefully examined to see what sort of a statement it is. In the Prosecutors Handbook, it is pointed out that the statement may amount to:
- a complete admission (in which case if X has not challenged the admissibility of the statement he or she probably will have pleaded guilty);
- a partial admission (as where X admits to stealing two of the four items which he or she is alleged to have stolen);
- a complete denial;
- a partial denial;
- an innocent explanation.
Partial admissions and partial denials need to be particularly carefully treated. If X is charged with attempted murder and says that he or she admits to having stabbed the complainant, this does not mean that he or she has admitted that he or she did so with the intention to kill; if a person is charged with assault with intent to do grievous bodily harm and denies that he or she used a knife as alleged, he or she may nonetheless have admitted in his or her statement to the police that he or she did have a fight with the complainant.
Note that the statement of X in reply to police questions is only evidence against the maker of the statement: s 259 CPEA. This is because there is no opportunity for cross-examination of the person who made this statement when he or she makes his or her statement. But if the maker goes into the witness box and repeats on oath what he or she said in the statement, he or she renders himself liable to cross-examination by an accused who is jointly charged with him or her and thus such evidence on oath is admissible against the co-accused.
The mere fact that there are some minor discrepancies in the testimony of a witness does not mean that that testimony must be rejected. The mere fact that there are some minor contradictions in the testimony of the various witnesses testifying for either the State or the defence does not mean that the testimony of all those witnesses must be rejected. The nature and extent of the discrepancies and contradictions must be probed, together with the overall question of credibility of the various witnesses.
A witness statement may be entirely or substantially true, entirely or substantially false, partially true or partially false. The witness may also have a reason for altering his or her testimony in favour of X or may have a reason such as a grudge for falsifying his or her evidence against X.
A State witness may have made a statement to the police which was against X. However, when the case gets to court, because of his or her relationship to X or for some other reason, he or she may depart from his or her statement and give evidence favourable to X and become a hostile witness against the State case.
When a witness has become hostile the State has the option to apply to impeach the witness in terms of s 316 CPEA. The object of impeachment is not to persuade the judge to accept the evidence he or she gave in his or her original statement, but to destroy his or her reliability for either side.
To have his or her own witness impeached by the court the correct procedure is as follows:
- The State must first produce the previous apparently conflicting statement from the witness and the prosecutor should give the witness sufficient particulars of the statement to identify the occasion on which it was allegedly made.
- The witness must then be asked whether he or she made the alleged statement. If the statement is signed, he or she may be asked to admit to the signature.
- If the witness admits to having made the statement on the specified occasion, the statement should be put to him or her and he or she should be asked to admit or deny using the words alleged.
- If the witness denies using the words alleged, the prosecutor may apply to adjourn the case so that he or she can call witnesses to prove that the statement was made by this witness. Where the statement has been interpreted, the interpreter must be called.
- If the witness admits using the words alleged, the statement may be used without further proof.
The witness must then be asked to explain the discrepancies between the statement on the occasion specified and his or her present testimony, and what the truth of the matter really is.
See Muhlaba & Ors 1973 (1) RLR 178 (GD) and C Goredema “Procedural aspects relating to the impeachment of witnesses” (1989) Legal Forum Vol. 1 No. 6, p 8.
The witness’ explanation of the apparent conflict may be entirely acceptable. If it is not, it may be appropriate to impeach him or her in terms of s 316  CPEA. In Chari 1989 (1) ZLR 231 (S), after a State witness had given evidence inconsistent with a previous sworn statement, the prosecutor had produced this statement as an exhibit. Without further ado the magistrate summarily dismissed the witness and excused him or her from further attendance. The Supreme Court held that this amounted to a gross irregularity. The prosecution should have laid a proper foundation for the impeachment and the defence should have had an opportunity to cross-examine the witness.
If the State decides to apply to have the witness declared hostile and the judge declares him or her to be adverse, the State can then proceed to cross-examine the witness. It is an irregularity for the State to be permitted to cross-examine its own witness before the court has declared the witness to be hostile. Before the witness has been declared hostile, the prosecutor may not go beyond putting the discrepancies and eliciting an explanation from the witness; he or she may not proceed with full blooded cross-examination of the witness.
Although the object of this cross-examination may be to discredit all aspects of his or her testimony, this witness may say some things under cross-examination which, in fact, implicate X. As seen below the State can then seek to rely on those portions of the testimony of this hostile witness which assist the State case.
In Bennett (1) 2010 (1) ZLR 42 (H) the State sought to have its principal witness declared hostile in order that it could cross-examine him. The witness had previously made a statement to the police, which had implicated the accused in the present trial. The witness had challenged the admissibility of the statement, on the grounds inter alia that he had not been correctly warned and cautioned. The prosecution in that trial did not attempt to introduce the statement in evidence. The witness was subsequently convicted and sentenced to a term of imprisonment, which he served. When notified that he would be subpoenaed to give evidence against the accused, the witness made it clear that he absolved the accused of any wrong doing. Nonetheless, the prosecution called him and his evidence was favourable to the accused and against the State case.
The court held that the witness's statement to the police was inadmissible against him and a fortiori inadmissible against the accused. In any event, being an alleged confession by the witness, it was, in terms of s 259 CPEA, inadmissible against any other person. It could not be used for the purpose of impeaching the witness.
The court ruled further that the basis for impeachment is not restricted to previous inconsistent statements. There are various ways of proving hostility and proof of previous inconsistent statement is only one of them. The basis of impeachment proceedings is adversity or hostility on the party of a witness against a party calling him. A witness can only be considered adverse, or hostile, if he is shown to bear a hostile animus towards the party calling him and so does not give his evidence fairly and with the desire to tell the truth. Hostility may be inferred through various considerations, which include his demeanour in the witness stand. In casu, the witness considered that he had been unjustly prosecuted, convicted and served a prison term at the instance of the State and he still viewed the State as an adversary. He would therefore be declared hostile and the State would be entitled to cross-examine him.
Where a witness has been impeached after departing from his or her statement to the police and it is obvious that the witness is favourably inclined towards X, it is permissible to accept and rely on that part of his or her evidence which tends to incriminate X: Millar 1971 RLR 159 (A) and Mpofu & Anor S-150-89. In Millar at 160 it is stated that “it is quite illogical to say that, because the witness is trying to help X to the utmost extent, he or she must not be believed when he or she gives evidence which does not help X, but which tends to incriminate him or her… There is no ground in law … for rejecting out of hand those portions of evidence … which implicate him.”
In Mutters & Anor S-66-89 defence counsel had been allowed to put to State witnesses previous inconsistent statements they were alleged to have made in a previous trial. However, the magistrate had refused to admit as evidence the record of the previous trial. The Supreme Court held that by refusing to admit the record as evidence, the magistrate had precluded the defence from proving inconsistencies in the testimony of the witnesses, and had disabled himself from adjudicating on their credibility. The record was perfectly admissible and should have been admitted in evidence.
If a State witness makes a statement which is apparently inconsistent with a previous statement made during investigations before trial, this witness must be asked if he or she made the previous statement and whether he or she has any explanation for the inconsistency.
The prosecutor can only cross-examine his or her own witness after the court has, on application, declared him or her to be hostile.
The entire testimony of an impeached witness does not have to be disregarded. The prosecutor may ask the court to take into account portions of the testimony which are prejudicial to X.
If a witness makes a statement which is apparently inconsistent with a previous statement made during investigations before trial, the party calling the witness may ask his or her witness if he or she made the previous statement and whether he or she had any explanation for the inconsistency.
The party calling a witness can only cross-examine its own witness if the court has on application first declared him or her to be hostile.
The entire testimony of an impeached witness does not have to be disregarded. The State may ask the court to take into account portions of the testimony which are prejudicial to the accused.
A statement made by a State witness in a criminal case to the police, whether as an affidavit or otherwise, is not normally admissible in evidence unless he or she departs from it in a material respect and is impeached. A witness, as a rule, is not permitted to confirm or strengthen his or her evidence by testifying that he or she had made a similar statement on a previous occasion.
There are, however some exceptions to this rule.
If X puts to the witness under cross-examination that his or her story is a recent fabrication, the witness’ previous statement becomes admissible in order to show that he or she had made a previous consistent statement at a time sufficiently early to be inconsistent with the suggestion that the present account was a recent invention.
Complaints in sexual cases are admissible to show consistency and to deny a defence of consent but not to prove its content or to corroborate the evidence of the complainant;
Other previous statements which are admissible are statements forming part of the res gestae, statements relating to previous identification, to show consistency in the identification and previous statements by accused persons.
(The res gestae consists of the facts constituting and immediately accompanying the matter which is in issue. It includes facts leading up to, explaining and following continuously from the facts in issue. Thus evidence by a hearer of what the victim shouted when assaulted is admissible as part of the res gestae.)
In Jesse v Attorney-General & Ors HH-213-94 it was pointed out that the police have a right to require a blood sample from an accused to ascertain any fact material to their investigations. This implies a right to use force if necessary to obtain it, if consent is withheld. The sample has to be taken by a medical officer at the written request of a police superintendent or above. X is entitled to be told who is to take the blood sample, where and when, and the facts sought to be established, i.e. why it is to be taken; and should be given a copy of the request to a named medical officer. Applicant had the right to ignore a demand that did not comply with this, and should not have been threatened with force until the correct procedure had been followed. However, his fear of false incrimination had not been justified.
In Mandwe S-142-93 it was stated that a blood test must be taken on the orders of an inspector or above. Thus, a blood test taken on the orders of a sergeant is incurably defective and is inadmissible.
Certain witnesses are not competent to give evidence according to the rules of evidence. For example, under s 246 CPEA “no person appearing or proved to be afflicted with a mental disorder or defect or labouring under any imbecility of mind arising from intoxication or otherwise, shall be competent to give evidence while under the influence of any such malady or disability”.
Where an allegation that a witness is mentally disordered is made during a criminal trial and the witness appears to be mentally disordered, the court must properly investigate whether the witness is competent in terms of this provision.
In Ndiweni S-149-89 the court failed to probe an assertion by the defence that a State witness was labouring under some mental disorder. The State did not challenge this assertion. The appeal court said that this was an irregularity.
Normally, it is totally impermissible for the State to prove the previous convictions of X before he or she has been convicted of the offence with which he or she is being charged. The prosecutor may not refer to the previous convictions of X prior to X being found guilty of the charge. Nor may the prosecutor ask X when he or she is testifying whether he or she has previous convictions: s 324 CPEA. The reason for this is obvious. The judge should be solely concerned with whether X has committed the present offence. If he or she knows that X has a string of previous convictions, this might mean that the judge will be biased against him or her and will find him or her guilty on the basis of his or her previous criminal tendencies rather than because his or her guilt on the current offence has been proved beyond reasonable doubt. For this reason, if the previous convictions are disclosed before he or she has been found guilty this will constitute a gross irregularity which will lead to the proceedings being set aside.
There are however some exceptions to this general rule.
1. Accused charged with receiving of stolen property.
If X is charged with receiving, in terms of ss 305 and 306 CPEA the State is permitted at any time during the trial to lead evidence that the person was found in possession of stolen property within the period of twelve months preceding the time when the person was first charged with the current offence and evidence that within the preceding five years he or she has been convicted of an offence involving fraud or dishonesty. X must be given three days written notice before such evidence is introduced.
This evidence can then be taken into account by the court in deciding whether or not X knew that the property he or she had in his or her possession on the present occasion was in fact stolen.
2. Although evidence of previous convictions or bad character of X may not normally be introduced before conviction, in terms of s 290 CPEA such evidence can be introduced before conviction if:
- X has given evidence of his or her own good character or he or she or his or her lawyer has asked a witness questions to try to establish his or her good character;
- aspersions are cast upon the character of the prosecutor or State witnesses by the defence; and
- X has given evidence against another person charged with the same offence;
In Jesse v Pratt & Anor 2001 (1) ZLR 48 (H) it was held that the magistrate had permitted a serious irregularity to take place, the adduction of evidence of X’s bad character, even though X’s cross-examination of the complainant had not even exposed himself to cross-examination as to his character. Although s 290 CPEA allows X to be cross-examined as to his or her character, the scope of the section is limited. X must have some latitude to examine on credibility without exposing himself to the extremely damaging prospect of being examined on his or her character. He or she should only be vulnerable to such an attack where he or she has attacked the credibility of a State witness, by attempting to impeach character, on an issue not being an essential element of the charge or offence, and to a degree not adjudged warranted, and where the court in its discretion permits such an attack upon the accused.
In Tswangira S-184-95 the police had detained X for more than the permitted 48 hour period. He confessed to the crime 63 hours after his arrest. In such a case, the court has the discretion to decide whether to admit the evidence despite the illegality. Factors that the court would take into account included:
- whether the illegality was intentional or inadvertent
- whether the illegality was trivial or technical or whether it involved serious invasion of important rights, the recurrence of which would involve a real danger to such rights
- whether it was a situation of urgency or emergency which provided some excuse for the illegality.
In this case the only illegality shown was detaining X beyond the permitted period without taking him before a magistrate. The illegality was inadvertent as X had surrendered himself at one police station and was collected and dealt with by an investigating officer from another station, who did not notice the time of his arrest. The policeman’s dereliction of duty was condemned, but the Appeal Court decided that the trial court had properly admitted the statement.
See also Hammer & Ors 1994 (2) SACR 496 (C) where the court said that it has a discretion to exclude evidence that has been illegally or improperly obtained.
Section 269 CPEA provides that it is not lawful for a court to convict a person charged with treason “except upon the evidence of two witnesses where one overt act is charged in the indictment or, where two or more such overt acts are so charged, upon the evidence of one witness to each such overt act.”
This rule is a safeguard against false accusations of treason.
In S v Tsvangirai & Ors HH-119-03 Garwe JP set out the special evidential rules relating to treason as follows:
“It is not competent for a court to convict a person of treason except on the evidence of two witnesses for each ‘overt act’ charged or, where two or more overt acts are charged, one witness for each overt act. There would be no compliance with the Act if, in a case where more than one overt act is charged and there is only one witness, the same witness were to give evidence on each of the overt acts. Where one overt act is charged at least two witnesses must give evidence on the overt act, although the evidence need not overlap. Where the court is relying on the evidence of only two witnesses to prove the whole overt act, the evidence of each of those witnesses must be such that, standing alone, it would, if believed, be adequate to establish that the accused committed the overt act of treason with which he is charged.
An overt act is any act manifesting the criminal intention and tending towards the accomplishment of the criminal object. It is generally a composite thing, passing through distinct stages and made up of various circumstances. Several witnesses speaking to those different stages and circumstances may be necessary.”
In Mupanedungu S-197-94 the court pointed out that traffic accidents happen suddenly and unexpectedly. Those involved, and the bystanders, are shocked. Their recollections are impressionistic and often inaccurate. The facts on the ground will often show that witnesses’ recollections of details are unreliable. This does not mean they are lying. It means only that their memory of a sudden, unexpected, fast-moving and fast-changing series of events is faulty. Judicial officers need to use common sense to sort out fact from imagination and self-justification. The parties in particular usually intended to act sensibly and are often convinced from their recollection or reconstruction of events that they did.
In Khupe & Anor HB 30-83 the court pointed out that penetration in legal terms means the slightest entry into the female body. It is not necessary that the hymen be ruptured. Doctors compiling medical reports understandably fail to appreciate this; and hence often state that there has been no penetration, while the complainant’s evidence shows that there was some. The court should not be unduly influenced by the medical report.
In Dube S-216-93 after giving evidence a young girl who had accused her father of 12 counts of rape had allegedly recanted her story in a letter to a church minister. The trial court refused to allow the letter to be introduced as hearsay and refused to allow the complainant to be recalled because it preferred finality in trials. This was wrongful exclusion of possibly material evidence, and the conviction was quashed.
In Zaranyika & Ors HH-41-95 the court said that magistrates should carefully compare the medical report against the evidence of the complainant in a rape case, not accept the report with an unquestioning mind unless it clearly and conclusively supports her evidence; and call the doctor if there is any possible ambiguity. This can have a significant effect on assessing the complainant’s credibility.
In Dube S-139-95 the court decided that two affidavits concerning medical examinations of the complainant were inadmissible, as X had not been given 3 days’ notice and had not consented to their production. The other evidence was insufficient to sustain the conviction, with the trial court ignoring accused’s requests for a medical examination to prove he did not have any sexually transmitted disease. If affidavits had been admissible they would still have needed to be elucidated by viva voce evidence because a number of questions arose from them, and it was not clear whether the doctors had been advised of facts that could have provided alternative explanations for the young complainant’s vaginal warts and injuries. Both doctors had left the country. The conviction was set aside.
In Rembani S-141-95 the court said that the medical report produced in this case was confusing; appellant argued that it suggested complainant was promiscuous and not a virgin as she claimed; but medical reports should not be taken at face value; there were other inferences that could reasonably be drawn from the report. Without oral testimony from the doctor, the court could not draw the inference suggested by the appellant. The complainant’s story was credible and corroborated by other evidence.
In Matekukamizora HH-192-94 the court said that in a case where X is charged with having sexual intercourse with an “imbecile” in contravention of the Criminal Law Amendment Act, there must be medical investigation and evidence essential to establish imbecility.
In Mbizi S-184-84 the court said that whether the requisite state of mental defectiveness has been reached in a particular case is a question of fact to be determined on the basis of expert medical testimony.
See also Chamukwanda HB-17-90.
In Chiunye HH-153-94 the court pointed out that even when an accused pleads guilty to abortion or unlawful termination of pregnancy, the court must establish that her actions actually caused the abortion and that it was not spontaneous or coincidental.
In Tamba S-81-91 the court criticised the “boxing match” approach especially in assault cases arising from an affray. In such cases the State tends to throw the two protagonists into the ring with the magistrate as referee. At the end of the bout, the magistrate awards points for demeanour and probability, and names the winner, who is usually the complainant. Usually, neither version is entirely true. Each party will tend to minimise his or her own role and exaggerate that of his or her opponent. So it is not just a question of who is telling the truth, but how much of the truth is being told by each of them. The complainant in a case of this nature must be regarded in many ways as the courts regard an accomplice. It is unsafe to rely on his or her unsupported word. In the present case the conviction was quashed because of the State’s failure to lead evidence from others present or involved in the incident.
In Chatyoka S-75-92 the Supreme Court pointed out that judicial officers should not forget that a complainant with a legitimate complaint may still exaggerate his or her own virtue and gloss over his or her faults; disbelieving X does not mean that the court should believe every word of the complainant. Usually, neither party has a monopoly of the truth.
In Chiweshe S-38-93 the court pointed out that in assault cases, often neither side has a monopoly of the truth, and the court must analyse the evidence to reconstruct the facts; not simply decide which story is more credible and accept it entirely.
In Moyo & Anor S-12-93 the court said that in assault cases the two common faults are:
- choosing and believing one or the other version of a fight, without considering whether both parties are improving their versions; and
- deciding the case against co-accused jointly - the court must weigh up the evidence against one accused and reach its conclusion for him, then weigh up the evidence against the next person etc.
A court also cannot decide on the basis of whose story is more probable and credible because of the burden of proof; if the defence version could reasonably possibly be true, although improbable, X must be acquitted.
A complainant is often a single witness with an interest strongly adverse to X’s. His or her evidence should be treated with great caution. Where it conflicts with the defence evidence it should be accepted only where it is corroborated or overwhelmingly probable.
In Chikasha S-94-94 appellant was a postal clerk; he recorded he had received $3 000 one day when he had received $3 200, and his books did not account for the extra $200, leaving him with a general deficiency for that amount that day. Appellant alleged some receipts were really carried over from the previous day, creating the appearance of a deficiency, and a surplus on that day plus the sale of stamps accounted for the apparent deficiency on the day in question. The page for the previous day was not produced in court; the postmistress simply said that the previous day balanced. The best evidence rule required that the book itself be produced so the court could judge for itself whether there was any truth in the appellant’s claim, not simply rely on a witness’s testimony and conclusion on the point. As with fingerprints, the court must look at the evidence itself on which the witness’s opinion is based. The appeal was allowed and the conviction and sentence were quashed.