---

CASE NO. 241/89

L J STEENBERG APPELLANT

and

DE KAAP TIMBER (PTY) LIMITED RESPONDENT

Judgment by: NESTADT, JA

CASE NO 241/89

/ccc

IN THE SUPREME COURT OF SOUTH AFRICA (APPELLATE DIVISION) In the matter between:

L J STEENBERG APPELLANT

and

DE KAAP TIMBER (PTY) LIMITED RESPONDENT

CORAM: BOTHA, NESTADT JJA et VAN DEN HEEVER AJA DATE HEARD: 19 AUGUST 1991 DATE DELIVERED: 8 NOVEMBER 1991

JUDGMENT NESTADT, JA:

The parties to this appeal are owners of adjoining farms in the Barberton district of the Eastern Transvaal. Appellant's farm (Snymansbult) lies roughly to the west of respondent's property (De Kaap). On the

2/

2. afternoon of Thursday, 12 September 1985 a fire, which emanated from Snymansbult, spread eastwards to De Kaap. It caused certain of respondent's timber plantations to be set alight and destroyed. Claiming that the fire had been negligently started by appellant or his servants (acting as such) and that they negligently failed to control it, respondent sued appellant for payment of the sum of R1,6 m. This represented the damages allegedly suffered by respondent as a result of the fire. The action was tried by DE VILLIERS AJ in the Transvaal Provincial Division. The learned judge granted judgment against appellant in the sum of R540 000 together with mora interest and costs. Against that order appellant now appeals, leave to do so having been granted by the trial judge. In what follows I refer to appellant as the defendant and to respondent as the plaintiff.

3/

3. On the pleadings, two broad issues arose for determination, viz, (i) whether defendant acted negligently and (ii) whether plaintiff proved its damages. I propose to deal with the latter issue first. This can be briefly done. During the course of the trial, plaintiff reduced its claim to R540 000 (the sum eventually awarded). At the same time defendant made two admissions. One was that on the day in guestion a fire spread from his property to that of plaintiff. The other was that plaintiff suffered damages in the amount of R540 000 as a result of certain of its forests having at the time been set on fire. Implicit in these admissions is an acknowledgment that part of plaintiff's damages resulted from the fire which originated on Snymansbult. There was, however, no admission that such f ire was the only cause of plaintiff's plantations having been set alight and

4/

4. destroyed. It was left open to defendant to contend that another fire (for which he was not responsible) contributed to plaintiff's damages. And this was indeed the case that defendant sought to make out. In support thereof, he testified that at about 2:45 pm on the Thursday (at a time when the fire which admittedly started on Snymansbult was raging) he actually saw certain workers of a Mr Pelser, the plantation manager of a farm (called Twello) to the north-west of De Kaap, burning a fire- break near the south-eastern boundary of such farm; that this fire then spread to De Kaap; and that it also caused trees of plaintiff to be burnt. The argument was that plaintiff had, in these circumstances, failed to establish what separate damages had been caused by each of the two fires; in particular it was not clear what (lesser) amount of damages was attributable to the fire which came from Snymansbult;

5/

5. and that plaintiff should therefore, on this ground alone, have been non-suibed. There is no merit in the argument. Pelser, who gave evidence for plaintiff, denied that his workers started a fire. The resulting dispute gave rise to a credibility issue. DE VILLIERS AJ resolved it in favour of plaintiff's version. The learned judge found that there was no second fire of the kind alleged. It is unnecessary to canvass his full reasons for so doing. Suffice it to say that they are cogent and, in my view, unassailable.

It foliows that it was the fire which originated on Snymansbult that caused the timber plantations on De Kaap to be destroyed. This being so, plaintiff was entitled to judgment in the sum of R540 000 - if defendant (or his servants) were legally responsible for it. This brings me to what was the main issue in the trial, viz, that of negligence. It will

6/

6.

be apparent from what nas already been said that this is not one of those cases (such as Van Wyk vs Hermanus Municipality 1963(4) SA 285(C) and Minister of Forestry vs Quathlamba (Pty) Ltd 1973(3) SA 69(A)) where liability rests on the failure of a land-owner to take reasonable precautions to prevent the spread from his property of a fire started on it or elsewhere by a third party. As I have said, plaintiff's case was that defendant's servants started the fire. This defendant denied (though, as indicated, he admitted that the fire originated on Snymansbult).

In seeking to prove that it was defendant's servants who started the fire and that they had acted negligently, plaintiff relied on the statutory presumption of negligence which applies inter alia to forest fires. It is contained in sec 84 of the Forest Act, 122 of 1984 ("the Act"). The section reads:

7/

7.

"When in any action by virtue of the provisions of this Act or the common law the guestion of negligence in respect of a veld, forest or mountain fire which occurred on land situated outside a fire control area arises, negligence is presumed, until the contrary is proved."

The date of commencement of the Act was 27 March 1986,

ie, after the fire, but before the issue of summons.

There was a dispute whether in these circumstances the

predecessor to the Act, namely, the Forest Act, 72 of

1968 did not apply. It too (in sec 23) provided for a

presumption of negligence. However sec 23, though in

similar terms to sec 84, did not require the land in

question to be situated outside a fire control area.

Initially Mr Zeiss, on behalf of defendant, contended

that it had not been shown that Snymansbult was outside

a fire control area (as defined by sec 1, read with sec

18(1), of the Act). If this was so and the Act applied,

then the presumption would not, for this reason alone,

operate. Counsel later, however, abandoned the point.

8/

8. It was conceded that the matter had to be decided on the basis that the fire occurred on land outside a fire control area. In the result, I (like the court a quo) find it unnecessary to decide which of the two Acts apply. I shall assume that it is sec 84 of the Act that must be looked to.

The term "forest...fire" in sec 84 is not

defined by the Act, but it was common cause that the fire in guestion was such a fire. This notwithstanding, defendant disputed that the presumption operated. Reliance was placed in this regard on what FANNIN J said in Quathlamba (Pty) Ltd vs Minister of Forestry 1972(2) SA 783(N) at 788 H, The learned judge, in dealing with sec 23, held that a "question of negligence" can only be said to "arise" where (i) negligence is alleged against the defendant and (ii) the plaintiff establishes a nexus or connection between the

9/

9. fire and the party against whom the allegation is made, which is consistent with such negligence. The submission on behalf of defendant was that such nexus had not been proved; plaintiff had in its summons and further particulars alleged that the fire had been started at a particular place on Snymansbult, namely, the maize lands; the allegations of negligence were limited to the burning of dried mealie stalks there; being bound by such allegations, the presumption could only be invoked if there was proof that the fire on 12 September originated in the maize lands; such proof was lacking; and the trial court was therefore incorrect in holding, as it did, that the presumption applied.

The Quathlamba case went on appeal to this Court (see Minister of Forestry vs Quathlamba (Pty) Ltd, supra). Though overruling FANNIN J's finding that the presumption had not been rebutted by the defendant,

10/

10. OGILVIE THOMPSON CJ affirmed the principle that the presumption cannot be invoked merely by averring negligence. The learned Chief Justice did not, however, adopt the approach of FANNIN J. It was simply held that the additional element required could be satisfied by proof that the fire originated upon iand owned and controlled by the defendant. On this basis the presumption would apply in casu. Snymansbult was not only owned by defendant, but was under his control. As will be seen, he was actively farming the property. And, of course, the fire admittedly originated on defendant's land. I shall, however, assume (in defendant's favour) that for the purpose of deciding whether the presumption created by sec 84 operates, it was necessary for plaintiff to prove a nexus or connection between the fire and defendant. I further assume that such nexus or connection had not only to be consistent (ie compatible) with defendant's negligence

11/

11.

as pleaded but must also have been of the nature and have arisen at the place so pleaded.

Proceeding on this premise, the first question that arises is whether, on a proper construction of the pleadings, plaintiff is confined to the case that defendant was negligent in starting the fire in the maize lands. DE VILLIERS AJ answered the question in the negative. I am of a different opinion. It would unduly lengthen this judgment were the relevant parts of the summons and further particulars to be quoted. I content myself with shortly stating my conclusions. The one is that the allegations that defendant failed "to provide adequate means for...controlling the fire or any fires which had been started on his property" and that defendant's workers "did not take all reasonable precautions to ensure that the fire which they had started did not spread to neighbouring properties" do not relate solely to what happened on the maize lands. They obviously embrace conduct or omissions pertaining

12/

12. not only to the maize lands but to that part of Snymansbult which lies to the east thereof. This is because the maize lands are not on the (eastern) boundary of Snymansbult and De Kaap. Between them and the western edge of De Kaap is an expanse of grazing land (containing a large donga) as well as a timber plantation of defendant. The fire had perforce to, and did, traverse these areas before it could reach plaintiff's trees. The following reproduction of one of the exhibits at the trial, namely, a drawing of the relevant part of Snymansbult, shows this.

See Original Judgement Image

13. The area marked 1 is the maize lands in question; 6 is the grazing section (the donga is the small part within it); and 7 is the plantation. (The boundary with De Kaap is along the line marked 13 on the right-hand side.) The other conclusion is that notwithstanding the width of the allegations referred to, defendant's interpretation of the pleadings must be upheld. On an analysis of the further particulars, I am satisfied that foundational to all the allegations of negligence is a fire which was allegedly started in the maize lands.

The next question is: did plaintiff sufficiently show that the fire which spread to De Kaap on 12 September began in defendant's maize lands? It follows from what has been said that only if the answer is in the affirmative, will the presumption operate. Most of the evidence at the trial was devoted to this issue. In terms of Supreme Court Rule 39(11), the Court

14/

14.

ruled that the onus of adducing evidence was on

plaintiff. This it proceeded to do. A number of

witnesses were called. The main one was Mr Pelser (to

whom I have already referred). He testified that on

Tuesday 10 September 1985 (ie two days before the fire)

he saw workers of defendant burning maize stalks in the

maize lands. He had previously noticed this being done

on various occasions during the preceding weeks. He

saw no further signs of burning on 11 September. At

about 1 pm the following day, he was at his house in the

vicinity when he received a report of the fire. He

hastened to the scene. His description of where the

fire was when he then, for the first time, observed it,

is of importance. It is necessary in this regard to

quote certain extracts from his evidence. He said:

"I noticed flames burning on the defendant's property next to his maize lands going in a south-easterly direction towards the gum plantation...

15/

15.

There was a small burnt area on the rear side

of the dongas towards the maize lands but it

was not a very big area that had already

burnt, where the fire had already finished

burning...

No I did not see the fire start, I just saw

flames burning in that area there.

In the area of the donga? -- In the area on

the top, the top part of the donga yes.

And that is 25 to 30 metres from the maize

fields? — That is correct, yes...

(F)rom there I saw the fire the first time and

there it was burning close to the maize fields

in a south-easterly direction towards the gum

plantation.

And how close to the maize field was it

burning? -- How close to the maize fields? It

was approximately five, ten metres,...

(T)here was already a burnt area.

Did the burnt area completely abut the maize

lands?-- Yes it went, bounded the maize

lands...

I saw the fire burning on the maize lands and

when I came ...

You saw it burning on the maize lands? -- No

burning a couple of metres away from the maize

lands, as I explained just now.

In the grazing? -- In the, it was burning in

the grazing.

Yes. — But it already backburnt into the

maize lands. I mean that area was already

burnt.

Did it burn into the maize lands? — Yes it

burnt into the maize lands...

16/

16.

The maize land itself could not burn, there

was nothing to burn in the maize land...

That area, in my opinion the fire started and

was burning from the maize lands towards the

gum plantation and down to the dongas, down

towards Twello property...

I only saw flames on top of the donga.

Yes, and you cannot exclude the possibility

that that is where it started, although you

did not see it start? — There could be a

possibility but I think it is highly

unlikely...

(W)hen I first saw the flames it was burning

on the top part of the donga and the area

between the donga and the maize lands had

already been burnt...

My opinion is that it started on top of the

donga. I did not see the fire start, I only

saw the flames. When I saw the flames it was

the flames were on the top part of the donga,

spreading down into the donga...

Yes it was burning into the donga from close

to the maize lands there, burning into the

donga down towards the gum plantation...

Approximately fifteen metres, ten, fifteen

metres away from the donga.

I see. Yes. -- That is the first, when I saw

the flames the first time...

No I saw the fire starting next to the maize

lands, you know in the grazing area. That is

where I saw the fire."

It cannot be gainsaid that this evidence

(which I think is fairly representative of what Pelser

17/

17. said) is somewhat inconsistent. Moreover, the allegation that workers of defendant started a fire in the maize lands on 10 September was denied by defendant. Though admitting that stalks in his land were burnt, his version was that this took place on 2 and 3 September; no burning took place after this date. The court a quo, however, accepted Pelser's evidence and rejected that of defendant. In the result it was held that defendant's workers did burn maize stalks on 10 September 1985. Here, too, there is, despite Mr Zeiss' earnest argument to the contrary, no warrant for interfering with what is essentially a credibility finding. Pelser made a favourable impression on DE VILLIERS AJ. Defendant, on the other hand, was regarded as an untruthful witness. The record discloses good reason for this assessment. There was acceptable evidence that shortly after the fire, he

18/

18.

admitted to an insurance assessor who interviewed him that he had indeed, two days before the fire, caused mealie stalks to be burnt in his lands. And, with justification, the testimony (on behalf of defendant) of his ex-wife that no burning took place on 10 September was found to be mistaken.

How then, on Pelser's evidence, do matters stand? The (alternative) submission on behalf of defendant was that in the absence of any evidence by Pelser that stalks in the maize lands were burnt after 10 September, plaintiff had to prove that it was the burning of maize on that day which resulted in the fire on the 12th; it had not discharged this onus; the presumption could therefore not be invoked. Reliance was placed on the evidence of a witness who testified for defendant. He was a Mr Kilian, a neighbouring farmer. He said that in his experience maize stalks normally do not smoulder for longer than 3-4 hours; it

19/

19.

was impossible that they could smoulder "vir dae". Furthermore, plaintiff had given a Rule 36(9) notice of its intention to call an expert witness regarding the length of time that burnt mealie stalks can smoulder. But no such witness testified. There was, therefore, so the argument went, no evidence of how the burning of maize stalks could, two days later, have caused the neighbouring grazing area to be ignited.

The approach of counsel for defendant was that in order to comply with the second requirement of FANNIN J in the Quathlamba case, it behoved plaintiff to establish as a fact that the fire on 12 September originated from the burning in the maize lands two days earlier. Only then would the necessary nexus between the fire and defendant exist. What LEON J said in Titlestad vs Minister of Water Affairs 1974(3) SA 810(N) may be said to support the argument. Dealing with the second requirement of FANNIN J, the learned judge held

20/

20. (at 815B) that (on the facts in that case) "(t)he enquiry is whether the Court is able to say that it is more probable than not that the fire broke out on land under the control of the defendant". I am, however, not sure that this quantum of proof was required in casu. Here, as I have said, it is common cause that the f ire originated on Snymansbult. It may be that in these circumstances the proposition that for the presumption to operate, plaintiff had to establish on a balance of probabilities that the fire of 12 September was caused by the burning that took place in the maize lands two days earlier, cast too heavy an onus on it. Perhaps, if a nexus or connection of the kind under consideration is required, it would have sufficed for this to have been only prima facie established. If this be so, evidence on which a reasonable man might find for plaintiff would suffice. On this basis, the

21/

21 .

test would be the same as that applied to determine whether there should be absolution from the instance at the close of a plaintiff's case. It is, however, unnecessary to express a firm view on the issue and I refrain from doing so. I shall assume (also in defendant's favour) that plaintiff had to prove the nexus or connection required by FANNIN J on a balance of probabilities.

Kilian's evidence loses some of its force when regard is had to the precautions that defendant himself says he took against the danger of fire spreading from smouldering maize stalks. Despite also asserting that maize stalks "brand feitlik onmiddellik uit", he conceded that it was nevertheless his practice to inspect his lands for a day or two after burning "om seker to maak dat alle brandende materiaal wel geblus is"; this was a reasonable precaution to take. Nevertheless, applying the postulated standard of proof,

22/

22.

I am inclined to agree that, for the reasons advanced, a

link between the burning which took place in the maize

lands on 10 September and the fire which Pelser saw on

the 12th cannot properly be inferred. That, however,

is not an end to the matter. In the further

particulars to plaintiff's summons, it is alleged that

burning of stalks in the maize lands occurred not only

on 10 September, but on the following two days as well.

In answer to a guestion: "Precisely when was the fire

started," it is alleged:

"The 10th September 1985, and/or the 11th September and/or the 12th September 1985. The Defendant commenced burning mealie fields on his property on the 10th September 1985. Fires were also burning on the Defendant's property on the 11th September and on the 12th September 1985."

So plaintiff was entitled to rely on a burning of the

maize lands by defendant's servants on 11 September or

the morning of the 12th as having been the source of the

23/

23.

fire which Pelser saw a little later.

Did plaintiff establish this on a balance of probabilities? It is true, as I have indicated, that there was no direct evidence that maize stalks were burnt on the 12th. The guestion then is whether this can be inferred. This brings me back to the facts. On 20 September 1985 a Mr Shewring (a so-called industrial surveyor experienced in assessing fire risks) inspected the area from the air. His evidence (on behalf of plaintiff) was that what he termed the apex of the fire, ie the point of its commencement, was " in the donga area". Defendant relied on this as indicating that the fire did not start in the maize lands. There might have been some merit in this contention were it not for the witness' later statement that "I cannot say where the f ire started". So one is lef t with Pelser' s evidence. His statement (at the end of the quotation)

24/

24.

that he saw the fire "starting next to the maize lands" may be said to indicate that the fire did not start in the maize lands. But it cannot be taken literally or out of context. As he explains in the next sentence, "that is where (he) saw the fire". The same applies to his earlier assertion that the fire "started on top of the donga". Indeed, he had already said that "in my opinion the fire started and was burning from the maize lands". Nor does his reference to what is called "backburning" assist defendant. Backburning is apparently a slow burning which tákes place against the wind. If this had occurred from the place where Pelser first saw the fire back to the maize lands, it would obviously indicate that the fire had not started in the maize lands. But this is not the true effect of what Pelser says. Immediately after his assertion that "it already backburnt into the maize lands" he corrects

25/

25.

himself by explaining "I mean that area was already burnt". It follows that the fire must have started from the maize lands. Pelser says as much. I have in mind his assertion that "the burnt area...bounded the maize

lands...It burnt into the maize lands". And,

consistent with this, is the proximity of the fire to the maize lands when, according to Pelser, he first saw the flames. His estimates of the distance vary between "next to", "25-30 metres", "close to", "approximately five, ten metres" and "a couple of metres away" (from) the maize lands. On 12 September a fierce north-westerly wind was blowing (ie in a south-easterly direction). This would, of course, have facilitated the fire in the maize lands spreading to the adjoining grazing area where Pelser saw the flames. It is true that plaintiff produced no evidence that the wind might have carried a smouldering part of a stalk to the grass

26/

26.

(called "spotting"); or that flames from the burning stalks spread directly to the grazing. Nevertheless, what has been stated was, in my judgment, strongly probative of the fire of 12 September having originated from the burning of stalks in defendant's maize lands within say a few hours before Pelser's observations. Defendant produced no evidence to counter this. The only other possible cause of the fire which was suggested by him was that unauthorised persons might have set fire to the grass in the donga when attempting to smoke out bees there (in order to extract their honey). But this can be discarded. It was based on mere speculation.

On a conspectus of all the evidence, I have come to the conclusion that the probabilities favour a finding that the fire on 12 September was caused by the burning of stalks in the maize lands by workers of defendant on 12 September shortly before Pelser came on

27/

27.

the scene. The nexus pleaded was therefore established. It was not in dispute that the fire was consistent with defendant's negligence. It follows that the court a quo correctly found that plaintiff proved that the presumption of negligence created by sec 84 applied.

The effect of this was that the onus then rested on defendant to show (on a balance of probabilities) either that in the particular circumstances harm to plaintiff was not, and cculd not reasonably have been, foreseen or, alternatively, that, notwithstanding the exercise by him of such care as the circumstances reasonably required, he could not prevent the fire from extending beyond the boundaries of his property and occasioning harm to plaintiff (see the AD judgment in the Quathlamba case at 84 H as also Clan Syndicate (Pty) Ltd vs Peattie and Others NNO 1986(2) SA 791 (A) at 796 G) . The . last question is whether he

28/

28.

discharged it. Reference has already been made to the

fact that a particularly strong wind was blowing on 12

September. Because of this, allied with the dry

condition then prevailing, it must, I think, be accepted

that once the fire took hold, at least on reaching the

plantation on Snymansbult, it became uncontrollable.

Indeed, that is the effect of the evidence. In

considering whether defendant rebutted the presumption

of negligence one must, therefore, look primarily to his

conduct in (via his servants) burning his maize lands so

that it thereafter spread to the grazing area to the

south-east. What was required of defendant was an

acceptable explanation, inconsistent with his negligence, of what precautions he took in this regard. More particularly is this so if account is had of the high degree of risk of a forest fire breaking out at that time of the year and the serious consequences to

29/

29.

the owners of adjoining farms if that happened. Under

cross-examination defendant conceded this in the

following terms:

"(I)n daardie wêreld in die winterseisoen, sê maar so van Julie tot Oktober bestaan daar 'n brand gevaar...Brand in 'n mielieland in September...kan...'n gevaarlike aktiwiteit wees ... as dit nie behoorlik geblus is nie...Brand is verbode daardie tyd...(Ek was) bewus dat 'n brand wat handuit ruk miljoen rande se skade kan berokken".

In my opinion, defendant wholly failed to lay

any factual foundation which could serve to discharge

the onus resting on him. His suggestion that the fire

could have been caused by passengers from buses that

used a road which skirted the plantation on Snymansbult

was based on speculation only. It was also inconsistent

with the fact of the grazing area to the north-west

having been burnt. Defendant's evidence that from 3

September up to and including 12 September he carefully

inspected his lands "om te kyk of daar enige brandende

30/

30.

materiaal (was)" cannot be taken seriously. I agree with DE VILLIERS AJ's observation that "op sy weergawe van die gebeure sou dit natuurlik ook nie nodig gewees het nie aangesien hy reeds op 2 en 3 September sou gebrand het". None of defendant's workers were called to explain how they went about burning and extinguishing fires in the maize lands. So we do not know whether earlier on the 12th conditions were such that it was safe to burn; what steps were taken to ensure that all burning stalks were rendered harmless; whether there was an adequate fire break between the maize lands and the grazing area; and what precautions were taken to prevent the fire from taking hold. It is true that on the 12th conditions were abnormally windy and dry. But I do not believe that they were unforeseeably abnormal as was submitted. The exceptional often has to be

31/

31 .

anticipated. Defendant also relied on a fire-break, 50 metres in width, between the grazing and his plantation which he testified he had burnt in about June 1985. The overwhelming weight of evidence was that this assertion was untrue; that there was no such fire-break. Mr Zeiss was rightly constrained to concede this.

To sum up. The fire that resulted in plaintiff's damages not only originated on defendant's land, but was caused by his servants. The duty to prevent a fire from spreading when you yourself have lit it is a high one (Van Wyk vs Hermanus Municipality, supra, at 300 D). Defendant failed to show that he discharged such duty. He was therefore correctly held liable to plaintiff.

32/

32.

The appeal is dismissed with costs. Such costs are to include the fees of two counsel.

NESTADT, JA BOTHA, JA – CONCURS

CASE NUMBER: 241/89

IN THE SUPREME COURT OF SOUTH AFRICA (APPELLATE DIVISION)

In the matter between:

L J STEENBERG Appellant

and

DE KAAP TIMBER (PTY) LIMITED Respondent

CORAM: BOTHA, NESTADT JJA et VAN DEN HEEVER AJA HEARD ON: 19 AUGUST 1991 DELIVERED ON: 8 NOVEMBER 1991

JUDGMENT

VAN DEN HEEVER AJA

2

I agree that defendant did not succeed in rebutting the presumption that he was negligent in relevant respects alleged in plaintiff's pleadings and with the order proposed.

In my view the link between a defendant and a fire alleged to make him responsible in law for damage caused by it, and his negligence leading to that damage, are distinct and separate issues. It is only on the second of these that the statutory presumption assists a plaintiff. Support for such a view is to be found in the fact that there are various grounds on which a person may be held "responsible" for a fire: because he started it; or because he is the person who was in control of either the ground from which it escaped, or the fire itself - cf the provisions of section 21(1)(c)(ii)(a) of the 1968 statute. The issues may overlap in cases like the present, where the nexus relied on is that defendant caused the fire in question. Under the weather

3

conditions that prevailed proof of that fact would amount to res ipsa loquitur and plaintiff would not require statutory assistance in discharging the onus of proof. It is probably because of those weather conditions that plaintiff did rely on that specific nexus and not merely allege that the fire originated on and escaped from defendant's farm. (Plaintiff could not control another fire that originated on its own property.)

Although the separate issues of nexus and negligence overlap in the matter before us because of the facts of this case any suggestion that they may do so as a matter of principle should in my view be avoided. Were a plaintiff's case based, for example, on his labourer's report that "I saw Mr Smit knock out his pipe on your gate post as he passed, shortly before the fire started in that very area" and should Mr Smit deny having been in the vicinity at the relevant time, the normal approach to quantum of proof must in my view apply. That there may

4

be a lesser quantum, not of proof, but of evidence required in a case such as the present, should not be sought in the terms of the statute but again in the facts: an adverse conclusion may be more readily drawn against someone, like the defendant, who should be able to answer a prima facie case with ease but fails to do so by remaining silent or by telling palpable untruths. MARINE AND TRADE INSURANCE CO LTD v VAN DER SCHYPF 1972 (1) SA 26 (A); Hoffmann and Zeffertt, SA LAW OF EVIDENCE, 4th ed at 596 et seq. esp. at 598; Schmidt, BEWYSREG, 3rd ed, p 46-47 and cases in n.2.

Plaintiff succeeded in discharging the onus which burdened it by reason of its pleadings, of establishing that it was more probable than not that the fire that swept from Snymansbult across De Kaap resulted from defendant's activities in burning his maize fields during the relevant time: the 10th to the 12th September. It is in my view unnecessary to exclude as

5

being causally irrelevant his proven activities on the 10th. Defendant's concession that it was proper to inspect for at least two days after burning "om seker te maak dat alle brandende materiaal wel geblus is" leads to the inescapable inference that he knew that his maize lands could well contain some or other material - not necessarily cattle dung, although Pelser saw defendant's cattle in the maize fields and dung may smoulder - other than maize stalks and cobs, which would constitute a danger. No factual as opposed to speculative evidence was adduced of any other possible origin of a fire "coincidentally" commencing in the very vicinity where defendant had been recently seen burning.

VAN DEN HEEVER AJA

---