South Africa: Supreme Court of Appeal Support SAFLII

You are here:  SAFLII >> Databases >> South Africa: Supreme Court of Appeal >> 1990 >> [1990] ZASCA 95

| Noteup | LawCite

S v Nkwanyana and Others (52/90) [1990] ZASCA 95; 1990 (4) SA 735 (AD); [1990] 2 All SA 591 (A) (18 September 1990)

Download original files

PDF format

RTF format


CASE NO. 52/90

JOHN MEMFU NKWANYANA AND TWO OTHERS APPELLANT
and
THE STATE RESPONDENT

Judgment by: NESTADT, JA

c

CASE NO. 52/90

/ccc

IN THE SUPREME COURT OF SOUTH AFRICA (APPELLATE DIVISION) In the matter between:
JOHN MEMFU NKWANYANA FIRST APPELLANT
SIPHO THEMBA NHLABATHI SECOND APPELLANT
JACOB NDODA NDLELA THIRD APPELLANT
and
THE STATE RESPONDENT
CORAM: VAN HEERDEN, NESTADT, KUMLEBEN JJA et
NICHOLAS, NIENABER AJJA DATE HEARD: 16 AUGUST 1990 DATE DELIVERED: 18 SEPTEMBER 1990

JUDGMENT NESTADT, JA:

At about midnight on 16 July 1987, a hut in the

2/
2.

Sovana Reserve in the district of Nongoma, Zululand, was set on fire. Five persons were in the hut at the time, presumably asleep. They were 84 year-old Mrs Elsie Magagula ("Elsie"), her two adult grand-daughters and two boys aged nine and five. It would seem that the flames quickly spread and engulfed the hut. A neighbour helped the three adults escape through a window. They had, however, already suffered severe and extensive burns. They were taken to hospital. Two of them died there the following day. The third died the day after. In each case the cause of death was toxaemia and renal failure. These were complications of the injuries they sustained. They were such that according to the medical evidence there was no chance of survival. The two children were found dead in the hut. The doctor who performed the post-mortem examination on them testified that their bodies had been "charred beyond recognition".

3/

3.
These events led to the prosecution of the three appellants before SQUIRES J and assessors in the Natal Provincial Division on five counts of murder. The State case was that it was the first and second appellants who, at the instigation of the third appellant, had intentionally set the hut alight and thus caused the death of the five occupants. Appellants denied these allegations. Their evidence in support of what amounted to alibi defences was, however, rejected. Third appellant was found guilty on all five counts of murder. But first and second appellants were convicted on one count only, viz, the murder of Elsie. This was because, so it was held, the State had not (as it had in the case of third appellant) proved that they knew that the hut was occupied by anyone else. No extenuating circumstances were found. Accordingly each of the appellants was sentenced to death on the counts on which they were found guilty. This

4/
4. appeal is against their convictions and sentences.
I must refer to a preliminary, procedural matter. It arises from the fact that at the commencement of the trial first and second appellants declined the services of the pro deo counsel who had been appointed to represent them and who had prepared their respective defences. They each stated that he wished to conduct his own defence. SQUIRES J was understandably anxious that this should not happen. But despite his commendable attempt to persuade first and second appellants to change their minds, they persisted in their attitude. Counsel were, however, not then discharged from their duties. At the learned judge's invitation they remained in attendance and cross-examined certain of the State witnesses after first and second appellants had been given the opportunity of doing so. Furthermore, at the end of the evidence counsel were asked to and did, in addition to appellants themselves, present

5/

5. argument to the Court. There may not be any objection to the latter course but counsel should not have been allowed to cross-examine. Whilst an accused has a fundamental right to legal representation (S vs Mabaso and Another 1990(3) SA 185(A)) he also has a fundamental right to represent himself (S vs Tyebela 1989(2) SA 22(A) at 31 D). What happened detracted, so it seems to me, from such right. Furthermore, it involved the potential of answers adverse to appellants being elicited. In fact, however, this did not occur so that in the result they did not suffer any prejudice.
Elsie was third appellant's mother-in-law. What the State sought to prove was that he wished to kill her. The setting alight of her hut on the night in question was the means of achieving his nefarious purpose. He had hired first and second appellants to do this. In summary the evidence relied on by the State was the following:

6/
6.
(i) That of Amos Mathe who stated that about a week before Elsie's house burnt down he was approachéd by third appellant who said that he wished to hire him "to go and kill the Magagulas". He was prepared to pay Mathe R3 000 to do this. The reason he gave for wanting them killed was that his wife had deserted him and "the Magagulas are aware of (her) whereabouts". Moreover, so he further told Mathe, his wife had stolen R70,000 from him. The witness "assumed" that they had refused to tell third appellant where she was. A few days after the fire he spoke to first appellant who admitted that he was one of the persons who had set the hut alight.
(ii) Then there was the evidence of Absolom Gama. He worked for third appellant. His testimony was
7/

7. that on the night in question he was in the company of third appellant at third appellant's shop. Two persons arrived there. He later identified one of them as second appellant. Third appellant handed them two gas cylinders and a container of paraffin. He said to them: "Take these two gas cylinders and this container and you go and burn everybody who is there, with these gas cylinders". They then left. They later returned. They reported to third appellant that they had "fixed everything up". Third appellant thanked them. At third appellant's request the witness took a bag of money from the safe of the shop and handed it to third appellant. Gama then went to bed. (iii) The State also tendered written confessions allegedly made by each of first and second

8/
8. appellants to a magistrate shortly after their arrest about a year after the commission of the crimes. First appellant denied having made the confession. In addition he and second appellant testified (in a trial within a trial) that they had been assaulted. On this basis they contended that the confessions had not been freely and voluntarily made and were therefore inadmissible. The trial court, however, held that the onus (which by reason of Sec 217(1)(b) of Act 51 of 1977 rested on them) of proving the duress relied on, had not been discharged. In their respective confessions first and second appellants describe how, by means of the eguipment supplied to them by third appellant, they set the hut on fire. What they state to be petrol was poured under the door. The cylinders

9/

9.
containing gas were opened and thrown inside the hut. A match was lit and the blaze started. (iv) A further admission proved (against first appellant) was what he said at the sec 119 proceedings. They were held the day after he made his confession. First appellant, in pleading guilty to the five counts of murder, briefly confirmed the terms of his confession. (v) Certain witnesses confirmed that two empty gas cylinders were found at the scene of the fire. In addition there was evidence that an explosion emanating from the hut was heard at the time. That then was the case which appellants had to meet. In his evidence third appellant, admitted that his wife had about a month before the incident deserted him and that he was looking for her. He said that he suspected that she had misappropriated, not money, but some R900 worth of

10/

10. stock from his shop. But he denied that this had affected his relationship with his mother-in-law; on the contrary they continued to be on good terms with each other; he had not asked her the whereabouts of his wife; he did not hold her or her family responsible for his marital problems; he never entertained any thought of harming them. He admitted that on the night in question he was with Gama at his shop. He denied, however, that first and second appellants arrived there. He had never hired them to kill Elsie. Nor had he ever approached Mathe for this purpose. First and second appellants similarly denied their alleged participation in the crimes or their presence at third appellant's shop as Gama alleged. And first appellant denied the admission he was alleged to have made to Mathe.
As I have indicated, the trial court, in a comprehensive judgment, rejected appellants' evidence. Having admitted first and second appellants' confessions it

11/
11.

found them reliable. The discovery of the gas cylinders was an objective fact supporting what was stated in them. The court seems to have had some hesitation in accepting Mathe's evidence involving first appellant but a certain amount of weight does appear to have been attached to his implication of third appellant. Gama's credibility was, however, approached more positively. He impressed the court. After a careful analysis of his evidence, SQUIRES J rejected the argument that he was part of a conspiracy to falsely implicate second and third appellants. Though regarding him as a single witness, the learned judge accepted his version as the truth. Moreover his identification of second appellant as one of the two persons who was at third appellant' s shop was held to be reliable.
Before us, counsel for appellants launched a wide-ranging attack on the correctness of the convictions.

12/

12. (i) On behalf of first appellant it was contended that neither his confession nor his admissions in the sec 119 proceedings should have been admitted; he had on his arrest and for some time subsequent thereto denied his guilt; even when initially taken to a magistrate by the name of de Klerk (not the one to whom he later confessed), he still did not admit any personal participation in the crimes; it was therefore probable that he later confessed because of the assaults which he said were perpetrated on him by the investigating officer, Sgt Ntuli and others; Sgt Ntuli was a bad witness who had contradicted himself and who admittedly had questioned first appellant for a lengthy period shortly after first appellant's arrest; on the State evidence itself the opportunity for first appellant having been

13/
13. assaulted prior to him making the confession existed; and such evidence did not explain what happened to first appellant during this period. Alternatively it was said that the confession differed radically from the events described by Mathe and especially Gama and was accordingly unreliable. Finally, attention was drawn to the fact that Gama, who (so first appellant alleged) knew him, did not purport to identify him as second appellant's companion on the night in question. (ii) Leave to appeal against the admission of second appellant's confession was refused by the trial judge. Counsel nevertheless relied on second appellant's evidence that his confession was involuntarily made in support of an argument that it should have been held to have been unreliable and therefore deserving of no weight. The

14/
14. cogency of Gama's evidence implicating second appellant was also attacked on various grounds. The main ones were that he had not satisfactorily explained why he had delayed for about a year before reporting the matter to the police; he had contradicted himself in his evidence; there were discrepancies between his evidence and the contents of the confession; it was improbable that appellants would have discussed the contemplated setting alight of Elsie's hut in front of him; and the witness's identification of second appellant was in any event unreliable. (iii) The principal argument on behalf of third

appellant was that Gama's evidence should not have been accepted. Echoing the argument on behalf of second appellant, his counsel submitted that it was incredible that appellants with a total disregard for their own interest would plan

15/
15. the murders in Gama's presence and that third appellant would make him a party to the fact that he was going to reward the two persons he had hired. A further complaint was that the trial judge had in his approach to Gama's evidence misdirected himself in certain respects. Gama should have been held to have had a motive to falsely implicate third appellant. Mathe's credibility was also impugned; counsel submitted that it was improbable that third appellant would have approached Mathe to carry out the murders; having done so their arrangements in this regard would not have been as indefinite as they apparently were; and Mathe too had failed to satisfactorily explain his delay (also of about a year) in reporting the matter to the police; he was moreover akin to an accomplice and had been warned as such in terms

16/

16. of sec 204 of the Act. Another submission was that in the absence of evidence by third appellant's wife (who it was said was an available witness) the State had failed to establish that Elsie had refused to disclose the wife's whereabouts to third appellant; accordingly the State had not established a motive on his part to harm her.

I do not propose to deal with these arguments in any detail. Substantially the same contentions were urged upon the trial court. In my view its rejection of them is unassailable. First appellant's denial that he made a confession was manifestly false. And the State presented a strong body of evidence in rebuttal of his and second appellant's allegations of assault. In the case of first appellant it showed that he failed to complain to Mr de

17/

17. Klerk (the magistrate to whom I referred earlier) or to a district surgeon who, at the instance of Sgt Ntuli, examined him immediately after he had made the confession in issue. Nor did he say anything to the magistrate during the sec 119 proceedings about having allegedly been assaulted. First appellant was unable to satisfactorily explain these omissions. And, of course, he told the magistrate before whom he confessed that he had not been assaulted. Neither the two magistrates nor the doctor saw any signs of injury on first appellant. On his version they should have. Second appellant too failed to allege that he had been assaulted when he should reasonably have done so. Furthermore I agree with SQUIRES J that his version of how he was mishandled by the police was far-fetched. The contents of both confessions are such as to make it improbable that they were the product of what the police told appellants to say (which is what they alleged).

18/

18. The trial judge was alive to the criticisms of Sgt Ntuli. In the result the argument that the confessions should have been ruled inadmissible must be rejected. The same applies to the submission that they should have been regarded as unreliable. What discrepancies exist between them and the objective facts established by the State are unimportant. A confession need not and often does not accurately describe all the details of the events to which it relates. In any event the finding of the cylinders and the hearing of an explosion corroborate what appellants say in their confessions. This brings me to Gama's evidence. It is crucial to third appellant's conviction. The trial court realised this. It dealt at length with the attack on Gama's credibility. Valid reasons were given for rejecting the criticisms levelled at this witness. I do not believe that his version, which is a fairly detailed one, was a fabrication. In my opinion there is no warrant

19/
19.

for interfering with the conclusion that in implicating third appellant as he did, Gama was telling the truth and that his identification of second appellant was reliable.
The result, in my opinion, is that the State proved (i) by means of Gama's (and Mathe's) evidence that third appellant instigated the burning down of Elsie's hut; (ii) that first appellant confessed to his participation in the crime; and (iii) second appellant's guilt by means of his confession and Gama's identification of him as being one of the persons who, on the night in question, was at third appellant's shop when the events described took place. On these findings it is clear that appellants were correctly found guilty of the murder of Elsie. It was, however, submitted on behalf of third appellant that the trial court should not have found it proved that he had the necessary intent to kill the four other occupants of the hut and that he should therefore not have been found guilty

20/

20. on counts 2, 3 4 and 5. I am unable to agree. Gama's evidence that third appellant instructed first and second appellants to "go and burn everybody who is there" shows that he subjectively contemplated at least the possibility of the presence of others in the hut and accordingly the possibility of their deaths. It was a case of dolus indeterminatus.
This brings me to the appeal against sentence. When the trial was concluded, what I call the old sec 277 of the Act was still in operation. This meant that in the absence of a finding of extenuating circumstances, the death sentence was compulsory. Now, however, by virtue of the introduction by sec 4 of the Criminal Law Amendment Act, 107 of 1990 of a new sec 277 the position is, as will be seen, substantially different. And, despite the fact that the amending Act came into operation only on 27 July 1990, sec 277 as amended is to be applied (see the recent

21/
21.

judgments of this Court in Masina and Others vs S, case no 695/89, delivered on 13 September 1990 and Senonohi vs S, case no 691/89, delivered on 17 September 1990). This is because by virtue of sec 20(1)(a) of the amending Act an appeal has to be continued and concluded as if secs 4 and 13(b) had at all relevant times been in operation. I have already mentioned sec 4. Sec 13(b), by inserting a new sub-sec (2A) into sec 322 of the Act, brings about a fundamental alteration to the powers of the Appellate Division when dealing with appeals against the death sentence. It enables this Court to set aside the sentence if it is of the opinion that it would not itself have imposed the sentence of death. In this event it may impose such punishment as it considers proper. In other words, the previous limitation on the power of this court to interfere with the imposition of the death sentence has fallen away. Now this Court can on appeal impose whatever

22/
22.

sentence it considers should have been imposed by the trial court. And in deciding what such sentence should have been, regard must, as I have indicated, be had to the new sec 277.
I therefore turn to a consideration of the effect of the new sec 277 and more particularly sub-sec (2) thereof. The following are its essential features: (i) It abolishes the compulsory imposition of the

sentence of death. The presiding judge is given what amounts to a discretion in this regard. It is true that the word "shall" is still used. But the consequent obligation to impose the death sentence only arises if the presiding judge "is satisfied that the sentence of death is the proper sentence". (ii) In pondering this fateful guestion, the presiding

23/

23.
judge is enjoined to have "due regard" to "the presence or absence of any mitigating or aggravating factors". Indeed, a finding in this regard has to be made. It is doubtful whether the absence of mitigating factors could in itself be aggravating. But it may be that the absence of aggravating factors is mitigating. The use of "a finding" in sub-sec (2)(a) can possibly lead to misunderstanding. The section cannot be read as requiring a single finding. Usually both mitigating and aggravating factors will be present. Neither term is defined. Mitigating factors, whilst obviously including those which until now have been regarded as extenuating circumstances, is a wider concept (see the judgment of FRIEDMAN AJA in Masina and Others vs S, supra, at p 10). What constitutes aggravating factors in relation to a crime which

24/

24. is a very serious one in its nature, will have to be clarified by the courts. Suffice it to say at this stage that the degree of planning, the manner of the commission of the murder, its motive, the circumstances of the victim and an accused's previous convictions are some of the considerations which will no doubt be viewed as relevant in deciding whether there are aggravating factors. (iii) The question of who bears the onus of establishing the presence or absence of mitigating and aggravating factors arises. So does the degree of proof. Contrary to sub-sec (3)(b) (which places the onus of showing that an accused was 18 years of age or older on the State), there is no reference in sub-sec (2)(a) to onus. It has been held that the use of the term onus in relation to factors relevant to

25/

25.

sentencing is inappropriate; and that no rigid
rules governing the degree of proof can

satisfactorily be laid down (see R vs Chinyani

1969(2) S A 588 (RAD) and the two cases cited at

589 as also S vs Seleke 1976(1) S A 675(T) at 690
F and S vs Mogashoa 1978(1) S A 492(T) at 496
A). But the position created by the new sec
277(2) calls for a different approach. A

finding or findings on the presence or absence of
mitigating or aggravating factors has to be made.
There may be a dispute about this. In these
circumstances it would be difficult if not

impossible to make the necessary findings unless
the incidence of onus operates. As Hoffmann and ..

Zeffertt, The South African Law of Evidence, 4th
ed, at 495 state:

"Any rule of law which annexes legal consequences to a fact, ... must, as a

26/

26.

necessary corollary, provide for which party is supposed to prove that fact".

No difficulty arises in relation to the onus and
degree of proof of aggravating factors. In
accordance with principle it will be for the
State to establish their presence. And in order
to discharge such onus, proof beyond reasonable
doubt will be required. There is authority in
support of both propositions (S vs Shepard and
Others 1967(4) S A 170(W) and 180 B - D; see too
Schmidt: Bewysreg, 3rd ed, 60-1). In my view
a similar rule should apply in regard to the
mitigating factors referred to in the new sec
277(2). An accused will, of course (unless a
mitigating factor already appears from the
evidence), have to raise it and adduce whatever
evidence he can on the point. But, having done
so, the onus should be on the State to negative,

27/
27. beyond reasonable doubt, the existence of such mitigating factors as are relied on by an accused. It follows that if there remains a reasonable possibility that mitigating factors exist, the onus is not discharged. This may be said to run counter to the rule that it was for the accused . to establish extenuating circumstances. But that rule originated from the particular wording of the statute which created the concept of extenuating circumstances (see R vs Lembete 1947(2) S A 603(A) at 609). I accordingly do not think that it can be assumed that the Legislature intended that the same * principle should apply to the new sec 277. On the contrary, there is reason to think that it did not so intend. Placing the burden of proof on the State to negative the existence of mitigating factors will avoid a difference in the

28/

28.

incidence of the onus of proof (which would
otherwise occur) in regard to mitigating factors
on the one hand and aggravating factors on the
other; and also in regard to facts relevant to
conviction on the one hand and those relevant to
sentence on the other. The illogicality of there
being a different onus in this latter regard was
referred to by COLMAN J in S vs Shepard, supra.
Having cited overseas authority that it is for
the prosecution to disprove mitigating facts, the

learned judge (at 180 G) said:

"What is attractive about the approach is this: To an accused person the sentence is at least as important as the conviction, and it might seem, in a sense, anomalous to give him the benefit of all reasonable doubts before finding him guilty, and then, when dealing with a question which may make a vast difference to his sentence, to place an onus on him so that the Court, if it f inds the probabilities equally balanced in relation to some mitigating fact, will punish him as if that fact did not exist."

29/

29.

Schmidt, op cit, at 61-2 endorses this approach. It may be said to be unfair to the State that it be burdened with the onus of disproving mitigating factors especially where, as will often be the case, they depend on matters peculiarly within the knowledge of the accused. This is a consideration but it must bow to the factors referred to. In any event the problem will often be more apparent than real. It follows from what was said earlier that the State's onus will not generally arise unless the mitigating factor relied on has been "'genuinely raised in a genuine fashion', or 'unless a proper foundation (for it) is laid' by the accused in the sense that the court could find that the (factor) existed, or 'unless the

30/

30.

evidence reveals it as a possible factor in the case'". (Burchell and Hunt: South African Criminal Law and Procedure, vol 1, 2nd ed, 135). What is required is a factual basis for the mitigating circumstance. A speculative one will not suffice to raise it (Du Toit, Straf in Suid-Afrika, 51). (iv) In considering whether the death sentence is "the proper sentence" (an expression which the Legislature has understandably not defined), the findings as to mitigating and aggravating factors are not necessarily decisive. What the section provides is that "due regard" be had to them. This means "consideration in a degree appropriate to (the) demands of the particular

case" (Black's Law Dictionary, 5th ed, sv "due regard"). Inherent in the expression therefore

is a recognition that other matters may be

31/
31 . relevant. The absence of mitigating factors (or, as before, extenuating circumstances) will not mean that the death sentence should be passed. Conversely the presence of mitigating factors will not mean that the death sentence should not be passed. And when both mitigating and aggravating factors are present, their respective force or significance will have to be weighed in order to determine whether the death sentence is the proper one. In doing this I agree with the view of E M GROSSKOPF JA in Senonohi vs S, supra (at pp 18-19) that regard will be had to the main purposes of punishment, namely, deterrent, preventive, reformative and retributive. This means that in deciding whether the death sentence is the proper one, consideration will be given to whether these objects cannot properly be achieved by a sentence

32/

32.

other than the death sentence (generally a lengthy period of imprisonment). If they can, then the death sentence will not be passed. This is because "the proper sentence" (unlike "a proper sentence") must be interpreted to mean "the only proper sentence". It follows that the imposition of the death sentence will be confined to exceptionally serious cases; where (in the words of NICHOLAS AJA in S vs J 1989(1) S A 669(A) at 682 D, albeit in a different context) "it is imperatively called for." I do not think that any further attempt at defining when the impositon of the death sentence will be justified, can or should at this stage be made.
The principal mitigating factor advanced on behalf of first and second appellants was that they did third appellant's bidding out of fear for him. The fear was that

33/

33. he would kill them. Neither appellant gave any such evidence seeing that their version was a denial that they had set the hut alight. They relied, however, on their confessions. In that of first appellant there is the statement that third appellant said he would shoot him "like Khalazame Buthelezi ... if I refused" (to carry out the assignment). According to second appellant's confession third appellant threatened to "hire people to kill me ... if I tell anybody about this incident". The confessions having been admitted in evidence, appellants were, despite their repudiation of them, entitled to have these favourable portions considered (R vs Valachia and Another 1945 AD 826; S vs Yelani 1989(2) S A 43(A) at 49H -50C). But the question is: what weight should they carry? It was said that the State evidence itself lends some support to the contention that first and second appellants were intimidated by third appellant. It shows that third appellant (aged 48 and who was regarded by the

34/
34.

trial judge as a domineering personality) had threatened to harm Gama if he told anyone about what he had seen or heard on the night in question. Gama took the threat seriously. Mathe also feared third appellant. Thus when he was approached by third appellant to kill Elsie, he, as a subterfuge, agreed to do so; he felt that if he refused "I would be jeopardising my life". But in the result Mathe did not carry out what third appellant asked him to do. And the threat to Gama was merely to ensure his silence. Their situation is therefore hardly comparable to that of first and second appellants who were allegedly compelled to take positive action of a drastic kind. The State evidence relied on does not assist appellants. So I revert to their confessions. They were of course found to be reliable. But this finding was in respect of the incriminating portions of the statements. It does not follow that the exculpatory portions or those portions which might be regarded as evidence of extenuation must

35/

35. also be accepted (S vs Tovakepi 1973(1) S A 694 (RAD) at 695 D - H) . The confessions were not made on oath and were not subject to cross-examination. First appellant falsely denied making his confession. Second appellant falsely testified that he was told what to say. These are factors which very much detract from the cogency of those portions of the statement now relied on (S vs Yelani, supra, at 50 C - E). Besides, the confessions contain bald assertions. It is not stated in that of first appellant (who, like second appellant, is a person of mature years - both were aged 35) that he had or thought he had no option but to carry out his assignment. There is nothing about how imminent or real he regarded the danger to himself or whether he considered it impossible to extricate himself from his dilemma. A strong indication that he did not feel compelled to act as he did is the disclosure that third appellant promised to "give us

36/
36.

money". First appellant goes on to state that third appellant, when they returned to his shop paid him R400 and promised him a further amount of R1 100. Second appellant's confession, which is subject to the same criticisms, is even weaker on the issue of duress. It will be apparent from what has been quoted that the threat relied on was made after the incident and was that second appellant should not say anything about the murder. There is no reference to any threat of harm if he did not carry out the crime. He too was paid for what he did. The confession alleges that third appellant gave him R1 800 and promised to build a butchery for him. The law jealously circumscribes the defence of compulsion (R vs Samuel and Others 1960(4) S A 702 (SR) at 703 D). I think the same applies to where compulsion is sought to be used in mitigation. In the result, I do not think that either confession raised a sufficient factual foundation for the

37/

37. factor of coercion to operate in mitigation. First and second appellants must therefore be sentenced on the basis that they were willing participants in the crime. At the same time, however, the role played by third appellant cannot be overlooked. The fact that the crime was his brain-child and that first and second appellants became involved in it only because of him, is to some extent mitigating.
First and second appellants relied on certain other factors as being mitigating. One was that first appellant is a first offender and that second appellant has no relevant previous convictions. Plainly, these are mitigating features and must be taken into account. I think that the fact that they appear to be unsophisticated and poorly educated persons who have spent most of their lives in a rural area may also be regarded as mitigating. The only others worthy of consideration were the lure of financial reward which was promised and paid to them and

38/
38.

that in appellants' minds there allegedly was an "acceptable" reason for killing Elsie, namely, that she had hired people to kill third appellant. Both confessions state that this is the reason third appellant gave first and second appellant for him wanting Elsie murdered. In my view neither can have any effect favourable to first or second appellants in deciding whether the death sentence is the proper sentence. Both are prima facie inconsistent with the plea of coercion (and, in the case of first appellant, with his statement at the sec 119 proceedings that Elsie was killed because she had been stealing from third appellant's shop). Far from the promise of payment being mitigating it is, as counsel for first and second appellant conceded, an aggravating factor. The reason allegedly given by third appellant why Elsie was to be killed was by no means an acceptable one. Moreover, the confession contained no details of how appellants were

39/
39.

influenced by what third appellant told them, if they were influenced at all.
The issue of whether there were mitigating factors in the case of third appellant can be briefly disposed of. It was argued that his domestic tribulations constituted a mitigating factor; the State evidence showed that he believed that his mother-in-law was preventing him from finding his wife and thus redressing the situation; she caused him emotional stress which led to his decision to kill her; in the circumstances the reprehensibility of his conduct (as far as the murder of Elsie was concerned) was reduced. A similar argument was presented to SQUIRES J. The learned judge seems to have accepted that Elsie had refused to disclose to third appellant his wife's whereabouts and that she had thus frustrated third appellant. This was, I think, a finding over-generous to third appellant. It was founded on Mathe's evidence. But

40/
40.

all he said was "I assume (third appellant) had already
been to the Magagulas and the Magagulas were not prepared
to tell him where the wife was and that is why he wanted to
get rid of them". In any event I agree with the following
conclusion of the trial judge:

"(T)here is nothing in our view to show... that a response of the nature here involved and described above, was in any way appropriate to the situation. Had the accused done his reasonable best to find out from his mother-in-law his ex-wife's whereabouts, and been frustrated or refused in so doing, the circumstances calculated to make a reasonable person angry, one could understand such an argument and better receive it. The mere refusal to tell him of her daughter's whereabouts, accepting Mathe's evidence for this purpose, would not in our view reduce the moral blameworthiness of putting her to death in this way; and the situation is a fortiori in respect of the other deceaseds on count two to five. Their destruction was ordered for no other reason than that they were either somehow related to his ex-wife, or by living in the mother-in-law's house they were guilty by association."

As Mr Blomkamp, during the course of his able argument for

the State pointed out, the death of Elsie necessarily ended

41/

41. any hope of ascertaining from her where his wife was. His decision to kill her was an act of vengence.
There are many aggravating features about this case. And most of them apply to all three appellants. I have already referred to their motives. In the case of third appellant it was unreasoned vengeance. First and second appellents were hired killers acting for reward.
The crime was committed not impulsively but after a number

of days of plotting and preparation by third appellant.

First appellant too had a period of days to reflect on his

involvement; second appellant at least a few hours. What
is more, the burning down of the hut was the second attempt
made to murder Elsie that night. The initial intention
was to stab her to death. First appellant's confession
describes how third appellant provided him and second
appellant with so-called cane knives with which they were

to "chop" Elsie. Gama's evidence and second appellant's

42/
42.

confession corroborate this. It appears, however, that the plan was frustrated by Elsie remaining in her hut or by. the presence of "many people there". But this did not deter appellants. They persisted in their machinations. First and second appellants returned to third appellant's shop. They reported that they "were not successful". It was then that third appellant supplied them with the gas cylinders and paraffin. Their use in the manner contemplated was bound to and did achieve appellant's purpose. Appellants acted with dolus directus. The attack was a stealthy one. It took place on the victims' home in the middle of the night when they were in all probability asleep. The occupants must have been oblivious to and quite unprepared for it. As it turned out and as-appellants obviously realised, there was no chance of escape. The hut was a two-roomed one with only one outside door. It led out of not the bedroom but the

43/

43. kitchen (cum dining-room). It would seem that the fire was started in the kitchen. In short, the crime was ruthlessly executed in a cold-blooded manner. Consider too the callousness of appellants' subsequent conduct. Having escaped unobserved from the scene into the darkness, first and second appellants returned to third appellant's shop. There, as I have said, they reported what they had done. Third appellant's exact response, according to Gama, was "Thank you my boys, you have done the work" - and he then paid them for it. They have shown no remorse. I should add that third appellant has certain previous convictions but I think they can be left out of account.
Those then are the relevant mitigating and aggravating factors. What this Court has to decide is whether, with due regard to them, it is satisfied that the sentences of death are the proper sentences. I have come to the conclusion that the answer is in the affirmative.

44/

44. There are only aggravating factors as far as third appellant is concerned. This is particularly so in relation to the murders of the other two women and the two boys. The mitigating factors applying to first and second appellants are outweighed by the aggravating factors referred to. The cumulative effect of these aggravating factors is, in my opinion, such as to have made this a heinous crime. Of course, the element of heinousness must not, in the words of HOLMES JA in S vs Matthee 1971(3) S A 769(A) at 771 B - C, be emphasized out of perspective. The learned judge goes on (at 771 C - D) to set out what the relevant factors are in deciding whether, in a case of murder where extenuating circumstances had been found, the death sentence, as opposed to a (lengthy) period of imprisonment, should be imposed. Included in what has to be considered is whether a prison sentence would not be regarded by society as an adequate deterrent to others, the

45/

45. issue of rehabilitation and whether the evil of the accused's deed "is so shocking, so clamant for extreme retribution, that society would demand his destruction as the only expiation for his wrongdoing". This approach applies to the new sec 277. It takes into account the main purposes of punishment. I propose to confine the enquiry to the aspect of retribution. It is true that in recent times its importance has diminished. But it still features in deciding upon a proper sentence (S vs B 1985(2) S A 120(A) at 124 I; S vs J, supra, at 682 H). In my view, the circumstances of the present matter are such that the element of retribution must play a decisive role. They are such as to satisfy me that the above remarks of HOLMES JA apply and that in the case of each appellant the death sentence therefore was and is the proper sentence.

In the result, the appeals of all three appellants

46/
46.

against both their convictions and against their sentences are dismissed.

NESTADT JA

VAN HEERDEN, JA )
KUMLEBEN, JA ) CONCUR
NICHOLAS, AJA )
NIENABER, AJA )