Briers NO and Others v Salmon NO and Others (A63/2022) [2023] ZAWCHC 16 (14 February 2023)

IN THE HIGH COURT OF SOUTH AFRICA

(WESTERN CAPE DIVISION, CAPE TOWN)

Case No.:A63/2022

In the matter between:

GERALD HOWARD ROBERT BRIERS N.O. First Appellant

KATHLEEN JOY BRIERS N.O. Second Appellant

CARL HENRICUS BEEKMANS N.O. Third Appellant

DANIEL RETIEF VILJOEN N.O. Fourth Appellant

DANIEL ALBERTUS VILJOEN N.O. Fifth Appellant

ALWYN JOHANNES VERMEULEN N.O. Sixth Appellant

DANIEL ALBERTUS VILJOEN Seventh Appellant

and

HEDLEY JAMES SALMON N.O. First Respondent

PIETER ETIENNE DU TOIT N.O. Second Respondent

SUSAN JANE DU TOIT N.O. Third Respondent

and

MUTUAL AND FEDERAL LIMITED First Third Party

GVC INSURANCE CONSULTANTS CC Second Third Party

JUDGMENT DELIVERED ELECTRONICALLY ON 14 FEBRUARY 2023

MANGCU-LOCKWOOD, J

A. INTRODUCTION

[1] This is an appeal, with leave of the Supreme Court of Appeal (SCA), against a judgment of Papier J, in which he dismissed the appellants’ claim with costs. The appellants had brought a claim for damages caused by a fire which originated on the respondents’ farm called Lemoensdrif, and spread into their farms, namely The Oaks and Winterhoek-Wes.

[2] The appellants’ case was couched as follows in the amended particulars of claim:

“17. The fire spread to the Oaks and to Winterhoek-Wes as a result of the Defendant Trust’s negligence and/or breach of its legal duties as aforesaid, or the negligence of or the breach of its legal duties by its employees acting in the course and scope of their employment with the Defendant Trust, in that:

17.1 They started or allowed to be started an open fire or allowed an open fire to start in an uncontained and unsafe area of the farm;

17.2 They failed to immediately douse the fire once it had started and/or failed to have adequate measures at the time of starting the fire to prevent the fire from spreading.”

[3] As a result of the non-applicability of the presumption of negligence against the respondents in terms of section 34(1) of the National Veld and Forest Fire Act 101 of 1998, the appellants bore an onus to prove all the elements of the delictual claim.

[4] The cause of the fire is unknown. There was no direct evidence in that regard in the court a quo. Instead, the appellants relied on circumstantial evidence regarding the cause and place of origin of the fire, and sought inferences to be drawn from the evidence with regards to liability, negligence and causation. The court a quo held that the appellants failed to place before it sufficient objective facts in order for it to reliably and objectively draw inferences regarding the cause of the fire and the concomitant element of negligence.

[5] The issue arising in this appeal is whether the appellants placed sufficient objective facts for the court a quo to draw inferences in their favour.

B. THE LAW

[6] The general rule regarding the drawing of inferences is trite. The inference that is sought to be drawn must be consistent with all the proved facts; if it is not, then the inference cannot be drawn.¹ The position was summarised as follows in S A Post Office v Delacy and Another²:

‘The process of inferential reasoning calls for an evaluation of all the evidence and not merely selected parts. The inference that is sought to be drawn must be “consistent with all the proved facts. If it is not, then the inference cannot be drawn” and it must be the “more natural or plausible, conclusion from among several conceivable ones” when measured against the probabilities.’

[7] ‘Plausible’ in this context means ‘acceptable, credible, suitable’.³ It has also been stated that, where one or more inferences are possible, a court must satisfy itself that the inference sought to be drawn is the most plausible or probable, even if that conclusion may not be the only one. ⁴

[8] If there are no positive proved facts from which the inference can be made, the method of inference fails and what is left is mere speculation or conjecture. ⁵

C. THE APPEAL

[9] In essence, the appellants rely on the following as sufficient objective facts for the court a quo to have drawn inferences in their favour:

a. The respondents’ workers were working in the vicinity of the origin of the fire until approximately 15 minutes before the fire was detected;

b. The appellants’ witness, Mr Jan-Stephan Lombard (“Lombard”) encountered two ‘stragglers’ from the respondents’ workers, walking away from the area of the origin of the fire;

c. The respondents’ workers are smokers, and would sometimes relieve themselves in the veld.

d. Only the respondents’ workers were in the relevant area of the respondents’ farm on that day.

[10] It is common ground that until 12h30 on 17 March 2017 the respondents’ workers were working in two groups on Reierskop, an olive grove on the respondents’ property. One group was cutting branches on the far left of the grove and transporting them away. The other was straightening poles from the top of the field downwards.

[11] As was usual for a Friday, they all stopped working at 12h30 for lunch, and were picked up on a road along the middle of Reierskop, where Mr Manfred Carolus (“Carolus”) was present. His evidence was that ‘the last [he] saw before [he] left was everyone busy climbing on’ to one of the tractors to be taken home for lunch. The appellants dispute the evidence reproduced in the last sentence, and claim that the court a quo’s finding in this regard constituted an error. This is not so. The court a quo correctly found that all the workers had left the vicinity of Reierskop by 12h30.

[12] As for the so-called stragglers, Lombard’s account was that he encountered two men walking away from the direction of Reierskop, and he instructed them in Afrikaans to assist with the fire, but they continued to walk in their direction, effectively ignoring him. His evidence was that he gained the impression that they did not understand him. He also opined that they were casual workers.

[13] The two so-called stragglers were never identified in the evidence. Lombard’s description was that they wore top boots, were not wearing overalls or uniform, and in fact, that he could not say that they were farmworkers. Although it is not clear from the translated transcript, it was not disputed later that he also identified them as ‘black’ - apparently a distinction from the Coloured workforce which forms the majority on the farm.

[14] The evidence of Carolus was that the only two non-Coloured labourers on the farm, whose names are Mandizi and Thabo, were sitting under a tree behind another labourer’s house (Ashwell Luitjies (“Luitjies”)), some distance away from Reierskop, having lunch during the time that Lombard claims to have seen the two alleged stragglers. Carolus testified that he drove past them when he went to pick up Luitjies so that the latter could start the fire bakkie.

[15] Carolus was challenged during cross examination regarding his recollection that he saw the two workers behind Luitjies’ house. The thrust of the challenge was to question how he was able to remember events which occurred so long ago, and also given that the two workers continue to sit under the same tree during lunch. He remained consistent in his evidence, stating that this was the only fire that had befallen his employer in his eight years of employment there, and that his eyesight had been affected by the fire. He also stated that he had occasion to revisit the events of that day when the respondents’ attorney paid him a visit in preparation for the trial some months before the trial. There is no basis upon which to reject his evidence in this regard. In any event, the evidence was that the land at Reierskop had lain fallow for over two years, so the work that was conducted there during that week was not an everyday experience for him and the workers.

[16] The evidence from both Mr Christiaan Jacobus Joubert (“Joubert”) and Carolus was further that, although the two workers’ first languages are isiXhosa and seSotho, respectively, they understood Afrikaans, which is the language used on the farm to communicate with them. This is yet another suggestion that the two persons encountered by Lombard were not Mandizi and Thabo. Another is Joubert’s evidence that the two were in fact permanently employed on the farm - not casual workers as suggested by Lombard. In other words, whatever gave Lombard the impression that the two men he encountered were not permanent farmworkers means that, whoever they were, they were not Mandizi and Thabo who are permanently employed by the respondents. But in any event, even on Lombard’s evidence, the two people that he saw could not be identified as farmworkers, let alone as respondents’ workers. At most, they wore top boots and were on the respondents’ property.

[17] This raises another string to the appellants’ bow, namely that there was evidence that no other people were in that specific area of the respondents’ property on that day. It would be remarkable if anyone were aware of every person present on the respondents’ property throughout the day, it being a sizeable commercial farming enterprise. It would be even more remarkable if any workers were aware of who was present in the vicinity of Reierskop after they had left that area for lunch. There was no evidence regarding access and egress to the respondents’ farm on that day.

[18] Despite the pleaded case in the amended particulars of claim, the appellants mounted a case that the fire was probably started by a worker who was a smoker, and whose lit match or cigarette ember must have started the fire (“the lit match/cigarette ember theory”). An obvious observation is that this new case appears to be at variance with the pleaded case which suggests a fire that was deliberately “started or allowed to be started”.

[19] In an attempt to draw an inference that the fire was probably started by the respondents’ workers who were smokers, the appellants place great reliance on the cases of Barker v Venter⁶ and Viljoen v Smith⁷. However, both cases are distinguishable from the facts of this case. In Barker v Venter there were witnesses who saw the individual identified as Bossie arriving at a certain spot and sitting or lying down. Almost immediately, the three witnesses witnessed a fire next to where Bossie was sitting. They also saw him trying to douse the fire unsuccessfully. Thus there were sufficient facts placed before the court regarding the origin of the fire.

[20] The case of Viljoen v Smith is similarly distinguishable. There, the culprit named ‘Links’ admitted to having started the fire to the investigating officer, the Magistrate’s Court, and the insurance representative, and pleaded guilty to the charges against him. In all three investigations he gave the same version of how the fire had started, namely when the head of his lit match fell to the ground. He was convicted based on this version, but later tried to recant it, which the court refused. There was again clear evidence of how the fire had started and where it had started.

[21] That is not the case here. There are no direct witnesses as to how the fire started. It was not established who exactly amongst the workers was a smoker, and where those persons were at the time that the fire is estimated to have started. This includes the two so-called stragglers who remain unidentified, and regarding whom there remains no evidence that they were in fact smokers. There was also no evidence that any workers smoked on that day.

[22] The appellants claim that the fact that the respondents’ workers were working close by, in time and in proximity, is sufficient for the court a quo to have applied inferential reasoning regarding the cause of the fire instead of insisting on direct evidence. They sought to draw parallels with the cases of Van der Eecken v Salvation Army and Clan Syndicate (Pty) Ltd v Peattie and Others⁸ in this regard. However, neither case assists the appellants.

[23] In the case of Van der Eecken v Salvation Army, it was established that the first fire had erupted after an employee of the defendant (“White”) started a controlled fire to burn an area of the defendant’s property referred to as an embankment. After the wind picked up, the fire was uncontrollable, and the local fire services attended at the property and extinguished the fire. Or so they thought. After the fire brigade left the scene of the fire, the wind picked up considerably and a second fire erupted approximately 5 to 6 metres north of the fire line. The evidence was that, because they were aware of the unpredictable nature of fires, when the fire brigade personnel left the scene of the fire, they instructed White and assistants to keep an eye on the fire line. However, White had left the fire line and gone to a dam nearby. It was because of all this evidence that the court concluded that the spatial and temporal relationship between the two fires led to a conclusion that the second fire started as a result of the first fire not having been extinguished effectively and because of the increase in the velocity of the wind. There is no such similarity to be made to the facts of this case because it has not been established that anyone started a fire, whether by a lit match or the ember of a cigarette or any other manner. Those allegations remain theory. Rather, the evidence in this case established that the workers had already left the vicinity of the origin of the fire by the time that the fire was detected.

[24] As for the case of Clan Syndicate (Pty) Ltd v Peattie and Others, the starting point is that unlike the present case, the appellant there bore a statutory onus to prove that its employees were not negligent either in causing the fire to start or in failing to prevent it spreading onto a neighbouring farm. In discharging that onus, the appellant’s case was that the fire was caused by emissions of carbon particles from a chainsaw that was used by one of its labourers. The appellant’s case was supported firstly by the first eyewitness to detect the smoke, one Magagula, who was across the road watching a co-labourer (Ndlela) while the latter was felling trees with a chainsaw, when he detected smoke four or five paces behind Ndlela. Magagula testified that his first impression was that the fire had been started by the chainsaw because he had experienced this once before. His evidence was also that Ndlela was a non-smoker. The appellant also called an expert witness who explained that it was possible for carbon emissions from a chainsaw to cause a fire, and how this was possible. It was on the basis of this evidence that the court found that the most probable cause of the fire was the operation of the chainsaw.

[25] Again, the facts of the present case are not comparable with the facts in Clan Syndicate. The appellants here failed to establish their ‘lit match/ cigarette ember’ theory by any means. The appellant relies on a statement made by the court in Clan Syndicate, that there was no suggestion in the evidence that the fire could have been caused otherwise than through human agency. This is correct. However, the context for this statement is instructive. There was an eye-witness to the start of the fire who immediately gained a strong impression about the cause thereof. As a result, there were two possibilities postulated throughout that trial, namely that Ndlela was a smoker, and that the chainsaw was the cause of the fire. But Magagula’s suspicions were confirmed by expert evidence. Bearing in mind that the appellant bore the onus, it is for this reason that the court observed that: “…if it is postulated that the fire was caused neither by one of the appellant's employees having smoked in the forest (which would clearly have constituted negligence) nor by the operation of the chainsaw, it would inevitably follow that the appellant had failed to furnish any explanation at all concerning the cause of the fire, which, in turn, would lead inevitably to the conclusion that it had failed to rebut the statutory presumption of negligence.”⁹

[26] Given that it is the appellants in this case who bear an onus regarding liability, one would have expected them to similarly establish sufficient facts regarding their ‘lit match/cigarette ember’ theory. In response to this criticism the appellants’ counsel has repeatedly argued that it is not necessary to prove that matches or cigarettes can cause a fire. Given the gaps in evidence already discussed above, this response calls for conjecture. There was no expert evidence presented regarding any investigations conducted after the fire or regarding the cause thereof. There was no evidence relating to physics of ignition and fire propagation. In those circumstances, the court a quo correctly observed that the appellants were calling upon it to speculate. For the same reasons, there was no obligation upon the respondents to place facts before the court to dispel the appellants’ theory that it was a smoking worker who started the fire.

[27] The appellants’ ‘lit match/cigarette ember’ version faces particular difficulty when regard is had to where the fire is said to have originated. According to Lombard, when he arrived to inspect the fire, it had spread across the road, such that he could not drive past. He continued that all the vegetation nearest to the road, both to the west and the east, had already been burnt, including a ‘waboom’ which became a focal point during the evidence. He estimated the fire to have originated approximately 20 to 25 metres from the edge of the road. The implication of this evidence is that the fire had originated closer to the road separating the fynbos and Reierskop, and was accordingly close to where the workers had been working earlier. Close enough, it is suggested by the appellants’ counsel, for a worker to have thrown a lit match or smouldering cigarette or a ‘twak pil’.

[28] However, both Carolus and Joubert testified that they had to climb over a big boulder, and walk across very dense fynbos - pictures of which were in the record - in order to reach the fire. They otherwise could not see the fire from the road. They estimated that the fire was a distance between 25 metres and 30 metres from the big boulder, and the big boulder is some 5 metres from the road. Both testified that the fire was not burning against the road, and had not burnt the wild veld between the road and the fire. Neither had the fire reached the west side of the road.

[29] In reconciling the two versions, Lombard’s evidence is instructive. He volunteered that the fire he witnessed in the road and close to the road was backburn - a fire that was burning slowly against the wind. In other words, the fire could not have started where he witnessed it. It must have travelled there from elsewhere. In this respect, his evidence ties in with the evidence of Joubert and Carolus, that they allowed a backburn to occur towards the road from where they encountered the fire. The most probable origin of the fire is therefore the area where Carolus and Joubert encountered it, after they both went into the veld, over the big boulders and stood within approximately 2,5 metres of it.

[30] This, in turn, raises the question of when Lombard arrived at the scene of the fire. There are various indicators from the evidence that he arrived later than he suggests, bearing in mind that no witnesses were watching the time, and were all working on an emergency basis throughout the events of that day. The stage at which Lombard found the fire, namely of having burned backwards into the road, is later than the stage at which Carolus and Joubert found it. They did not find the fire to have advanced that far backwards, but were involved in allowing the backfire that Lombard found upon his arrival. Another indicator is the state in which he found the ‘waboom’ and surrounding vegetation. According to him, the ‘waboom’ and the vegetation surrounding it had been burnt completely, with no flames on it. By contrast, Joubert and Carolus did not find any vegetation close to the road as having been burnt, including the ‘waboom’.

[31] As for the broken down fire-fighting bakkie which contained a ‘bakkie sakkie’, Lombard’s evidence was that he came across it on his way down from Reierskop, although he could not remember at which point this was. It was established that this fire bakkie had taken the same route as Lombard to eventually reach the place where it broke down. In other words, the bakkie, which was burdened with a heavy load of a filled water tank, must have made it there by the time that Lombard came down from Reierskop, having traversed rough and steep uphill terrain to reach a point very close to the fire. The evidence was that tractors move incredibly slowly up the hill towards Reierskop. Carolus’s evidence, on the other hand, was that he encountered the broken down bakkie on his way back from Reierskop when he went to help the deceased Eben Burger with his tractor. By then, according to Carolus and Joubert, they had climbed over the big boulders at Reierskop and navigated through the thick and tall vegetation to get as close as possible to the fire; moved back to the road; and agreed that Carolus could go and investigate the delay with the fire bakkie and to assist Eben Burger with his tractor. All of this occurred while Joubert remained at Reierskop.

[32] The balance of probabilities is therefore that Lombard arrived at the point of origin of the fire well after Carolus and Joubert had first inspected it. This conclusion is supported by the evidence of Joubert, who states that he, like Lombard, initially only saw smoke from his house when he was first alerted about the fire. Contrary to the criticisms levelled during his cross examination, there is nothing illogical about his view that he first wanted to conduct his own inspection of the fire before he sounding an alarm and mustering support from the neighbourhood. After all, Lombard himself adopted the same approach after first sighting smoke from his driveway. The probabilities are that he must have phoned Lombard after his inspection of the fire at Reierskop.

[33] This conclusion is also consonant with Lombard’s evidence that he took the long route usually used by tractors to reach the fire, after receiving a phone call from Joubert, because he was not familiar with the respondents’ farm. By contrast, Joubert and Carolus took a shortcut which is only accessible to motorbikes to go and inspect the fire.

[34] What is more, Lombard and Carolus testified that when Lombard arrived at Biesiesbult, the workers of Lemoendrif were already there with their fire equipment, and amongst them was Carolus. In other words, the respondents had already made a determination, ahead of Lombard, to go to Biesiebult to fight the fire. In my view, this puts paid to the appellants’ version that Lombard arrived at the point of origin of the fire ahead of Carolus and Joubert. His evidence was that he did not stay at Reierskop, but that he immediately turned around when he encountered the fire in the road. Like him, the respondents would have had to inspect the fire at Reierskop, and realise that it was headed towards Biesiesbult, in order to advance to Biesiesbult with their resources.

[35] It was suggested that Carolus and Joubert tailored their version, implying that it was fabricated or manipulated. Unfortunately, the court a quo did not make any observations regarding any of the witnesses’ demeanour or regarding their credibility. From a reading of the record, there is no reason to believe this accusation. Both Carolus and Joubert readily admitted that some of the respondents’ workers were smokers and that they sometimes smoke while at work. They both gave their evidence in a cogent and reliable manner. In the case of Carolus, it was not disputed that his eye-sight was permanently affected by the smoke on that day. There was no basis to conclude that he had reason to fabricate a story in favour of the respondents. By contrast, it was clear from the evidence that Lombard was not well-orientated with the respondents’ farm, and at times, could not say where he drove. As a result, his evidence in relation to his time of arrival at the point of origin of the fire is not reliable. In any event, as I have found above, the balance of probabilities favours the version proferred by the respondents.

[36] Lastly, there remains the issue of vicarious liability. The appellants bore an onus to also prove that the persons who started the fire were acting within the course and scope of their employment. The evidence did not even come close to discharging this aspect of the onus.

[37] In the circumstances, the appellants’ appeal is dismissed with costs, including costs of counsel.

_________________________

N. MANGCU-LOCKWOOD

Judge of the High Court

I agree and it is so ordered.

_________________________

R ALLIE

Judge of the High Court

I agree.

_______________________

C M FORTUIN

Judge of the High Court

APPEARANCES:

For the appellants : Adv R van Riet SC

Adv A D Brown

Instructed by : Mr G van Niekerk

De Klerk & Van Gend Inc

For the respondents : Adv Le Grange SC

Instructed by : Mr J Theron

Werksmans Attorneys

For the first third party : Adv W Duminy SC

Instructed by : BDP Attorneys