Van Deventer v Botha (1523/2014) [2019] ZAFSHC 110 (4 July 2019)

IN THE HIGH COURT OF SOUTH AFRICA,

FREE STATE DIVISION, BLOEMFONTEIN

Case number: 1523/2014

Appeal No: A211/2018

In the matter between:

ADRIANA

SALOMINA VAN DEVENTER Applicant

and

HENNIE BOTHA Respondent

CORAM: DAFFUE, J et MATHEBULA, J et LOUBSER, J

HEARD ON: 24 JUNE 2019

JUDGMENT BY: DAFFUE, J

DELIVERED ON: 4 JULY 2019

I

INTRODUCTION

[1] Two questions have to be decided in this appeal. First, whether a Boer mastiff (boerboel in Afrikaans) acted contrary to the nature of domesticated animals and dogs in particular when it attacked and bit an adult man several times and if so, whether the attack and resultant injuries caused or materially contributed to the man’s death in hospital ten days later.

[2] The matter came before an acting judge of this Division who dismissed the plaintiff’s action. It was firstly found that the dog did not act contrary to its nature and secondly, that the injuries sustained (“the wounds”) did not cause the death of the deceased given his medical condition prior to the attack.

[3] Lekale, J, who was allocated the unsuccessful plaintiff’s application for leave to appeal in the absence of the acting judge, granted leave to appeal, costs to be costs in the appeal.

II

THE PARTIES

[4] Mrs Adriana Salomina van Deventer, a widow residing at Brakpan, Gauteng when action was instituted, previously from Bloemfontein, is the appellant in this appeal, she being the unsuccessful plaintiff in the trial court. She was represented before us by Adv PJJ Zietsman as was the case during the trial.

[5] Mr Hennie Botha, the respondent and successful defendant in the trial court is a major male, previously employed and resident in Bloemfontein, but now employed at Halfway Ford, Kuilsrivier, Western Cape. He was duly represented by legal representatives during the trial and the application for leave to appeal. Notwithstanding being fully informed of the appeal and served with the appeal records on 5 September 2018 after withdrawal of his attorneys of record and informed of the appeal hearing in that the notice of set down was served upon him on 29 March 2019, he failed to give notice to oppose the appeal or to file heads of appeal. The deputy sheriff of Kuilsrivier South on both occasions served the relevant documents on respondent personally. Respondent did not appear when the matter was called and subsequently heard by us.

III

BRIEF FACTUAL MATRIX

[6] Several aspects are either common cause or have been proven on a balance of probabilities. Contrary versions will be highlighted.

[7] During the afternoon of 8 April 2011 Mr David van Deventer (“the deceased”) was bitten several times by the dog, Capone (also spelt Kepone in the record) (“the dog”). Respondent was the owner of the dog.

[8] The attack on the deceased, to be referred to as “the incident,” occurred at respondent’s smallholding in

Bainsvlei, Bloemfontein in the presence of the deceased’s employees, the appellant, the deceased’s sister, respondent’s

wife at the time, as well as respondent’s mother in law and possibly the children of the parties.

[9] Mrs Susan van Deventer, the deceased’s sister and the deceased’s wife, Mrs A S van Deventer, the appellant in the

appeal, testified about the incident on behalf of appellant. Dr A van Aswegen testified as expert on behalf of the appellant.

Mrs Karen Botha, respondent’s wife at the time (the parties became divorced after the incident, but before the trial) testified

for respondent about the incident and Prof JBC Botha presented expert testimony on behalf of the respondent.

[10] Respondent’s smallholding was fenced-off with devil’s fork and an iron sliding gate provided entrance to the premises. This was locked at night, but merely closed during the day. Respondent had two dogs, a bitch named Lulu and her off-spring, Capone. Both dogs were described by Mrs Botha as “very loving, playing with my children. I have never had any aggression with any of those two dogs.” According to Mrs Susan van Deventer, the deceased’s sister to whom I shall refer as Susan to avoid confusion, Capone was well-known to her. She and Mrs Botha were friends for many years and she generally attended “DVD movie evenings” at respondent’s premises on Friday evenings. She went to the Botha’s smallholding early the particular Friday afternoon and took appellant’s son, Juandré, with her. Capone and Lulu usually roamed free on the smallholding as was the case that day while the Botha’s children and Juandré were playing outside.

[11] Appellant and the deceased arrived at the respondent’s smallholding at about 17h00 in order to pick up Juandré.

The deceased was driving his LDV, commonly known as a bakkie. He towed a trailer. Two of his employees were conveyed on the back of the bakkie. They entered through the sliding gate and parked close to the house. The children apparently opened the gate. Appellant and the deceased disembarked while Susan, Mrs Botha and her mother exited the house to meet them according to the evidence of Susan and appellant. When the dog (Capone) jumped up against the bakkie the deceased reprimanded it according to Susan. Hereafter the dog attacked and bit the deceased on his hands, arms and lower leg. The deceased eventually received a rope from one of his employees which he managed to put around the dog’s neck to contain him. Susan assisted him and later held the rope whilst the deceased’s leg wounds were bandaged by appellant. According to appellant she was talking to Mrs Botha at the time, did not observe the incident from the start, but only heard a “commotion,” as she called it, behind her. When she turned around the deceased was already injured. As the deceased was bleeding profusely from his lower left leg, appellant ran into the house in order to look for bandages to dress the wounds and stop the bleeding. Mrs Botha’s mother handed her a bandage which she used to dress the wounds. Hereafter they left for the hospital, but on their way they stopped to offload the employees and to disconnect the trailer, leaving it in possession of a friend. Susan remained behind on the smallholding.

[12] Mrs Botha’s version contradicts that of Susan and appellant. According to her, she came running out of the house, screaming at the visitors not to disembark. She testified that the deceased hit the dog several times with a rope from behind whilst it was standing against the bakkie and this caused the dog to attack him. On her version the dog was seriously provoked by the deceased.

[13] The deceased was immediately treated at the hospital, but his condition worsened to such an extent that he passed away on 18 April 2011, ten days after admission. His death certificate indicates that he died from natural causes with liver failure as the immediate cause and portal hypertension as secondary cause. The experts agreed that the objective evidence indicated this conclusion to be incorrect. More will be said about this later.

IV

GROUNDS OF APPEAL

[14] The first ground of appeal deals with the trial court’s finding that the dog did not act contrary to its nature when it attacked the deceased, notwithstanding several common cause facts, alternatively the proven facts, inter alia (a) that the dog was not provoked, (b) the dog had never bitten anyone before, (c) the dog did not have a propensity to bite people, (d) the dog was present when the children were playing on the premises the particular day. It should have found that the dog indeed acted contrary to its nature.

[15] Secondly, the trial court’s finding that the wounds did not result in the death of the deceased, given his pre-existing medical status, is challenged to be incorrect and in this regard appellant raised several factual and legal arguments arising from the evidence of the two experts, some of which will be considered hereunder. In short, it is appellant’s case, based on the evidence of Dr Van Aswegen, that the wounds caused bacteraemia which in turn caused peritonitis which triggered Type 1 Hepatorenal Syndrome (“HRS”) and this ultimately caused multiple organ failure and the deceased’s death. Type 1 HRS is characterised as a rapid and progressive hepatorenal impairment triggered inter alia by a bacterial infection. As Dr Van Aswegen testified, HRS “is a syndrome where you have a patient with a liver that is not working well and then despite a normal kidney you have a functional kidney failure” and “something precipitates something and the kidney fails without the liver actually becoming worse.” In most cases, according to the expert, Type 1 HRS is fatal and virtually all patients die in about ten weeks and many much earlier.

[16] Consequently, two issues need to be adjudicated and those are the findings of the trial court pertaining to the contra naturam sui generis principle, i.e. whether the dog, a domesticated animal, acted contrary to its domesticated nature, and secondly, whether the attack on the deceased and the wounds sustained by him caused his death. I need to point out immediately that liability based on delict, i.e. the alleged negligence of respondent as pleaded in the alternative in the particulars of claim, does not have to be considered. The trial court considered the actio legis acquiliae in paragraphs [21] to [24] of the judgment, but it is not required to deal with the issue of negligence. Fact of the matter is that appellant does not rely on respondent’s liability based on the actio legis acquiliae in this appeal. There is no evidence that respondent was negligent.

V EVALUATION OF THE TRIAL COURT’S JUDGMENT

[17] I make some comments on the trial court’s findings under this heading, but shall deal further with specific issues later under the next two headings. The trial court did not evaluate the evidence of the two witnesses, Mrs Van Deventer and Mrs Botha in order to conclude which version, particularly that of Susan or that of Mrs Botha, was preferred and for what reasons. However, a study of the judgment tends to lead the reader to the inevitable conclusion that Susan’s version of the events was accepted. The trial court held that the dog was not provoked by the deceased, but its reasoning is unconvincing. It was found in paragraph [16] that Mrs Botha “did not come forth as an untruthful and dishonest witness nor had any dishonest intention.” This witness testified how she ran out of the house after observing the approaching vehicle and shouted “stop” several times in order to allow her an opportunity to lock the dog away. She did that as the dog had a propensity to bite vehicles’ tyres. For this reason Susan had to phone her brother to inform him to make contact when he arrives at the gate to enable the witness to lock away the dog. She also testified that the deceased hit the dog repeatedly with a nylon rope whereupon the dog attacked and injured him. However, in paragraph [24] and when dealing again with Mrs Botha’s evidence, it was held that the “deceased’s conduct in

attempting to control the dog by using a rope cannot be attributed as provocation; the deceased was merely attempting to contain the dog.” The trial court confused the evidence. On Susan’s version the deceased tried to contain the dog after it had started attacking him and on that version there was no provocation. Mrs Botha’s evidence is to the contrary. Bearing in mind the conclusion at which the trial court arrived, it is unnecessary to do an in-depth evaluation of the evidence, save to state that Susan’s version is not only credible, but more probable and reliable than Mrs Botha’s version. Mrs Botha did not present a reliable and truthful version of the incident. Crucial aspects of her version presented to the court were never put to appellant and particularly Susan for them to respond. There was no reason to contain the dog in the manner described by Susan if the dog was not aggressive towards the deceased and had already started attacking him. I would have expected Mrs Botha to reprimand the deceased for repeatedly hitting her dog, but that was not her version.

[18] The trial court’s view point in paragraph [20] that if the dog had the propensity to attack, it would have bitten appellant’s

son who was on the premises for the first time that day, does not correlate with the finding that the dog “must be regarded as not having acted contrary to his nature.” Precisely the opposite finding should have been made based on the evidence of Susan and Mrs Botha pertaining to the good nature of the dog. A dog that never had any propensity to bite people acts contrary to its nature if it bites a person for no reason or without being provoked.

[19] The trial court considered the evidence of the two experts and without properly evaluating the evidence concluded in paragraph [37] that “the plaintiff to strengthen her case would have called the three doctors who treated the deceased to come and testify. Thus this makes it further difficult to have an overall view of the cause of death …., as well as the absence of blood cultures and the post-mortem results. Further having regard to the poor state of health the deceased was in before the dog bite, it makes it difficult to be persuaded that the dog bite caused the death of the deceased.” In paragraph [38] the trial court continued: “A normal healthy person can survive a dog bite. The deceased was sick even before the dog bite.” It concluded in the next paragraph that it was “not persuaded that the dog bite caused the death of the deceased given his pre-existing medical history.” I shall deal with the expert evidence later, but merely wish to say at this stage that the fact that an old and frail person of 80 or 90 years dies of an attack such as in casu, while a young, healthy and strong person would have survived the attack is no reason to dismiss a claim for lack of proving causation. Similar is the case where a baby is killed by a dog in circumstances where an adult would have survived a similar attack, save for a few wounds being sutured and antibiotics administered. The adage “you must take your victim as you find him,”

comes to mind and should be applied.

VI

ACTIO DE PAUPERIE

[20] An owner of a domesticated animal, such as the dog in casu is liable for damage caused by the animal without requiring proof of negligence.[1] In O’Callaghan Innes, CJ concluded as follows:

“By our law, therefore, the owner of a dog that attacks a person who was lawfully at the place where he was injured, and who neither provoked the attack not by his negligence contributed to his own injury, is liable, as owner, to make good the resulting damage.”[2]

[21] It is trite that, bearing in mind the quoted judgments, appellant had to prove that

(a) the ownership of the dog vested in respondent at the time the damage was inflicted;

(b) the dog was a domesticated animal;

(c) the dog acted contrary to the nature of domesticated animals and in particular dogs and

(d) the conduct of the dog caused the appellant’s damage.

Clearly, the respondent’s liability as owner of the dog is independent of any fault on his part.

[22] It is common cause that respondent was the owner of the dog when the deceased was attacked. This allegation was never seriously attacked. It also appears from the evidence that when Mrs Botha vacated the smallholding after their divorce, the dog remained on the premises which respondent continued to occupy. Respondent did not testify to deny his ownership. It is also common cause that the dog was not kept as a vicious watch dog with the prime purpose of securing the premises from potential criminals. It was used to visitors and children in particular. Its only threat to visitors, according to Mrs Botha, was the propensity to bite vehicles’ tyres. This was never witnessed by her friend, Susan, who often visited the Bothas and it also did not occur the fatal afternoon when the deceased and appellant arrived to fetch her child.

[23] Paragraphs (c) and (d) above are in contention as previously said. Causation will be dealt with under the next heading. Paragraph (c) requires immediate attention. There is no onus on appellant to show that the dog acted out of perversity or inherent viciousness, but merely that the conduct was foreign to the domesticated nature of the particular class of animals which has done the damage, to wit dogs in casu.[3] Appellant was also not required to explain the peculiar behaviour of the dog. The contra naturam sui generis requirement requires some unpredictable action from the domesticated animal. Action that is to be expected as part of the animal’s natural behaviour does not qualify. In Da Silva v Coetzee[4] the plaintiff’s wife, whilst window-shopping, passed close to the defendant’s employee, a security guard, who had a dog on a chain. The dog jumped up, attacked the woman from behind and injured her. The defendant’s plea of provocation was dismissed by the High Court. That court found that the trial magistrate incorrectly placed an onus on the plaintiff to prove why the dog behaved as it did, whilst the onus rested on the defendant to prove that the dog acted in the said manner because of the fault of the plaintiff’s wife as held in O’Callaghan supra.

[24] Certain defences are available to the owner of a domesticated animal. I referred to Da Silva supra[5] pertaining to provocation. The latest judgment in this regard is Van der Westhuizen v Burger,[6] the interesting case of the plaintiff who provoked an ostrich, then ran away when he was chased, fell over a piece of wood, torn his Achilles tendon and as a result suffered damages. The SCA held, correctly with respect, that the plaintiff’s provocation

constituted a defence. I am mindful of the fact that the ostrich was a wild animal and not domesticated, but Ponnan, JA in his minority judgment (whilst agreeing with the majority) dealt with the defence of provocation in respect of the actio de pauperie with reference to inter alia O’Callaghan.[7]

[25] In the present case respondent pleaded provocation, but failed to prove the defence. The trial court correctly found that provocation had not been proven, albeit incorrectly on Mrs Botha’s version. This witness was the only witness who testified that the deceased repeatedly hit the dog with a nylon rope before he was attacked. The court should have accepted the version of Susan above that of Mrs Botha based on credibility, reliability and the probabilities. The appellant’s evidence was not helpful in this regard. She heard what she called a “commotion” behind her back and her husband saying that the dog was busy biting him. When she turned the deceased had already fastened a rope around the dog’s neck. If she wanted to make up a version to strengthen her case, she could have done so, but she refrained from testifying about the incident which she did not fully observe from the beginning.

[26] The negligence of a third party in charge of or in control of a dog or other domesticated animal may present a possible defence to the owner of the domesticated animal.[8] Mrs Botha was apparently in control of the dog in the temporary absence of respondent. It was not averred in the plea that she was negligent in any manner; consequently, it is not surprising at all that no evidence was led by or on behalf of respondent in this regard. The facts in Lever v Purdy[9] are distinguishable from the facts in casu. This judgment was relied upon by respondent’s legal representatives in opposition of the application for leave to appeal. It is accepted with respect to be the law that if Mrs Botha or another third party provided the dog with an opportunity to injure the deceased and failed to prevent it from doing so, respondent could have relied upon such facts in defence of the claim. Such a defence was not pleaded, but more importantly, would be without merit within the accepted factual matrix.

[27] A final possible defence in casu is that the deceased’s presence on the smallholding was unlawful. Respondent pertinently pleaded that the deceased entered the smallholding without consent and that his presence was unlawful; alternatively he pleaded that the deceased’s presence was unlawful in that he was requested to phone before entering the smallholding to allow the person in control of the dogs an opportunity to lock them away. This defence is without merit. Respondent’s counsel unsuccessfully made a serious attempt during cross-examination of Susan for her to concede that she had phoned the deceased before his arrival at the smallholding. It was put to her that both Mrs Botha and her mother, Mrs Russell, would testify that Susan was instructed to phone the deceased to warn him not to enter the premises unannounced. He had to phone upon his arrival at the gate. An order was even requested and granted for Susan’s cellphone records to be obtained from the relevant cellphone company. Nothing came of this. Respondent’s legal representative went so far as putting it to Susan that Mrs Botha and her mother would testify that when they saw the deceased’s vehicle entering the smallholding, both of them ran out of the house, shouting (loosely translated from Afrikaans) “don’t get out of the car” and “just stay in the car.”

This was also denied. The impression created by the cross-examiner was that the dogs were vicious and would certainly attack the strangers if they dare exiting the vehicle. When Mrs Botha testified, she stated that the reason for the request was merely because she was afraid that her dog might be run over by the vehicle as it was used to bite vehicles’ tyres.

VII

CAUSATION

[28] I mentioned earlier that the fourth element to be proven to be successful with the actio de pauperie is causation. In Skosana[10] Corbett, JA (as he then was) defined causation in the law of delict. The first requirement is a factual one relating to the question whether the negligent act or omission in question caused or materially contributed to the harm giving rise to the claim. The so-called “but for” test applies. If factual causation is not proven, it is the end of the matter. The second requirement is a sufficient link between the negligent act or omission and the harm suffered, or put otherwise, legal causation. A flexible approach is followed in this regard as set out in Standard Chartered Bank of Canada[11] in which “factors such as reasonable foreseeability, directness, the absence or presence of a novus actus interveniens, legal policy, reasonability, fairness and justice all play their part.”

[29] In Van Duivenboden[12] the SCA held that a plaintiff is not required to establish the causal link with certainty, but merely that the wrongful conduct was probably a cause of the damage. This calls for “... a sensible retrospective analysis of what would probably have occurred, based upon the evidence and what can be expected to occur in the ordinary course of human affairs rather than an exercise in metaphysics.” (emphasis added).

[30] The sentiments in Van Duivenboden were repeated in Za v Smith[13] by Brand, JA and the learned judge of appeal proceeded as follows:

“Unlike the court a quo, I therefore do not think it can be found as a fact that the warning measures proposed by Tromp (appellant’s expert) would be of no consequence. On the contrary, in my view, they would probably have been effective. This means that, but for the respondents’ wrongful and negligent failure to take reasonable steps, the harm that befell the deceased would not have occurred.”

[31] In Fourway Haulage SA (Pty) Ltd v SANRAL[14] Brand, JA cautioned that the factors normally applied to consider legal causation “should not be applied dogmatically, but in a flexible manner so as to avoid a result which is so unfair or unjust that it is regarded as untenable.”

[32] In SA Hang and Paragliding Association v Beswink[15] Brand, JA again had an opportunity to consider and adjudicate a claim based on delict. He explained the “but for” test in respect of factual causation and found in favour of the appellants. He also dealt with legal causation or remoteness and stated that this is determinded by considerations of policy, it being a measure of control or as he called it “a long-stop where right-minded people, including judges, will regard the imposition of liability in a particular case as untenable, despite the presence of all other elements of delictual liability.” The ultimate question to be asked in casu, based on Brand, JA’s reasoning further on in the paragraph as well as in Za v Smith supra, is whether the attack of the dog and the wounds sustained by the deceased as a result caused or materially contributed to his death.

[33] In Lee[16] the Constitutional Court rejected the approach of the court a quo that failed to follow the approach in Van Duivenboden. Nkabinde, J, the scribe of the majority judgment (there was a 5/4 split) dealt with causation and the flexible approach to determine factual causation and accepted in paragraph [47] that the “most recent, post-constitutional affirmations of that flexibility are to be found in Van Duivenboden and Gore.” The learned judge concluded in paragraph [73] as follows: “A court ultimately has to make a finding as to whether causation was established on a balance of probabilities on the facts of each case. Causation will not always follow whenever a wrongful and negligent omission is shown.”

[34] I am of the view that there is no reason why the “but for” test for factual causation should be based on metaphysics. It is rather a matter of common sense as Brand, JA held,[17] taking into account the background of everyday-life experiences and considering the matter in a practical manner. The appellant

merely had to prove on a balance of probabilities that the deceased died on 18 April 2011 as a result of the wounds, or put differently,

that it was more likely than not that the deceased would not have died on 18 April 2011, but for the wounds. The causal link does not have to be established with certainty. In order to consider causation it is necessary to evaluate the expert testimony. In order to do so, it is useful to refer to the principles to be applied.

[35] Experts are frequently called in to assist our courts, but courts are not bound by the opinion of an expert. An expert must be called as a witness on matters calling for specialised knowledge. It is the duty of the expert to furnish the court with the necessary scientific criteria for testing the accuracy of the expert’s conclusions so as to enable it to form an independent judgment by the application of these criteria to the facts proved in evidence. See Coopers (South Africa) (Pty) Ltd v Deutsche Gesellschaft Für Schädlingsbekämpfung MBH.[18] In the evaluation of the evidence of experts it is required to determine whether and to what extent their opinions are founded on logical reasoning.[19] This approach has been consistently followed. In Linksfield[20] the SCA warned presiding officers against adjudicating evidence by applying the yardstick used by expert scientific witnesses who tend to assess likelihood in terms of scientific certainty. Presiding officers should not be seduced to apply to the expert

evidence the standard which the expert applies, but instead, the balance of probabilities must be ascertained based on a review of the totality of the facts.

[36] In Oppelt v Department of Health[21] Molemela AJ (as she then was) criticised the court a quo for deviating from the approach earlier adopted in Linksfield and I quote:

“[38] First, Dr Newton testified that the dislocation of the spinal cord causes pressure and obstruction in the spinal canal which, when left unattended, results in the secondary ischaemic injury to nerve cells, to the extent that the cells cease to function. This evidence passes the reasonable and logical requirement for the acceptance of expert evidence set in Linksfield. The Supreme Court of Appeal erred when it concluded that the scientific evidence that supports his theory is 'questionable'. The conclusion deviates from the Linksfield principle that where the logic of a medical approach is not in dispute, the court must not assess the cogency of scientific evidence by scientific standards, but by the legal standard of the balance of probabilities.

[41] Fourth, the Supreme Court of Appeal fell into the trap of focusing on scientific proof instead of assessing where the balance of probabilities lies, based on an evaluation of the whole evidence.”

[37] In Medi-clinic v Vermeulen[22] Zondi, JA, writing for a unanimous SCA bench, stated the following with reference to the judgment of that court in Michael and another v Linksfield Park Clinic (Pty) Ltd:

“An opinion expressed without logical foundation can be rejected. But it must be borne in mind that in the medical field it may not be possible to be definitive. Experts may legitimately hold diametrically opposed views and be able to support them by logical reasoning. In that event it is not open to a court simply to express a preference for the one rather than the other and on that basis to hold the medical practitioner to have been negligent. Provided a medical practitioner acts in accordance with a reasonable and respectable body of medical opinion, his conduct cannot be condemned as negligent merely because another equally reasonable and respectable body of medical opinion would have acted differently.” (emphasis added).

[38] The deceased was not a healthy man. He suffered from liver failure and liver cirrhosis, an irreversible condition which would have led to his death ultimately, although the experts were in agreement that he could live for a number of years. He started to feel ill two weeks before the incident, but still continued with his normal business activities. He consulted a doctor who arranged for blood tests and on the day of the incident the doctor informed him telephonically that according to the results of the blood tests he was suffering from jaundice. Notwithstanding his illness, he was not bed-ridden. The Friday prior to the incident he and appellant were busy with restoration of a building and in particular the painting thereof. The deceased was even on the roof, doing some repairs. Her version in this regard was not challenged. The deceased was apparently fit enough to do manual labour and drive around.

[39] I do not agree with the trial court that it should be held against appellant for failing to call the three doctors that treated the deceased after the incident. The same applies to the absence of blood cultures and a post-mortem examination and results. The critical question to be asked is whether the deceased would have died on 18 April 2011, but for the wounds sustained ten days earlier, or put differently: but for the wounds, would Type 1 HRS have ensued and would the deceased have passed away as a result of that on 18 April 2011. Although the deceased was diagnosed with jaundice, the fact remains that he would not have been rushed to hospital for treatment on 8 April 2011, let alone emergency treatment, was it not for the wounds sustained.

[40] The experts were in agreement that the available tests indicated signs of renal failure 24 hours after admission at the hospital and a progressive worsening of the deceased’s renal functions thereafter without a significant worsening of his liver functions.

Prof Botha conceded during cross-examination that the cause of death was not liver cirrhosis and portal hypertension, but Type 1 HRS. It was also agreed by the experts, based on the available evidence, that the deceased’s white cell count increased and as conceded by Prof Botha, this was most probably as a result of an underlying bacterial infection. He also conceded that the deceased’s white cell count was within normal limits upon admission and thus excluding an underlying infection.

[41] Prof Botha conceded that the most probable cause of the infection was the wounds, but later tried to backtrack on the concession in saying that it was possible, but improbable. Despite his concession he was not prepared to concede that the bacterial infection of the wounds caused the Type 1 HRS; instead he insisted that spontaneous bacterial peritonitis was the most common infection that induces Type 1 HRS. The professor did not at all deal with spontaneous bacterial peritonitis in his written report, neither provided a basis for such conclusion in that report, nor did he do so in his evidence in chief. Dr Van Aswegen was neither cross-examined on this aspect, nor was the professor’s proposition put to him for his response. In my view the professor resorted to speculation and his version should have been rejected as not based on a logical foundation where he disagreed with the logical version of Dr Van Aswegen.

[42] Although Prof Botha’s wife suffered from the Type 1 HRS syndrome, he never treated patients suffering from such illnesses.

His opinion that the onset of peritonitis following bacteraemia within a 36 hour period was unlikely is based on unsubstantiated

facts and speculation as to when Augmentin was administered to the deceased. Dr Van Aswegen, relying on multiple documented case studies and his experience, testified that bacteraemia follows quite rapidly after a dog bite which leads to subclinical peritonitis which in turn leads to Type 1 HRS. This can occur within 24 hours of a bacterial infection. His opinion that the onset of bacterial peritonitis triggering Type 1 HRS within a 24 hour period was both possible and likely was not seriously attacked in cross-examination.

[43] I agree with Mr Zietsman that Prof Botha tendered viva voce evidence on aspects which was far removed from his written opinion. Although Prof Botha made several concessions, his unwillingness

to concede that the admitted infection caused by the wounds induced the Type 1 HRS, but rather relying on spontaneous bacterial

peritonitis without any evidence to support such contention, does not take account of the probabilities of this case and reflects

negatively on the witness’ objectivity. As mentioned, the deceased was actively involved with his business activities and even doing manual work immediately before the incident. On all probabilities he would not be hospitalised that day, was it not for the wounds sustained.

[44] In summary, Dr Van Aswegen’s opinion was not properly considered by the trial court. Liver failure on its own does

not cause HRS. Furthermore, biochemically and clinically, the deceased did not suffer from an acute deterioration of his liver functions. Dr Van Aswegen testified with reference to objective evidence that by the time of his death, the deceased’s liver did not significantly decrease in its function, but his kidneys totally decreased in function. This was in essence common cause between the experts. The trial court should have found, based on the evidence of Dr Van Aswegen, that the wounds caused bacteraemia which in turn caused sub-clinical peritonitis which induced Type 1 HRS and this ultimately caused multiple organ failure and the deceased’s death. Therefore, the trial court erred in concluding that causation was not proved because the deceased was ill, his immune system compromised and the wounds sustained did not cause his death, “given his pre-existing medical history.” I refer to my earlier criticism which I do not intend to repeat.[23] The trial court was unfortunately persuaded by the following statement by Prof Botha pertaining to the deceased’s injuries:

“These would not normally be regarded as being serious or life threatening and I would have expected a healthy patient with normal reactions that the wounds would heal without any …(inaudible) if they were treated adequately and with combative care.”

VIII

CONCLUSIONS

[45] The two issues to be considered in this appeal, to wit liability based on the actio de pauperie, specifically whether the dog acted contra naturam sui generis and causation, have been dealt with in detail. I am satisfied that the trial court erred in finding against appellant on these two issues and therefore the appeal must succeed.

[46] Appellant as the successful party is entitled to the costs of the appeal as well as the costs in the trial court.

IX

THE ORDERS

[47] Consequently the following orders are made:

1. The appeal succeeds with costs.

2. The order of the trial court is set aside and substituted with the following orders:

“1. The defendant is liable for payment to the plaintiff of all her proved or agreed damages.

2. Defendant is liable for plaintiff’s party and party costs of the merits trial, such costs to include the reasonable preparation,

qualifying, reservation and attendance fees of Dr A van Aswegen.”

_____

J P DAFFUE, J

I concur

M MATHEBULA, J

PJ LOUBSER, J

On behalf of Appellant: Adv PJJ ZIETSMAN

Instructed by: Honey Attorneys

BLOEMFONTEIN

On behalf of Respondent: No appearance

[1] The leading cases are O’Callaghan NO v Chaplin 1927 AD 310 at 329 and Coetzee & Sons v Smit 1955 (2) SA 553 (A) at 558A.

[2] At 329 and approved in numerous judgments, inter alia Solomon and another NNO v De Waal 1972 (1) SA 575 (AD) at 581D and further.

[3] Van Niekerk v Jantjes [1996] 2 All SA 517 (E) at 519f – 520a; Loriza Brahman en ‘n ander v Dippenaar 2002 (2) SA 477 (SCA) at paras [18] – [20].

[4] 1970 (3) SA 603 (TPD) at 604E-G.

[5] Ibid.

[6] 2018 (2) SA 87 (SCA) at par [20].

[7] At par [36].

[8] Lever v Purdy 1993 (3) SA 17 (A) at 25G – 26A.

[9] Ibid.

[10] Minister of Police v Skosana 1977 (1) SA 31 (A) at 34E-G.

[11] Standard Chartered Bank of Canada v Nedperm Bank Ltd [1994] ZASCA 146; 1994 (4) SA 747 (A) at 764I – 765A.

[12] Minister of Safety & Security v Van Duivenboden 2002 (6) SA 431 (SCA) at par [25].

[13] 2015 (4) SA 574 (SCA) at par [32].

[14] [2008] ZASCA 134; 2009 (2) SA 150 (SCA) at par [34].

[15] 2015 (3) SA 449 (SCA) at par [37].

[16] Lee v Minister of Correctional Services 2013 (2) SA 144 (CC)

[17] Za v Smith op cit at par [30]. See also Van Duivenboden par [25] and Lee par [41].

[18] 1976 (3) SA 352 (A) 370H – 372A.

[19] Michael and another v Linksfield Park Clinic (Pty) Ltd and another 2001 (3) SA 1188 (SCA) paras [36] – [40] and Medi-clinic v Vermeulen 2015 (1) SA 241 (SCA) at paras [5] – [8] & [25], [26] & [31].

[20] Op cit at par [40].

[21] 2016 (1) SA 325 (CC) paras [38] – [41].

[22] Op cit at par [5].

[23] Par [19] supra.