Crafford v South African National Roads Agency Ltd (215/2012) [2013] ZASCA 8 (14 March 2013)



THE SUPREME COURT OF APPEAL OF SOUTH AFRICA


JUDGMENT




Case No: 215/12

Reportable



In the matter between:


DIRK JOHANNES CRAFFORD Appellant



and



THE SOUTH AFRICAN NATIONAL ROADS AGENCY LIMITED Respondent





Neutral citation: Crafford v South African National Roads Agency Limited (215/2012) [2013] ZASCA 8 (14 March 2013)


Coram: Brand and Leach JJA and Schoeman, Plasket and Saldulker AJJA


Heard: 21 February 2013


Delivered: 14 March 2013


Summary: Claim for damages arising out of a collision between a motorist and a kudu at night ─ appellant alleging collision due to respondent’s negligent failure to mow the grass in the road reserve ─ no direct evidence as to how the collision occurred ─ appellant failing to prove that the respondent’s alleged negligence the cause of the collision.





___________________________________________________________________


O R D E R

___________________________________________________________________



On appeal from: North Gauteng High Court, Pretoria (Louw J sitting as court of first instance):

The appeal is dismissed with costs, including the costs of two counsel.

___________________________________________________________________

J U D G M E N T

__________________________________________________________________


LEACH JA (BRAND JA AND SCHOEMAN, PLASKET AND SALDULKER AJJA concurring):


[1] The superstition that Friday the 13th is an unlucky day proved to be true in the case of the appellant on 13 January 2006. Late that Friday night he was driving home along the R510 national roadway between Thabazimbi and Ellisras when his motor vehicle collided with a kudu cow. The impact was considerable, the kudu having been propelled through the windscreen into the interior of the vehicle. Most unfortunately, the appellant was severely injured in the process.

[2] The respondent was at the time responsible for the maintenance of both the road on which the collision occurred and the road reserve immediately adjacent thereto and, in due course, the appellant sued the respondent for damages suffered as a result of his injuries. His claim was essentially founded on an averment that the respondent or its employees had negligently failed to cut the grass alongside the road and had allowed it to grow so high that it had prevented him from seeing the kudu until it entered the road, at which stage it was too late to avoid a collision. This the respondent denied.

[3] When the matter came to trial in the North Gauteng High Court, Pretoria, the hearing proceeded solely in respect to the question of liability with the issues relating to the appellant’s damages standing over for later determination. The high court held that the appellant had failed to establish that the condition of the grass alongside the road had caused the collision and dismissed the appellant’s claim. With the leave of the high court the appellant appeals to this court against that order.

[4] Before October 2004 the R510 and its road reserve had been the responsibility of the Limpopo Provincial Government. However, on 22 October 2004 the road was declared to be a national road under s 40(1) of The South African National Roads Agency Limited and National Roads Act 7 of 1998, whereupon the respondent assumed responsibility for the maintenance and care of both the road and its reserve. In the 15 month period that elapsed from then until the incident giving rise to the appellant’s claim, the grass in the road reserve at the scene of the collision had not been mown but been left to grow wild (mowing the road reserve alongside the R510 had commenced but had not yet been completed). It is upon this omission that the appellant founded his claim for damages.

[5] It is trite that an injured party who seeks to recover damages in a situation such as this must establish that the omission sued upon was (a) negligent, (b) wrongful – in the sense that it was a failure which as a matter of public and legal policy should be regarded as actionable – and (c) caused the alleged loss. This third element requires proof of the omission complained of having been in fact a cause of the loss suffered – commonly known as factual causation – and, once that is established, that the loss is not too remote but sufficiently linked to the loss to attract liability – so called legal causation.1

[6] All three of these elements, negligence, wrongfulness and causation, were placed in issue, both in this court and in the court a quo. The high court’s decision to dismiss the appellant’s claim as the condition of the grass on the road reserve had not been shown to have caused the plaintiff’s collision with the kudu, is solely one of factual causation.

[7] The classical formulation for deciding whether an omission caused the loss allegedly suffered is the so-called ‘but for’ test. This involves an enquiry as to whether, but for the omission, the loss probably would not have occurred. In this regard, the question is not one of mathematical percentage but, rather, what is more likely. As was explained by this court in Minister of Finance & others v Gore NO,2 a judgment recently referred to with approval by the Constitutional Court:3

‘[A]pplication of the “but for” test is not based on mathematics, pure science or philosophy. It is a matter of common sense, based on the practical way in which the ordinary person's mind works against the background of everyday-life experiences. Or, as was pointed out in similar vein by Nugent JA in Minister of Safety and Security v Van Duivenboden:

“A plaintiff is not required to establish the causal link with certainty, but only to establish that the wrongful conduct was probably a cause of the loss, which 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 metaphysics.”’4


[8] Bearing these principles in mind, I turn to consider whether the court a quo correctly held that the appellant had failed to prove the element of factual causation. Although it was common cause that the collision had occurred on the R510 national road some 50 kilometres from Thabazimbi, the precise scene was a matter of some dispute at the hearing, the respective parties having identified two places a few hundred metres from each other as being where the impact had occurred. In my view, nothing really turns on this and it can be accepted that it was on a straight stretch of tarred road running approximately east to west. The road surface was approximately 7,6 metres in width while the road reserves both to the north and to the south of the edges of the roadway were approximately 15 metres wide. After the collision the appellant’s motor vehicle was found stationary on the trafficable surface of the roadway, and the inference is irresistible that the kudu cow was on the road at the time of impact.

[9] There was considerable dispute at the trial in regard to the condition of the road reserve at the material time, in particular in regard to the length and density of the grass. It can be accepted that there were clumps of a grass known as hyperthelia dissoluta that can grow to at least two metres in height but, unfortunately, no photographs were taken shortly after the incident nor was there any detailed evidence from which it is possible to reliably determine the length and density of the grass at the place where the impact occurred at the relevant time. Despite this difficulty, it can be accepted that the grass had been allowed to grow tall and was, in places, in clumps sufficiently large and dense to have seriously affected an approaching driver’s ability to see kudu on the road reserve ahead.


[10] It must be mentioned that photographs handed in at the trial show that the road passes through fairly thick bushveld where kudu are likely to abound. The road reserve to the north of the road was at the time separated from the surrounding veld only by a standard stock-proof fence which, so the evidence establishes, kudu are capable of clearing with no difficulty at all. Moreover, not only were there nearby traffic signs warning of the possible presence of kudu, but the appellant, who used the road regularly to travel between his home on a game farm near Ellisras and his work in Pretoria, had on several occasions seen kudu next to the road, both by day and at night.

[11] The plaintiff unfortunately sustained severe head and facial injuries which has given rise to retrograde amnesia, and he has no recall of the circumstances under which the collision occurred. However, despite the appellant’s inability to describe what happened, it was argued on his behalf that the most probable inference to be drawn from the known facts is that the kudu with which he collided was not visible to him by reason of the long uncut grass until it emerged into the roadway at a time when it was too late to take effective avoiding action.

[12] It takes little imagination to think of various circumstances in which the collision may have taken place. The kudu may have been in the road as the appellant approached but was not seen by him until it was too late; the appellant may have seen the kudu crossing the road at a safe distance ahead and relaxed his vigilance, only for it to turn back when the vehicle was close by; the kudu may have come bounding over the stock fence and across the road reserve, either in fright or because the bull of the herd on the other side of the road barked a call. These are but a few scenarios that readily spring to mind, but the list is truly endless. What has to be considered is whether the scenario advanced by the appellant, namely, that the kudu emerged into the roadway from a position in which it had been obscured by the long grass in the road reserve at a time when it was too late to take effective avoiding action, has been factually established as being what probably occurred

[13] The appellant’s argument is that he proved that the collision occurred in these latter circumstances as a result of the necessary inferences which are to be drawn from the other known facts. Of course it is necessary at all times to distinguish between speculation on the one hand and factual inference legitimately drawn from proven objective facts on the other. An inference of fact can only properly be drawn if the other proven facts justify doing so. If they do not, the inference sought to be drawn becomes mere speculation.

[14] Where, as here, a court is asked to draw inferences of fact in relation to an alleged negligent omission in determining whether factual causation has been established, the enquiry is further complicated by requiring ‘the substitution of a hypothetical cause of lawful conduct for the unlawful conduct of the defendant and the posing of the question as to whether in such case the event causing harm to the plaintiff would have occurred or not’.5 As was observed by this court in Gore NO, inferential reasoning is required in such circumstances to determine ‘what would have happened if the wrongful conduct is mentally eliminated and hypothetically replaced with lawful conduct?’6 Accordingly it becomes necessary to consider whether it has been shown by inference drawn from the known facts that the kudu with which the appellant collided would probably have been visible to him early enough for effective avoiding action to be taken had the road reserve been mown short.


[15] In considering that issue, it must be remembered that the incident occurred late at night at a time when the appellant’s ability to see the road and the adjacent road reserve ahead of him was considerably compromised, his range of vision having been restricted to the area illuminated by his headlights. Of course drivers do not necessarily have to be able to stop within the range of their headlights at all times, but a kudu which ventures into a main road at a place beyond the lights of an approaching vehicle might well be difficult to avoid if a driver cannot stop in time once it becomes visible. In addition, the evidence in this case established that even if the grass had been cut, the appellant’s lights would not have illuminated the whole of the road reserve but only about that half closest to the road ie up to about eight metres from the edge of the road. Thus neither a kudu on the road reserve beyond that range nor one beyond the fence would have been visible in the appellant’s lights.

[16] It must also be remembered that notwithstanding their large size, kudu are extremely difficult to see at night. Not only is this a matter of common experience but the parties’ experts were agreed that ‘kudu are by way of skin colouring and habits well camouflaged and are difficult to see’ and that they are ‘more difficult to see at night, even when light sources like headlights of cars are present’. This was also borne out by the evidence of Mr Du Toit, called on behalf of the plaintiff, and the information contained in the annexures to his expert’s report to which he referred, which established that, as kudu are difficult to see at night and often move into and across roadways despite the presence of fences, collisions with them are all too common an occurrence on roads traversing rural areas in which kudu are regularly found.

[17] A further consideration is the speed at which kudu are capable of travelling. The evidence established that a kudu trots at about 30 kph and can run at 50 kph or faster. Consequently, during the 1.5 seconds which the evidence established it takes the average driver to visualise, perceive and react to the sudden appearance of danger, kudu will travel approximately 12.5 metres if it is trotting and approximately 20 metres if running. Over the same period, a motor vehicle driven at 100 kilometres per hour (as the appellant testified he would have been driving) will cover about 40 metres. It is readily apparent from this exercise that a kudu on a road reserve trotting towards the adjacent road from a position beyond the range of an approaching vehicle’s lights will only become visible at best a second or so before it reaches the road itself, and that a motorist approaching at 100 kph, then 50 metres or so away at that stage will hardly have the opportunity to react, let alone to take effective avoiding action, before reaching the animal.

[18] While I am acutely aware of the difficulties attendant upon attempting to make mathematical calculations in matters of this nature, based as they are on estimates as to positions, speeds and motions which may not be at all reliable, it is nevertheless an exercise ‘useful as a check’7 that shows how difficult it may be to avoid colliding with a moving kudu while driving along a road at night, even where there is nothing to obstruct visibility alongside the roadway.

[19] The truth of the matter is that even had the grass alongside the road been short at the time, one does not have sufficient information to determine how the collision probably took place. The list of imponderables is infinite. We do not know whether the kudu came from the northern or southern side of the road, nor whether it was trotting or running. Even accepting that the appellant was driving at about a 100 kph, one has no idea how, in what manner and at what speed the kudu moved as the gap between it and the motor vehicle closed. It may have moved slowly into the road from a position in which it was standing behind a large clump of grass close to the road but, equally possibly, it may have come at a run from the bushveld beyond the road reserve, clearing the fence and charging towards the road into the roadway directly in front of the vehicle, giving the appellant no chance to see it. To find that any one of these scenarios is in fact what probably occurred would be to indulge in impermissible speculation.

[20] Moreover, even if the grass in the reserve had been short, it would also be a matter of speculation to find that the kudu would have become visible when there was still sufficient space between it and the approaching appellant for successful avoiding action to be taken. Possibly it could have been seen in time but, bearing in mind the restricted view of the road reserve that the appellant would have had in his headlights and the difficulty one has of seeing a kudu at night even when they fall within the range of a vehicle’s lights, the kudu may well not have been visible to the appellant until a very late stage even if his ability to see had not been compromised by the long grass there was on the night of the collision. Certainly that is no less probable than the kudu becoming visible when there was sufficient time and space to avoid it ─ or, for that matter, that the collision occurred in any other way.


[21] Without knowing where the kudu came from, how it moved, the manner in which it came to be in the road, and where it and the appellant’s motor vehicle were in relation to each other at any material time, it is really impossible to determine solely from the fact of a collision where the kudu would have been and at what stage it would have become visible to an approaching motorist, irrespective of the length of the grass alongside the road. In my view there are thus insufficient objective facts from which it can be inferred that if the grass alongside the road had been kept short the appellant would have seen the kudu earlier than he did, let alone that on seeing it he would have had sufficient time and space to have reacted and slowed his vehicle sufficiently to avoid a collision. The appellant therefore failed to establish on a balance of probabilities that if the grass had been kept short the collision would not have occurred.

[22] I am accordingly of the view that the trial court correctly concluded that the appellant had failed to discharge the onus of establishing that the state of the road reserve caused the collision. This renders it unnecessary to consider the elements of negligence and wrongfulness as, even if they were to be determined in the appellant’s favour, his claim would still fail.

[23] I need to mention one final matter. It is important for the administration of justice that the roll of this court is not clogged by cases which should properly have been referred to a full bench of the high court. The inappropriate granting of leave to appeal to this court has been adversely commented on regularly in the past.8 Section 20(2)(a) of the Supreme Court Act 59 of 1959 requires a high court granting leave to appeal to direct that the appeal be heard by a full court ‘unless it is satisfied that the questions of law and of fact and the other considerations involved in the appeal are of such a nature that the appeal requires the attention of (this court)’. This is clearly not such a case. Although the damages claimed by the plaintiff were substantial, it is a relatively straightforward factual matter involving well-settled legal principles. Leave to appeal should clearly have been granted not to this court but to a full court.

[24] Be that as it may, the appeal is dismissed with costs, including the cost of two counsel, including the costs of two counsel.





_____________________

L E Leach

Judge of Appeal


APPEARANCES:


For Appellant: J F Mullins SC (with him J D du Plessis)

Instructed by:

Van Zyl Le Roux Inc, Pretoria

Honey Attorneys, Bloemfontein

For Respondent: J D Maritz SC (with him T Potgieter)

Instructed by:

Savage Jooste & Adams, Pretoria

Webbers Attorneys, Bloemfontein

1 See in this regard eg Siman & Co (Pty) Ltd v Barclays National Bank Ltd 1984 (2) SA 888 (A) at 914F-H and Delphisure Insurance Brokers v Dippenaar 2010 (5) SA 499 (SCA) para 28.

2Minister of Finance v Gore NO 2007 (1) SA 111 (SCA) para 33.

3 See Lee v Minister of Correctional Services [2012] ZACC 30 para 47.

4Minister of Safety and Security v Van Duivenboden 2002 (6) SA 431 (SCA) ([2002] 3 All SA 741) para 25.

5Siman & Co at 915E-F.

6Gore NO para 32.

7 Per Ogilvie Thompson AJ in Van der Westhuizen v SA Liberal Insurance Co Ltd 1949 (3) SA 160 (C) at 168, a comment subsequently approved by this court – see eg Netherlands Insurance Co of SA Ltd v Brummer 1978 (4) SA 824 (A) at 831E-F.

8 See eg Shoprite Checkers (Pty) Ltd v Bumpers Schwarmas CC 2003 (5) SA 354 (SCA).

▲ To the top