Safire Crop Protection Co-Operative Ltd v Normandien Farms (Pty) Ltd (AR246/2023) [2024] ZAKZPHC 115 (29 November 2024)

IN THE HIGH COURT OF SOUTH AFRICA

KWAZULU-NATAL DIVISION, PIETERMARITZBURG

Case no: AR246/2023

In the matter between:

SAFIRE CROP PROTECTION CO-OPERATIVE LTD APPELLANT and NORMANDIEN FARMS (PTY) LTD RESPONDENT

Coram: Olsen J, Mossop J and Nicholson AJ

Heard: 25 October 2024

Delivered: 29 November 2024

ORDER

On appeal from: KwaZulu-Natal Division of the High Court, Pietermaritzburg (P Bezuidenhout J sitting as court of first instance):

1. Save to the extent set out in paragraph 2 of this order, the appeal against the judgment of the trial court handed down on 26 January 2023 is dismissed with costs.

2. Paragraphs 1(i) and 1(ii) of the order of 26 January 2023 are set aside and are replaced with an order declaring the appellant to be liable to the respondent for such damages as the respondent may prove at a hearing in due course arising out of the fire that occurred on the farm Albany, situated in the Newcastle district, KwaZulu-Natal, on 7 November 2015.

JUDGMENT

MOSSOP J:

Introduction

[1] The respondent in this appeal is a private company that owns several farms. One of those is the farm Albany (Albany), which is situated in the district of Newcastle, KwaZulu-Natal and which is primarily planted to timber. On 7 November 2015, there was a fire on Albany (the fire) and just over 300 hectares of trees were consumed by the conflagration that swept through portions of the farm.

[2] The appellant is a co-operative company and is the insurer of the respondent, having issued it with a document titled ‘Plantation Fire Insurance Certificate’ (the certificate) and which covered Albany. The appellant consequently received the respondent’s insurance claim for the damages to Albany that were occasioned by the fire. After having the claim for approximately eight months, and after the period of cover of the certificate had lapsed, for it was an annually renewable certificate, the appellant repudiated the respondent’s claim.

[3] The respondent consequently sued the appellant for the alleged value of the loss that it sustained. The trial came before P Bezuidenhout J, who, after hearing evidence over a protracted period for reasons that will be mentioned shortly, delivered a judgment in favour of the respondent, and directed the appellant to make payment to it as claimed. The appellant now appeals against that judgment.

The pleadings

[4] In its particulars of claim, the respondent pleaded that the certificate indemnified it against any loss by fire to standing

timber on Albany. It pleaded that there was such a fire on Albany on 7 November 2015, that standing timber to the value of R14 385 720.84

was destroyed by the fire and that despite complying with all its contractual obligations, the appellant had unlawfully repudiated its claim.

[5] The appellant denied the occurrence of the fire in its plea,[1] but admitted issuing the certificate to the respondent. It pleaded that any material misrepresentation, misdescription or non-disclosure

by the respondent rendered the certificate voidable by it at its sole discretion. Such a misrepresentation had occurred when the respondent represented to it that the fire had originated in compartment A13a (compartment A13a) of Albany when, according to the appellant, it had originated in, and spread from, a sawdust and timber waste area situated on the north-western boundary of that same compartment. The sawdust and timber waste area is, in fact, not a single area, but is comprised of two sawmill waste sites, which I shall refer to collectively as ‘the sawmill waste sites’ and, where a single waste site is referred to, I shall refer to it as either sawmill waste site 1 or 2, as the case may be. Therefore, by virtue of that misrepresentation, the appellant contended that it was entitled to reject the respondent’s claim arising out of the fire on 7 November 2015.

[6] In addition, the appellant pleaded that the respondent had permitted the dumping of sawdust and timber waste in the sawmill waste sites, thereby substantially increasing the load of combustible material in the vicinity of the insured area and increasing the fire risk. The appellant pleaded that the existence of the sawmill waste sites was not disclosed to it by the respondent and this non-disclosure materially affected the risk indemnified by it and afforded it a further ground upon which it was entitled to reject the respondent’s claim.

[7] Two additional grounds allegedly entitling the appellant to avoid the certificate were pleaded, but as they were later not relied upon by the appellant at the trial, there is no need for them to be considered. Further pleadings in the form of a replication,

rejoinder, and a surrejoinder were delivered by the parties but it is also not necessary to consider them either given the tight focus of the grounds of appeal.

The trial and the judgment

[8] The trial commenced on 13 November 2018 and judgment was finally delivered on 26 January 2023, the length of the trial being unnaturally prolonged by the occurrence of the ruinous COVID-19 pandemic. During the period over which the pandemic held this country in thrall, both the presiding judge and senior counsel appearing for the respondent contracted the COVID-19 virus, became seriously ill, and had to be hospitalised. There was thus a forced interregnum in the trial that covered a period of approximately three years.

[9] The judgment ultimately delivered by the trial court (the judgment)[2] reads as follows:

‘1. Defendant is ordered to pay to Plaintiff

(i) The sum of R14,385,720.84 (14 (sic) million, three hundred and eighty-five thousand, seven hundred and twenty rand and eighty-four cents)

(ii) Interest on the said sum at the rate of 9 per cent per annum from date of service of summons to date of payment.

(iii) Costs of suit such costs to include the costs of senior counsel where appropriate.’

[10] The judgment thus upheld the respondent’s claim in its entirety.

The error

[11] There is an unfortunate error in the judgment. When the trial commenced, the parties requested the trial court to grant an order in terms of the provisions of Uniform rule 33(4), separating the issues of liability and quantum. Such an order was duly granted. Only liability was thereafter to be determined by the trial court.

[12] However, as appears from the judgment, the trial court ordered the appellant to pay the respondent the exact amount claimed by it in its particulars of claim. It clearly ought not to have done so, nor could it have done so, for it had heard no evidence at all on the quantum of the loss allegedly sustained by the respondent and the appellant did not admit in its plea the quantum of the damages claimed by the respondent.

Leave to appeal

[13] The appellant consequently sought leave to appeal against the judgment on the grounds that the trial court had erred in three principal respects:

(a) It determined quantum when it ought not to have done so;

(b) It incorrectly found that the alleged misrepresentation by the respondent’s servants as to the point of origin of the fire on Albany did not entitle the appellant to repudiate the respondent’s claim; and

(c) It incorrectly concluded that the existence and presence of the sawmill waste sites on Albany were not material to the risk insured against by the appellant and therefore the respondent was not required to disclose their existence to the appellant.

[14] Leave to appeal was granted to the appellant by the trial court on these defined grounds.

[15] The respondent itself sought leave to cross-appeal from the trial court, notwithstanding the fact that judgment was granted in its favour. It contended that the trial court had misdirected itself in finding that the fire had, indeed, originated in the sawmill waste sites and had spread from there into compartment A13a of Albany and contended that it ought to have found that the fire started in compartment A13a. Leave to cross-appeal was granted by the trial court. However, before us, counsel for the respondent, Mr Roberts SC, indicated that the respondent no longer intended pursuing its cross-appeal. This was, in my view, the correct decision as no appeal lies against reasons for a judgment. Nothing further need be said about the cross-appeal.

The layout of Albany

[16] It may be helpful to provide a brief description of the location of the principal features and landmarks of Albany that have relevance to the determination of this appeal to which some reference has already been made, and to which more will be made.

[17] The sawmill owned by the respondent is situated on the extreme eastern flank of Albany and was constructed there in either 2003 or 2004. The two sawmill waste sites previously mentioned are located on the western flank of the farm, almost opposite the sawmill, but separated from it by a distance of approximately 1.75 kilometres,[3] as the crow flies. Sawmill waste site 1 is located north of sawmill waste site 2 but is very close to it, approximately 60 metres

away.[4] Both sawmill waste sites are separated from the main part of Albany by an unnamed dirt road that runs along the western boundary of the farm (and elsewhere) in a roughly north/south direction at that point. During the trial, this road was referred to as the ‘blue road,’ a blue line having been applied to it on the large aerial photograph to highlight its sinuous presence. I shall continue to refer to it as ‘the blue road.’

[18] The sawmill waste sites in the west are connected to the sawmill in the east by an entanglement of dirt roads that have been incised out of the compartments that sit between the western and eastern flanks of Albany. Immediately to the east of the two sawmill waste sites, and thus between them and the sawmill, is compartment A13a, which covers approximately half the distance between the two sawmill waste sites and the sawmill itself. Beneath compartment A13a, that is to the south of it, is compartment A13b, which also stretches approximately half the distance between the sawmill waste sites and the sawmill.

[19] As the ubiquitous crow flies,[5] approximately 800 metres due south of sawmill waste site 2,[6] being the southernmost of the two sawmill waste sites, is a fire tower (the fire tower), from where the first observation of the

emergence of the fire was made. The fire tower permits observations to be made northwards into the plantations on Albany.

The sawmill waste sites

[20] A central feature of the trial, and the appeal, was the sawmill waste sites. The sawmill generated sawdust from its operations and whilst some of that sawdust was disposed of either through sales or by incineration at the sawmill, not all of it was. The residue of the sawdust, together with other materials mentioned below, were transported westwards from the sawmill to the sawmill waste sites for disposal.

[21] Mr Dennis Pretorius (Mr Pretorius), who was at the time of the fire the respondent’s head of forestry harvesting on Albany, testified at the trial that he was also in charge of transporting a mixture of materials from the sawmill to the sawmill waste sites for disposal. As to what was deposited there, Mr Pretorius said that it was comprised of:

‘… saagsels, bas en dan grond en quarry klip’ (sawdust, bark, soil, and quarry stone).

These materials were collectively referred to by him as ‘the scrapings’. I shall also use that term to describe this mixture of materials. His evidence was corroborated by the evidence of the chief executive officer of the respondent, Mr Lawrence Hoatson (Mr Hoatson), who stated that the materials deposited in the sawmill waste sites comprised:

‘… clinker, sawdust, wood chips, log yard scrapings which should include bark and so on. In summer, a huge percentage of mud would be in there.'

[22] Mr Pretorius described the area in which the sawmill waste sites were located as being a ‘vallei’ (valley). Mr Hoatson used the term ‘donga,’ which is a steep-sided gully created by soil erosion usually caused by the action of running water, but which is now dry,[7] meaning that the area of land was craggy and irregular in form and, essentially, not otherwise productively usable. Mr Pretorius explained that the scrapings would be placed in a heap in the valley but would then be flattened with a grader. The stated purpose behind doing so was to rehabilitate that piece of land by filling in its deformities, gradually levelling the land out and, finally, collapsing the sides of the valley inwards to form a flat, usable surface upon which more trees could be planted in the future.

[23] After a load of scrapings was placed in the sawmill waste site and was graded flat, it would then be sealed with a further layer of soil and the whole deposit would be compacted, according to the evidence of Mr Pretorius. This procedure was followed with each deposit of scrapings. This evidence was not seriously challenged by the appellant.

[24] I turn now to consider the three grounds of appeal.

The appellant’s first ground of appeal

[25] The appellant submitted that the trial court incorrectly gave judgment on the issue of quantum. This excites no controversy because the respondent acknowledges this to be the case, and it abandoned the money judgment granted in its favour by the trial court when the application for leave to appeal was argued.

[26] The trial court clearly erred in giving judgment as it did, and the first ground of appeal must therefore be upheld and the order directing the appellant to pay the respondent the amount of R14 385 720.84, together with interest, must be set aside.

[27] Counsel for both sides agreed before us that the true issue before the trial court was whether the appellant was liable in law to the respondent. Counsel were further agreed that the judgment that ought to have been delivered should have been in the form of a declaratory order determining the appellant’s liability, if any, to the respondent.

[28] Implicit in the judgment of the trial court directing the appellant to make payment to the respondent is an unexpressed finding by the trial judge that the appellant was, indeed, liable to the respondent, as payment would not have been ordered if there was no liability on the part of the appellant to do so. It is upon the understanding that the trial court intended to answer the issue of liability by finding the appellant to be liable to the respondent that the remaining issues in this appeal will be considered.

The appellant’s second ground of appeal

[29] The appellant contends that the respondent represented to it that the fire originated in compartment A13a and that such representation was false. It contended in its plea that the fire, in fact:

‘… originated in and spread from a sawdust and timber waste area situated on the north western boundary of compartment A13a.’

[30] From this, it is evident that the appellant did not distinguish between the separation of the two sawmill waste sites and regarded them as a single waste site. However, from the evidence that the appellant itself led at the trial, about which more later, it is apparent that what was alleged was that the fire emanated from sawmill waste site 1, the northernmost of the two sawmill waste sites.

[31] How the misrepresentation was made was not specifically pleaded by the appellant. It appears to me to be reasonably safe to assume that the representation relied upon by the appellant is to be found in the respondent’s claim form submitted to it. The respondent’s claim for indemnification for its losses was submitted to the appellant on 13 November 2015, six days after the fire occurred. This was done in the form of a written claim on a pre-printed document prepared by the appellant that featured the appellant’s letterhead and logo on the first page, and the claim form appeared to contain five pages. The document was comprised of a mixture of questions that had to be answered, in some instances by ticking either a box marked ‘yes’ or ‘no’ or one or more boxes showing multiple options, and questions which had blank lines where explanations by the respondent had to be inserted, as well as questions which were a hybrid of these two types of questions.

[32] Paragraph 5 of the claim form required the respondent to state the cause of the fire and its reasons for that statement. No answer was inserted in the claim form submitted to the appellant by the respondent. Paragraph 12 required the respondent to identify the compartment first affected or threatened by the fire. The answer inserted by the respondent was ‘A13a’. At page 4 of the claim form, the question was asked as to the cause of an uncontrolled fire. The fire in question was undoubtedly such a fire. Many potential answers were listed in boxes in relation to this question, which had to be identified and ticked if appropriate. Under the grouping ‘accuracy of ignition’ a number of alternatives were identified. The choices available were ‘definite’, ‘probable’, ‘unknown’ and ‘alleged arson’. The respondent marked the box reading ‘unknown’.

[33] The certificate issued by the appellant contained clauses that would permit the appellant to avoid liability in certain circumstances. Clause 1 of Section A of the certificate reads as follows:

‘Misrepresentation, misdescription or non-disclosure in any material particular shall render voidable the particular item, section or subsection of the Certificate, as the case may be, affected by such a misrepresentation, misdescription or non-disclosure.’

[34] In addition, clause 8 of the same section of the certificate reads as follows:

‘If any claim under this Certificate is in any respect fraudulent or if any fraudulent means or devices are used by the Member or anyone acting on their behalf or with their knowledge or consent to obtain any benefit under this Certificate, or if any event is occasioned by the wilful act or with the connivance of the Member, the benefit afforded under the Certificate in respect of any such claim shall be forfeited.’[8]

[35] It appears to me that clause 8 refines the concept of a misrepresentation and requires it to be a fraudulent misrepresentation. To establish a fraudulent misrepresentation, there must be evidence of a statement[9] that was false[10] and that such misrepresentation was intended to induce the party receiving it to act.[11] A fraudulent misrepresentation is thus a false assertion of a fact that most commonly takes the form of spoken or written words, although in some instances it may be inferred from conduct. Whether a statement is false depends on the meaning of the words in all the circumstances, including what may fairly be inferred from them.

[36] Whether the respondent misrepresented where the fire first commenced burning is a factual question that requires an assessment of the evidence called by both sides at the trial. If the fire did start in the sawmill waste sites and the respondent said, falsely, that it had started in compartment A13a to claim a benefit under the certificate that it might otherwise be denied, then there may well have been a misrepresentation by the respondent. On the other hand, if the fire did start in compartment A13a and this is what the respondent told the appellant then, clearly, there could have been no misrepresentation. In considering this issue I bear in mind that the findings made by a trial court are presumed to be correct and will only be disturbed by an appeal court if the evidence shows them to be clearly wrong.[12]

[37] At the commencement of the appeal, the proposition was put to Mr Troskie SC, who appeared for the appellant both at the trial and before us, that for the appellant to succeed on this ground, it would have had to have established the true point of origin of the fire, on a balance of probabilities. In other words, it would not be sufficient to merely suggest that the fire did not start in compartment A13a: the appellant would have to establish that it started in the sawmill waste sites. Mr Troskie agreed with the proposition.

[38] As a matter of fact, the trial court found that the fire did originate in the sawmill waste sites. Its reasoning in this regard was encapsulated in this portion of its judgment:

‘[114] ... The char marks and the angle thereof is indicative that the fire burnt in a westerly direction[13] from west to east and therefore it would appear on the probabilities that it commenced at the sawdust heap. There was no dispute about the direction of the wind and no credible evidence that contradicted the conclusions drawn from the char marks. It entered the compartment A13(a) and from there due to the wind of at least 20 km per hour in a westerly direction kept on burning up the compartment away from the sawdust heap. I accept for the said reasons that there was no back burning as this does not appear from the angle of char. If the fire had started in compartment A13(a) as testified to by Pretorius the fire must have burnt against the wind. There was no evidence to support this possibility as the wind speed was about 20 km per hour or more at times. There was no evidence direct or indirect that the fire burnt from inside compartment A13(a) towards the dump site. Although Pretorius stated that the fire started in A13(a) he did not testify that he saw it burning against the wind. The V shape also did not show burning towards the west.

[115] Considering all these factors and especially the direction of the wind the charring on the remaining trees and the fact that immediately after the fire plaintiff commenced extensive work on the sawdust heap, it has been proved on a balance of probabilities that the sawdust heap must have been the origin of the fire.’

[39] Before considering this finding, it is necessary to make a general observation about the evidence led at the trial. For obvious reasons, only the respondent led witnesses at the trial concerning the events of, and the observations made on, 7 November 2015. This is perfectly understandable because the appellant had no way of predicting that the fire would occur and only had a representative visit Albany, for the first time, on Monday, 9 November 2015, two days after the occurrence of the fire. But that did put the appellant at somewhat of a disadvantage regarding what was alleged to have occurred and what was alleged to have been observed on the day of the fire. The appellant could in such circumstances only rebut the evidence of direct observations made by witnesses who were present with the ex post facto evidence of an expert advancing theories about what could, might, or should have happened. While such expert evidence is not inadmissible, the theories presented must, however, account for, and accommodate, all the known facts. The appellant led such evidence, and I shall consider it shortly.

[40] The respondent’s case was that the fire had commenced inside compartment A13a, burnt eastwards away from the blue road but also burnt back towards the blue road. Its case was that the fire had not first commenced in the sawmill waste sites and had then been carried from there into compartment A13a.

[41] To establish that the fire commenced inside compartment A13a, the respondent led the evidence of three eyewitnesses at the trial. I do not deal with this evidence in the sequence in which it was presented at the trial. Mr Paul Saayman (Mr Saayman), was the manager of forestry culture activities[14] on Albany and lived on the farm. He testified that on the morning of the fire he was on another farm owned by the respondent, namely

Buffelshoek, inspecting some grass, when he received a report of a fire on Albany. He returned to Albany and went to compartment A13a, where he observed the fire and tried to extinguish it, but was not able to do so. He believed that the fire started in that compartment and marked the spot where he believed it began on the large aerial photograph used at the trial.

[42] The next witness called by the respondent on this issue was Mr Pretorius, who testified that he was attending to some cattle elsewhere on the day of the fire when he first learnt of it over the open channel radio in his motor vehicle. He hastened to Albany and when he arrived there, he described his observations in this fashion:

‘Toe ek op of by A13A kom toe kon ek sien waar die vuur onstaan het. Dit was omtrent van die pad af ongeveer 10 tot 15 meter van waar ek kon fisies sien waar die vuur begin het. Ek het toe van daar af met my voertuig in beweeg in die kompartement in om die vuur te begin blus.’

[43] Mr Dumisani Mfusi (Mr Mfusi), who is the head of the respondent’s sawmill operation, was the third witness to testify about where he first saw the fire. His home is situated on Albany, and that is where he was on the morning of Saturday, 7 November 2015, when he also heard the report of the fire on the open channel radio that he kept at his home. He quickly gathered a firefighting team of 10 persons and approached the area where the fire was said to be by driving on the blue road that passed by the sawmill waste sites. The following exchange occurred when he was led in his evidence in chief by Mr Roberts:

‘Did you see any fire? --- Yes, there is that I saw.

Where did you see the fire? --- It was inside the plantation.

When you saw it, where were you, were you on the road or were you in the plantation or what? --- I was on the road.

Where in relation to the sawdust were you? --- It was a bit far, about 20 metres.

Where in the plantation did you see the fire? --- It was inside.’

[44] The appellant had no evidence to rebut the observations of these three witnesses for the reason previously mentioned. That, however, did not necessarily mean that the fire did not first exhibit itself in the sawmill waste sites. Notionally, it could have first burnt there and then, through the occurrence of spotting,[15] ignited compartment A13a where Mr Saayman, Mr Pretorius and Mr Mfusi first observed its naked flames. But the respondent also led the evidence of four eyewitnesses who testified that there was no fire, in any form, in the sawmill waste sites shortly before, or when, the fire was first observed in compartment A13a.

[45] The first of those witnesses was Ms Ntombifuthi Kubheka (Ms Kubheka), an employee of the respondent, who was assigned to duties on the fire tower on the morning of 7 November 2015. She testified that she had walked past the sawmill waste sites at 05h35 on the morning of the fire, en route to the fire tower to commence her duties at 06h00. Her evidence in chief on what she observed then was the following:

‘The route that you took that particular morning from where you were walking, could you saw (sic) the sawdust there? --- Yes, I could see it.

Did you see any smoke? --- No, I did not see anything.’

[46] Ms Kubheka first raised the alarm about the fire via a radio broadcast that she made from the top of the fire tower at approximately 08h45 that morning. This was the radio message that Mr Pretorius and Mr Mfusi both heard and to which they both responded. She first reported seeing white smoke from atop the fire tower, which shortly thereafter turned to black smoke.

[47] The second witness led by the respondent on this issue was Mr Saayman. He testified that he drove past the sawmill waste sites between 06h00 and 06h45 on the morning of the fire on his way to the farm Buffelshoek. It appears that he used the blue road for that purpose. He observed nothing amiss in the sawmill waste sites and saw no smouldering of any material occurring. Had he seen that occurring, given his responsibilities on Albany, it is reasonably certain to assume that he would have acted.

[48] Mr Pretorius also testified about this issue. He was emphatic that the fire’s origin was not in the sawmill waste sites. When Mr Roberts put the following question to him, he responded thus:

‘As daar beweer word dat hierdie brand wat u daar opgemerk het in kompartement A13A het sy oorsprong by die waste site, wat se u

daarvan? --- Dit is onwaar, meneer.’

[49] Mr Pretorius did, however, testify that later, during the afternoon of the fire and after it had been brought under control, he returned to the area of the sawmill waste sites. He described what happened then as follows:

‘Het u iets toe opgemerk by die sawmill waste site? --- Ek het dit opgelet by nommer 1 was daar tussen ’n halwe meter tot by ’n meter wat gesmeul het.

Het u enige vlamme daar gesien? --- Daar was geen vlamme gewees meneer.’

He testified that he extinguished what was smouldering.

[50] The fourth, and final, witness who testified on the state of the sawmill waste sites at the time of the observation of the fire in compartment A13a was Mr Mfusi. Mr Roberts asked him the following question:

‘When you went past the sawdust, did you see any fire or any smouldering in the sawdust? --- No.’

[51] Mr Mfusi said that he, too, had gone back to the sawmill waste sites after the fire had been brought under control and had seen Mr Pretorius extinguishing ‘something little’ there, which he described in cross-examination as being approximately 50 centimetres by 50 centimetres in size.

[52] There was accordingly direct evidence from the respondent’s witnesses that the fire was first observed in compartment A13a, and that there was no sign of any form of fire in the sawmill waste sites immediately before, or when, that occurred. On a factual level, none of that evidence could be challenged by the appellant. There was, moreover, no dispute at the trial that compartment A13a was, indeed, severely burnt by the fire.

[53] The appellant met this direct evidence, essentially, with the evidence of three witnesses, one of whom, Mr Abraham du Preez (Mr du Preez), was qualified as an expert. The other two witnesses were Mr Michael Mullins (Mr Mullins) and Ms Ruth Bezuidenhout (Ms Bezuidenhout). The evidence of Mr du Preez must be considered first.

[54] It appears that while Mr du Preez was qualified as an expert by the appellant and had previously given evidence at trials involving fires, he had never given evidence as an expert. He explained that his mandate was to establish, based upon the evidence at his disposal, the cause and origin of the fire.

[55] Mr du Preez candidly acknowledged that the best possible scenario would be for him to be at the site of a fire as soon as possible after its occurrence. His early arrival at the scene of a fire would permit him to consider the typical signs left behind in the aftermath of a fire. However, in this instance, he found himself at Albany some two years and eleven months after the fire, a considerable period after the occurrence of the fire and which period can only be described as less than ideal for his specific purposes. Indeed, as will be expanded upon shortly, this made things very difficult for Mr du Preez, as he himself acknowledged.

[56] Mr du Preez testified that when he first observed compartment A13a, it had changed since the fire, as most of the trees had been harvested and only about half a hectare of trees that had survived the fire on 7 November 2015 remained in that compartment. He acknowledged that this was a ‘small area of trees.’ He testified that he inspected these trees and used what he found there as a building block to develop his theory about the origins of the fire.

[57] Fully fleshed out, Mr du Preez’s theory was that the sawmill waste sites were the most probable place where the fire first started. It could have ignited, according to him, either by spontaneous combustion within the sawmill waste sites or from a previous burning operation, such as the burning of a firebreak, during which operation a fire could have entered the sawmill waste sites and smouldered there underground, unseen, for months. He believed that the fire came from the roadside on the west of Albany and spread eastwards into the farm, entering the first compartment, which was compartment A13a.

[58] Mr du Preez stated that he ascertained the direction of the spread of the fire by considering the angle of char on the few remaining standing trees in compartment A13a. The angle of char is a pattern of burning left behind on a standing tree after a fire has passed through that enables the direction of the spread of the fire to be ascertained. According to Mr du Preez, a fire typically leaves a low level of scorching on the side of the tree from which the fire approached and a greater scorching on the other, or leeward, side of the tree. Based upon what was left on the remaining trees in compartment A13a, Mr du Preez was of the view that the fire had spread from west to east. The following extract from his evidence in chief is illustrative of his evidence:

‘Before we move on to the next aspect can there be any other explanation for this angle of char, or the char that you observed, can there be any other explanation other than what you have told this court? --- No, sir, in my view, in my experience, definitely not. There are no signs of a so-called back fire burning towards the origin of the fire.’

[59] On the issue of back burning, Mr du Preez was asked the following question by Mr Troskie:

‘Is there any indication, or what you could see on the ground that indicated that the fire might have moved in the opposite direction, from east to west? --- No, no signs whatsoever were observable at the time of the field visit.’

[60] Mr du Preez was later cross-examined by Mr Roberts and was asked whether the side of the tree from which the fire had approached could never exhibit charring to the same extent as the charring on the leeward side of the tree. His answer was:

‘It depends on the velocity of the wind and the residence time of the fire in that particular area. If there was little wind and if it burnt severely it could be equal on the other side.’

[61] Mr du Preez was compelled to further acknowledge that the undergrowth present in a compartment could also play a role in how a tree was burnt by an approaching fire. This emerged from the following interaction when Mr Roberts cross-examined Mr du Preez:

‘So, the undergrowth would have affected the extent to which a fire would go up a tree. --- That is normally the case, correct.

So, even if it is situated on the leeward side of the tree. --- Yes.’

[62] The point of contestation between the two parties was neatly summed up in the following exchange between Mr Roberts and Mr du Preez:

‘Because the factual evidence is that the fire started in the plantation and it then burnt back towards the road and you are trying to refute that. I am putting it to you that depending on the fuel load inside the pattern could have been same as you found. --- But the fuel load was not the same, that was not what I observed.’

[63] Given the fact that Mr du Preez went to Albany two years and eleven months after the fire, he did not, nor could he reasonably have expected to, observe the fuel load. Mr Roberts consequently put it to Mr du Preez that the interval of time between the date of the fire and the date of Mr du Preez’s visit to Albany made:

‘… it difficult for any expert to really express a firm view without having been on the scene shortly after the fire, correct? --- That is correct.’

[64] The difficulty in the time lapse between the date of the fire and the date of Mr du Preez’s visit to Albany was highlighted, and magnified, when Mr du Preez stated that when considering the origin of the fire:

‘I had to rely on macro and micro indicators of which only two were present three years after the fire, namely, the angle of char and the V pattern. So, curling, freezing, cupping, all those other indicators were not possible to determine because of the time which passed was just too long, they were no longer present, so they were not reliable.’

[65] This, so it appears to me, meant that Mr du Preez’s theory of the origin of the fire was based on very few of the indicators that he would ordinarily have utilised to develop a comprehensive theory on the inception and spread of a fire.

[66] While there is no doubt a possibility that the theories propounded by Mr du Preez as to the origin of the fire may be correct, they suffered from an absence of facts that demonstrated that this is how the fire commenced. Recognising this, Mr Roberts put the following to Mr du Preez:

‘But you will agree with me that you have no factual evidence to base this on. --- I was not there on the day of the fire.

So, you have no factual evidence to base this theory on. --- I can only reconstruct on physical evidence, which I have already covered in my statement, as well as today, and that is the angle of char which is a very clear indication of the fire spreading from west to east.’

[67] This, in my view, is the high-water mark of Mr du Preez’s evidence: he could only confidently assert that the fire spread from west to east. Nothing more. He could not say how the fire started nor could he say where the fire had started. Moreover, what he could say, namely that the fire burnt eastwards, was not really in issue at the trial, for the respondent accepted that the fire had moved from west to east. The real point of dispute was whether the fire had started inside compartment A13a and had burnt back for a short distance to the blue road. For this, the appellant had no countervailing evidence to refute the evidence adduced by the respondent that has already been considered.

[68] Concerning whether spontaneous combustion could have occurred within the sawmill waste sites, Mr du Preez provided these answers to questions that were put to him in cross-examination:

‘You are not an expert on spontaneous combustion, are you? You have not been qualified as such in any event. --- I am not qualified but I have seen several examples of spontaneous combustion elsewhere, so I know it is possible.

But you have not seen it here. --- No.

You have not seen the evidence of that. --- I was not there on 7 November 2015.’

[69] As regards his proposition that a fire could have entered the sawmill waste sites when fire breaks had been burnt in the vicinity of the sawmill waste sites earlier in the year and had thereafter simmered away underground, undetected, Mr du Preez answered a question about this in the following way:

‘... Now you are saying the fire if it went into the sawdust heap it remained there unnoticed, dormant, for four, five months. --- I would not use the word dormant, it was just not visible, it was slowly but surely creeping and it was sustained but it was not visible to the naked eye above ground.’

[70] Thereafter followed a volley of questions directed to Mr du Preez on this issue:

‘… Now, you do not know when it crept in there. --- No I cannot say with certainty.

You do not know how it crept in there. --- No, I did not witness it personally.

You do not know where on the heap it crept into. --- No, I cannot point out a specific area.’

[71] Thus, Mr du Preez’s evidence was merely theoretical. As to why he believed the fire to have started in the sawmill waste site, he stated that:

‘The photographs which were presented to me clearly show signs of smoke and burning inside the sawdust heap.’

This then was the basis for Mr du Preez’s assertion that the fire started in the sawmill waste sites.

[72] That brings me to the evidence of Ms Bezuidenhout. She was an employee of the appellant and is the person who had largely attended to all the respondent’s insurance requirements with the appellant for approximately 14 years. She attended Albany on Monday, 9 November 2015, two days after the occurrence of the fire, and observed the destruction wrought upon, inter alia, compartment A13a and also went to the sawmill waste sites. Mr Troskie asked her whether there were any signs of fire in the sawmill waste sites when she went to Albany. She responded that:

‘When we drew up to the top end of waste sawmill dump one, before I could even get out the car, I observed that there was smoke, coming out of the sawmill waste, in an area that had been bulldozed into almost, if I could call it a tipi.’[16]

[73] Ms Bezuidenhout took several photographs that day, and she also made a video recording of what she observed in the sawmill waste sites. These must have been the photographs to which Mr du Preez referred as convincing him that the fire had started in the sawmill waste sites, for the respondent introduced no photographs showing this. Of the photographs taken by Ms Bezuidenhout, only two allegedly depicted the smoke that she stated that she observed in sawmill waste site 1 (the contentious photographs).

[74] The contentious photographs were included in the appeal record but were not of a particularly high standard, both as to their resolution and their print quality. I was not able to discern the smoke that was allegedly present in those two photographs, nor could I identify the so-called ‘tipi.’ The court drew this to the attention of both counsel at the appeal and it was agreed that bigger, better prints of the contentious photographs would be prepared by the appellant’s attorneys and would be submitted to us after the appeal. This was later then done, and we received the larger, allegedly improved photographs.

[75] I am by no means certain that the reception of these apparently improved photographs has resulted in an improved appreciation on my part of what is allegedly depicted therein. In one of the contentious photographs, I can discern no smoke whatsoever. In the other, I think that I can discern a change in colour of a small portion of the image that may be indicative of the presence of smoke. But I cannot be sure.

[76] My misgivings about what the contentious photographs depict are compounded by the fact that Ms Bezuidenhout admitted in her evidence that the video recording that she had made on 9 November 2015 did not depict any smoke at all in the sawmill waste sites. Why this should be the case is not clear. If it could allegedly be captured in a photograph, why could it not be recorded in a video recording? The answer may be that what was observed and captured in the contentious photographs was fleeting in nature. Because it was not sustained, it defied being recorded in the video that was taken.

[77] It appears that the idea of the fire having its origin in sawmill waste site 1 was not initially that of Mr du Preez but was rather that of Ms Bezuidenhout. She herself did not speak to this but Mr du Preez, who chronologically testified after Ms Bezuidenhout had testified at the trial, was asked if she had informed him of her theory of how the fire began. There then followed a most uncomfortable and obvious attempt by Mr du Preez to avoid answering what was, essentially, a very simple question:

‘And she must have informed you what her version is. --- She just showed me the photograph.

Did she tell you what she thought happened there? --- It was quite clear what I could see from the photographs.

The question is, did she tell you? You are evading the question. --- Yes, she did tell me.

Yes, why did you evade the question? Do you not want to answer this question? Will it incriminate you, or what? --- ... [no reply]

So, she told you what her view is on how the fire started. --- She showed me the photographs and I could make my own assumptions from it.

No, but she told you what she thought how it started. --- Yes, she did.’

[78] Mr Mullins was called as a witness by the appellant. He has been a professional loss adjuster for approximately 40 years and specialises in timber adjustments. He attended Albany on 11 November 2015 at the request of Ms Bezuidenhout and he prepared a written report on the fire. Whilst he was extremely experienced in the performance of his job, estimating that he had been involved in over 1 000 fires during his career, Mr Mullins readily conceded that he was not an expert on fires.

[79] In his report, Mr Mullins recorded, inter alia, that:

‘The fire appears to have started in or adjacent to compartment 13b, the fire spreading rapidly in the south easterly direction due to the strong north-westerly wind with a wind speed of 19 KMP.’

However, he later theorised in the same report that:

‘There is a possibility that, due to the prevailing continuous weather conditions, that the saw dust/waste dump spontaneously ignited,

caused by the low humidity, high temperatures and strong winds.’

[80] A consideration of his evidence leads me to the conclusion that Mr Mullins was not necessarily a witness whose testimony advanced the appellant’s case. He conceded, when placed under pressure while being cross-examined, that he may have incorrectly referred to compartment A13b in his report. Challenged on this, he indicated that he may have intended to refer to compartment A13a. He also conceded that his theory about the fire occurring first in the sawmill waste sites was not based upon any scientific exercise that he undertook to determine whether that was, indeed, physically possible. He further conceded that the observation by Ms Kubheka of first observing white smoke from atop the fire tower could have been the result of grass between the trees in the plantation igniting. He, finally, answered a question about the colour of the smoke from Mr Roberts as follows:

‘So, you will agree with me that it is not improbable that the white smoke that she saw could have emanated from within a compartment in amongst the trees. --- That is quite a possibility.

Well, it could be a probability. --- As well.’

[81] It is upon an assessment of this evidence that the trial court came to its conclusion that the fire first started in the sawmill waste sites. In its judgment, it did not define in which of those two waste sites the fire originated but Ms Bezuidenhout’s evidence suggested that it had to be sawmill waste site 1. I am, however, not satisfied that the evidence adduced by the appellant was sufficiently compelling to permit the trial court to come to the conclusion that it did.

[82] The theory about the fire having its origin in the sawmill waste sites was, as the evidence already considered reveals, the idea of Ms Bezuidenhout. There was, however, no objective evidence to establish this. The trial court had the advantage of seeing and hearing her evidence and it is therefore important to consider what it thought of her performance as a witness. It appears that it did not think much of her, for the trial judge found that:[17]

‘She was not the best of witnesses. At times she could not answer the questions posed to her or explain her conduct. Her conduct does not seem to accord with what she found and in my view no reliance can be placed on her evidence as to the origin of the fire.’

It is clear that she did not impress the trial court.

[83] Ms Bezuidenhout shared her theory of the origin of the fire with Mr du Preez. He accepted that theory and endorsed it through his evidence. It, however, appears that the only evidence that he considered as to where the fire commenced was the theory itself and the contentious photographs that Ms Bezuidenhout took of smoke allegedly rising out of sawmill waste site 1, two days after the fire. Mr du Preez apparently believed that the photographs clearly showed ‘signs of smoke and burning inside the sawdust heap.’ As already stated, I am by no means convinced that smoke is to be seen in either of those photographs.

[84] However, assuming for a moment that there was smoke depicted in both photographs, I am not sure what inference, if any, could properly be drawn from that assumption. At best, it could be accepted that two days after the fire there were vestiges of smoke in sawmill waste site 1. What would that prove? It could not be inferred therefrom that the source of that smoke was present two days earlier when the fire occurred or that it was a remnant of the original fire that sparked the inferno. There was no evidence adduced that sawmill waste site 1 ever burnt or that if it did burn, it burnt continuously from the day of the fire until observed by Ms Bezuidenhout two days later. All this while work was being performed in the sawmill waste sites by employees of the respondent, who did not observe such smoke, or having observed it, simply allowed it to continue burning. That seems entirely improbable to me.

[85] Indeed, it was argued at the trial by the appellant that the work that was done at the sawmill waste sites immediately after the fire was intended to obscure the fact that the fire had originated there. The trial court in its judgment accepted the work done in the sawmill waste sites as one of the factors that persuaded it that the fire had, indeed, started there. If the work done in the sawmill waste sites was intended to hide the origins of the fire, it seems unlikely to me that a burning fire producing smoke would in such circumstances not immediately be extinguished by the respondent’s servants, particularly when the respondent knew that Ms Bezuidenhout would be at Albany on 9 November 2015, as she had told Mr Hoatson that she would be there then.

[86] There are other explanations that may account for the smoke, if it is accepted that it does appear in the contentious photographs (it was also suggested by Mr Hoatson that what was observed and captured in the contentious photographs may have been dust or water spray). Mr Lucas Mthanti (Mr Mthanti) testified at the trial that the day after the fire, he burnt a firebreak around the sawmill waste sites for fear that the wind may turn and blow what he referred to as ‘charcoal’ from compartment A13a in the direction of the sawmill waste sites. If smoke was captured in the contentious photographs, it could, conceivably, have been a remnant of that activity. Additionally, depicted in the range of photographs taken by Ms Bezuidenhout were several heavy-duty earthmoving machines being put to work in the sawmill waste sites. It was suggested to Ms Bezuidenhout in cross-examination by Mr Roberts that if there was smoke in the contentious photographs, it may have been the exhaust emissions of those machines. Ms Bezuidenhout claimed to be able to distinguish between the two. Perhaps she can. But the presence of that equipment raises the possibility that this is what she observed and what was captured on the contentious photographs. Whatever she observed, it appeared to be momentary and could only be captured on two photographs and not on a video camera at all.

[87] In my view, even if there is smoke captured in the contentious photographs, that does not establish that it is probable that the fire started in sawmill waste site 1, and I believe that Mr du Preez was incorrect to attach the significance that he did to the contentious photographs.

[88] The evidence of Mr du Preez, as an expert, was utilised to add gravitas to the theory conceived of by Ms Bezuidenhout. In doing so, sight must not be lost of when he made his own observations. As previously stated, he only went to Albany some two years and eleven months after the fire. Much had changed since the fire. Thus, regarding the issue of back burning, not much weight, if any, can be attached to his evidence that he observed no signs ‘at the time of the field visit’ of the fire burning from east to west. Had he been there earlier, there may have been such evidence to observe, for despite the finding of the trial judge that he accepted that there was no back burning as that did not appear from the angle of char on the few remaining trees, there was direct eyewitness testimony that the fire had, indeed, back burnt.

[89] That evidence came from the testimony of Mr Mfusi. In his evidence in chief, he was asked by Mr Roberts about the direction in which the fire that he observed was burning. His answer was the following:

‘The one that I saw was moving backwards towards the road.’

The reference to ‘the road’ is a reference to the blue road. Mr Mfusi was adamant that while the head of the fire had moved on towards the east by the time that he arrived at the scene, the fire was contemporaneously burning back towards the blue road on which he was and from where he made his observations. The following was his evidence in this regard:

‘Is there a reason why are you reason (sic) why you left the fire creeping towards the road? --- Yes.

What is the reason? --- The wind was not pushing it hard towards the road.

Was it burning with the wind or against the wind towards the road there? --- It was against the wind.

Was it moving fast or slow what was the situation? --- It was moving slowly.’

[90] Critically, in my view, Mr Mfusi’s evidence on this point was not challenged. It was not suggested to him when he was cross-examined

that there was no evidence that the fire burnt back, as Mr du Preez later would assert when he testified. Not a single question was put to him in that regard. His observations consequently remain unchallenged. The trial judge did not lose sight of his evidence and summarised it as follows:[18]

‘Mfusi indicated that he saw the fire in the plantation burning against the wind. He admitted that the wind was in a westerly direction. Once again his evidence as to where the fire was seen by him has to be considered against the probabilities and the other factors such as the wind direction. He could not provide any explanation as to how this fire would have burnt against the prevailing wind which it was common cause was at least 20 km per hour at the time. Therefore just as in the case of Pretorius and Simon (sic) the other factors such as the wind direction and the other evidence which I will deal with later makes one to conclude that no reliance can be placed on their evidence.’

[91] The trial judge appeared to find that it was not possible for a fire to burn against the wind, contrary to what Mr Mfusi claimed to have observed. It is so that where the evidence of a witness is not challenged, it does not necessarily follow that it must automatically be accepted because the fact that it has not been contradicted does not make it true.[19] As was stated in MacDonald v Young:[20]

‘It is settled that uncontradicted evidence is not necessarily acceptable or sufficient to discharge an onus. In Kentz (Pty) Ltd v Power, Cloete J undertook a careful review of relevant cases where this principle was endorsed and applied. The learned judge pointed out that the most succinct statement of the law in this regard is to be found in Siffman v Kriel, where Innes CJ said:

“It does not follow, because evidence is uncontradicted, that therefore it is true . . . The story told by the person on whom the onus rests may be so improbable as not to discharge it.”

[92] But there was nothing inherently improbable about what Mr Mfusi said. In coming to its conclusion, the trial court appears to have lost sight of the evidence of an expert witness called by the respondent, namely Mr Willem Vorster (Mr Vorster). Mr Vorster mentioned in his report, which formed part of the appeal record, that:

‘All fires burn back a little as well.’

Asked to explain this, Mr Roberts elicited the following response from Mr Vorster:

‘You make your point here on page 106 that all fires burn back a little bit as well. What do you mean by that? --- If a fire start (sic) in the vegetation and if there was no winds, it would have burnt in a circle outwards. Now you put of (sic) wind so it burnt to the one side and it also burnt back into the wind.

Against the wind? Against the wind.’

[93] In my view, the trial court erred in simply dismissing the reliability of Mr Mfusi’s evidence and the evidence of Mr Pretorius and Mr Saayman. The evidence of Mr Mfusi was both possible, given the evidence of Mr Vorster, and uncontradicted. It should therefore have been accepted.

[94] The trial court accepted that a further indicator that demonstrated that the fire had started in the sawmill waste sites was the fact that work commenced on those sites on the day following the fire. I am afraid that I do not see that the same way. There was evidence from Mr Mthanti that the wind direction rapidly changed in that area, which was also not challenged, and in working on the sawmill waste sites to ensure that they did not later burn if there was another fire driven in their direction by a changed wind, I take the view that the respondent acted proactively and sensibly. I cannot see any basis therefrom to suppose that what was really being done was to hide evidence of the true origin of the fire nor that this fact, either alone or when viewed with others, established that the fire commenced in sawmill waste site 1.

[95] The trial court was faced with direct eyewitness testimony of what occurred on 7 November 2015 and the countervailing theory primarily advanced by Mr du Preez. It preferred the evidence of the expert over the evidence of the eyewitnesses. In my view, it erred in doing so.

[96] In PriceWaterhouseCoopers Inc and others v National Potato Co-operative Ltd and another,[21] the Supreme Court of Appeal referred with approval to the Canadian decision in Widdrington (Estate of) v Wightman,[22] where the following was held:

‘[326] “Before any weight can be given to an expert’s opinion, the facts upon which the opinion is based must be found to exist.”

[327] “As long as there is some admissible evidence on which the expert’s testimony is based it cannot be ignored; but it follows that the more an expert relies on facts not in evidence, the weight given to his opinion will diminish.”

[328] An opinion based on facts not in evidence has no value for the Court.’

[97] In AM and another v MEC for Health, Western Cape,[23] Wallis JA remarked that:

‘The opinions of expert witnesses involve the drawing of inferences from facts. The inferences must be reasonably capable of being drawn from those facts. If they are tenuous, or far-fetched, they cannot form the foundation for the court to make any finding of fact. Furthermore, in any process of reasoning the drawing of inferences from the facts must be based on admitted or proven facts and not matters of speculation. As Lord Wright said in his speech in Caswell v Powell Duffryn Associated Collieries Ltd:

“Inference must be carefully distinguished from conjecture or speculation. There can be no inference unless there are objective facts from which to infer the other facts which it is sought to establish … But 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.”’ (Footnotes omitted.)

[98] In Representative of Lloyds and others v Classic Sailing Adventures (Pty) Ltd,[24] Lewis JA considered the direct evidence of eyewitnesses and the theoretical constructs of expert witnesses and concluded thus:

‘… I must emphasise that where there is eyewitness or direct evidence of an occurrence, this may render the reconstructions of experts less relevant or even irrelevant …’

[99] In the earlier matter of Van Eck v Santam Insurance Co Ltd,[25] the court observed that the evidence of expert witnesses:

‘… is inevitably based on reconstruction and cannot conceivably bear the same weight as direct, eye-witness testimony of the event in question.’

[100] In my view, these views remain valid, more so when the reconstruction of the expert is attempted nearly three years after the event in question. When these authorities are considered, it is apparent that theories and inferences drawn therefrom must rest on a solid foundation of facts. There must be some factual basis linking the theoretical possibilities proposed by the theory to the known facts.

[101] It appears to me that Mr du Preez considered very few facts in coming to his theory. Those facts appear to comprise of the wind direction and weather conditions prevailing on the day of the fire, the state of a few trees remaining in compartment A13a nearly three years after the fire, and the content of the contentious photographs. In my view, that was insufficient to permit a proper conclusion on a balance of probabilities that the fire started in sawmill waste site 1. Of those factors just mentioned, only one of them, the contentious photographs, served to satisfy Mr du Preez that they showed evidence of where the fire actually commenced. That reasoning was not justified because it did not account for all the known facts.

[102] It seems to me that Mr du Preez was not aware of the evidence of Mr Pretorius or the evidence of Mr Mfusi for he did not mention either of them or what they claimed to have observed. Of course, it is possible that he knew of this evidence but chose to discount it. His theoretical explanation does not sit easily with the direct observations of the respondent’s witnesses. And, it appears to me, that insufficient consideration was afforded, in particular, to the evidence of Mr Mfusi by the trial court for his unchallenged evidence conclusively established that the fire had back burnt.

[103] The appellant’s version of the fire having its origin in the sawmill waste sites is simply a theory. It may be possible that a fire could have smouldered underground for months before sending out a flame together with a burning ember, to be carried aloft by the wind into compartment A13a and that then ignited the inferno that resulted. It may be possible that spontaneous combustion occurred in the sawmill waste sites. But there are no facts to give either of these theories oxygen. They conflict with the evidence of the eyewitnesses, and they accordingly merely amount to speculation, in my view.

[104] The consequence is that it may not be possible on the evidence adduced at the trial to conclude precisely how the fire started. It may have been arson.[26] It may have been through human error. It may have been from a carelessly discarded cigarette thrown into compartment A13a. It may have been through any one of the theories proposed by the appellant. But the fact of the matter is that there were at least three witnesses who observed the fire first burning in compartment A13a. No witnesses saw the sawmill waste sites smouldering or burning before the fire in compartment A13a was observed.

[105] In my opinion, and with the greatest respect to the trial court, it erred in finding that the fire commenced in the sawmill waste sites and it ought to have found that it commenced in compartment A13a of Albany.

[106] It accordingly becomes readily apparent that in submitting its claim form to the appellant, the respondent did not misrepresent anything. None of the responses by the respondent considered earlier appear to me to be false or misleading. In fact, they appear to me to be truthful and accurate. It is important to note that the claim form did not ask the respondent to specify where the fire had first started. The nearest that it came to such a question was the question that asked which compartment had first been affected. There was therefore no misrepresentation and certainly no fraudulent misrepresentation on the part of the respondent. And the appellant was not therefore entitled to avoid the terms of the certificate on this basis.

[107] The appellant, consequently, in my opinion did not establish that the fire originated in the sawmill waste sites or that the respondent made any misrepresentation to it about its origins. The second ground of appeal must consequently fail in my view.

The appellant’s third ground of appeal

[108] The court’s finding on the issue of the alleged failure by the respondent to disclose the existence of the sawmill waste sites and the alleged increase in the fire risk posed by those waste sites was the following:[27]

‘In the circumstances due to the factors mentioned it does not appear that there was a misrepresentation by Plaintiff nor that there was a duty to disclose the waste dump site to Defendant as an issue which would affect the insurance policy or Defendants decision to insure the said property. Defendant has failed to prove that the non-disclosure or representation induced it to conclude the contract. The evidence of its witnesses that it increased the fire risk in itself is not sufficient to prove such materiality.’

[109] It needs to be mentioned that the appellant’s case was not that it had been induced to issue the certificate because of the respondent’s alleged failure to mention the existence of the sawmill waste sites. Its case was that it was entitled to avoid the respondent’s claim, not the contract. This is affirmed by the fact that after the appellant concluded, erroneously in my view, that the fire had started in the sawmill waste sites, it did not revoke the certificate in its entirety because of that alleged non-disclosure. Instead, it merely issued a qualification to the certificate to the effect that any future fire that was established to have commenced in the sawmill waste sites would not be covered by the certificate. The certificate otherwise remained in full force and effect and was not cancelled by the appellant.

[110] The respondent submitted that the existence of the sawmill waste sites was not material on the facts of this matter, whereas the appellant claimed that it was. On the level of causation, it is clearly not material because it has been found in this judgment that the fire did not commence burning in sawmill waste site 1, as alleged by the appellant. The presence of the sawmill waste sites thus played no active part in what occurred on 7 November 2015.

[111] Whether an issue is material in the context of a short-term insurance contract is explained by the wording of s 53(1) of the Short-Term Insurance Act 53 of 1998, which reads as follows:

‘(a) Notwithstanding anything to the contrary contained in a short-term policy, whether entered into before or after the commencement of this Act, but subject to subsection (2) -

(i) the policy shall not be invalidated;

(ii) the obligation of the short-term insurer thereunder shall not be excluded or limited; and

(iii) the obligations of the policyholder shall not be increased,

on account of any representation made to the insurer which is not true, or failure to disclose information, whether or not the representation or disclosure has been warranted to be true and correct, unless that representation or non-disclosure is such as to be likely to have materially affected the assessment of the risk under the policy concerned at the time of its issue or at the time of any renewal or variation thereof.

(b) The representation or non-disclosure shall be regarded as material if a reasonable, prudent person would consider that the particular

information constituting the representation or which was not disclosed, as the case may be, should have been correctly disclosed

to the short-term insurer so that the insurer could form its own view as to the effect of such information on the assessment of the relevant risk.’

[112] This legislation has been enacted:

‘… to preclude insurers from treating misrepresentations that are trivial, and more recently non-disclosures that are trivial, as grounds for avoiding insurance contracts and rejecting claims.’[28]

[113] The test when considering misrepresentations and non-disclosures is an objective one in both instances.[29] The insurance company bears the onus of establishing materiality and must establish that the non-disclosure or misrepresentation induced it to come on risk.[30]

[114] On a factual level, it is difficult on the evidence led at the trial to decide whether the existence of the sawmill waste sites was ever disclosed by the respondent to the appellant. I appreciate that in making this statement, I am at odds with the finding of the trial court that found that it was common cause that the respondent did not inform the appellant of their existence.[31] My hesitation in accepting that finding is based upon two factors.

[115] The first is that the sawmill waste sites were first created in either 2003 or 2004. The appellant was already at that time the insurer of the respondent and Albany was consequently the subject of inspections performed by the appellant to consider such insurance. The inspector was not Ms Bezuidenhout at that stage, but a Mr Norris who was never called as a witness at the trial. But in completing his annual inspection reports, he recorded that the areas adjoining Albany in the years 2003, 2004, and 2005 posed no risk to the insured area and were thus scored by him as a zero. The sawmill waste sites, as already described, adjoined the insured area, and existed at that time. In preparing his report, Mr Norris could conceivably either have been told of the sawmill waste sites or he could personally have observed them. While there is a need for a party seeking insurance to make a full and frank disclosure of all the facts that may affect the risk against which it seeks indemnification, there is no obligation to disclose facts of which the insurer is already aware.[32]

[116] The second reason for my hesitation regarding the proposition is that Ms Bezuidenhout, whose evidence did not leave a favourable impression upon the trial judge, claimed that she had never seen the sawmill waste sites in the 14 years that she had done business on Albany. She had, on her own admission, been granted unfettered access to Albany and could visit and traverse the farm as she desired. On her own version, she did so. She agreed that she had travelled on the blue road that would have taken her past the sawmill waste sites but claimed that she never drove so far along that road as to reach the area where she would have seen the sawmill waste sites. Given that she did not impress the trial court, her evidence on this aspect seems contrived to me. In addition, the appellant was, until 2015, the insurer of the sawmill. It thus knew of the sawmill and what it did and produced. I find it simply too fanciful to accept that Ms Bezuidenhout, and therefore the appellant, had no knowledge of the sawmill waste sites.

[117] Whether the sawmill waste sites were disclosed to the appellant, or were known by it to exist is, in the final analysis, of no moment because their existence played no part in the fire. However, accepting for the purposes of argument that the appellant did not know of the sawmill waste sites, I would nonetheless agree with the trial court when it found that this was not material. It follows that I have reached the same conclusion as the trial court, albeit for a different reason.

[118] Finally, the appellant contended that the existence of the sawmill waste sites substantially increased the load of combustible material in the vicinity of the insured area and thereby increased the fire risk. No evidence was led by the appellant to establish this proposition. But the respondent did lead evidence on the mixture of materials deposited in the sawmill waste sites. Mr David Butt (Mr Butt) testified that he was employed by the respondent as its head of planning and took several samples from a sawdust waste site at the Franklin sawmill on the south coast of KwaZulu-Natal as well as from the sawmill at Albany, sawmill waste site 1, and from another farm in the area of Albany, namely the farm Kendal. These other samples appear to have been taken for comparative purposes. He personally flew to Gauteng and handed over the samples to a laboratory in Pretoria known as ‘Firelab’ (Firelab) for analysis. This exercise was done approximately six weeks before the trial commenced in November 2018 and was thus not done in close proximity to the occurrence of the fire.

[119] Mr Butt testified that the sample that he took from sawmill waste site 1 on Albany:

‘… had soil in it, bark, sticks, pieces of rock, gravel, mixed material. Maybe a little bit of sprinkling of sawdust.’

[120] The samples were received by Firelab and came to the attention of Mr Jacobus Strydom (Mr Strydom), its proprietor. One of the functions that Firelab performs is the fire testing of materials. It is not necessary to narrate how materials are tested in any detail save to say that a particular furnace is preheated to 400 degrees Celsius and then the specimen to be tested is inserted and the increase or decrease in the heat within the furnace is then observed. In essence, what is being measured is the heat contribution of the specimen to be tested.

[121] The samples given to Firelab were tested and a report compiled. As far as the sample from sawmill waste site 1 was concerned, Mr Strydom testified that the temperature of the furnace dipped slightly when the sample was introduced and then the temperature began to rise, which then led to the sample beginning to ‘glow’. Significantly, there was no flaming ignition, merely smouldering. A sample taken from the plantation floor of Albany comprising pine needles and forest debris, ignited within six seconds of being introduced into the furnace but terminated within approximately 30 seconds, whereafter the material continued to glow for approximately four and a half minutes. A grass sample flamed within nine seconds and increased the temperature in the furnace to 520 degrees Celsius, before decaying. A sample of fresh sawdust began glowing after approximately 30 seconds in the furnace and then began charring superficially but did not flame.

[122] Mr Strydom explained that the presence of sand would impact upon the ignitability of a sample, as the sand absorbs heat when exposed to it and the heat temperature will be reduced because:

‘… it cannot ignite. So, it will absorb the heat. So, if it is mixed there, it would be very difficult for something to burn.’

And he also explained that there is a difference between ignition and flaming ignition. The former results in slight glowing and the emission of smoke as the cellulose in the sample decomposes. Flaming combustion, as the term would suggest, results in flames.

[123] Mr Strydom concluded that the samples taken from Albany did not constitute any special fire risk because it could not burn, as oxygen, an essential ingredient for burning, would not be able to penetrate the interior of the dump. The result would be a smouldering, but flameless, combustion of the external surface of the dump. As to the ease of ignition, he explained that when a cellulose-based substance burns, it is the gas that is released from the substance that burns, not the substance itself. Sawdust is a fine substance comprised of small particles and does not burn easily because it does not emit sufficient gas to combust when heated. Mr Strydom believed the presence of the scrapings may well have decreased the fire load, contrary to what the appellant pleaded, because the most that could have occurred was smouldering on the surface of the sawmill waste sites.

[124] Under cross-examination, Mr Strydom fairly conceded that all that his tests revealed was how a test substance had performed under controlled circumstances. He conceded that he did not know how representative the samples taken from Albany were of the conditions

in the sawmill waste sites or at the sawmill itself. And he conceded that it was not impossible for flaming combustion to occur but remained of the view that it would have been very difficult for this to occur. It could occur if there was the introduction of a strong wind, which would then feed oxygen to the combustible material.

[125] It seems to me that the strong wind would have to exist at the moment that the fire first existed. It would have to be present to permit the flame to develop, and from that, spotting into compartment A13a would have occurred on the theory advanced by the appellant. However, that did not accord with the evidence of Ms Kubheka, who was asked the following question by Mr Troskie regarding her first observation of the smoke:

‘Now you have noticed that there are (sic) smoke and you are looking at the smoke somewhere in the distance, is that correct? --- Yes.

Do you remember whether the smoke was moving from left to right or right to left or not at all. Do you remember if it was moving in that direction? --- I saw smoke going up.

You saw smoke going up? --- Yes, it was going up.’

[126] That indicates to me that there was either no wind or, if there was a wind, that it was not sufficiently strong enough to dissipate the smoke in a particular direction. In other words, the strong wind that Mr Strydom testified about as being necessary for a flame to develop was not present.

[127] After a consideration of this evidence, I am of the view that the appellant did not establish that the existence of the sawmill waste sites created an increased fire risk.

[128] It should be observed, however, that even if that conclusion is wrong, the appellant faces another obstacle. This is not a case in which the insurer has established that the non-disclosure would have led to no policy being issued, or only being issued at a higher premium. On the contrary, by its own conduct the appellant has shown that if the disclosure had been made, it would have qualified its obligations under the policy, by excluding the obligation to compensate if a fire originates in the sawmill waste areas. If, as held above, this fire was not one of those, the issue of non-disclosure becomes academic.

[129] In my view, the third ground of appeal consequently must fail.

Costs

[130] The appeal must succeed on the first ground raised by the appellant, but given that the error underpinning that ground of appeal was that of the court, it can have no bearing on the issue of costs. The fact that the respondent abandoned the money judgment apparently does not result in the extinction of the order, but merely renders it unenforceable.[33] In this case the order must accordingly be set aside to open the way to the final determination of the action. The appellant has failed on the two substantive grounds of appeal. Costs must follow that outcome.

Order

[131] The following order is accordingly granted:

1. Save to the extent set out in paragraph 2 of this order, the appeal against the judgment of the trial court handed down on 26 January 2023 is dismissed with costs.

2. Paragraphs 1(i) and 1(ii) of the order of 26 January 2023 are set aside and are replaced with an order declaring the appellant to be liable to the respondent for such damages as the respondent may prove at a hearing in due course arising out of the fire that occurred on the farm Albany, situated in the Newcastle district, KwaZulu-Natal, on 7 November 2015.

MOSSOP J

OLSEN J

NICHOLSON AJ

APPEARANCES

Counsel for the appellant: Mr A Troskie SC Instructed by: Hay and Scott Attorneys Top Floor, 3 Highgate Drive Redlands Estate Pietermaritzburg Counsel for the respondent: Mr M G Roberts SC with Ms E Roberts Instructed by: Vinnicombe and Associates Locally represented by: Stowell and Company Incorporated 295 Pietermaritz Street Pietermaritzburg

[1] Despite such denial, it was common cause at the trial that the fire had occurred on 7 November 2015.

[2] Normandien Farms (Pty) Ltd v SAFIRE Crop Protection Co-operative Limited [2023] ZAKZPHC 6.

[3] This distance is gleaned from the scale measuring tool that forms part of a large aerial photograph of Albany (the large aerial

photograph) that was used at the trial and which forms part of the appeal record.

[4] This distance has been calculated using the previously mentioned scale on the large aerial photograph.

[5] I have twice used the phrase ‘as the crow flies.’ This is intended to describe the shortest possible distance between two points. In fact, the crow does not fly straight to its destination. The bird that does so is the rook, but as they are both black in colour, they have been popularly identified with each other: S Dent Interesting Stories About Curious Words: From Stealing Thunder to Red Herrings (2023) at 19.

[6] The distance has again been calculated with reference to the scale measuring tool on the large aerial photograph.

[7] Dictionary of South African English: https://dsae.co.za/entry/donga/e02052.

[8] Clause 8 reappears in identical form in Section B of the certificate as clause 16.

[9] Feinstein v Niggli and another 1981 (2) SA 684 (A).

[10] Ruto Flour Mills (Pty) Ltd v Moriates and another [1957] 3 All SA 28 (T).

[11] Bill Harvey’s Investment Trust (Pty) Ltd v Oranjegezicht Citrus Estates (Pty) Ltd 1958 (1) SA 479 (A). See also generally L T C Harms and M Townsend Amler’s Pleadings 10 ed (2024) at 204-205.

[12] Mkhize v S [2014] ZASCA 52 para 14; S v Hadebe and others 1997 (2) SACR 641 (SCA) at 645E-F.

[13] This should have read ‘an easterly direction’ and not ‘a westerly direction.’

[14] Forestry culture is, essentially, the planting of trees and their nurturing and development until they are ready for harvesting.

[15] ‘Spotting’ refers to the transport of burning pieces of firebrand by wind, which, at the time of landing, may ignite new fires beyond the direct ignition zone of the main fire.

[16] ‘Tipi’, or ‘teepee’, is a tent used by the indigenous first peoples of the United States of America, and which generally has a conical shape.

[17] The judgment para 112.

[18] The judgment para 110.

[19] Siffman v Kriel 1909 TS 538 at 543.

[20] Mcdonald v Young [2011] ZASCA 31; 2012 (3) SA 1 (SCA) para 6.

[21] PriceWaterhouseCoopers Inc and others v National Potato Co-operative Ltd and another [2015] ZASCA 2; [2015] 2 All SA 403 (SCA) para 99.

[22] Widdrington (Estate of) v Wightman 2011 QCCS 1788 (CanLII) paras 326-328.

[23] AM and another v MEC for Health, Western Cape [2020] ZASCA 89; 2021 (3) SA 337 (SCA) para 21.

[24] Representative of Lloyds and others v Classic Sailing Adventures (Pty) Ltd [2010] ZASCA 89; 2010 (5) SA 90 (SCA) para 60.

[25] Van Eck v Santam Insurance Co Ltd 1996 (4) SA 1226 (C) at 1229H-I.

[26] The occurrence of the fire was reported to SAPS Normandien, and a CAS number was issued to the respondent.

[27] The judgment para 127.

[28] Regent Insurance Co Ltd v King’s Property Development (Pty) Ltd t/a King’s Prop [2014] ZASCA 176; 2015 (3) SA 85 (SCA) (Regent) para 20.

[29] Regent para 23.

[30] Qilingele v South African Mutual Life Assurance Society 1993 (1) SA 69 (A) at 74C-H; Regent para 23.

[31] The judgment para 116.

[32] Carter v Boehm (1766) 3 Burr 1905 at 1910.

[33] Coetzer v Wesbank t/a FirstRand Bank Ltd 2022 (2) SA 178 (GJ) at paras [26] to [27].