T.M v S (Appeal) (CA&R58/2024) [2025] ZAECMHC 72 (19 August 2025)

Case Note

T[...] M[...] v The State, Case No: CA&R 58/2024 (High Court of South Africa, Eastern Cape Division, Mthatha), 19 August 2025

Reportability

This judgment is reportable because it clarifies the approach an appellate court must adopt when reviewing convictions for rape that turn primarily on the testimony of a single child witness. It underscores the continued relevance of the cautionary rule in relation to such evidence while emphasizing that the ultimate test remains whether the State has proved the case beyond a reasonable doubt. The court’s careful integration of principles from leading authorities on appellate restraint, single-witness testimony, and the evaluation of mixed medical and lay evidence gives the judgment broader significance for criminal adjudication involving vulnerable witnesses.

It is also significant in its treatment of the interplay between forensic indications of sexual assault and the statutory elements of rape. By interrogating the probative value of a J88 conclusion that “sexual assault probably occurred” alongside a DNA result showing “no semen detected,” the court articulates a principled evidential threshold for proving sexual penetration. The decision therefore provides valuable guidance on how trial courts should evaluate medical evidence, expert elucidation, and first disclosures in sexual offence cases, particularly where complainants are minors.

Finally, the case has sentencing significance. It demonstrates that where the trial court has materially misdirected itself on conviction, any sentence—however carefully considered or tailored—must fall away. In doing so, the judgment reinforces the primacy of the conviction inquiry and the necessity that the State’s case meets the standard of proof beyond reasonable doubt before sentence considerations arise.

Cases Cited

R v Dhlumayo and Another 1948 (2) SA 677 (A).

S v Hadebe and Others 1997 (2) SACR 641 (SCA).

S v Monyane and Others 2008 (1) SACR 543 (SCA).

S v Francis 1991 (1) SACR 198 (A).

S v Sauls and Others 1981 (3) SA 172 (A).

S v Webber 1971 (3) SA 754 (A).

Woji v Santam Insurance Co Ltd 1981 (1) SA 1020 (A).

S v M 1999 (2) SACR 548 (W).

S v Trainor 2003 (1) SACR 35 (SCA).

S v Van der Meyden 1999 (1) SACR 447 (W).

R v Difford 1937 AD 370.

Legislation Cited

Criminal Law Amendment Act 105 of 1997, section 51(1).

Criminal Procedure Act 51 of 1977, section 208.

Criminal Law (Sexual Offences and Related Matters) Amendment Act 32 of 2007, section 5.

Rules of Court Cited

No specific Rules of Court were cited in the judgment.

HEADNOTE

Summary

The appellant appealed against his conviction for rape and a sentence of 15 years’ imprisonment imposed by the Mthatha Regional Court. Although the prescribed minimum sentence for rape of a child under 16 is life imprisonment under section 51(1) of the Criminal Law Amendment Act 105 of 1997, the trial court found substantial and compelling circumstances and imposed 15 years. The High Court granted leave to appeal and heard the matter.

The central complaint was that the trial court erred in accepting the complainant’s evidence—she was both a child and a single witness—without adequate caution, and that the State failed to prove sexual penetration beyond a reasonable doubt. The first report suggested sexual assault rather than rape, the J88 noted that sexual assault “probably occurred,” and DNA analysis detected no semen. The appellant argued that these features, together with inconsistencies in the disclosure to the first confidante, undermined the State’s case on penetration.

The High Court (Mhambhi AJ, Rusi J concurring) held that the trial court materially misdirected itself by discounting critical evidential gaps and by failing to engage with the suggestibility of the child’s first report and the inconclusive medical evidence. The State had not proved the charge of rape beyond a reasonable doubt. The appeal succeeded, the conviction and sentence were set aside, and the appellant was ordered to be released forthwith.

Key Issues

The court considered whether the trial court correctly applied the law governing the evaluation of a single witness, particularly when that witness is a child, and whether, on the totality of the evidence, the State discharged its burden of proof beyond a reasonable doubt. The court also confronted whether the medical evidence (J88 and DNA) supported the allegation of rape, as distinct from sexual assault.

It addressed the significance of the complainant’s first report to a confidante, which used the language of “sexual assault,” and whether the subsequent decision to lay a rape charge after the confidante’s suggestion undermined the reliability of the allegation of penetration. The court weighed this against the accepted cautionary approach to child testimony and the risk of suggestibility.

Finally, the court determined whether the trial court’s handling of the medical evidence, including its failure to call the doctor to elucidate the J88 conclusion in light of a negative semen finding, constituted a misdirection justifying appellate interference with factual findings.

Held

Held, the trial court materially misdirected itself by not adequately assessing the complainant’s evidence with appropriate caution given her status as a single child witness, by over-emphasizing her credibility in isolation, and by under-valuing significant countervailing evidence, including the first report and the forensic results.

Held, the J88 notation that sexual assault “probably occurred” and the DNA result that “no semen was detected,” viewed with the complainant’s first report and the absence of expert elucidation, failed to establish the element of sexual penetration necessary to prove rape. The State accordingly did not prove the case beyond reasonable doubt.

Held, the appeal against conviction succeeds; the conviction and sentence are set aside; and the appellant is to be released immediately. The sentence appeal is rendered moot by the setting aside of the conviction.

THE FACTS

The appellant, the biological father of the complainant, was charged in the Regional Court with the rape of his daughter, then approximately 14 years old. They lived together in a single rented flat with the complainant’s mother, and the alleged incident occurred while they were all present. The complainant testified that the appellant first touched her, including on her breasts, then assaulted her physically, forced her to lie on her back, and inserted his penis into her vagina. Her description of the penetration was that she “felt something inside” and that her thighs were “wet.”

The complainant’s first report was made to a confidante, Pamela, a friend of her mother. The record shows that the complainant asked Pamela how long it takes to open a case of “sexual assault.” Pamela testified that the complainant did not provide details of how the rape allegedly occurred, and that she, Pamela, did not know the difference between “sexual assault” and “rape.” Pamela accompanied the complainant to the police station where a rape charge was laid, reportedly after Pamela suggested that a rape case be opened.

Medical evidence consisted of a J88 completed by Dr Linda Mtshulana. He observed a whitish discharge and took swabs for DNA analysis. The DNA analysis concluded that no semen was detected. The J88 recorded that sexual assault “probably occurred.” Dr Mtshulana was not called at trial to clarify the reconciliation between the negative DNA finding and the J88 conclusion, nor to explain the probative value of the discharge in relation to sexual penetration.

THE ISSUES

The first issue was whether the trial court properly applied the cautionary approach to the evidence of a single child witness, and whether it adequately weighed the complainant’s testimony against the totality of the evidence, including first reports and medical findings. The question was not whether the complainant was competent to testify, but whether her evidence was reliable and sufficiently corroborated to establish rape beyond a reasonable doubt.

The second issue was whether the State proved the element of sexual penetration, required for a conviction of rape, as opposed to sexual violation underpinning sexual assault under section 5 of the Criminal Law (Sexual Offences and Related Matters) Amendment Act 32 of 2007. The J88’s notation that sexual assault “probably occurred,” coupled with DNA results indicating “no semen detected,” necessitated careful scrutiny of whether penetration had been established.

The third issue was whether the trial court’s discounting of the complainant’s first report, the role of the confidante’s suggestion to lay a rape charge, and the failure to call the medical practitioner to elucidate the J88 conclusion amounted to misdirections that warranted appellate interference with the factual findings and credibility assessments.

ANALYSIS

The court commenced by restating the appellate standard: absent material misdirection, a trial court’s findings of fact and credibility attract deference and are presumed correct. However, where misdirection is demonstrable on the record, an appellate court may intervene. This framework draws from R v Dhlumayo and Another and its progeny, including S v Hadebe and Others, S v Monyane and Others, and S v Francis. Against this yardstick, the court examined the Regional Court’s treatment of the evidence.

On the evaluation of a single witness, the court acknowledged section 208 of the Criminal Procedure Act 51 of 1977, which permits a conviction on the evidence of a single competent witness, and rehearsed the approach in S v Sauls and Others and S v Webber that there is no rigid formula; rather, the court must assess trustworthiness in light of merits and demerits. In relation to a child witness, the court referenced Woji v Santam Insurance Co Ltd on the determinants of trustworthiness—powers of observation, recollection, and narration—while endorsing the more holistic evaluation suggested in S v M, namely that caution must not supplant the ultimate test of proof beyond reasonable doubt.

Applying these principles, the court held that the Regional Court failed to consider critical features that undermined proof beyond reasonable doubt. The complainant’s first report to Pamela referenced “sexual assault,” and Pamela’s testimony revealed that the complainant did not describe the mechanics of rape at that stage. Pamela further admitted not knowing the difference between sexual assault and rape, yet it was on her suggestion that a rape charge was laid. These features raised legitimate concerns about the complainant’s suggestibility—a well-recognized risk when dealing with child witnesses—and should have prompted a more exacting evaluation of reliability and consistency.

The medical evidence did not cure the deficiencies. The J88’s conclusion that sexual assault “probably occurred” does not equate to proof of sexual penetration, an essential element of rape under the statutory scheme. Moreover, the DNA analysis reflected “no semen detected,” which, while not dispositive of penetration, was squarely at odds with the complainant’s assertion that her thighs were “wet,” especially in circumstances where the State sought to rely on bodily fluids as corroborative of penetration. The court held that the trial court should not have discounted this conflict by remarking that “somebody failed to do his duty.” Instead, the proper course was to call the medical practitioner to elucidate the clinical basis for the J88 conclusion and to reconcile it with the DNA result.

Synthesizing the evidence as required by S v Trainor and S v Van der Meyden, the court emphasized that reliable and unreliable components must be weighed together and not in isolation; the State’s case can succeed only if, on the conspectus, there is no reasonable possibility of innocence. On that conspectus, the court found that the complainant’s evidence—given her age, the first report, the role of Pamela’s suggestion, and the inconclusive medical evidence—did not, without more, establish penetration beyond reasonable doubt. At best, the evidence suggested the possibility of sexual violation as contemplated in section 5 of Act 32 of 2007, which is distinct from rape because it specifically excludes penetration.

In the result, the court concluded that the trial court had materially misdirected itself by failing to treat the evidence with the necessary caution and by overlooking material inconsistencies and forensic limitations. The State therefore failed to discharge its burden, and the appellant was entitled to the benefit of the doubt. The conviction could not stand.

REMEDY

The High Court upheld the appeal against conviction. Having found that the State failed to prove the charge of rape beyond a reasonable doubt, it set aside both the conviction and the sentence imposed by the trial court. The sentence appeal thus became moot as a matter of course.

The court ordered the appellant’s immediate release from incarceration. The order reflects the principle that once a conviction is set aside for want of proof, no further considerations of sentence or mitigatory factors are required; the accused must be restored to liberty without delay.

Given the dispositive nature of the evidential misdirections and the insufficiency of proof on penetration, the court did not remit the matter for further evidence or a retrial. It considered the record complete for purposes of determining the appropriate outcome and granted final relief accordingly.

LEGAL PRINCIPLES

The judgment reaffirms that the ultimate and only test in criminal adjudication is whether the State has proved the accused’s guilt beyond a reasonable doubt. Cautionary rules, including those pertaining to a single witness and a child witness, are aids to evaluation and must not displace the primary standard. Where there is a reasonable possibility that the accused might be innocent, the court must acquit. This principle is anchored in S v Van der Meyden and R v Difford and echoed across subsequent appellate authority.

It clarifies the correct approach to assessing the evidence of a single child witness. Trustworthiness depends on the witness’s powers of observation, recollection, and narration, as per Woji v Santam, and the court should evaluate the evidence contextually, considering first reports, potential suggestibility, internal consistency, and the interplay with objective or scientific evidence. While section 208 of the Criminal Procedure Act permits conviction on the evidence of a single witness, courts must apply caution and assess the evidence in its totality.

The decision distinguishes between the statutory elements of rape and sexual assault under the Sexual Offences and Related Matters Amendment Act 32 of 2007. Rape requires proof of sexual penetration, whereas sexual assault concerns sexual violation and expressly excludes penetration. Medical evidence such as a J88 noting probable sexual assault does not, without more, establish penetration. Where forensic results (such as DNA indicating no semen) and clinical observations are in tension, trial courts should consider calling the medical practitioner to elucidate conclusions before drawing adverse inferences. Failure to properly engage with such evidence can amount to a material misdirection justifying appellate intervention.

Finally, the court reiterates the principle of appellate restraint and the circumstances warranting interference with factual findings and credibility assessments. Guided by R v Dhlumayo and S v Hadebe, an appellate court will defer to the trial court’s advantages unless material misdirection is shown. Where misdirection is demonstrated—such as ignoring salient features of first disclosure, suggestibility, and inconclusive or unexplained medical findings—appellate courts are entitled to substitute their own conclusions on the merits and, where appropriate, set aside the conviction and sentence.