Ngwadla v S (936/2020) [2026] ZASCA 77 (25 May 2026)

THE SUPREME COURT OF APPEAL OF SOUTH AFRICA
JUDGMENT

Not Reportable
Case no: 936/2020

In the matter between:

HLALANATHI NGWADLA APPELLANT

and

THE STATE RESPONDENT

Neutral citation: Ngwadla v State (936/2020) [2026] ZASCA 77 (25 May 2026)
Coram: KGOELE, KEIGHTLEY and COPPIN JJA and MABESELE and
BASSON AJJA
Heard: 04 March 2026
Delivered: 25 May 2026
Summary: Criminal Law and Procedure – evidence of a single witness – whether
sufficient to sustain a conviction of murder – doctrine of common purpose –
requirements for liability thereof – whether reliance on common purpose was
competent in the absence of an allegation thereof in the charge sheet – circumstantial
evidence – proper approach to inferential reasoning – proper approach to factual
findings.

2



ORDER

On appeal from: North West Division of the High Court, Mahikeng (Hendricks
DJP and Nonyane AJ) sitting as a court of appeal:
1 The appeal succeeds.
2 The full bench’s order is set aside and is replaced with the following order:
‘1. The appeal succeeds. The appellant’s conviction and sentence are set aside.
2. The order of the trial court is set aside and is replaced with the following
order:
“Accused 2, Hlalanathi Ngwadla, is found not guilty and is acquitted.”’
3. The appellant is to be released promptly.

JUDGMENT

Kgoele JA (Mabesele AJA concurring) (dissenting):

[1] Mr Hlalanathi Ngwadla ( the appellant) stood trial in the Regional Court,
Stilfontein (the trial court), together with his brother, Mr Namhla Ngwadla, who was
accused 1 during the trial, on a charge of murder read with the provisions of s 51(1)
and Part I of Schedule 2 of the Criminal Law Amendment Act 105 of 1997 (the CLA
Act), as amended. They were convicted as charged and sentenced to life
imprisonment. His brother subsequently passed away.

[2] The appellant appealed his conviction and sentence. The appeal was heard by
the North West Division of the High Court, Mahikeng (the full court). The full court
confirmed the conviction but reduced the sentence from life imprisonment to

3

18 years. Special leave was granted by this Court on both conviction and sentence.
However, the appellant chose to appeal only the conviction.

The evidence
[3] The evidence tendered for and on behalf of the State can be succinctly
summarised as follows. During the early hours of 23 November 2014, Mr Isaac
Papie Riet (Mr Riet) was on his way home when a man he knew only by sight
accosted and threatened him with a knife. He was coming from a Tavern walking
alone and his friends were walking ahead about four streets away from him. This
gentleman was wearing a vest. Upon realising that Mr Riet was not the person he
was looking for, he stopped whatever he had planned to do and released him. Curious
about what the gentleman was up to , Mr Riet followed him by walking along a
nearby street, maintaining the man’s direction.

[4] As Mr Riet approached the appellant’s residence along the street he was
traversing, he heard music playing. At first, it looked like people were dancing in
the appellant’s yard , but upon closer inspection, he saw three men surrounding
another person. Part of his evidence was that , while surrounding that person, the
three men raised their hands and legs high in the air and performed a dance -like
movement. The person who was surrounded then fell to the ground, after which the
three men picked him up. Two of the men held each of his legs, while the third held
his head. They carried him out of the yard and dropped him in the street. In the
process, he recognised this person as his friend , Klaas Matsipe (the deceased). Mr
Riet could only identify the appellant and his brother amongst the three men, as he
knew them be fore. He observed that the appellant was in poss ession of a knife in
one of his hands whilst carrying the deceased out of the yard; his brother had nothing

4

in his hand s and the third person, who was unknown to him (the unidentified
assailant), had a pick-handle.

[5] Mr Riet yelled at them to leave the deceased . This, unfortunately, diverted
their focus to him . The gentleman who accosted him in earlier the street appeared
from the opposite street and joined them. He threw stones at them to fend off their
attack. His friend Nombulelo and other people arrived and helped throw stones at
them. The assailants then left the deceased outside on the street and ran back into
their house and continued dancing. Mr. Riet observed that the deceased had a wound
on his head that was bleeding. The knife’s blade was sticking out from the side of
the deceased’s head, and the handle was broken and detached from the blade. More
people in the area also assembled and the police were summoned. The appellant and
his brother jumped over the fence and ran away when they saw the police. They were
later arrested and charged with murder.

[6] Although Mr Riet confirmed during cross-examination that he did not see the
actual assault or the stabbing, he stated that he saw the appellant’s brother’s hands
raised in the air, seemingly making a stabbing motion, even though he did not see a
knife at that moment in the air. He explained that this was the gesture he referred to
as dancing and can only assume that this was likely when the deceased was injured,
which explains his use of the word ‘injure’. He also explained that he saw a knife in
the appellant’s hand only after the deceased had fallen, when the appellant was
lifting him. I will discuss this point further later in the judgment.

[7] Mr Riet further testified that he heard two utterances made by the appellant
and his brother when he tried to intervene. The appellant’s brother said to him: ‘What
do you want here, do you also want to feel ?’ At that stage , he and another

5

unidentified assailant were running towards Mr Riet, and the appellant’s brother was
busy throwing stones at him. The appellant then shouted at them and said : ‘leave
Vasspotty, because this person has not yet died ’. Mr Riet also admitted that he and
the deceased had previously been members of a gang known as Alaska but had
ceased their membership before the incident occurred.

[8] The second witness for the State was the investigating officer. He testified that
the deceased was the only eyewitness in a murder case in which the appellant and
his brother, together with two others, were charged. They were acquitted in terms of
s 174 of the Criminal Procedure Act 51 of 1977 (the CPA) because the State could
not present sufficient evidence against them as a result of the death of the deceased
in this matter.

[9] The appellant ’s testimony is that while he and his friends were inside the
house, they heard stones being thrown from outside. He went out with h is friend
Lulu to find out what was happening and the reason for the behaviour. As he was
going out, they realised that a group o f about 17 to 20 people were busy throwing
stones at their house. An empty beer bottle struck him on the head in the process. At
that moment, Lu lu was walking ahead of him. Lulu stabbed the deceased with a
knife, without the deceased saying anything to him. They ran back into the house
and immediately came out again, now accompanied by two friends. They found the
deceased lying on the ground and decided to carry the deceased outside their yard.
According to him, his brother was not with them; he remained in the house. The mob
of people continued hurling stones at them , whilst retreating to the outside of the
yard. After putting the deceased outside the yard , they ran back to the house.
Eventually, he and his brother went to the police station to report the matter.

6

[10] To their surprise, the police didn’t listen to them upon arrival; instead, they
arrested them, saying they had been looking for them. Before the trial court, t he
appellant denied that he was involved in causing the death of the deceased. He also
denied being in possession of a knife and stabbing the deceased. When asked during
cross-examination why he was carrying the deceased out of the yard, seeing he was
badly injured, he indicated that he wanted the ambulance to find him outside his
yard.

[11] The testimony of the appellant’s brother was not consistent. He first testified
that he was inside the house and heard that a sibling had been stabbed. His version
changed as he said that they went out and stood at the front door when they saw a
mob throwing stones. The appellant was stabbed, although he did not see how. Since
the attackers were many, they quickly went back to the house, exited through the
side door with the appellant, and rushed to the police station to report the incident.
The police refused to allow them to go to the clinic first and instead, arrested them.
In addition, they assaulted them. The stabbing event also evolved at a subsequent
stage, and the revised account suggested that the appellant might have been struck
by a stone or cut by a broken piece of a windowpane that was hit by a stone. This
version also became transient when challenged w ith further questions about the
differing accounts of the event, as he later insisted he saw nothing at all, including
the presence of the deceased. He reasoned that it was dark outside as the Apollo light
was not working . He later changed and said that he saw Nombulelo among the
attackers and the deceased but only heard at the police station that the deceased was
stabbed.

7

The findings of the trial court
[12] The trial court found both the appellant and his brother guilty of murder. They
were sentenced to life imprisonment pursuant to s 51(1) of the CLA Act (minimum
sentence regime) because the trial court found that the motive for killing the
deceased was that he was a state witness who was expected to testify in their case in
another proceeding.

[13] The following findings of the trial court are crucial and need to be restated, as
the appeal rests mainly on the factual findings made from the evidence before it:
‘(a) It is also not in dispute that the [two] accused, together with two other persons, were charged
in a matter in the Regional Court in Klerksdorp, where the deceased in this matter, Mr Ma tsepi,
was a state witness.
(b) It is not in dispute that th is incident for which the accused were charged together with other
persons would have taken place in the Klerksdorp Regional Court on 12 April 2014 seven months
before these events for which the accused stand trial today.
(c) It is also not in dispute that that case was finalised in the Regional Court in Klerksdorp and that
the accused in that case were apparently acquitted.
(d) The court is satisfied with the evidence of the witness despite him being a single witness.
(e) As regards the testimony of Warrant Officer or Sergeant Kotze, it is not in dispute. And this ,
once again, corroborates the evidence Mr Riet presented to the court. There was apparently an
incident in which accused 1 and accused 2 [appellant] were involved in relation to the deceased.
(f) As regards the evidence that accused 1 and accused 2 put before the court, the court found that
they were extremely poor witnesses. Their testimony differs in essential respects regarding the
events of the day in question, as if they were not at the same place.
(g) The court has no doubt that accused 2 is presenting a fabricated version to the court.
(h) Mr Riet did not see who stabbed the deceased.

(h) Mr Riet did not see who stabbed the deceased.
(i) The evidence in this matter, which is important to infer intent with respect to both the accused,
is the following: Papi Riet, that when he asked the accused what they were doing when Klaas was
on the ground, accused 1 said to him: “What do you want here, do you want to follow? ” At that
stage, accused 1, as well as the unknown person, came running towards him, whereupon accused

8

2 shouted to accused 1 and the other person: “Leave Van Spotty because this person has not yet
died”.
(j) It is therefore clear to the court when looking at the actions of accused 1 and accused 2 that
evening in question together with a third person, that the intention was to kill the deceased.’

The findings of the full court
[14] In upholding the conviction of the appellant , the full court found : the trial
court made very strong credibility findings in favour of Mr Riet and against the
appellant and his brother; the trial court was correct to make a finding that the
deceased was eliminated because he was a material witness which brings the case
against the appellant within the confines of s 51(1) read with Part I of Schedule 2 of
the CLA Act ; and the finding that the evidence of Mr Riet was clear, reliable,
satisfactory and also corroborated by other evidence aliunde cannot be faulted.

[15] The full court, referencing the case of S v Musingadi and Others (Musingadi),1
found that the appellant never disassociated himself from the conduct of his brother
and the unidentified assailant. During the analysis of the sentence, the following
remarks were made by the full court, which seemingly influenced it to reduce the
sentence: ‘Chances are that it was not the appellant that actually landed the blow to
the head of the deceased that caused the blade of the knife to stick in the head of the
deceased because he was still in possession of his knife’.

Before this Court
[16] This Court is called upon to determine whether the full court was correct in
upholding the trial court’s order, which found the appellant guilty of murder read

1 S v Musingadi and Others [2004] ZASCA 128; [2004] 4 All SA 274 (SCA); 2005 (1) SACR 395 (SCA) paras 34-35.

9

with s 51(1) of the minimum sentence regime. The question ordinarily entails a
thorough examination of both the trial court’s and the full court’s judgments.

[17] To this end, it is trite that an appellate court applies a deferential approach to
appeals on findings of fact , giving significant weight to the trial court’s findings.
The appellant must therefore show that the trial court’s findings are clearly and
demonstrably wrong or not supported by the evidence. This approach to the factual
findings of the trial court upon review or on appeal was restated, inter alia, in S v
Kekana2 where this Court , with reference to the earlier case of S v Monyane ,3
emphasised that:
‘…This court’s powers to interfere on appeal with the findings of fact of a trial court are limited.
It has not been suggested that the trial court misdirected itself in any respect. In the absence of
demonstrable and material misdirection by the trial court, its findings of fact are presumed to be
correct and will only be disregarded if the recorded evidence shows them to be clearly wrong . . .
This, in my view, is certainly not a case in which a thorough reading of the record leaves me in
any doubt as to the correctness of the trial court's factual findings. Bearing in mind the advantage
that a trial court has of seeing, hearing and appraising a witness, it is only in exceptional cases that
this court will be entitled to interfere with a trial court's evaluation of oral testimony.’

[18] In this appeal, the appellant relied primarily on the ground that the appellant’s
conviction was based on the doctrine of common purpose, whereas the appellant was
not charged under that principle. I pause to note that this ground mutated in form
before this Court, as the ground of appeal before the full court was not articulated in
this manner. It was couched as follows:
‘the [trial court ] erred by not indicating whether the appellant herein was acting in pursuit of

common purpose with the other accused or not and whether the actions of others can be attributed

2 S v Kekana [2012] ZASCA 75; 2013 (1) SACR 101 (SCA) para 8.
3 S v Monyane and Others [2006] ZASCA 113; [2006] SCA 141 (RSA); 2008 (1) SACR 543 (SCA) para 15. See also
S v Hadebe and Others 1997 (2) SACR 641 (SCA); 1998 (1) SACR 422 (SCA) at 645E -F and S v Francis 1991 (1)
SACR 198 (A); [1991] 2 All SA 9 (C) at 204E.

10

to him, especially the stabbing of the deceased as it was evident that he never stabbed the deceased
on the morning in question.’

[19] The other grounds are that the facts presented by the State do not support a
finding of guilt for murder and that the conviction is based on the evidence of a
single witness. To support these grounds, the state contested the reliability and
credibility of Mr Riet’s evidence. In addition, the appellant submitted that, even if
the evidence of the single witness is accepted, it is not proven beyond a reasonable
doubt that the appellant caused the death of the deceased. Further, that the full court
incorrectly relied on the principles of common purpose to sustain the conviction for
murder when the charge sheet was silent on it.

Evaluation of the evidence
[20] At the onset, it is important to note that the presence of the appellant at the
scene is not in issue, as it is common cause. The manner in which the deceased was
carried out of the yard as described by Mr Riet is also common cause , but the
appellant provided a reason for his action. The bone of contention relates to whether
the appellant stabbed the deceased, as the deceased suffered stab wound (s) that
caused his death. On this aspect , there is no direct evidence as Mr Riet did not see
the appellant stabbing the deceased. He, however, stated that he saw a knife in the
appellant’s hands at the time he picked up and carried the deceased, and also after
they dropped him outside the yard, a version which the appellant vehemently denied.
From these facts alone, it is apparent that the State relied on circumstantial evidence,
but their mutually destructive versions on this part of the evidence need to be
evaluated in light of the evidence as a whole that was before the trial court.

[21] As to the cautionary rule regarding Mr Riet’s evidence, the trial court gave a
detailed account of this aspect, which the State also acknowledged . The trial court

11

warned itself against the dangers not only of Mr Riet being a single witness, but also
of his friendship with the deceased, including their prior membership in a gang. In
my view, the trial court adopted a careful approach. Likewise, it ultimately found
Mr Riet’s evidence reliable despite the identified shortcomings . I have no reason,
having perused the record, to differ from the trial court’s findings on these
conclusions. There is no demonstrable material misdirection in this regard, and I did
not find anything in the record that shows that they are clearly wrong.

[22] The full court echoed the same observation. As previously noted, it
highlighted that the trial court made a strong credibility determination in favour of
Mr Riet and, additionally, against the appellant. I entirely concur with the full court’s
assessment of the evidence.

[23] One of the circumstantial pieces of evidence accepted by the trial court is that
the appellant was seen with a knife in his possession at the time they picked up and
carried the deceased outside. The trial court accepted the evidence that there was no
obstruction in the way of Mr Riet that could have prevented him from observing
what he saw from the vantage point where he was. The trial court made the latter
finding after observing the proximity of where Mr Riet was when he saw the
appellant during the inspection in loco. Of note, an Apollo light was also pointed out
at the scene. It is significant to mention at the outset that, as far as circumstantial
evidence is concerned, the appellant’s proposition remained that the trial court and
the State did not rely on it to prove the appellant ’s participation. The second
judgment also accepted this proposition. I deal with this aspect more fully hereunder.

[24] Although the trial court did not explicitly mention the term ‘circumstantial
evidence’ in its judgment, there are several instances in its judgment indicative of its

12

reliance on it. Firstly, it drew inferences from the evidence of Mr Riet that it accepted
regarding what each of them did at the scene. Secondly, it drew inferences from their
utterances during the incident. The reliance on circumstantial evidence became more
pronounced when it addressed the issue of whether the appellant and his brother
intended to kill the deceased, as already quoted in para 13 above. Thirdly, during its
conclusion regarding the motive for the killing. Lastly, one need not be a rocket
scientist to conclude that it also drew an inference from the type of weapon used that
caused the deceased’s death, as confirmed by the post-mortem report. The criticism
cannot, therefore, hold water.

[25] It is worth noting that it is trite that the fact that a court did not mention a
particular aspect in its judgment does not necessarily mean that it did not deal with
it at all. Recently, this Court, in Van der Nest NO v Minister of Police,4 remarked as
follows:
‘Also, no judgment can be al l-embracing, and the omission of a specific factor from a judgment
does not necessarily imply that the court a quo failed to consider that factor in exercising its
discretion.’
This approach was reaffirmed a month later by this Court in MEC for Health,
Gauteng Provincial Government v AAS obo CMMS,5 that:
‘It is trite law that judgments are never all -embracing, and it does not necessarily follow that
because something has not been mentioned, therefore, it has not been considered.’

[26] To emphasise this point, I can do no better than quote the following remarks
made by this Court in Venter v S:6

4 Van der Nest NO v Minister of Police [2025] ZASCA 42; [2025] 2 All SA 655 (SCA); 2025 (5) SA 152 (SCA) para
27. See also Rex v Dhlumayo and Another 1948 (2) SA 677 (A) at 702.
5 MEC for Health, Gauteng Provincial Government v AAS obo CMMS [2025] ZASCA 91; 2025 (6) SA 152 (SCA)
para 31.
6 Venter v S [2021] ZASCA 21 para 93.

13

‘…whilst the Magistrate can be criticised for not having given sufficient reasons in respect of each
charge, the conclusions that he arrived at, as borne out by the record, were correct.’
The sentiments are apposite in our matter especially because the trial court delivered
an ex-tempore judgment.

[27] Whilst dealing with the issue of inferential reasoning, it is opportune to deal
with the issue of the motive to kill , as the appellant criticised the finding the trial
court made in this regard. The trial court found that the deceased was eliminated
because he was a state witness . The submission made by the appellant is that no
evidence was presented for the foundational reason that the deceased was killed to
eliminate him. It has been accepted in our law that motive, like intention, is, in most
instances, inferred because cases with direct evidence of motive and intention are
rare. Inferences can only be drawn from proven facts stemming from circumstantial
evidence. The trial court found that the evidence of Mr Riet and the investigating
officer corroborated each other on this aspect. The trial court even remarked that ‘it
is only too coincidental that accused 1 and accused 2 are involved in the death of the
deceased and that he was indeed a witness in another case before a court in
Klerksdorp’. This finding is, in my view, correct and cannot be disturbed. The
evidence regarding the motive, although circumstantial, serves as another piece of a
puzzle in the mosaic of inferential reasoning upon which the trial court based its
decision. The criticism is therefore not justified.

[28] The appellant’s approach to the trial court’s evaluation of evidence is
piecemeal and sometimes inaccurate. This Court, in S v Reddy and Others7 warned
against this, where it stated as follows:
‘In assessing circumstantial evidence one needs to be careful not to approach such evidence upon
a piece-meal basis and to subject each individual piece of evidence to a consideration of whether

7 S v Reddy and Others 1996 (2) SACR 1 (A) at 8C-E.

14

it excludes the reasonable possibility that the explanation given by an accused is true. The evidence
needs to be considered in its totality. It is only then that one can apply the oft -quoted dictum in R
v Blom 1939 AD 188 at 202-3, where reference is made to two cardinal rules of logic which cannot
be ignored. These are, firstly, that the inference sought to be drawn must be consistent with all the
proved facts and, secondly, the proved facts should be such “that they exclude every reasonable
inference from them save the one sought to be drawn.’

[29] To demonstrate the above, the observations from the arguments can be
summarised as follows: First, the appellant submitted in his heads of argument that
Mr Riet, along with the investigating officer, testified that the case involving the
appellant and his brother could not proceed due to the deceased’s death . This is
factually incorrect. The investigation officer testified that the matter was finalised
and the appellant and his brother were both acquitted in terms of s 174 of the CPA.
Second, the appellant su bmitted that the trial court failed to properly consider that
the single witness was a friend of the deceased. This is a clear compartmentalisation
of the evidence before the trial court. In accepting the reliability and honesty of the
single witness evidence, as indicated earlier , the record depicts that the trial court
took this issue into consideration in conjunction with other factors namely: that he
did not falsely implicate the appellant by lying that he saw him stabbing the
deceased; that he did not see any weapon on his brother; that he was honest about
his involvement in the gang called Alaska and that the deceased had previously also
been a member of the gang.

[30] Thirdly, the appellant submitted that one blow does not accord with the
intention to kill. It is a well-established principle that intent cannot be deduced solely
from the number of blows; the evidence regarding the type of instrument employed

from the number of blows; the evidence regarding the type of instrument employed
also provides sufficient grounds for inferring intent. This, the appellant lost sight of

15

when it made this submission. I will revert to this issue of one blow more fully later
in the judgment.

[31] Lastly, the appellant also argued that the fact that he and his brother
voluntarily went to the police station to report the incident is not indicative of an
intention on their part to kill the deceased, but rather an indication that they were not
involved in the attack on the deceased. This is a nother failure by the appellant to
consider the evidence holistically. The evidence before the trial court, which was not
disputed, is that, upon their arrival at the police station, the matter had already been
reported. They also acknowledged this fact. Another piece of evidence is that his
brother testified that they intended to report the incident of the stabbing or assault
on the appellant and then go to the clinic, but the police refused and instead arrested
them. It was then that he heard that the deceased had been stabbed. This is not only
a contradiction of their evidence on the issue of going to the police station, but it also
does not support the appellant’s arguments at all.

[32] Returning to the issue of one blow, this issue demonstrates a lack of proper
reading of the record; unfortunately, this time around, by both the appellant’s legal
representative and the State. A careful examination of the evidence before the trial
court shows that the trial court called Dr Shelendra Lala (Dr Lala), a doctor who
conducted the post -mortem to testify . Although she initially stated there was one
blow, she later admitted that the possibility of two blows could not be excluded. The
changed opinion came after she was made aware of the uncontested evidence
indicating that the blade of a knife was still embedded in the deceased ’s skull. Her
answer to this was that ‘[t]hen you have a possibility of two wounds, that could be a
second blow as well, that could be a second injury’. She also opined that because the

16

wound(s) were sutured, this is an indication that when the deceased arrived at the
hospital, he was still alive.

[33] The possibility of a two-blow injury is in keeping with her description of the
injuries in exhibit ‘A’ that:
‘there was a six-centimetre incision on the left side of the head, which was penetrating. There was
a fracture of the skull on the left frontal parietal which is the front and the side of the skull on the
left hand side. There was a two-centimetre penetrating incision also on the left side of the brain’.
Dr Lala described the two incisions as a ‘longitudinal’ one and a ‘vertical’ one.
Therefore, the possibility of two blows cannot be ruled out or ignored in the holistic
evaluation of the evidence that was before the trial court . The fact that everyone
appears not to have referred to this piece of evidence is of no moment. It is therefore
not justified to only refer to one blow throughout the evidence because the doctor’s
final answer was that she cannot say whether there had been one or two blows. Once
again, th is constitutes another piece of the puzzle that completes the mosaic of
circumstantial evidence from a holistic evaluation, which this Court cannot ignore.

[34] The appellant also sharply criticised the evidence of Mr Riet that he saw the
appellant and the other assailant raise their hands or legs high as if they were dancing
because it was elicited through cross-examination, and not during his evidence in
chief. I am of the view that this issue does not taint the evidence of Mr Riet, which
the trial court found to be reliable. Firstly, whatever he said in cross -examination
about the assault cannot be classified as a contradiction. At best, it can be classified
as an addition to his earlier statement, meaning, it supplemented it, as he said he saw
them surrounding the deceased, but did not explain how they were dancing at that
time. It should be noted that he was being led by the State at that time and responding

time. It should be noted that he was being led by the State at that time and responding
to the questions as they were posed to him. Equally , during cross-examination, he

17

was merely answering questions as posed to him by the appellant’s legal
representative.

[35] Secondly, and more importantly, the averments bemoaned of were made
under oath and when the appellant’s legal representative was testing his evidence. It
is trite that the only statement or averment to a person that carries less weight is one
not made under oath because it lacks the inherent safeguards of truthfulness and
accountability; its weight is diminished.

[36] In Musingadi,8 relied on by the appellant but for the argument relating to
common purpose, this Court, after quoting the remarks made by Marais J, confirmed
this approach to a statement made during cross-examination, albeit dealing with
admissions:
‘Marais J quoted at 425c – e from the judgment of Greenberg JA in Valachia at 837:
“Naturally, the fact that the statement is not made under oath, and is not subject to cross -
examination, detracts very much from the weight to be given to those portions of the statement
favourable to its author as compared with the weight which would be given to them if he had made
them under oath, but he is entitled to have them taken into consideration, to be accepted or rejected
according to the Court’s view of their cogency.”
The upshot is that there is a clear distinction between a statement made under oath
and one that has not, in particular, in respect to the weight to be attached thereto.
The latter lacks the inherent safeguards of truthfulness and a ccountability, and its
weight is accordingly diminished. By contrast, evidence given under oath carries a
solemn assurance of veracity, reinforced by the legal consequences attendant upon
falsehood. It is for this reason that sworn evidence is generally re garded as more
reliable and deserving of greater weight.

8 Musingadi fn 1 above para 44. See S v Cloete 1994 (1) SACR 420 (A) at 425C -E and Rex v Valachia and Another
1945 AD 826 at 837(Valachia).

18


[37] What is more telling on this point is that the record reveals that the interpreter
provided clarification on this aspect. For a full understanding of the clarification, it
is best to start the quote from what transpired before it. The record reads:
‘MR LANGA: So he joined them from where? Coming from where? --- Your Worship he
appear from the street that I have emerged from, the street in that street that is where now the
accused they were busy injuring the deceased. This person appeared down the street on the very
same street.
You see now you are amending your previous story. You did not say you saw them assaulting him
or beating him. You did not say that during that. --- Your Worship I am saying like that because
when I was at house number two I saw accused 1 having in the air as if now he was stabbing like
this motion to the Court. Now presumably that was the time when now they injure him, the
deceased person so that is where now I started to scream.
MNR BURGER: Agbare jammer ek val in die rede en ek wil nie die Tolk sy werk leer nie
maar die manhet iets gesê van ’n mes ook. Iemand aan die einde van sy ... (onduidelik) ... mes ook.
HOF: Ja klaar dit net op.
INTERPRETER: As the Court pleases.
HOF: Kan jy dit net opklaar Mnr Metswi?
INTERPRETER: As the Court pleases Your Worship. --- Your Worship as the Court pleases Your
Worship. As I am saying that they were using those gestures as if they were dancing Your Worship
and then I did not see a knife at that stage like being on air. …(Inaudible) …but now I saw accused
2 holding a knife in his possession.
That is the explanation that we got now from the witness. As the Court pleases.
MR LANGA: That is why I am saying like I put it that you are changing. This is where maybe let
me rather say it about the knife you said after you disturbed them with stones accused 1 was not
carrying anything, accused 2 was carrying a knife, the other guy was carrying a pick handle. They

were coming to you that now you were close. That is the time accused 2 was carrying a knife. Is
that correct, you said that? --- I confirm that Your Worship.
So at the time when they were appearing to be jiving, to be dancing you did not see who was
having the knife, who was carrying it you were far? --- Your Worship I saw he was having a knife
because the visibility or the illumination there was clear. Your Wo rship I saw them as they pick

19

up the deceased person. The accused persons was wearing a knife together with the leg of the
deceased when they pick him and then they throw him down.
COURT: Who is that, which accused are you referring to? --- Accused 2. ’ (Emphasis
added).

[38] To demonstrate that he even stuck to his version when h is evidence was so
tested, the following transpired:
‘MR LANGA: I want you to listen. I am not talking about the time, the time you say they were
carrying the deceased. I am talking about the time when you were saying they were dancing, they
were playing music, their hands were up, people were dancing. --- Yes.
And then you agree that you did not say he was wearing a knife at that stage because he was far.
--- Your Worship I was not far I did see that but I did not explain that I did see, I did see a knife. I
am answering, I am now answering the question that was conducted to me.
Okay you told the Court that you turned at the street, the music was being played in the fifth house?
--- Correct.’ (Emphasis added).

[39] It is significant to re-emphasise that cross-examination is a recognised method
of placing evidence before a court. Statements made by a witness or the accused
under cross-examination are, therefore, part of the evidential material that any court
must assess . If regard is had to the principles emphasised by this Court in S v
Trainor,9 that courts must evaluate evidence holistically rather than in isolation , it
will then be wrong for a court to disregard evidence simply because it arose during
cross-examination. What matters is the weight thereof, which depends on its quality
and reliability after being weighed against a conspectus of all evidence and the
applicable burden of proof. This is what was also recognised by the quote in
Valachia.10 The criticism by the appellant, therefore, does not have merit at all.

9 S v Trainor [2003] 1 All SA 435 (SCA); 2003 (1) SACR 35 (SCA) para 9. The dictum in para 9 was also applied in

S v Kapa [2023] ZACC 1; 2023 (4) BCLR 370 (CC); 2023 (1) SACR 583 (CC) para 104 and S v Ntshongwana [2023]
ZASCA 156; [2024] 1 All SA 345 (SCA); 2024 (2) SACR 443 para 65.
10 Ibid fn 8.

20

[40] The credibility findings of the trial can be addressed briefly. The trial court
rejected the account s provided by the appellant and his brother. The appella nt’s
version in particular, was found to have been fabricated. That of his brother amounts
to no version at all, as it mutated several times. In my view, these findings relate to
factual matters that should not be disturbed by this Court, as there appears to be no
demonstrable or material misdirection regarding the credibility determinations in the
context of the whole evidence that was before the trial court.

[41] Regarding the reliability of their account, the trial court concluded that their
version is fraught with improbabilities. Sight should not be lost that the version of
the appellant was that they were attacked by stones by a mob of about 17 to 20 people
for no apparent reason. He and Lulu went outside twice to confront the aggressive
people. He came out allegedly with only one injury, which he could not prove with
a medical report in court. In addition, they managed to pick up the deceased and take
him out of the yard unscathed when, according to their version, this mob continued
to hurl stones at them. No evidence was presented to suggest that the mob suddenly
stopped and waited for them to remove the deceased from their yard. All of these
factual averments seem highly unlikely. Mr Riet’s account, which states that the first
person to join him was Nombulelo and another person and that many other people
joined later, is the most plausible version.

[42] It is also improbable that h e was so unfortunate on that day , that after being
so attacked without a reason by this mob ; the police also apparently refused (a) to
listen to him and his brother; (b) to give him permission to go for medical attention;
and instead, arrested and assaulted them. It is also improbable that he did not get a
chance to tell the police about Lulu upon arrival at the police station, or at any time

chance to tell the police about Lulu upon arrival at the police station, or at any time
thereafter. Not only is this version improbable, but, on the contrary, it strengthens

21

the version of Mr Riet, which the trial court accepted, that the appellant and his
brother jumped over the fence and ran away when the police arrived at the scene, as
the other people around there had already called the police. This begs the qu estion
as to why they ran away when they saw the police if they were innocent.

[43] In addition to the several improbabilities noted above and those identified by
the trial court in its judgment, their conduct further weakens the reliability of their
version. To cite a few examples: the appellant claimed that he did not want the
deceased in their yard, even though, according to his version, he was not injured by
him but by one of his friends, Lulu. The explanation he gave, namely that he wanted
the ambulance to find him outside the yard is highly improbable, especially in view
of the fact that the deceased was the appellant’s neighbour. He also realised that the
deceased was badly injured, as the knife was still embedded in his skull, threatening
his life. Despite this, he showed no concern and carried the deceased outside, leaving
him helpless in the street instead of calling an ambulance, while at the same time
claiming to have been innocent.

[44] While acknowledging the appellant’s right to remain silent and that he had no
obligation to inform the police, his decision not to disclose the information about
Lulu to the police immediately before he was arrested raises some concerns. A right
to remain silent is a constitutional protection afforded to an arrested person against
self-incrimination during police interrogations and court proceedings. The evidence
about what Lulu did does not incriminate him; rather, it is exculpatory. As to why
he kept quiet until somewhat later during cross-examination, when there were
limited actions the police could take , is problematic and prompts important
questions. This concern is further amplified by his late revelation that Lulu had

questions. This concern is further amplified by his late revelation that Lulu had
passed away. In my view, the trial court was justified in rejecting it, as it clearly

22

appears to be a deliberate and convenient defence, especially because the police
would not have been able to investigate further.

[45] The appellant furthermore submitted that, even if the evidence of Mr Riet that
he saw the a ppellant in possession of a knife can be accepted as circumstantial
evidence, there is more than one inference that could be drawn from his conduct.
This argument does not salvage the appellant’s case. The reason can be found in the
decision of this Court S v Saul s and Others (Sauls),11 wherein the following was
stated when this Court was faced with a similar submission:
‘No doubt many inferences can be drawn from the fact that appellant No 3 stood at the window. It
may be inferred that he was a sleepwalker or that he had got up out of bed to get a breath of fresh
air; all sorts of fanc iful motives for his conduct may be suggested. The State is, however, not
obliged to indulge in conjecture and find an answer to every possible inference which ingenuity
may suggest any more than the Court is called on to seek speculative explanations for conduct
which on the face of it is incriminating. And when the accused misleads the Court by lying,
arguments based on improbable inferences are not calculated to impress a trial Judge. A passage
in a minority judgment given by MALAN JA in R v Mlambo 1957 (4) SA 727 (A) at 738 is
apposite. I may add that two paragraphs in this passage were cited with approval by RUMPFF JA
in S v Rama 1966 (2) SA 395 (A) at 401:
“In my opinion, there is no obligation upon the Crown to close every avenue of escape which may
be said to be open to an accused. It is sufficient for the Crown to produce evidence by means of
which such a high degree of probability is raised that the ordinary reasonable man, after mature
consideration, comes to the conclusion that there exists n o reasonable doubt that an accused has
committed the crime charged. He must, in other words, be morally certain of the guilt of the
accused.

accused.
An accused’s claim to the benefit of a doubt when it may be said to exist must not be derived from
speculation but must rest upon a reasonable and solid foundation created either by positive
evidence or gathered from reasonable inferences which are not in conflict with, or outweighed by,
the proved facts of the case.

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

23

Moreover, if an accused deliberately takes the risk of giving false evidence in the hope of being
convicted of a less serious crime or even, perchance, escaping conviction altogether and his
evidence is declared to be false and irreconcilable with the proved facts a court will, in suitable
cases, be fully j ustified in rejecting an argument that, notwithstanding that the accused did not
avail himself of the opportunity to mitigate the gravity of the offence, he should nevertheless
receive the same benefits as if he had done so”.’12

[46] Borrowing from the principles espoused in Sauls, many inferences can be
drawn from the fact that he was seen in possession of a knife. To cite a few: the fact
that the appellant was just in possession of a knife and did not use it ; the appellant
delivered the first blow that caused the first incision and the second blow relating to
the second incision was delivered most possibly by his brother, almost on the same
spot that is why this blow went deeper and got stuck; the knife probably broke when
the deceased’s head touched the ground and hence his brother was not armed at all
after the deceased fell; including that the two incisions were caused by the blows
delivered solely by his brother. Of particular importance is that all of these inferences
stem from the proven facts of the version accepted by the trial court . Thus, the
appellant’s version that Lulu stabbed the deceased cannot be counted in the
inferences to be drawn in his favour because the trial court rejected it . The same
applies to the inference that he was merely in possession of a knife, as he denied any
such possession.

[47] Secondly, no inference can be drawn that the third unidentified person could
have stabbed the deceased, as, according to the evidence the trial court accepted, he
was seen in possession of a pick-handle. Even if an effort may be made to interpret
the appellant’s version charitably, it cannot be inferred or concluded that this

the appellant’s version charitably, it cannot be inferred or concluded that this

12 Ibid fn 12 at 182G-H and 183A-C. The dictum was also applied in Sentrachem Bpk v Wenhold 1995 (4) SA 312
(A); [1995] 2 All SA 524 (A) at 327E-F.

24

unidentified individual was Lulu, as the appellant state d that Lulu used a knife to
stab the deceased.

[48] The principles in Sauls are applicable on the facts of this matter if regard is
taken of the fact that there is incriminating evidence showing that the appellant was
seen surrounding the deceased, raising his hand and leg shortly before the deceased
fell, and was in possession of a knife when he picked up and carried the deceased
outside his yard . The incriminating evidence is strengthened by the evidence of
Dr Lala that the possibility of two blows cannot be excluded . The fact that the trial
court found that the appellant had fabricated his story and deliberately given false
evidence to exonerate him self and his brother tilts the scale s in favour of the
incriminating evidence. The principles laid down in S v Rama13, which were relied
upon in Sauls are still binding. If the appellant had an innocent explanation about his
possession of a knife at that awkward time when the deceased was stabbed by his
friend Lulu, as alleged by him, he should have given it at his earliest opportunity
instead of placing the cause of the deceased’s death on a person who had passed
away. He also does not have an explanation as to what happened to the knife his
friend Lulu used to stab the deceased , despite having observed the stabbing . The
coincidence of the stabbing of the deceased, who was an eyewitness in a case he was
charged with, occurring in his yard, when he was seen in possession of a knife and
having dropped the deceased outside his yard , leaving him fighting for his life, is
more telling but also, remains unexplained.

[49] In this matter, there is sufficient evidence of high probability that the ordinary
reasonable man, after mature consideration, can come to the conclusion that there
exists no reasonable doubt that the appellant and his brother committed the crime

13 S v Rama 1966 (2) SA 395 (A).

25

they were charged with. It is trite that where there are two mutually destructive
versions, both cannot be true. Only one can. If the appellant’s version is rejected as
false, as it has to be, then any conviction would, of necessity, have to rest solely on
the State’s case.14 Similarly, as it was held in Sauls, there was, therefore, no
obligation upon the State to close every avenue of escape which may be said to be
open to the appellant. The proven facts are so overwhelming that they all point to
the logical conclusion that the appellant also stabbed the deceased.

[50] What then could be the proven facts from which the inference can be drawn
that the appellant stabbed the deceased? It is not hard to fathom that the trial court
based its decision on the following proven facts, which were derived from the
common cause facts and the circumstantial evidence it accepted:
(a) the appellant and his brother were identified by Mr Riet as two of the three people
surrounding the deceased when he fell to the ground;
(b) at the time they surrounded the deceased, they were raising their hands and or
legs high as if they were dancing;
(c) after the deceased fell to the ground, they picked him up and carried him to the
outside of the yard, where they dropped him;
(d) the appellant was seen by Mr Riet holding a knife in his hand when he picked up
the deceased and carried him, and after dropping him outside;
(e) there was enough light for Mr Riet to be able to observe what was happening;
(f) Mr Riet was at a distance of 8 metres when he saw the appellant and the other
people surrounding and carrying the body of the deceased;
(g) the deceased was still alive at that time;
(h) Mr Riet did not see the actual stabbing;

14 Doorewaard and Another v S 2021(1) SACR 235 (SCA) para 116.

26

(i) a blade of a knife was still stuck and protruding from the deceased’s head, and
the handle appeared to have broken off;
(j) two sutured incisions that looked like one stab wound were observed by the doctor
who performed the post -mortem on the deceased body; one was vertical and the
other horizontal;
(k) the possibility of two separate blows causing one wound with two incisions could
not be ruled out by the doctor;
(l) the decease d was an eyewitness in the case the appellant and his brother were
charged with, which was, at the time of his death, still to be tried; and
(m) the appellant and his brother were subsequently acquitted in terms of s 174 of
the CPA.
I therefore found no fault in the reasoning of the trial court regarding the fact that
the State proved beyond a reasonable doubt that the appellant and his brother killed
the deceased.

The doctrine of common purpose
[51] Having reached the conclusion above, I now turn to deal with the doctrine of
common purpose, which is the mainstay of the appeal. The appellant relied heavily
on this doctrine to support its argument that the appeal should succeed . It was
submitted that the full court did rely on this doctrine to confirm the trial court’s
decision, and that this was a misdirection because no mention of the doctrine was
made in the charge sheet. The appellant referred, correctly so, to several judgments
of this Court, where it was decided that it would be unfair to the accused and could
not be allowed. While there may be a risk of repetition, it is essential to reiterate that
the trial court did not convict the appellant based on this doctrine. Clearly, the full
court misconstrued the trial court’s judgment and some of its findings . It did not

27

fully appreciate that the trial court’s judgment was based purely on circumstantial
evidence relating to the murder itself, including the motive to kill.

[52] Nowhere in the judgment can a reference to common purpose be deduced.
The trial court appears to have been alive to the fact that it was not mentioned in the
charge sheet. Unlike the full court, the trial court never used the sentiments found in
the two judgments (conviction and sentence) of the full court to wit: ‘ appellant never
disassociated himself from the conduct of Namhla and the third person’; ‘there was no
disassociation of the appellant from what was happening to the deceased’; ‘Chances are that it was
not the appellant that actually landed the blow to the head of the deceased that caused the blade of
the knife to stick in the head of the deceased because he was still in possession of his knife’.

[53] The bulk of these remarks made by the full court were made during
sentencing. The lack of pro per comprehension of the import of the trial court’s
judgment is clearly depicted where the full court remarked that ‘the absence of a
clear and satisfactory evidence that it was indeed the appellant who landed the fatal
stab wound in the head of the deceased to the extent that the blade broke off from
the handle is in my view, compelling enough to warrant a deviation fro m imposing
life imprisonment as a sentence’. This is in contrast with its finding (the full court)
that the murder fell within the purview of s 51(1) read with Part 1 of Schedule 2 of
the CLA Act as a result of the motive to kill that was found by the trial court, which
the full court endorsed.

[54] Clearly, as it will be shown in this paragraph, when a further analysis of the
doctrine of common purpose is made with specific reference to the judgment of the
trial court, the full court ’s reasoning, although it confirmed the verdict of the trial
court, is flawed. Fortunately, on the facts of this case, this flawed reasoning does not

28

vitiate the verdict of the tria l court, as it is trite that the appeal lies in the court’s
verdict or substantive order, not in its reasoning.

[55] Having accepted the version of the single witness as credible, satisfactory, and
reliable, the trial court embarked on an inferential inquiry and inferred the intention
of the appellant and his brother from their direct participation on the scene and their
actual actions or conduct. This posits a clear positive actus reus that is not dependent
on an individual’s association or dissociation. The liability of the action of the
appellant was not imputed by the trial court from the action of the unidentified
perpetrator or his brother. It was based purely on circumstantial evidence.

[56] The law is trite that the doctrine of common purpose is used to establish
whether the actions of a participant(s) can be imputed onto the others to hold them
liable. When applied, the conduct of each of them is impute d to the others , and
liability arises from their common purpose to commit a crime . What is key is that
the trial court did not rely on this doctrine. It just inferred the intention of the
appellant from the circumstantial facts adduced. The intention the trial court inferred
is also not in the form of dolus eventualis or dolus indirectus, but a clear, direct
intention to cause the death of the deceased. Thus, an accused person can be
convicted of murder on circumstantial evidence without resorting to the doctrine of
common purpose. The doctrine does not create a separate offence but addresses the
criminal liability of perpetrators and, in addition, may be relevant to sentencing.
Fortunately, this Court is not seized with the appeal on sentence.

Conclusion
[57] We now know that the doctor did not rule out the possibility of a second blow.
Whether the first blow was delivered by the appellant’s brother is of no moment, as

29

the possibility that the appellant also stabbed the deceased cannot be excluded in
light of the cumulative weight of the proven facts against him. There is, after all, no
explanation for the possible second blow or incision described by the doctor, or at
the very least, for the stab wound described as consisting of two incisions in the same
area, one horizontal and the other vertical. The most compelling inference is that the
appellant also stabbed the deceased and that his brother thereafter delivered the
second blow, causing the knife to become lodged in the deceased’s skull, which
explains why t he appellant was seen without a knife. In my view, the trial court
correctly inferred that the appellant and his brother intentionally killed the deceased.

[58] I conclude by answering the questions that were raised by the State as issues
in this case:
(a) Issue 1: Could the appellant be convicted on the evidence of a single witness?
Indeed, the trial court was alive to this; it made strong credibility findings in favour
of the single witness and also, against the appellant. There was no demonstrable or
material misdirection on its factual findings ex facie the record, and they cannot be
disturbed.
(b) Issue 2: Could the appel lant be convicted on the doctrine of common purpose
despite the State not averring it in the charge sheet? Yes, the trial court did not rely
on it at all. To the extent that the full court appears to have approached the matter in
that manner, it erred, but this is not material to disturb the conviction of the trial
court.
(c) Issue 3: Did the appellant cause the death of the deceased? Yes, the only logical
inference that can be drawn from the proven facts is that he also stabbed the
deceased. It is immaterial whether his blow was the fatal one or not.

30

(d) Issue 4: In the event of circumstantial evidence, could the only inference be that
the appellant was the one who stabbed the deceased? The court having rejected the
version of Lulu being the one who stabbed the deceased, the evidence of the State
remains of a high degree of probability that the only reasonable conclusion that can
be made is that he also stabbed the deceased.

[59] Therefore, the trial court arrived at the correct decision in finding the appellant
guilty of murder , read with the provision s of s 51(1) of the minimum sentence
regime, given that a state witness was killed. The misdirection committed by the full
court is not material to vitiate the findings of the trial court and its order on the
conviction of the appellant. Another reason is that the full court’s substantive order
did not change the trial court’s substantive order. The same is true for the conclusion
reached by this Court. Thus, whether or not a court of appeal agrees with a lower
court’s reasoning would be of no consequence if the result would remain the same.15
This is the case in this appeal.

The second judgment
[60] The second judgment places great emphasis on the full court ’s judgment ,
which referred to the doctrine of common purpose. On the same breath, it accepted
the fact that the trial court never mentioned the doctrine, let alone relied on it, in
convicting the appellant and his brother. The two analyses cannot coexist, as they
create a dichotomy. It is obvious that the two judgments differ. The difficulty I find
is that the second judgment does not explain how the doctrine was imported by the
full bench judgment into the trial court’s judgment. The interpretation exercise
undertaken by the second j udgment on the full court’s decision is not justified

15 Masstores (Pty) Limited v Pick n Pay Retailers (Pty) Limited (CCT242/15) [2016] ZACC 42; 2017 (1) SA 613
(CC); 2017 (2) BCLR 152 (CC); [2017] 1 CPLR 1 (CC) para 59 . See also Tecmed Africa v The Minister of Health

[2012] ZASCA 64 (21 May 2012).

31

because the full court accepted the credibility and factual findings of the trial court.
With the risk of repetition, an appeal lies against the substantive order of the full
court and, by necessary implication, to that of the trial court. This simply means that
the trial court’s judgment and order cannot be simply ignored, including the record
of the proceedings. In para 77 the second judgment also stated:
‘Having so concluded, the first judgment goes on to hold that, in any event, all the full bench’s
remarks concerning common purpose ‘were made during sentencing’. The record does not support
that proposition.’ [ Emphasis added]
This is yet another misinterpretation of what I stated in paragraph 53 above. The
word ‘bulk’ was used intentionally, not to mean ‘all,’ but to refer to ‘the largest or
greatest portion of something.’

[61] The other difficulty that stems from this acceptance of the second judgment is
this: on what basis did the trial court convict the appellant and his brother if it did
not rely on the doctrine ? The ineluctable logical conclusion is that it relied on
circumstantial evidence. It inferred the stabbing of the deceased from the direct acts
of the appellant and his brother; borne out of the evidence it accepted of Mr Riet. It
is worth emphasising that sight should not be lost that the appellant’s evidence that
Lulu stabbed the deceased was found to be a fabrication and is not one of the proven
facts. In Doorewaard16, this Court reaffirmed the approach to circumstantial
evidence as espoused in Sauls that speculative or conjectural hypotheses, which are
not logical conclusions arising from the established facts , cannot pre vail over a
reasonable inference consistent with the proven facts.

[62] The second judgment finds that it was clear from Mr Riet’s evidence that the
fatal wound had already been inflicted when the appellant was seen with a knife. A

16 Ibid fn 15 para 126-127.

32

correct summary of the sequence of the evidence of Mr Riet is important in this
regard. He testi fied that he just saw the appellant, his brother , and a third person
surrounding the deceased, raising their hands and legs as if they were dancing, and
then the deceased fell to the ground. They picked him up and carried him to the
street. He saw the appe llant having a knife when he was picked up and whilst so
carrying the deceased, even after the deceased was left in the street.

[63] According to Mr Riet , t he events unfolded in a continuous manner that
followed each other. Contrary to the evidence that was rejected by the trial court,
correctly so in my view, there was no point where the appellant was seen by Mr Riet
going back into the house, which could have created an opportunity for him to fetch
it from the house at that time. The only logical conclusion is that he was having it
all along when he did those dancing-like movements and when the deceased fell.

[64] Lastly, the second judgment’s further criticism relates to the finding I made
that there was sufficient circumstantial evidence to conclude that the appellant also
stabbed the deceased. It reasoned that ‘there is no direct evidence that the appellant
did so’, and ‘he inflicted one of the two stab wounds’, further that ‘that was not the
case the appellant had to meet’. The record of proceedings is crucial to the proper
analysis of the judgment appealed against ; hence, there is a recognised rule that
specifically provides that it must be filed with an appeal. The second judgment seems
to suggest that th is Court should disregard the doctor’s evidence because no one ,
including the trial court, referred to it. This cannot be. It is evidence before the trial
court that must also be evaluated . It is trite that in our law, courts rely on
circumstantial evidence to prove the guilt of an accused person, mainly because there
is no direct evidence. The same applies when the court relies on the doctrine of

is no direct evidence. The same applies when the court relies on the doctrine of
common purpose. I dealt with what the doctrine of common purpose is and ho w it

33

operates and there is no need to repeat. What is left is to indicate that circumstantial
evidence, on the other hand, is information pointing towards a conclusion without
direct proof. The Oxford dictionary defines the word circumstantial as ‘an adjective
describing information or evidence that strongly suggests, but does not definitively
prove, that something is true’. Simply put, it means indirect evidence.17

Order
[65] If I commanded the majority, I would have made the following order
The appeal against the conviction of the appellant is dismissed.


_______________________
A M KGOELE
JUDGE OF APPEAL



Coppin JA (Keightley JA and Basson AJA concurring):

[66] I have read the judgment of Kgoele JA (the first judgment) and, respectfully,
do not agree with its reasoning and conclusion, thus this dissenting judgment. In
terms of the first judgment, the appellant was correctly convicted of the murder of
the deceased, because it can be inferred that he had inflicted one of possibly two stab
wounds on him. There is no evidence to support this conclusion; and that was not
the case the State made against the appellant, or which the appellant was called on

17The Shorter Oxford English Dictionary 7 ed (2007) defines ‘circumstantial’ as:
‘depending on, or derived from, circumstances; indirect; not directly proving a fact. ’

34

to defend. Further, neither the trial court, nor the full bench made such a finding.
While the first judgment places great reliance on the salient principle that an appeal
court will not lightly interfere wit h a trial court’s factual findings, it overstates or
misconceives the principle as one of deference, which it is not. The factual findings
of a trial court are not cast in stone. Those findings may be interfered with where
there is a demonstrable misdirect ion, or where they are clearly wrong 18, as in this
instance.

[67] The main issue before the full bench of the North West Division of the High
Court in Mahikeng (the full bench), hearing the appeal from the regional court (the
trial court), was whether the appellant was correctly convicted of murder in
circumstances where there was no direct evidence that the appellant administered
the fatal stab wound to the head of the deceased. It effectively found that, even in
the absence of evidence of the appellant having done the stabbing, he was guilty of
the charge because of common purpose. It found that even though the State had
never invoked or relied on the principle of common purpose in the charge sheet or
in the trial court.

[68] The appellant (as accused 2) and his brother (accused 1) were arraigned in the
trial court on a charge of murdering the deceased by stabbing him with a knife. Both
were convicted on the charge and sentenced to life imprisonment. They appealed
against their respective convictions and sentences. In the interim accused 1 died. The
appellant noted an appeal to the full bench in respect of his conviction and sentence
for murder. On 28 February 2023, the full bench confirmed his conviction but set

18 Bernert v ABSA Bank Ltd [2010] ZACC 28; 2011 (3) SA 92 (CC); 2011 (4) BCLR 329 (CC) para 105-106; See also
President of the Republic of South Africa and Others v South African Rugby Football Union and Others [1999] ZACC

11 (CC); 2000 (1) SA 1 (CC); 1999 (10) BCLR 1059 (CC) para 58-65 and 78-80:Mashongwa v Passenger Rail Agency
of South Africa 2016 (3) SA 528 (CC) para 45; [2015] ZACC 36; 2016 (2) BCLR 204 (CC); 2016 (3) SA 528 (CC);
Rex v Dhlumayo and Another 1948 (2) SA 677 (A) at 705-706.

35

aside the life sentence that had been imposed on him and replaced it with a sentence
of 18 years' imprisonment. On 9 September 2023, this Court granted the appellant
special leave to appeal against both his conviction and sentence. The appellant
elected to only pursue the appeal in respect of his conviction.

[69] It is not disputed that at no stage in the trial court did the State indicate a
reliance on or invoke the doctrine of common purpose. The charge sheet never
referred to it, nor did the trial court. In convicting the appellant, the trial court
accepted that the State witness (Mr Riet) never saw who stabbed the deceased.
Without finding who could have done so, the trial court nevertheless found both the
appellant and accused 1 guilty on a basis that is not articulated in its judgment. It
found that ‘it is cle ar when looking at the actions of accused 1 and accused 2 that
that evening in question together with a third person, that the intention was to kill
the deceased.’ It is trite that mens rea in the form of intention is but one of the
elements of the crime of murder. There are also other elements that had to be proved
before the appellant could properly have been found guilty of the crime, notably the
actus reus, or unlawful act (or conduct) on the part of the appellant, as well as
causation.

[70] Even though the full bench confirmed the appellant’s conviction by the trial
court, the basis on which it did so is tenuous. The full bench confirmed the following
concerning the evidence regarding the stabbing: ‘The evidence of Mr Riet is to the
effect that you can not say with certainty who amongst the three men stabbed the
deceased. The appellant still had his knife in his hand. The deceased was stabbed in
his head which caused trauma to his brain. The knife that was used to stab the
deceased broke and the blade wa s stuck in the head of the deceased. Chances are
that it was not the appellant that actually landed the blow to the head of the deceased

36

that caused the blade of the knife to stick in the head of the deceased because he was
still in possession of his knife.’

[71] However, despite this finding, elsewhere in its judgment the full bench found
that the appellant was guilty of the charge. The only basis on which it could have
reached this conclusion was common purpose. It specifically found so, as is apparent
from its judgment. After having referred to the appellant’s version which, inter alia,
was that another person, one ‘Lulu’, had stabbed the deceased with a knife, the full
bench stated: ‘Mr Riet did not see the actual stabbing blow that caused the death of
the deceased. However, on the evidence of Mr Riet, although he is a single witness,
there was no disassociation of the appellant from what was happening to the
deceased. As already alluded to, the appellant was armed with a knife and he helped
to carry the deceased. He also he also added words to the effect that the deceased is
still alive and did not die yet.’ (Emphasis added.)

[72] The full bench then went on to explain what it meant by referring to this
Court’s decision in Musingadi,19 which deals with common purpose and the
relevance to that inquiry of whether, by his conduct, an accused dissociated himself
from the common criminal purpose. Having found that: first, there was no direct
evidence of the appellant delivering the fatal blow; second, it was unlikely that the
appellant had delivered that blow; and, third, the full bench’s reference to, and
application of the principles stated in Musingadi,20 the only sensible interpretation
of the full bench’s judgment is that it found that the appellant was g uilty of murder
on the basis of common purpose.


19 Op cit fn 1 para 34-35.
20 Ibid.

37

[73] One of the difficulties with the full bench’s decision is that at no stage did the
State in any way invoke the doctrine of common purpose. The trial court also never
mentioned, let alone relied, on the doctrine in convicting the appellant and his co -
accused. This Court in S v Msimango21 held that an accused may only be convicted
on the basis of that doctrine if the State makes known its reliance on that doctrine in
the charge sheet at the outset. Or, alt ernatively, where the charge sheet is amended
in terms of s 86 of the CPA, to indicate such reliance, or where the material elements
establishing common purpose are properly established through the evidence at the
trial. Accordingly, a conviction on the ba sis of common purpose, where none of
those requirements are met, cannot stand because it constitutes an infringement of
an accused’s right to be informed of the charge with sufficient detail to answer it,
and may prejudice an accused in the conduct of his or her defence.

[74] A further difficulty with the full bench’s judgment is that before there can be
an enquiry into whether an accused disassociated himself with the common criminal
purpose, there must first be a finding that there was association with t he purpose,
based on fact. If the State relies on ‘active association’ to establish common purpose
to prove the guilt of an accused for murder, it is required to prove beyond a
reasonable doubt that the accused was:
‘… present at the scene where the violen ce has been committed; must have been aware of the
assault on the victim; must have intended to make common cause with those who were actually
perpetrating the assault. Secondly, he must have manifested his sharing of a common purpose with
the perpetrators of the assault, by himself performing some act of association with the conduct of
the others. Thirdly, he must have intended to make common cause with those who were actually

the others. Thirdly, he must have intended to make common cause with those who were actually
perpetrating the assault. Fourthly, he must have manifested his sharing of a common purpose with
the perpetrators of the assault by himself performing some act of associatio n with the conduct of
the others. Fifthly, he must have had the requisite mens rea; so, in respect of the killing of the

21 S v Msimango [2017] ZASCA 181; 2018 (1) SACR 276 (SCA).

38

deceased, he must have intended them to be killed, or he must have foreseen the possibility of their
being killed and performed his own act of association with recklessness as to whether or not death
was to ensue.’22

[75] The facts that the appellant was allegedly seen carrying a knife, assisted in
carrying the deceased, and uttered words to the effect that the deceased was still
alive, are insufficient, without more, to establish active association. Significantly, it
is clear from Mr Riet’s evidence that the fatal wound had already been inflicted when
the appellant was seen with a knife, helping to carry the deceased and was heard to
utter that the deceased was not dead. Mr Riet did not see the fatal blow being
inflicted. This must have been inflicted before he saw the appellant. On his version,
he was also not able to testify as to the events that led to the stabbing and how it
transpired. It was never put to the appellant that even though he never administered
the single, fatal stab wound, he had associated himself with the actions of the person
who did, or what those actions were. Instead, the State was content with insisting
that he ha d administered the single, fatal stab wound. In the circumstances, the
appellant’s conviction could not be confirmed based on common purpose.

[76] The first judgment finds in effect that there was sufficient circumstantial
evidence to conclude that the ap pellant did stab the deceased. There is no direct
evidence that the appellant did so. Nevertheless, the first judgment finds that the
pathologist, Dr Shelendra Lala, (Dr Lala) admitted the possibility of two blows, or
two stab wounds having been inflicted on the deceased’s head. The first judgment
proceeds from the premise, based on a misconception of the principle regarding
factual findings by the trial court, that this evidence must also feature in the mosaic

22 S v Thebus and Another [2003] ZACC 12; 2003 (6) SA 505 (CC); 2003 (10) BCLR 1100 (CC); 2003 (2) SACR 139

(CC) para 20. Confirming what was held in, inter alia, S v Mgedezi and Others [1988] ZASCA 135; 1989 (1) SA 687
(A).

39

of evidence that has to be considered in drawi ng an inference of guilt. Relying on
the evidence of Mr Riet, that he saw the appellant in possession of a knife, and
having referred to what this Court held in S v Sauls and Others 23 regarding the
drawing of inferences, the first judgment concludes the following:
‘… many inferences can be drawn from the fact that he was seen in possession of a knife. To cite
a few, the fact that the appellant was just in possession of a knife and did not use it; the appellant
delivered the first blow that caused the first incision and the second blow relating to the second
incision was delivered most possibly by his brother, almost on the same spot that is why; this blow
went deeper and got stuck; the knife probably broke when the deceased’s head touched the ground
and hence his brother was not armed at all after the deceased fell; including that the two incisions
were caused by the blows delivered solely by his brother. Of particular importance is that all these
inferences stem from the version accepted by the trial court. The appellant’s version that Lulu
stabbed the deceased cannot be counted in the inferences to be drawn because the trial court
rejected it.’

[77] Against that background the first judgment finds that the trial court correctly
held that the appellant stabbed the deceased. The judgment then goes on to criticise
the full bench’s judgment saying that it misconstrued the trial court’s judgment and
some of its findings. According to the first judgment, the full bench did not
appreciate that the tr ial court’s judgment was based on circumstantial evidence
relating to the murder, and the motive to kill. Having so concluded, the first judgment
goes on to hold that, in any event, all the full bench’s remarks concerning common
purpose ‘were made during s entencing’. The record does not support that
proposition.

[78] The inferences drawn by the first judgment cannot be permissibly drawn.

proposition.

[78] The inferences drawn by the first judgment cannot be permissibly drawn.
They are based on conjecture, and supposition, and not even on a true probability
that there were two separate blows or two distinct stab wounds inflicted. Dr Lala’s

23 Op cit fn 12 at 182G-183C.

40

supposed concession of a ‘possibility’ of two distinct wounds or blows cannot be
accorded any weight for the reasons I shall shortly deal with. Further, the first
judgment accords too much credit to the trial court. That court did not expressly find
that the appellant stabbed the deceased. It also did not find that there were two stab
wounds or two blows, and that the appellant had inflicted one of them. In fact, the
trial court made no finding in that regard. I nstead, it was pre -occupied with the
question of intention. The full bench did not misconstrue the trial courts findings. It
was acutely aware of the yawning lacuna in that court’s reasoning and conclusion.
It realised that there was no basis for finding t hat there were two stab wounds, or
that the appellant inflicted one of them. It, therefore, purported to invoke the doctrine
of common purpose, albeit not very explicitly, to justify the conviction of the
appellant for murder.

[79] Dr Lala who gave evidence, after both the State and defence had closed their
cases, and according to the record, who was probably called by the magistrate,
testified that according to her post -mortem findings there are ‘two injuries, two
incisions.’ With reference to a photograph, she pointed out ‘a longitudinal incision’
and the one above it as ‘a vertical incision.’ She testified that these wounds had been
sutured and were ‘characteristic of injuries made by sharp objects.’ Questions were
put to Dr Lala concerning the tw o incisions. The question(s) and answer(s)
proceeded as follows:
‘… the court just want[s] to enquire there, you indicated two injuries on the brain, two incisions.
So would that be two separate stabbing or motions that the deceased had been struck by shar p
objects.
[Dr Lala] if I look at the photograph it looks like one laceration but with two aspects to that injury.
The one is the longitudinal one and the other in the horizontal one. So, to me it looks like a single
injury.

41

[The court] Single injury. Would it be possible for a person standing facing the deceased and the
deceased in a standing position to stab the deceased and for the deceased to sustain that injury?
[Dr Lala] Yes, it is possible.’

[80] The prosecutor then questioned Dr Lala. He wanted to know if the long, six-
centimetre wound was where the blade penetrated the skull, and what caused the
‘second wound’. Dr Lala responded as follows:
‘… you know it is difficult because when [I] examined I did not find any blade in the brain so I
did not hav e that information. This looks like, ja it looks like a single wound. My assumption
would be and based on my interpretation would be before you told me this information was that
there was an initial stab and with the hand motion it is usually a stab and a drag down. And my
assumption is that the bottom long six centimetre would be the single wound and the original
penetration would be two centimetres. That is what I assume it to be. You know if a blade, a piece
of the blade breaks there is still, it depends on where it breaks. If it is still, there is still blade
attached to the knife it can cause the longitudinal injury as it comes out, with the piece stuck inside
the brain. So that can be a possibility but when I did the post mortem I did not find any residual.’

[81] This exchange followed:
‘…[Prosecutor] so if we include the evidence that a piece of blade stayed within the skull can it
then be a possibility of two wounds as well?
[Dr Lala] then you have [a] possibility of two wounds, that could be a secon d blow as well, that
could be a second injury. It is also consistent with depending like I said if the knife, if only the tip
of the knife broke off and there is still knife left on the handle that piece is also still sharp. That
can also cause this becaus e your normal action on a right handed person which I assume this is,
would be, [a] roundabout stab and drag down. That is usually how people, it is a stab and a recoil.

So my assumption is that [the] recoil would be that six centimetre thing. And the initial penetration
would be the initial blow.’

[82] After the magistrate questioned Dr Lala, the appellant’s legal representative,
Mr Burger, questioned her. After establishing from Dr Lala that there had been a

42

surgical intervention in which the blade had b een removed and the wound sutured,
the question and response was as follows:
‘[Mr Burger] Doctor there was some indication that it might have been two blows. But if it was
two you must have found at least indication on the skull of the second blow as well. A mark on
the skull, on the bone. The flesh at the side of the bone. You must have found something up until
the bone is it not?
[Dr Lala] Okay blows vary in the intensity when they struck the skull depending on the force they
[indistinct]. Depending on th e force and the person’s or the victim’s skin response you do not
always find wounds on the skin but you can still get trauma inside the skull’s contents of the brain.
Ok most of the time we do not even find any bruising or lacerations but we find quite a bit of
trauma in the brain. You do not have to have a wound on the skin for you to have trauma inside
the brain, that is the first thing. Second thing is these injuries like I said my assumption is that it is
a single wound but it is also possible to be a second blow. It is very difficult for me to tell you
whether it is two blows or a single blow. Like I said my assumption based on this here would be
that it is a single blow based on what I described earlier. But it is also possible to be a two blow
injury.’

[83] Mr Burger asked: ‘obviously a much better indication, much, just looking at
the photos but a better indication if we get the records of his treatment.’ Dr Lala
answers (tellingly); ‘[Y]es, no definitely’. There should be a chain of medical
record.’ After Dr Lala’s evidence the magistrate enquired from the prosecutor
whether the State wanted to re -open its case in light of that evidence. The same
request was extended to the legal representatives of the accused. The record reflects
that all of them, except for Mr Burger, gave an unequivocal, negative answer.
Mr Burger indicated that he did not intend requesting the reopening of the appellants

Mr Burger indicated that he did not intend requesting the reopening of the appellants
case but insisted that it was necessary for the chain of the deceased treatment to be
placed before the trial court. According to Mr Burger it was necessary because ‘…we
do not know what happened here. We do not know whether it is a single or two
blows.’

43

[84] Seemingly, the State was happy with the uncertainty created by Dr Lala’s
speculative opinion based on the incomplete chain of evidence. Maybe it knew that
the medical record might paint a completely different picture that was not favourable
to its case. That the history of the treatment was crucial, cannot be overstated.
Dr Lala’s opinion on that aspect was s peculative, and her views were not based on
fact. She had no insight into the deceased’s treatment history following the stabbing.
Surgery also involves the use of sharp instruments. To assign any significant weight
to Dr Lala’s alleged concession in those circumstances, would be irregular and
unjust.

[85] The more serious concern with this aspect of the case in the trial court is that
the magistrate seems to have overlooked his duty to see to it that justice was done,
and in particular concerning the exer cise of its discretion in terms of s186 of the
CPA. The section empowers a court, at any stage in criminal proceedings, to
subpoena or cause any person to be subpoenaed if such a person’s evidence appears
to the court to be essential to the just decision of the case. After both the prosecution
and defence indicated their reluctance to reopen their cases, or to call evidence
regarding the surgical treatment following the stabbing, the trial court should have
exercised that power. The trial court seems to have exercised it when calling Dr Lala,
and nothing seems to have prevented it from doing so regarding the chain of the
medical treatment. If the possibility of two blows was relevant to the question of the
appellant’s guilt, the trial court ought to have ens ured the necessary evidence was
adduced. Worryingly, instead of exercising that power, the magistrate seems to have
‘declined to pursue the issue further’ concerning that aspect of the matter. The
appellant bore no onus in that regard. And there was a failure of justice. There would

appellant bore no onus in that regard. And there was a failure of justice. There would
be a further failure of justice if this Court were to dismiss the appeal by relying on
the unverified conclusion that it was possible that two blows were inflicted and that

44

the appellant delivered one of them with the knife in h is hand. It is concerning that
the first judgment appears to rely on and to use the fruit of the trial court’s failure of
justice against the appellant.

[86] Like the trial court and the full bench, the first judgment assumes the
reliability and veracity of Mr Riet’s evidence. And prefers his version instead of that
of the appellant, because amongst others, it regards the factual findings of the trial
court as cast in stone and immutable. For example, it accepts Mr Riet’s evidence
that he saw the appellan t having a knife and rejects the appellant’s version that he
never had a knife. Mr Riet’s evidence was uncritically accepted without considering
very crucial aspects. In many respects his evidence was unsatisfactory. It is almost
impossible to glean from t he record of Mr Riet’s testimony exactly what he saw,
when and where. As he was a single witness, in order to be accepted his evidence
should have been clear in all material respects. It was not. The only aspect that was
clear from his evidence was that he did not see the appellant, or anyone for that
matter, stabbing the deceased.

[87] Moreover, Mr Riet was not an impartial, unbiased and independent witness to
the events surrounding the deceased’s death or injury. According to his own
testimony, he was a member of a gang, called Alaska (the gang), for fifteen years
and the members of that gang continued to be his friends. The gang was involved in
nefarious activities, and the evidence clearly shows that members of the gang went
to the appellant’s house, wh ere the appellant was socializing with accused 1 and
friends, to accost them. Because of his relationship to the gang, Mr Riet was clearly
not above fabricating bits of evidence to suit a narrative implicating the appellant in
the killing of the deceased. This is consistent with the lack of clarity in his testimony.

45

[88] The deceased was also a member of the gang. His presence at the house of the
appellant is not explained. Further, Mr Riet’s version as to how he and his other
friends from the gang got th ere, if closely analysed, is most improbable. On
Mr Riet’s version, he, his friends and the deceased were at the tavern. When they
left, they were walking separately, but in a direction opposite to where the
appellant’s house was. Mr Riet was walking alone, and his friends were four streets
ahead of him. He was accosted by a person with a knife who was looking for a
member of the gang. Mr Riet responded by picking up stones. When this person left
him, he told his friends that they should follow him because the person was looking
for a gang member. He did not explain how he had contacted these friends, given
that they were four streets away from him. His version was that they followed the
person, but he went ahead of the rest. Mr Riet did not know where this person was
going to, and because he did not want the person to see that he was following him,
Mr Riet took a different route to that person. By happenstance, Mr Riet got to, what
turned out to be the appellant’s, house. According to Mr Riet, he heard music and
saw people dancing and, perceiving that these people were dangerous, he stood still
to watch what was happening. That is where he saw a person, whom he identified as
the deceased, fall and being carried from the yard and put in the street. This version
is improbable. It is more probable that Mr Riet and the other gang members,
including the deceased, went to the appellant’s house to attack the appellant and
those that were there.

[89] In his evidence in chief, Mr Riet testified to the effect that it is at the stage
when he saw the deceased being carried out of the appellant’s yard that he saw the
appellant with a knife in his hand. The appellant’s co-accused, according to Mr Riet,
was not armed, but the third person had a pick handle. Mr Riet later ch anged his

was not armed, but the third person had a pick handle. Mr Riet later ch anged his
testimony, saying that he saw the knife in the appellant’s hand when they were

46

dancing. He also later adapted his version and said that when they danced, they were
assaulting the deceased. According to Mr Riet, he identified the deceased by his
clothing, further confirming that they must have been together shortly before the
events he claims to have seen, otherwise it is difficult to understand how he could
have known what the deceased was wearing.

[90] Mr Riet’s version further was that he asked the accused what they were doing
because the person they were carrying was ‘Klaas Matsipe’. That is when the
appellant’s co-accused came towards Mr Riet and Mr Riet testified that the appellant
said that his co-accused should leave the deceased because he was not dead. In order
to ward them off, according to Mr Riet, he pelted them with stones. His friends were
not there to see what he saw. Seemingly they only appeared during the stone
throwing and joined him. They even remarked amongst each other that the accused’s
group must have been using muti because they did not get tired. They would attack
Mr Riet’s group and would retreat to the appellant’s place to enjoy and dance to the
music, apparently, repeatedly. It is only after someone had summoned the polic e,
and they arrived, that the appellant’s group ran away. Mr Riet then saw the injury
sustained by the deceased. A knife had penetrated his skull. The knife handle was
broken but the blade was protruding from the side. The deceased was taken from the
scene by an ambulance that had been summoned by the police.

[91] Mr Riet’s version was not coherent. The summary above is pieced together
from what he said at various stages of his testimony. He adapted his version as he
was confronted with conceptual difficul ties. These were obvious from, amongst
other things, his version that he followed the unknown person who had first
confronted him by a different route despite not knowing where the person was
heading; when, precisely, he allegedly saw the appellant with a knife; his belated

47

inference that the appellant and the other two people were assaulting the deceased
when it appeared that they were dancing; and his feigned ignorance of the gang
members’ attack of the appellant’s group. It is improbable that Mr Riet wo uld on
his own have followed a person armed with a knife. Since he did not see where the
person was heading, it is too much of a co -incidence that he just stumbled on the
scene where a member of the gang was being assaulted. It is unlikely that he would
have approached the scene alone, knowing, on his version, that the people there were
dangerous.

[92] Mr Riet’s version bears all the hallmarks of having been contrived. The
appellant’s version on the other hand, could not be lightly rejected as not being
reasonably possibly true. It is eminently probable that while the appellant and his
friends were socialising, drinking and listening to music at his house they were
attacked by members of the gang. It is more probable that Mr Riet was also with
them and that is how he came to be on the scene, rather than how, on his version, he
came to be on the scene. The members of the gang were his friends and he in all
probability still owed allegiance to the gang that he, on his version, had belonged to
for 15 years. Th ere is no other reasonable explanation for the presence of the
deceased and those others at the appellant’s place. Mr Riet’s explanation as to how
he and his friends got there is palpably false. He also had a motive and reason to
falsify facts to suit the narrative, inter alia, that the appellant was involved in the
stabbing of the deceased. Mr Riet was a single, unreliable and mendacious witness.
His version, insofar as it differed from that of the appellant, could not be accepted
as true.

[93] The appellant not only unequivocally identified his own friend, Lulu, as the
person that stabbed the deceased, but frankly testified about removing the deceased

48

from his premises and the reason for doing so. He denied that either he or accused 1
had stabbed the dece ased, or that he was armed with a knife. Considering all the
admissible evidence, properly evaluated, his version could not be rejected without,
more.24 Although the appellant is closely linked to events surrounding the deceased’s
death, the evidence, properly evaluated, does not exclude other probable inferences,
including that another person stabbed the deceased fatally and that the accused did
nothing to cause the deceased’s death. 25 The appellant bore no duty to prove his
innocence. The State had to pro ve his guilt beyond a reasonable doubt. It did not
meet that threshold. The findings of the trial court, the full bench and the first
judgment to the contrary, are wrong. Accordingly, the appellant’s conviction for
murder and the sentence imposed for it cannot stand and he should be immediately
released.

[94] In the result the following is ordered:
1 The appeal succeeds.
2 The full bench’s order is set aside and is replaced with the following order:
‘1. The appeal succeeds. The appellant’s conviction and sentence are set aside.
2. The order of the trial court is set aside and is replaced with the following
order:
“Accused 2, Hlalanathi Ngwadla, is found not guilty and is acquitted.”’
3. The appellant is to be released promptly.

_______________________
P COPPIN
JUDGE OF APPEAL

24 S v Shackell [2001] ZASCA 72; [2001] 4 All SA 279 (A); 2001 (4) SA 1 (SCA); 2001 (2) SACR 185 (SCA) para
30; S v Phogole ) [2025] ZASCA 54 paras 22-24.
25 S v Komane [2022] ZASCA 55 para 12-15.

49


Appearances

For appellant J Henzen-Du Toit
Instructed by: G A Mokaa Attorneys, Mmabatho
Kgoelenya Attorneys, Bloemfontein

For respondent: T B S Kalakgosi
Instructed by: Director of Public Prosecutions, Mmabatho
State Attorney, Bloemfontein.