THE SUPREME COURT OF APPEAL
OF SOUTH AFRICA
Case no: 055/05
REPORTABLE
In the matter between:
Pieter Hendrik GROENEWALD Appellant
and
The STATE Respondent
Before: Mpati DP, Scott JA, Cameron JA, Van Heerden
JA and Mlambo JA
Appeal: Tuesday 16 August 2005
Judgment: Thursday 8 September 2005
Criminal law – Evidence – Admissi ons – Effect and interpretation
of – Accused tendering admission cannot bind state to a meaning
radically at variance with state’s case
JUDGMENT
_____________________________________________________
CAMERON JA:
[1] The appellant w as convi cted i n the Pretoria High Court
(Pretorius AJ and assessors) of the murder of Prince Makena
2
and Simon Chuene Kgobo, and the attempted murder of Xavier
Lekgoate. He was sentenced to an effective term of twenty
years’ imprisonment. He appealed to the Full Court, where a
majority (Seriti J, Mbha AJ c oncurring) dismissed his appeal; R
D Claassen J dissented. This court granted special leave for a
further appeal against conviction and sentence.
[2] The case is unusual because of the ti me between the i ncident
and the prosecution. The i ncident occurred on Saturday night
6 May 1990. The appellant wa s arrested three days later.
Despite initial opposition by t he state he was granted bail.
Some months later, in late 1990, he l eft South Africa i n breach
of his bai l conditions for Portugal , where attempts to extradi te
him on the present charges fail ed. He returned for unrelated
reasons in 2000, and was arrested some two years later. His
trial commenced i n the Pret oria Hi gh Court on 10 February
2003, nearly thirteen years after the fatal events.
[3] The case is singul ar also because of the nature of the state’ s
evidence. Two eyewitnesses test ified against the appellant –
the complainant in the atte mpted murder charge, Mr Xavier
Lekgoate, and M r Brian Chester-Browne, who was in the
appellant’s company at the time of the events. Lekgoate and
Chester-Browne testifi ed to si milar effect regarding the
3
shooting that led to the deaths and to Lekgoate’s injuries, but to
radically different effect about what preceded it.
[4] The appellant pleaded not guilty and put all the elements of the
case against him in issue. Hi s counsel informed the court that
there had been discussions between him and the state and that
‘the necessary admissions’ regarding the post-mortem reports
‘etcetera, etcetera’ would be formulated and handed up later to
save time. In the result the state called no medical or ballistic
evidence. It reli ed instead on agreed admissions that defence
counsel formulated. These admi ssions became pivotal to the
defence argument at the cl ose of the trial, and in the appeal to
the full court, where they form ed the basis of the di ssenting
judgment of Claassen J.
State case
[5] Chester-Browne – whom the state undertook formal ly not to
prosecute – testi fied that on the night in questi on he was
driving the appellant, his passenger, from Cullinan to Mamelodi
(a Pretoria township) in his Alfa Guilietta, when they were
‘pulled over’ by a Nissan Skyline that cut in front of them. The
occupants, three black m en, approached. One of them
accused the occupants of the Al fa of throwing a stone at their
4
vehicle and breaki ng the windscreen . The appellant ali ghted.
His pistol was drawn, pointing at the black men. Chester-
Browne’s account of what then happened was terse: ‘I didn’ t
take part in much of the conv ersation. We inspected the
vehicle, the accused suggested we go to the Mamelodi Police
Station. There seemed to be some discussi on and then the
accused shot them.’
[6] The appellant, Chester-Brow ne sai d, was at this stage a little
more than an arm’s length from the others. He pursued one
who ran away, and later returned to the vehi cle. He asked
where the first person he had shot was, but he ‘had in the
meantime got up and run away’. Since there were headlights
approaching, ‘we got in the car and drove away’.
[7] In cross-examinati on, Chest er-Browne stated that no stone
was thrown from his vehicl e at the Nissan; that the Nissan
forced hi s vehi cle to a halt; t hat the actions of the occupants
were aggressive; and that ‘a very tense situation’ had resulted.
He was firm that he himself had fired no shots, and that the
appellant alone had. There was no attack upon him or the
appellant, and he was aware of ‘no attempt to arrest’ anyone.
[8] Lekgoate gave a completely different account of events leading
to the shooting. He and his companions, the deceased Simon
5
Kgobo and Prince Makena, were t ravelling to Cullinan in the
Nissan. He was in the f ront passenger seat; Kgobo was
driving. A white vehicl e passed them. It fli cked i ts l ights at
them, but Kgobo declined to stop in the dark. The vehicl e
passed them agai n. This time the front passenger threw an
object at the Nissan, whi ch bro ke its windscreen. (Police
photographs after the inci dent showed a damaged
windscreen.) Kgobo slow ed dow n and stopped. T he other
vehicle stopped ahead. Tw o white men alighted and
approached. The person from the passenger side had a
firearm in his hand. The men ord ered them from the vehicl e.
The driver chal lenged Kgobo for not stoppi ng when ordered to
do so. Lekgoate noti ced that the vehicl e w as not a poli ce
vehicle – but the armed man told him that the registration plate
was incorrect. The men alleged that cards they displayed were
police appointment certifi cates; but Lekgoate could see they
were not. The driver asked them if they knew of the Wit
Wolwe.1 Kgobo and Lekgoate deni ed knowing of them. The
1Cognisance may be taken of this as a reference to a white supremacist organisation, one of
6
driver then asked for their i dentity documents. Lekgoate and
Kgobo produced theirs. W hen Makena wished to retrieve hi s
from the vehicle, someone said: No, leave it, old chap (Nee, los
dit, ou kerel ). The driver then asked his com panion: Shouldn't
we just let them go? The other replied: Why?
[9] Lekgoate was then shot. He suffered two gunshot w ounds to
the abdomen, and one to the left elbow. He fell to the ground.
The shooting conti nued. When he heard footfalls, of the men
running away from where he was, he managed to crawl and
hide himself i n the grass next to the road. Shot s continued.
Then he heard the men close by, asking each other where h e
had hidden himself. One said: No, man, leave them, l et us get
in the vehicle. The doors slammed and they drove off.
Lekgoate dragged himself from t he grass to the road surface.
A passing motorist stopped but did not assist. He crawled to
the vehicl e, where he found K gobo lying close to the driver’s
door. K gobo proved unre sponsive to efforts to rouse him.
Lekgoate then drove the Nissan to the Mamelodi police station.
whose members, Barend Strydom, gunned down a number of black bystanders in Pretoria in
1988.
7
He was taken to Mamelodi day clinic and thence by ambulance
to Kalafong hospital, where he was operated upon.
[10] Lekgoate insisted that he was the first person to be shot, and
that only one person fired. This was the passenger of the other
vehicle. The driver, he stated, was not armed and did not fi re.
On 16 May, a week after the appellant’s arrest, Lekgoate
identified the appellant at an identification parade. However, at
the same parade he al so made a second, mistaken,
identification, whil e omitting to identify Chester-Browne, who
was present.
[11] On Monday 7 May, two days after the shooting, the
investigating officer, detecti ve warrant offi cer Oosthuizen,
visited Lekgoate i n hospital. Later that day or the next day,
Oosthuizen made a note on the case dossier. This recorded
that Lekgoate had described the man who shot as blond, with
longish and slightl y curly hair. It w as not di sputed at the tri al
that this description would apply to Chester-Browne, but not to
the appellant. Oosthuizen’s note also recorded that the blond-
haired person was wearing khaki (Chester-Browne testified that
he could not recall whether he was wearing khaki, or what the
appellant was wearing).
8
[12] Later, after his di scharge from hospital, Lekgoate made a
formal written statement to the police. In this, his description of
the man who fired the shots (sho rter than his companion; black
hair) accorded with his identifi cation of the appell ant at the
identity parade. At the tri al Lekgoate did not recognise the
appellant in court as hi s assailant, but confirmed that the man
he identified at the identity parade was he.
[13] The investigating officer, Oosthuizen, who had the dossier
from Monday 7 May unti l he le ft the police force at the end of
June 1990, testified that s pent cartridges (doppi es) were
collected at the scene by w arrant officer Viljoen, who w as first
at the scene, and who handed them to him on Sunday 6 May.
Oosthuizen stated that he too was at the sce ne on the night of
the events, though only brief ly. After taking over the
investigation on the Monday, he visited Lekgoate i n hospital.
Lekgoate was in shock, but coul d speak to him. H e did not
read the notes he later made on the dossi er back to Lekgoate
nor did he confirm them with him.
Admissions recorded during state case
[14] At an advanced stage of Lekgoate’s cross-examination, the
appellant tendered certai n formal admissions. Their purpose,
9
his counsel said, was to assi st the state to dispense with
certain witnesses who woul d not be necessary. The
admissions concerned the doppies and the fi rearm used at the
scene, and the identity parade. Counsel recorded that it was
common cause between the state and the defence, and that
the accused consented in terms of s 220 of the Criminal
Procedure Act 51 of 1977 (the Act), 2 that certain facts be
placed on record. These i ncluded that the doppies from the
scene w ere di scharged from a firearm the appell ant used
during the incident on 5 May 1990. It was also recorded that:
It is common cause that only five doppies were on the scene of the
incident and the parties admit that the five doppies were picked up and
handed to Wolmarans a ballistic expert for analysis.
(Dit is gemeensaak dat daar slegs vyf doppies op die betrokke toneel van
die voorval was en erken die par tye dat die vyf doppies opgetel is en aan
Wolmarans ‘n ballistiese deskundige oorhandig is vir ontleding.)
[15] Later, during the cross-examination of the third state witness,
investigating officer Oosthui zen, further admissions were
2 Section 220 of the Criminal Procedure Act 51 of 1977 reads:
Admissions An accused or his or her legal representative or the prosecutor may in criminal
proceedings admit any fact placed in issue at such proceedings and any such admission shall
be sufficient proof of such fact.
10
made. These concerned the deaths of Makena and Kgobo and
the correctness of the state’s medi co-legal reports concerning
the shooting of Makena, Kgobo and Lekgoate.
[16] The post-mortem reports on Makena and Kgobo revealed
that, aside from the three shots that apparently struck Lekgoate
[in the absence of medical evidence, it was not clarifi ed
whether one of Lekgoate’s two abdominal wounds mi ght not
have been coi ncident with hi s elbow wound], the dead men
each had three gunshot wounds. At least six further shots had
therefore been fired. This rendered a total of eight or perhaps
nine shots.
[17] Kgobo’s post-mortem report recorded also that in his thorax
there was ‘a semi-circular piece of plasti c, with fine pi eces of
metallic material and a copper c artridge casing’. It was not
disputed at the tri al that thes e were the rem nants of a ‘safety
slug’. T his Chester-Browne explained in cross-exam ination
was a bullet in a frangibl e jacket which tended ‘not to perform
at longer ranges’. It was suggested to him that the appellant
always loaded a ‘ safety slug’ (if he had one) first, and that it
would therefore be fired first. Chester-Browne granted that this
sounded ‘intelligent’, but could not say if that was either ‘normal
procedure’ or the appellant’s habit. ‘Where one puts i t in your
11
weapon would be over to personal preference and over to
personal experience’.
[18] A further recordal was made. This concerned a police
statement a constable, Petrus Marthi nus Beukes, made on 14
January 1991. B eukes was apparent ly unavailable to testify.
He had been instrumental i n the arrest of the appell ant, in that
he had furnished criti cal in formation regarding a shooting
practice at the Premier Shooting Range, Culli nan, on Saturday
5 May 1990, in which the appe llant had parti cipated. Doppies
from the range were analysed and linked to those at the scene.
[19] Beukes’ statement averred that on the day of t he incident at
the shooting range the appell ant had ‘a big full beard and
moustache’. The state, in accepting that Beukes was out of the
country and unavailable to testif y (and that he in any event no
longer had an independent recoll ection), accepted al so that
Beukes’ averment about the appellant’s beard and moustache
had been made in good faith: in other words that if he erred, he
had no motive to fabricate and had not deliberately lied.
[20] The person Lekgoate described, as also the appell ant when
he identified him at the identity parade on 16 May 1990, was
clean-shaven. In his evidence Lekgoate mai ntained that the
person who had fired was clean-shaven.
12
Appellant’s evidence
[21] The appellant testified th at he had recei ved training in
offensive action, inter alia as second in command of the South
African Defence Force’ s VIP protection uni t, as a means to
forestall attack. He left the f ormal operations of the SADF in
1988 to work full-time for ‘covert intelligence’, but continued to
receive remuneration in cash f rom the commander of his unit,
commandant De Castro.
[22] On the evening of the incident he was a passenger in
Chester-Browne’s father’s vehicle, which Chester-Browne was
driving from the Cullinan shooti ng range to Mamelodi. A
vehicle from behi nd forced them off the road. It was an
extremely dangerous situati on. Both he and Chester-Browne
were armed: Chester-Browne with a revolver (which does not
eject doppies); he with his o fficial Browning Hi-Power pistol
(which does). Chester-Brow ne was dressed in khaki and he in
black jeans and a russet leather jacket.
[23] As they were forced off the road, Chester-Browne shouted to
him to ‘be ready’ (staan reg). The three occupants of the other
vehicle – black men – storm ed their vehicle. He, the appellant,
immediately alighted, pointing his firearm at them, and warning
13
that he was armed. He told t hem to back off. One of the men
accused them of throwing an objec t at thei r vehicl e, causing
the windshield to break. Though he k new this to be a total li e,
it merely made the situation more tense.
[24] He ordered the men back to their vehicle, to ascertain if there
was damage to its windscreen. There was – though no pieces
of shattered glass inside. This made him even more uneasy.
He told the men that he didn’t wish to argue, but suggested that
they all proceed i mmediately to the Mamel odi police station.
On his way back to Chester-Brow ne’s vehicle, one of the men
expressed reluctance. The appellant insisted. At that stage he
saw a movement to his left, and as he turned shots were fired.
He did not know immediately who had fi red, but at the very
moment that hi s attenti on wa s distracted he was physically
attacked. A tussl e ensued. He struggled to free his weapon,
which was stil l in his hand. M ore shots were being fi red. He
managed to free his weapon, a nd fired three shots at hi s
attacker. After the third shot, his assaila nt fel l to the ground
and no longer constituted a threat to him.
[25] The first round in his magazi ne, he testified, was always a
Glazer safety slug. He loaded his magazine thus in
preparedness for short-ran ge situati ons, where he wished to
14
avoid the risk of ‘over-penetrat ion’ (ie, not to hit someone
behind the target). The rest of his magazine was loaded with
fully-cased military bullets. He therefore inferred that the
person he had shot was K gobo – not Lekgoate. It was
Chester-Browne who had been tal king to Lekgoate. W hat was
more, he had fired a total of only five shots: three at Kgobo (his
attacker after the shooti ng to his left started); and a further two
at the third person – who was fleeing as Chester-Browne was
firing at him. He tried to shoot low at this person, at hi s upper
legs. He fired no shots at the pe rson struck by the first shots
fired.
[26] He saw a vehicl e approaching from Mamelodi. Since he did
not know if the vehicle was wi th those who h ad just attacked
him, he and Chester-Brow ne deci ded to get as far away as
possible from any further potential problems. They drove to his
home. There he informed his com manding offi cer,
commandant De Castro, of the i ncident by telephone. De
Castro told him that it was quite l ate, and that he shoul d go to
sleep. He ordered th e appellant to report to him the next
morning with his firearm, which he did.
[27] He denied referring to th e ‘Wit Wol we’: if Chester-Browne
did, he did not hear it. He had a reasonably thi ck full beard on
15
the day of the incident. He shav ed this off the day after the
incident, after he had seen De Ca stro. For this he could give
no account, other than to state that it was ‘probably the
dumbest thing that I could have done’.
[28] After he obtained bail, De Castro instructed him to l eave the
country with him as soon as po ssible, giving him two passports
and $10 000 in cash.
[29] The appellant attributed hi s actions during the shooti ng to
self-defence. His work within the townships for the military was
against radical or extremi st elements withi n Umkhonto we
Sizwe (the armed wing of the Afri can National Congress) and
the Azani an Peopl e’s Liberation Army (the armed wi ng of the
Pan Africanist Congress), and he lived in a state of constant
fear for his life. It had not been his intention to kil l or injure
anyone. The road was dark, visi bility was limited. He wished
to ward off the attack on hi m and to ‘anchor’ the fleeing third
person so as to arrest him.
Findings of trial court and of full court
[30] Pretorius AJ regarded C hester-Browne as a possibl e
accomplice, and approached hi s evidence with caution. From
the difference between his account of the events preceding the
16
shooting and that of Lekgoat e she inferred absence of
collusion. What was important was what happened after the
initial events: and on thi s th eir evidence was substanti ally
corroborative. She accepted Lekgo ate’s identification of the
appellant and rejected the appellant’ s versi on as not
reasonably possibly true and f ound that he had the necessary
intention to kill.
[31] The majority of the full court in substance endorsed these
findings. In his dissenting ju dgment, Claassen J reasoned tha t
the court was obliged to accept as a proven fact the admission
that there ‘were’ only fi ve doppies on the scene. The state,
though admitting to an error in a greeing to the formulation, had
not sought to withdraw the admission. It foll owed that where
this fact confli cted with other evidence on behalf of the state,
the admitted fact had to prevail. In consequence, the evidence
on behalf of the state that onl y the appell ant had fi red shots,
that the appellant had fi red first at Lekgoate, and that only the
appellant had a fi rearm had to be rejected. By contrast, the
admitted fact accorded with the essence of the appellant’ s
case, and rendered his accoun t of the shooting reasonabl y
possibly true. The appell ant had also conducted hi s case on
the basis of the admitted fact – for instance i n not seeking to
17
call De Castro – and to ignore it now would impeach his right to
a fair trial . Despite a ‘very st rong possibility that the appellant
in fact fired all the shots’, he therefore had to be acqui tted of
the murders and the attem pted mu rder. On his own versi on,
however, though he acted reasonably in killing Kgobo, he acted
without justification in shooti ng at Makena. In respect of that
killing he was guilty of assault with intent to commit grievous
bodily harm and should be sentenced to imprisonment with the
option of a fine.
The effect of the admission concerning the doppies
[32] Before this court, the appe llant pressed substantiall y the
argument that found favour with Claassen J: that the state was
bound by the admission that there ‘were’ only five doppi es on
the scene; that the inevi table corollary was that more than one
firearm had been used, since ot herwise the number of gunshot
wounds did not square; that th is im pugned Chester-Browne’s
self-exculpatory version, w hile casti ng doubt on Lekgoate’ s
recollection; that the statement of Beukes about the appellant’s
beard, and the investig ating offi cer’s notes about Lekgoate’ s
hospital description of the k iller added consi derable doubt to
the mix, while Chester-Browne’ s account of the preceding
18
events supported appell ant’s version of self-defence; and that
in consequence the versi on of the appell ant coul d not be
rejected as false beyond reasonable doubt.
[33] The keystone of these contentions is the admission. But the
argument is fallaci ous. It seeks to i mpute to the admission a
meaning and effect it was clearly never intended to have. An
admission is an acknowledgment of a fact. When proved or
made formally during judicial proceedings, it dispenses with the
need for proof in regard to that fact. Wigmore on Evi dence
calls it ‘a method of escaping from the necessity of offering any
evidence at all’: a ‘waiver relieving the opposing party from the
need of any evidence’. 3 Section 220 of the Act accordingly
makes it possi ble for a contested fact to be put beyond issue,
since once made the admission cons titutes ‘sufficient proof’ of
it.
[34] But what was acknowledged when the parti es recorded that
it was ‘common cause’ that ‘only fi ve doppi es were on the
scene of the incident’? Appellant contends that this constituted
3 Evidence in Trials at Common law by John Henry Wigmore, revised by James H Chadbourn
(1972), vol 4, § 1058.
19
an admi ssion by the state that as a matter of absolute truth
there ‘were’ at all times onl y five doppies on the scene. But
this i s an ambitious interpretat ion, which asks us to read the
admission in isolation, as th ough disembodied from the events
at the tri al. An admission does not stand in isol ation, and
cannot be interpreted in isolation. Li ke all other recordals or
documents admitted at the tria l, its proper meaning and effect
must be determined in its setting.
[35] An admission may be equi vocal or ambiguous, permitti ng of
more than one interpretation: th is has prompted the beneficent
rule that where the defence ma kes an ambiguous admission,
the construction more favourable to the accused should be
adopted in relation to his case: S v Maweke 1971 (2) SA 327
(A) at 329H. And as Phipson on Evidence points out, the
weight of an admi ssion ‘depends on the ci rcumstances under
which it was made’. 4 This applies w hether the admission i s
made inside or outside court.
4 Phipson on Evidence 15ed 2000 (general editor, MN Howard) para 28-09 page 712
(referring to informal admissions; but the point is more general).
20
[36] Context and background, as elsew here, thus inform both
meaning and effect. Counsel ri ghtly conceded this. Here,
dealing first with effect, the tri al record shows that the
admissions were made ‘namens die beskuldigde’ (on behalf of
the accused). The acknowledgement of fact was thus tendered
and recorded as an admission, not by the state, but by the
appellant. The Act was am ended in 1996 to make it possibl e
for the state – and not only the accused – to m ake admissions
in terms of s 220: hence the appell ant’s em phasis that the
state, too, was bound. Bu t, though the admi ssion recorded a
fact that was ‘com mon cause’ between the partie s, the state
did not proffer it. The appellant di d. Its effect must therefore
be assessed in the light of the fact that its declared object when
tendered was to relieve the state of part of its duty in relation to
the case it set out to prove.
[37] The state’s case was that fi ve doppies linked with those the
appellant had fired earl ier on the fa tal day ‘were on the scene’.
It was no part of what it was tr ying to prove that ‘only fi ve’
doppies were on the scene. And the appellant by proffering an
admission to this effect could not and did not relieve the state
of any duty in relati on to the case it w as seeking to prove. Still
less could he tie the state to a p roposition that was radi cally at
21
variance with its case: an a ccused cannot by purporting to
relieve the state of duties in relat ion to its case burden it with
millstones in relation to his own. This is therefore a converse
case to Maweke, since here the accus ed seeks to impose on
the state an unfavourable construction of an admission.
[38] In any event, I am disincli ned to accord the admissi on the
meaning appell ant’s counsel soug ht to ascribe to it. The
admission does not specify at what time there ‘ only five
doppies were on the scene’. It co uld mean either that there
were at all times only five, or there were only five when Viljoen
visited the scene. Since the state did not call Viljoen – no
doubt in the light of the adm ission – we do not know precisely
when he retrieved the doppies . Oosthuizen said (without
stating the time) that both he and Viljoen visited the scene that
night, and that Viljoen handed the doppies to him the next day.
But an exact time specification wa s crucial to the far-reaching
meaning appellant’s counsel sought to extract from the
admission, since without it other possibilities, such as that
doppies may have been removed f rom the scene, or that they
were dispersed by passing traffic, l oom large. Without a time
specification, the more obvious meaning is that there ‘were
only five doppies’ when Viljoen retrieved them.
22
[39] In determining what the admission meant, it is moreover
relevant that it was formulated by appellant’s counsel. The rule
that ambi guous written provisions are interpreted agai nst the
party drafting them is not m erely app licable in contract law; it
embodies a sensible practical approach also in determining the
extent to which the state is bound by a meaning an accused
who formulates an admission seeks to impose on it.
[40] Here it is clear that the state intended merely to acknowledge
that only five doppies were found on the scene when Viljoen
visited it. At no stage did i t intend to acknowledge that at all
times there ‘were’ no more than five. When appellant’s counsel
handed up the admission, Chest er-Browne had compl eted his
testimony, and the cross-examination of Lekgoate was close to
completion. Both testifi ed that onl y the appellant had fired
shots: neither specified or estimated how many. But it was
apparent from both that a subst antial number was fired. At no
stage did appellant’s counsel put to either Chester-Browne or
Lekgoate, or even imply, that the appellant fired only five shots.
The suggestion that the admissi on ineluctably entailed that
further shots were fired by another person was alien to the
proceedings at that stage.
23
[41] The further implication aros e for the first tim e during cross-
examination of the third state witness, Oosthui zen. And when
counsel suggested the meaning now sought to be imposed, he
received the obvious response: ‘It is hard for me to say how
many shots were actually fired. The possibility surely also
exists that not all the doppies were picked up on the scene.’
[42] Counsel for the state complained in his written argument that
the state had been trapped in to making an admission whose
ambit and implications it had nev er intended to endorse. That
may be so. But as pointed out in the dissenti ng judgm ent in
the full court, the state never sought to retract the admission.
That has proved to be the correct course, since, interpreted in
its context, the admission can not support the meaning or effect
the appellant seeks to extract from it.
The strength of the evidence against the appellant
[43] Without the meaning sought to be ascribed to the admission,
the state’s case against the appellant was strong. Chester-
Browne and Lekgoate corroborated ea ch other on the critical
details: that the appellant alone was armed, that he fired first at
Lekgoate – the survivor who escaped and managed to hide –
24
and that he alone fired all the shots that struck the Ni ssan’s
three occupants.
[44] It is true that the two princ ipal state witnesses contradicted
each other radical ly as to what preceded thi s. But i t is cl ear
from the record that Chester-B rowne had a motive to distort
those events – namely to minimise his own invol vement in
them, si nce on Lekgoate’s account and without the state’s
indemnity he was l iable to be prosecuted as an accom plice or
accessory to the shooti ng. By contrast to C hester-Browne –
who rebuffed his cross-examiner with patent hostility –
Lekgoate volunteered a ful l, co herent and candi d account of
those events.
[45] What is more, while Chester-Browne had a motive to
minimise his involvement in the entire sequence, and thus to lie
about whether he shot – P retorius AJ for thi s reason rightl y
approached his evidence with caution – Lekgoate had no such
motive. It may have been po ssible, without C hester-Browne’s
evidence, to infer that Le kgoate might have erred about the
number of firearms wielded. But the possibil ity that he could
have erred about the other cri tical detail s is remote. These
included the fact that the man who alighted from the passenger
seat of Chester-Browne’s vehicl e was the one who shot him ;
25
that hi s attention was on this man when the shooti ng started
(and not on Chester-Browne); and, critically, that he, Lekgoate,
was the first of the three to be shot. All thes e features, on
which it is highly unlikely that Lekgoate could have erred, point
strongly to the appellant’s guilt.
[46] And if Chester-Brow ne had been armed, or if he had
participated in the shooti ng, Lekgoate had no conceivable
reason to exculpate him: nor wa s any reason put to him, or
suggested to us in argument. Hi s evidence on this score must
therefore be accepted as reliable.
[47] It is pl ain that Lekgoate cannot be held responsibl e for the
description of the killer in Oost huizen’s dossier notes. They
were not read back to him , nor did he confi rm them. As he
stated in cross-examination, hi s statement then was confused
‘because at that stage I also was confused’ (volgens my is die
verklaring deurmekaar om dat ek ook deurmekaar was op
daardie stadium ). Oosthui zen stated that Lekgoate w as ‘not
one hundred percent com fortable at that stage in making a
statement’: and this is why he wa ited until his discharge from
hospital before securing a formal statement.
[48] Lekgoate’s later sworn stat ement depicted the appel lant
accurately; and hi s identification of the appell ant at the parade
26
eleven days after the shooting points powerfully to the
accuracy of his recall – particularly when it is borne in mind that
at this ti me the appellant deni ed that he had fired any shots at
all. As Lekgoate stated i n cr oss-examination: ‘The person I
identified is the person who sh ot at me. He and I looked at
each other during the in cident. That is why I identified him.’
And his failure to i dentify Chester-Browne, together with the
mistaken identification, is exp licable on the same basi s – that
he was not looki ng at Chester-Brow ne, but was concentrati ng
on the man who was armed and dangerous.
[49] I would add that Lekgoate was a careful and candid witness
who readily conceded uncertainty, doubt or inability to recall.
He gave a chil lingly detailed and consistent overall account of
the events. The fact that Chester-Browne corroborated him on
the essential details of the appell ant’s role poi nts powerfully to
the appellant’s guilt.
[50] By contrast, the appell ant was a poor witness. He was
unable to give a pl ausible account for his failure on the night of
the incident to go to the polic e – on his own version he, after
all, suggested thi s before t he shooting started – nor for hi s
failure to enlist the assistance of the police thereafter. If he had
indeed been attacked, as he claimed, he surely had little to fear
27
in 1990, with his security-apparat us connections, from making
a clean breast. Instead, w hen Oosthuizen i nformed hi m that
his fi rearm had been linked to the shooting, he deviousl y
replied that it must be the w rong firearm. This may have been
technically correct, since the B rowning Hi-Power pistol seized
at his home was apparently that of his father: but it was
nonetheless a tell ing evasi on; one made more signi ficant by
the fact that he al so initially denied to Oosthuizen that he had
fired any shots at all.
[51] The appellant’s account of the sh ooting itself was
unconvincing: he was for instance unable to account plausi bly
for the fact that Kgobo – whos e frontal attack he claimed to
have warded off – had a bul let wound in the back of hi s head.
And he could give no coherent account for pursui ng and
shooting Makena. On the appella nt’s account there was, after
all, no reason to try to arrest him.
[52] The statement by Beukes, in January 1991, that on the day
of the shooting practice i n January 1990 the appel lant had full
facial hair, though made in good faith, was obviousl y mistaken.
Lekgoate was cl ear that the appellant was cl ean-shaven: he
would hardly have been abl e to i dentify him, clean-shaven, at
the identity parade if he had a full beard nine days before.
28
[53] I would add that there is subs tance to the state’s suggestion
that the appellant appeared to have cobbled his version
together on the basis of inventive scrutiny of the police dossier.
This appl ies to hi s averment about the ‘safety slug’. But his
attempt by this means to escape liability must fail in the face of
the damning direct testimony against him.
[54] The appellant’s guilt on the charges of murder and attempted
murder was clearly establ ished. There is also an appeal
against sentence, but in my view Pretorius AJ, both in imposing
20 years for each murder, and te n years for the attempted
murder, as al so i n deci ding on an effecti ve term of only 20
years’ imprisonment, blended her di scretion with a m easure of
mercy that renders her sentence immune to criticism.
[55] The appeal is dismissed.
E CAMERON
JUDGE OF APPEAL
CONCUR:
MPATI DP
SCOTT JA
VAN HEERDEN JA
MLAMBO JA