---

IN THE HIGH COURT OF SOUTH AFRICA

CAPE OF GOOD HOPE PROVINCIAL DIVISION

Case no 9578/04

In the matter between

MARK HARRINGTON N O First plaintiff

SIYAVUMA NGALEKA Second plaintiff

and

TRANSNET LIMITED

trading as METRORAIL First defendant

THE SOUTH AFRICAN RAIL

COMMUTER CORPORATION LIMITED Second defendant

JOHANNES CHRISTOFFEL HUMAN Third defendant

KUFFS SECURITY SERVICES CC Third party

JUDGMENT DELIVERED ON 22 AUGUST 2006

BLIGNAULT J:

[1] On the night of 3 February 2002 Mr Sijongile Ncaza (“first plaintiff”) and Mr Siyavuma Ngaleka (“second plaintiff”) were on duty as security guards. They were patrolling the railway lines from Cape Town to Woodstock on foot. Whilst walking along a railway line they were struck by a train that was moving from Cape Town to Woodstock. They sustained serious bodily injuries in the accident and as a result thereof they instituted this action for the recovery of damages.

The parties

[2] First plaintiff is represented in this action by Mr Mark Harrington, an advocate of this court, as curator ad litem. First and second plaintiffs are adult males residing in Khayelitsha and Delft, Cape, respectively. At the time of the accident they were employed as security guards by the third party, Kuffs Security Services CC (“Kuffs”).

[3] First defendant is Transnet Limited, a public company established pursuant to the provisions of section 2 of the Legal Succession to the South African Transport Services Act 9 of 1989 Act (“the SATS Succession Act”). Metrorail is a business unit of Transnet Limited that operates, inter alia, the Cape Town suburban railway system. I shall refer to first defendant herein as Metrorail. Second defendant is the South African Rail Commuter Corporation Limited, a corporation created in terms of section 22 of the SATS Succession Act. It is the owner of the property where the accident took place and the owner of the train in question. Third defendant is Mr Johannes Christoffel Human (“Human”). He was the driver of the train, acting in the course and scope of his employment with Metrorail.

[4] The third party, Kuffs, is a close corporation which carries on business as a provider of security services and has its principal office at Grassy Park, Western Cape.

The pleadings

[5] First and second plaintiffs particulars of claim contain, inter alia, the following allegations of negligence on the part of defendants:

“8. The said collision was caused by the negligence of First and/or Second and/or Third Defendant, in that:

8.1 they failed, notwithstanding being aware that the railway

line was patrolled at night after 22h00 by security guards such as First and Second Plaintiffs, to warn Plaintiffs of the approach of the train; and/or

8.2 they failed, notwithstanding being aware of the facts and circumstances set out in sub-paragraph 8.1 above, to warn and/or advise Plaintiffs and/or their employer, Kuffs Security Services (‘Kuffs’) of the unscheduled journey of the said train; and/or

8.3 the Third Defendant failed to keep a proper lookout; and/or

8.4 the Third Defendant failed to apply the brakes of the said train timeously, adequately, or at all; and/or

8.5 the Third Defendant failed to warn the Plaintiffs of the approach of the train by flashing the lights of the train; and/or

8.6 the First and Second Defendants failed to act with due care in that they:

8.6.1 failed to take any or reasonable measures to ensure that security personnel working on their premises received adequate safety training;

8.6.2 failed to take any or reasonable measures to ensure that the employees of Kuffs completed the First Defendant’s in-house test and induction training before commencing their duties on the First and Second Defendant’s premises;

8.6.3 failed to take any or reasonable measures to ensure that Kuffs provided its employees with adequate safety training for their duties on the First and Second Defendant’s premises;

8.6.4 failed to take any or adequate measures to ensure that Kuffs undertook adequate Rail Safety Awareness Programmes, appointed Safety Representatives, and held safety meetings of which proper minutes were kept;”

[6] In their plea defendants denied the allegations of negligence levelled against them. They pleaded that the collision was caused by plaintiffs’ own negligence. Metrorail issued a third party notice in terms of rule 13 against Kuffs. Metrorail alleged that plaintiffs were performing their duties pursuant to a contract between Metrorail and Kuffs in terms of which Kuffs agreed to provide security services to Metrorail. Metrorail claimed that in terms of an indemnity provided under the contract between them, Kuffs was obliged to indemnify it against the claims in question. Kuffs admitted the indemnity but pleaded that neither first nor third defendant was negligent. Kuffs later amended its plea in order to allege that second defendant was negligent in that it failed to take adequate steps and precautionary measures that would have enabled Human to avoid the accident.

[7] The parties agreed that the issues of negligence and causation would be determined separately from those relating to the quantum of plaintiffs’ damages. That agreement was sanctioned by the court.

The evidence

[8] At the hearing various plans and photographs were tendered in evidence. The court also attended an inspection in loco. The material facts regarding the place of the collision and the surrounding circumstances are not in dispute. Railway lines leave Cape Town station in an easterly direction from about 24 platforms. Some of these are main lines and others are suburban lines. On the extreme southern side is the line known as the Simonstown up line (leading from Simonstown to Cape Town) and next to it the line known as the Simonstown down line (leading from Cape Town to Simonstown). These lines pass through Woodstock station which is about 2 kilometres from Cape Town station and Salt River station which is about two kilometres further. At a point about 200 metres to the west of Woodstock station the Simonstown down line crosses over another railway line by way of a single bridge. It is common cause that the accident occurred close to this bridge on the Cape Town side thereof, on the Simonstown down line.

[9] Mr Julian Gounder was called as a witness on behalf of plaintiffs. At the time of the accident he was employed by Kuffs. There was in existence a contract between Kuffs and Metrorail in terms of which Kuffs rendered certain security services to Metrorail. The contract came into operation on 1 February 2001 for an initial period of 12 months. It was then extended by agreement for an additional period of 6 months. Gounder said that he was the operations manager in respect of this contract. with Metrorail. As such he was responsible for supervising Kuffs’ performance of its contractual duties. His normal working hours were during day time but he carried out night time visits from time to time.

[10] In terms of the contract Kuffs provided a number of security services to Metrorail. One of these was the provision of two security guards to patrol the railway line from Cape Town to Woodstock in order to guard against the theft of signal cables or overhead cables. This provision of guards to prevent cable theft did not from the outset form part of the contract. It was introduced a few months later. The individual guards were required to pass Metrorail tests for the use of firearms but no specific training or tests were provided by Metrorail for cable patrol. The arrangement was that the guards would perform cable patrol from the time when the last scheduled train had passed in the evening (about 22:00) until the first scheduled train passed through the next morning, which was at about 04:00. Metrorail had an operations room at Cape Town station. Kuffs had their own control room at Cape Town station which could make radio contact with the guards on duty. Metrorail, he said, was supposed to inform Kuffs when a train was about to pass on the lines in question during that period. This arrangement had been conveyed to him by Mr Apollis, an official of Metrorail in charge of security.

[11] On the night in question first and second plaintiffs were responsible for cable patrol. According to the check sheets completed that evening they were both clothed in Kuffs’ uniform. The ordinary uniform at that time included the wearing of a reflective orange vest over their clothes. He received a call at home that night informing him that two persons had been knocked over by a train. When he arrived at the scene of the accident both plaintiffs were lying on opposite sides of the railway line. Both were unconscious. Second plaintiff regained consciousness whilst he was on the scene. A strong wind was blowing that evening. The two plaintiffs were removed by ambulance. He (Gounder) completed a standard Metrorail incident form in respect of the accident. According to him he also completed a written report of the accident which he submitted to Metrorail but he never saw that report again. In preparing that report he spoke to the driver of the train who told him that he did not see anyone on the track until it was too late to stop without colliding with them.

[12] Mr Bayetha Bidli testified that he worked as a security guard for Kuffs from February 2001 until he resigned in 2004 in order to join the SA Police Service. He joined Kuffs as part of a large group of security guards that were employed for purposes of the contract that had been awarded to Kuffs by Metrorail. The whole group received some basic training from Kuffs and they were taken for an orientation at the Cape Town station where they were shown various points of interest. They did not, however, receive any specific training or tests on cable control by Kuffs or Metrorail. Bidli was normally stationed at Cape Town station. He performed cable patrol duty between Cape Town and Salt River on about ten occasions. It formed part of the night shift which usually started with a parade at 17:45 and extended from 18:00 until 06:00 the next morning. After about three to four months the wearing of reflective vests was introduced as a compulsory element of their uniform. If a security guard on duty was found without a vest, Metrorail could impose a penalty on Kuffs. He understood that the purpose of the cable patrol was to prevent the theft of the overhead and signal cables. The patrols covered the entire area from Cape Town station to Woodstock station and beyond to Salt River station. During the period from 22:00 to 04:00 there were no scheduled trains moving along these lines. He did on occasion see a train moving along the railway lines during those hours but that was something unusual.

[13] Mr Siyavuma Ngaleka (second plaintiff) testified that he started working for Kuffs as a security guard in February 2001. He was part of the same intake as Bidli. He received the same training as Bidli and the same orientation at the Cape Town station. He did not receive any specific training in regard to cable patrol. On 3 February 2002 he was on cable patrol from Cape Town to Woodstock station. This was not the first occasion that he did cable patrol. On his first patrol he had been accompanied by a colleague, Mr Mkhabe, who showed him what the work entailed. Mkhabe did not tell him that he was not allowed to walk on the railway tracks. At times he and Mkhabe walked on the tracks and at times they walked on the paths between the lines. Thereafter he performed cable patrol with various other guards. He never saw any trains moving around in the patrol area between the hours of 22:00 and 06:00 except the Shozoloza Express which was a mainline train and which did not follow the line to Woodstock. He said that he could not perform the work properly without crossing railway lines in the course of the patrol.

[14] On Sunday 3 February 2002, second plaintiff said, he reported for the parade at Woodstock station but he was then told that he was required to do cable patrol that night as there was a shortage of guards at Cape Town. His partner that night was first plaintiff. He wore a vest that evening over his jacket and so did first plaintiff. First plaintiff carried the radio. They set out from platform 19 at the Cape Town station that night. There was a strong wind blowing that evening and they were walking against the wind. At one stage they were walking on the footpath on the immediate right hand side of the Simonstown down line. As they were approaching the bridge they started to walk on the sleepers that were laid under the railway tracks. He walked on the sections of the sleepers to the right of the right track and first plaintiff walked on the sections of the sleepers to the left of the left track. They were fairly close to the bridge when they were struck by a train from behind. He could not recall hearing or seeing the train before the collision nor could he recall first plaintiff giving any indication that he had heard or seen a train. He recalled that he regained consciousness whilst lying on the ground next to the railway line. He said that he had no reason to think that there would be a train running on that line at that time of night. Second plaintiff testified that the had done cable patrol on eight occasions only but under cross-examination he accepted that according to Kuffs’ records he had done so on 22 occasions from 23 December 2002 to 3 February 2002.

[15] Mr Johannes Human (third defendant) testified that he joined the South African Railways in February 1975. He qualified as a train driver in 1981. Since 1983 he was employed as a train driver on suburban lines in the Western Cape. He recently resigned from Metrorail. He was the driver of the train that collided with first and second plaintiffs. That train left at about 22:30 on the evening in question. It was not unusual, he said, for trains to move around after 22:00. Some of these trains were moved to other platforms or shunting sidings. Others were taken for mechanical repairs or the repairs of equipment. The train in question comprised eight coaches of which the first and last coaches were motor coaches. The length of each coach is about 20 metres. There is a driver’s cab at each end. There are two windows in front of the cab. His own seat was behind the window on the left. On the left side in the front part of the cab is the brake handle and in the middle is the accelerator handle. The siren button is on the ground and is activated by foot. If the accelerator handle is not depressed the train will not move. If that handle is released it will revert to its original position. In that event the accelerator will cut out and the brakes will apply immediately and automatically.

[16] On the evening in question, Human said, he left from platform 3 at Cape Town station on the Simonstown down line. His supervisor had instructed him to move the train to the Salt River repair yard. Before his departure he performed routine safety checks of the headlights, the siren and the brakes. He said that in his experience pedestrians did not often walk between the railway tracks. It is more comfortable to walk on the flat areas between lines. His headlights were on bright that evening. Just after his departure he tested the brakes. They worked normally. He then proceeded along the line. At a bend he saw two black figures on the tracks in front of him. They were walking between the tracks. He activated the siren but the two figures did not react to the siren. He then pulled the brake handle and he closed the accelerator handle, releasing it at the same time. The train continued to move towards the two figures whilst braking. He saw them looking back at the train at the last moment and then jumping away – the one to the left and the other to the right. His train collided with them as they were in the process of jumping away. It was still moving forward at that point. The train came to a halt shortly thereafter.

[17] After the train had come to a standstill Human climbed out of the cab on the right hand side. The front of the train had stopped before reaching the bridge. He walked back to the person on the southern side of the tracks. It was second plaintiff. He was in a sitting position with his hands on both sides of his head. He (Human) asked him whether he was all right and he replied yes. He realised that he was a security guard because he was wearing a uniform. He did not see him wearing a reflective vest. Second plaintiff’s position on the ground, he said, was about 15 to 20 metres behind the front end of the train. He then went to look for the other person. He found him on the other side of the train, slightly further away as second plaintiff from the front of the train. He was lying on his side. When he (Human) asked him whether he was all right he just mumbled. He formed the impression that he was more severely injured than second plaintiff. He was also wearing a uniform but no vest. He went back to the driver’s cab in the train and he contacted the control station at Windermere by radio and reported the incident. He then proceeded on his journey towards Salt River. He explained that according to his standing instructions he was obliged to proceed on the journey unless a person could be injured by the further movement of the train. Human said that his speed when he saw plaintiffs was considerably less than 60 kilometres per hour. On the other side of the bridge he would have approached Woodstock station. The prevailing wind, he said, was a strong southeaster. Human testified that there was nothing that he could have done to avoid the accident.

[18] Defendants called Mr Brian Carver, a mechanical engineer, to give evidence as an expert. He obtained the degree of Bachelor of Science in Mechanical Engineering in 1969. During the period from 1969 to 1980 he was employed as a mechanical engineer in various positions by first and second defendants’ predecessor, the South African Railways. From 1980 to 1983 he was the technical manager of Knorr Bremse, a company involved in the marketing and development of rail and automotive braking systems. From 1983 to 1988 he was the sales manager of SCAW Metals, a company which carried ion the business of the marketing and development of railway products. From 1991 to 1994 he was the rolling stock manager of second defendant and from 1996 to 1999 the executive manager (operations) of Metrorail. Since 1999 he has been practising as a private consultant. In this capacity he has been employed on a number of projects for first and second defendants.

[19] Carver investigated the circumstances of the accident in which plaintiffs were injured. He inspected the scene of the accident and he prepared an accurate plan of the scene. For this purpose he measured the distances between various relevant points by pacing them. For purposes of describing the relevant distances in this judgment I propose to take the western edge of the side-wall of the bridge on the northern side of the railway line, as starting point, ie as Point zero, and then describe all points along the line with reference to their distance from the starting point. The masspole 1/11D, for example, is 23 metres from the western edge of the bridge. I shall describe its position as Point 23 m. Using that system of reference the positions of some of the main landmarks and the inferred points where first plaintiff and second plaintiff were struck by the train, are as follows:

Head of train at standstill Point 15 m

Mass pole 1/11D Point 23 m

The second (eastern) palm tree Point 33 m Collision with second plaintiff Point 34 m

Collision with first plaintiff Point 44 m

The first (western) palm tree Point 136 m

Signal box WDC 98 Point 229 m

[20] Carver prepared a table of stopping distances which shows the distance, at different speeds, which the train would have travelled after the brakes had been applied, to the point of standstill. These calculations were performed by him by means of a formula based on the actual characteristics of the train in question. The formula incorporates the 1/50 gradient which the train would have ascended in approaching the point of collision. The formula also allows for the time taken for the brakes to become fully operational after the brake handle had been applied, described as the train reaction time. Carver explained that the distances in his table do not allow for human reaction time. Depending upon the assumption made in that regard an additional distance must be allowed for human reaction time by using the applicable figure for speed in metres per second. The table of stopping distances reads as follows:

Speed km/h	Speed m/s	Time to stop	Dist to stop	Reaction time (3 sec)	Stopping Distance
		Seconds	Metres	Distance	Distance
5	1.39	1.26	0.88	4.17	5.04
10	2.78	2.53	3.51	8.33	11.84
15	4.17	3.79	7.89	12.50	20.39
17	4.72	4.29	10.14	14.17	24.30
20	5.56	5.05	14.03	16.67	30.70
25	6.94	6.31	21.92	20.83	42.75
30	8.33	7.58	31.57	25.00	56.57
35	9.72	8.84	42.96	29.17	72.13
40	11.11	10.10	56.12	33.33	89.45
45	12.50	11.36	71.02	37.50	108.52
50	13.89	12.63	87.68	41.67	129.35
55	15.28	13.89	106.10	45.83	151.93
60	16.67	15.15	126.26	50.00	176.26

[21] Carver made certain informed assumptions regarding the visibility of plaintiffs to the driver of the train. All Metrorail motor coaches are fitted with the same type of headlight. Two sealed beam, narrow spot lamps are fitted in the middle above the cab in the front of the coach. The manufacturer provided the following rating information in regard to this kind of lamp:

Designation PAR56

Wattage 300W

Voltage 120V

Light output 3840 Lumens

Centre beam candle power 68 000 Candela

Beam spread 10˚ horizontal 8˚ vertical

Field Angle 20˚ horizontal 14˚ vertical

“Beam spread”, he explained, is the angular dimension of the cone of light encompassing the central part of the beam out to the angle where the intensity is 50% of the maximum. “Field angle” is the angular dimension of the cone of light encompassing the central part of the beam out to the angle where the intensity is 10% of the maximum.

[22] As two of these lamps are fitted on each motor coach they would produce 132 000 Candela. Carver compared this with a standard, issued by the USA Federal Railroad Administration, of 200 000 candela which is derived from the requirement that the headlight should be strong enough to illuminate a dark object the size of a man at a distance of at least 800 feet (244 metres). On this basis he estimated that a dark object such a person would only have become reasonably visible at night to the driver of a Metrorail train at a distance of approximately 160 metres directly ahead of the train when the headlights are on bright. The intensity of the light will reduce to 50% at an offset of 5 degrees (14 metres) and to 10% at an offset of 10 degrees (28 metres).

[23] In his report and evidence Carver expressed the view that plaintiffs were possibly unaware of the approach of the train. The train is an electric train and it is silent when it runs. It was fitted with an electric siren and not an air horn as air horns were prohibited on the Simonstown line. Electric sirens are not as loud as air horns. The windy conditions could have drowned out or masked the siren noise to a certain extent. The headlight beam would not have been directly behind plaintiffs as the curvature of the track would have partly off-set it.

[24] In seeking to reconstruct the manner in which the accident occurred, Carver presented two scenarios which he described as the 40 kph scenario and the 55 kph scenario. The following assumptions underlie both scenarios:

1. the front of the train came to a halt (after colliding with plaintiffs) at Point 15 m;

2. second plaintiff came to rest at Point 30 m;

3. first plaintiff came to rest at Point 40 m;

4. each plaintiff was struck by the train at a point 4 (four) metres to the west of the point where he came to rest;

5. prior to the collision the plaintiffs were walking between the two tracks in an easterly direction at a speed of one metre per second;

6. neither plaintiff was aware of the approach of the train until very shortly before the accident; both tried to get out of the way at the very last moment but each was struck a glancing blow which caused his body to be thrown forwards and away from the tracks;

7. the reaction time of the train driver from the moment in time when he observed plaintiffs, until the brake handle was actually applied, was three seconds.

[25] According to Carver’s 40 kph scenario the driver saw the plaintiffs at Point 136 m, the brakes were applied at Point 103 m and the first plaintiff was struck at Point 44 m. The speed of the train at Point 44 m was 29 kph. At Point 136 m (which is at the first palm tree) the plaintiffs would have been visible to the train driver. The 40 kph scenario could therefore be described as a feasible scenario. In his view 40 kph would have been a reasonable speed in the circumstances.

[26] According to the 55 kph scenario the driver should have seen the plaintiffs at Point 210 m. Brakes were applied at Point 167 m and the collision with first plaintiff was again at Point 44 m. At a distance of 210 m plaintiffs would, however, not yet have been reasonably visible to the driver. The first palm tree would also have obscured them from the train driver’s view at that point. This means that the accident probably did not occur in accordance with the 55 kph scenario.

[27] Under cross-examination of Carver the effect of alternative assumptions regarding the driver’s human reaction time were debated with him. He pointed out that the table of stopping distances shows what distance in metres per second the train would have travelled at any given speed. At a speed of 40 kph, for example, the train travels a distance of 11 m per second. During each second of human reaction time allowed for, the train would therefore have travelled a distance of 11 m. It follows that the saving of each second of wasted reaction time would have caused the train to have come to a halt at a distance of 11 m closer to Cape Town station. On that assumption, had the driver saved three seconds of reaction time by sooner applying the brakes, the train would have come to a halt at a point 33 m closer to Cape Town station which is four metres before the point where it struck first plaintiff.

[28] Carver explained that the speed at which the train was travelling at any particular point whilst the brakes were fully operative, could be calculated by means of the same formula. By referring to the information contained in the table of stopping distances, he said, the approximate speed of the train at any distance from the presumed point of impact, could be estimated with a reasonable measure of accuracy. Thus, in terms of his 40 kph scenario the train was moving at a speed of approximately 29 kph at the point where it struck first plaintiff.

[29] At the request of the court Carver prepared a separate calculation in order to determine the fastest speed at which the train could have been travelling if the brakes were applied at the same position and still have come to a halt before the point where first plaintiff was struck. He prepared a graph to illustrate the method of calculation. The answer is 32,5 kph. It follows that at such speed and any slower speed at the same point of braking, the train would have come to a halt before striking second plaintiff.

[30] Mr Hendry van Reenen is a security officer in the employment of Metrorail. On 3 February 2002 he was called out to attend to the accident in question. When he arrived at the scene there were already other persons present. He saw both plaintiffs and he noted their positions. At the inspection in loco held by the court he pointed these positions out. The two plaintiffs were clothed in their blue uniforms. He did not see anyone of them wearing a reflective vest. It was not at that time, he said, a standard requirement for the security guards to wear vests.

[31] Mr Gareth Apollis testified that he is a security officer employed by Metrorail. He became the chief security officer for the Ikapa area in June 2001. The contract between Kuffs and Metrorail was already in operation. At some stage after he had joined the cable patrol was introduced in order to prevent the recurring theft of signal cables in that area. He pointed the cables out to Gounder at that time. The cable running along the southern boundary of the Metrorail yard was pointed out to Gounder. He said that the wearing of reflective vests was never a requirement of Metrorail. He also attended the scene of the accident on the evening of 3 February 2002. He saw first and second plaintiffs at the scene of the accident. Neither one of them wore a reflective vest. He testified that he did not convey to Gounder that Metrorail would inform Kuffs every time a train was about to cross the area between Cape Town and Woodstock. He denied that there was any such practice in existence. He said that Kuffs knew that unscheduled trains were moving around in that area at night after 22:00.

[32] Kuffs closed its case without adducing any evidence.

[33] The parties agreed that the evidence of Mr Johan Stander, a meteorologist and Head of the Cape Town Weather Office, be admitted in written form. The evidence concerns the weather conditions in the Cape Peninsula on 3 February 2002 between 22:00 and 23:00. The wind speed at the Cape Town Harbour area averaged 45 kph with gusting up to 72 kph. This wind could be regarded as reasonably representative of the area between the Cape Town and Woodstock stations. This kind of wind regime is described as a Deep South Easter. Typical visibility under such a regime in this area would be described as good.

The inspection in loco

[34] The court attended an inspection in loco in the course of the trial. Counsel and the attorneys for all three parties were present, as well as various witnesses. In the minutes of the inspection the following observations were recorded:

“1 There were overhead cables above all of the lines, and signal cables in various places in the entire service area.

2 The electric siren of a passing train was activated for the benefit of the inspection.

3 In order to walk across the bridge on the Simon’s Town down line, one would have to move onto and walk on or along the tracks.

4 It was not generally uncomfortable to walk on the sleepers between the tracks.

5 If one walks along the Simon’s Town down line in the direction of Woodstock station, on the approach towards the bridge (and particularly from the palm tree closer to the bridge) there is a sleep slope on either side of the rails. These slopes are not a comfortable area for walking.

6 A footpath was visible on the southern side of the Simon’s Town down line at a level below the level of the track.

7 There is a signal cable alongside the Simon’s Town down line which commences from just past the bridge when one is travelling in the direction of Woodstock station.

8 There are various obstructive poles and other objects in the passage between the vibracrete wall and the signal cable.

9 Mr van Reenen pointed out where, he said, he found the First and Second Plaintiffs. These were measured by Mr Carver as follows:

9.1 First plaintiff was 7 metres before the second palm tree travelling from Cape Town to Woodstock and on the left hand side.

9.2 Second plaintiff was 6-7 metres before mast pole 1/11D and on the right hand side travelling from Cape Town to Woodstock.

3. They were approximately 10 metres from each other.

… … …

13 A barbed wire fence runs the length of the service area and demarcates the northern boundary of the service area. On the northern side of that boundary is the mainline.

14 There is a signal cable running on the Metrorail side of the northern boundary alongside the barbed wire fence.”

General principles of delictual liability

[35] Wrongfulness, negligence and causation are the three requirements for delictual liability that are relevant at the present stage of the proceedings. There is no dispute as to the general principles of law that apply in this regard.

[36] The principles of negligence were restated as follows in Ngubane v South African Transport Services [1990] ZASCA 148; 1991 (1) SA 756 (A) at 776D/E-I/J:

“Liability in delict based on negligence is proved if:

'(a) a diligens paterfamilias in the position of the defendant -

(i) would foresee the reasonable possibility of his conduct injuring another in his person or property and causing him patrimonial loss; and

(ii) would take reasonable steps to guard against such occurrence; and

(b) the defendant failed to take such steps.

This has been constantly stated by this Court for some 50 years. Requirement (a) (ii) is sometimes overlooked. Whether a diligens paterfamilias in the position of the person concerned would take any guarding steps at all and, if so, what steps would be reasonable, must always depend upon the particular circumstances of each case. No hard and fast basis can be laid down. Hence the futility, in general, of seeking guidance from the facts and results of other cases.'

(Kruger v Coetzee 1966 (2) SA 428 (A) at 430E - G.)

As regards the requirement in para (a) (ii) above in this judgment, it is acknowledged that reasonable steps are not necessarily those which would ensure that foreseeable harm of any kind does not in any circumstances eventuate. The contributor (Prof J C van der Walt) in Joubert (ed) The Law of South Africa vol 8 sv 'Delict' para 43 at 78 comments in this regard that:

'Once it is established that a reasonable man would have foreseen the possibility of harm, the question arises whether he would have taken measures to prevent the occurrence of the foreseeable harm. The answer depends on the circumstances of the case. There are, however, four basic considerations in each case which influence the reaction of the reasonable man in a situation posing a foreseeable risk of harm to others: (a) the degree or extent of the risk created by the actor's conduct; (b) the gravity of the possible consequences if the risk of harm materialises; (c) the utility of the actor's conduct; and (d) the burden of eliminating the risk of harm.'”

[37] The theoretical and practical differences between the elements of wrongfulness and negligence are well established. See Gouda Boerdery BK v Transnet 2005 (5) SA 490 (SCA) at 498G-499E (footnotes omitted):

“[12] It is now well established that wrongfulness is a requirement for liability under the modern Aquilian action. Negligent conduct giving rise to loss, unless also wrongful, is therefore not actionable. But the issue of wrongfulness is more often than not uncontentious as the plaintiff's action will be founded upon conduct which, if held to be culpable, would be prima facie wrongful. Typically this is so where the negligent conduct takes the form of a positive act which causes physical harm. Where the element of wrongfulness gains importance is in relation to liability for omissions and pure economic loss. The inquiry as to wrongfulness will then involve a determination of the existence or otherwise of a legal duty owed by the defendant to the plaintiff to act without negligence: in other words to avoid negligently causing the plaintiff harm. This will be a matter for judicial judgment involving criteria of reasonableness, policy and, where appropriate, constitutional norms. If a legal duty is found to have existed, the next inquiry will be whether the defendant was negligent. The test to be applied will be that formulated in Kruger v Coetzee, involving as it does, first, a determination of the issue of foreseeability and, second, a comparison between what steps a reasonable person would have taken and what steps, if any, the defendant actually took. While conceptually the inquiry as to wrongfulness might be anterior to the enquiry as to negligence, it is equally so that without negligence the issue of wrongfulness does not arise for conduct will not be wrongful if there is no negligence. Depending on the circumstances, therefore, it may be convenient to assume the existence of a legal duty and consider first the issue of negligence. It may also be convenient for that matter, when the issue of wrongfulness is considered first, to assume for that purpose the existence of negligence. The courts have in the past sometimes determined the issue of foreseeability as part of the inquiry into wrongfulness and, after finding that there was a legal duty to act reasonably, proceeded to determine the second leg of the negligence inquiry, the first (being foreseeability) having already been decided. If this approach is adopted, it is important not to overlook the distinction between negligence and wrongfulness.”

[38] The test for factual causation is also well established. See International Shipping Co (Pty) Ltd v Bentley 1990 (1) SA 680 (A) at 700E-701:

“The first [enquiry] is a factual one and relates to the question as to whether the defendant's wrongful act was a cause of the plaintiff's loss. This has been referred to as 'factual causation'. The enquiry as to factual causation is generally conducted by applying the so-called 'but-for' test, which is designed to determine whether a postulated cause can be identified as a causa sine qua non of the loss in question. In order to apply this test one must make a hypothetical enquiry as to what probably would have happened but for the wrongful conduct of the defendant. This enquiry may involve the mental elimination of the wrongful conduct and the substitution of a hypothetical course of lawful conduct and the posing of the question as to whether upon such an hypothesis plaintiff's loss would have ensued or not. If it would in any event have ensued, then the wrongful conduct was not a cause of the plaintiff's loss; aliter, if it would not so have ensued. If the wrongful act is shown in this way not to be a causa sine qua non of the loss suffered, then no legal liability can arise. On the other hand, demonstration that the wrongful act was a causa sine qua non of the loss does not necessarily result in legal liability.”

Main submissions on behalf of plaintiffs

[39] Mr G M Budlender appeared on behalf of first and second plaintiffs. He relied on three main grounds of negligence, two on the part of Metrorail and the third on the part of Human for which Metrorail is vicariously liable. He did not rely on any negligence on the part of the second defendant – it was, so it would appear, cited ex abundante cautelae. Mr Budlender submitted that the court should make the following findings in regard to four key underlying factual issues. The first is that first plaintiff and second plaintiff were in an area where they were supposed to be working, when the collision occurred. The second is that neither first nor second plaintiff received any training from Kuffs or Metrorail in cable patrol work. The third finding, he suggested, is that neither first nor second plaintiff knew that an unscheduled train was about to move down the railway line at the time that they were performing cable patrol work. The fourth suggested finding is that Metrorail did not give any specific warning to Kuffs or plaintiffs that the train was about to proceed down the line in question.

[40] Mr Budlender accepted that it was not necessary for the court to make any findings in regard to two other issues that featured in the evidence, namely (i) whether plaintiffs wore reflective vests outside their clothes and (ii) whether the curve in the railway line or the dispersion of the train’s headlights had any effect on the visibility of plaintiffs at the time when the train approached them. These two issues fell away as he was prepared to accept Human’s version, as reconstructed by Carver, as to the position where the train was when he first saw plaintiffs.

[41] The first ground of negligence is that Metrorail was negligent in failing to warn Kuffs and plaintiffs that an unscheduled train was about to move down the line in question. The security guards, he pointed out, were working in an inherently dangerous situation, particularly as it was night time. It is not in dispute, he submitted, that Metrorail could have issued such a warning. The movement of trains was controlled from the Windermere control room and one of Kuffs’ officials was present in Metrorail’s control room at the Cape Town station. It would not have been difficult or costly to issue such a warning. Had such a warning been issued plaintiffs would in all likelihood have kept a proper lookout for the train and they would then have been able to avoid the accident.

[42] Metrorail was also negligent, Mr Budlender submitted, by failing to take reasonable measures to ensure that security personnel working on their premises received proper safety training. Such training should have included a specific warning to the security guards to be on the lookout for unscheduled trains travelling down that line at night. The need for such training was increased by the fact that it did not issue any specific warnings before the departure of each train.

[43] The third ground of negligence relied upon is that Human could and should have applied the brakes of the train immediately after he observed plaintiffs on the railway line in front of him. Had he done so, it was submitted, the collisions would probably not have occurred.

Main submissions on behalf of defendants

[44] Mr A de V le Grange SC appeared on behalf of the three defendants. He did not dispute three of the findings on the key factual issues suggested by Mr Budlender, namely that plaintiffs were in an area where they were supposed to be working, that neither first nor second plaintiff received any specific training from Kuffs or Metrorail in cable patrol work and that Metrorail did not give any warning to Kuffs or plaintiffs that the train was about to proceed down the line in question.

[45] In regard to the third factual issue, however, Mr le Grange submitted that both first and second plaintiffs knew that an unscheduled train might move down the railway line at the time that they were performing cable patrol work. He submitted that second plaintiff’s evidence that he did not know that there would be trains moving through the area after 22:00, should be rejected. He argued that the uncontested evidence of defendants’ witnesses (Human, van Reenen and Apollis) established that the moving of trains through the area in question during that period was a regular occurrence. The conduct of Bidli, he submitted, embodied the standard of the reasonable security guard. Bidli was aware of the fact that trains moved through the area at night and he testified that he never crossed a railway line without first paying attention to the possibility of a train passing through.

[46] In considering the foreseeability of the harm in this case, he submitted, it is necessary to bear in mind that both plaintiffs were qualified security guards who could have been expected to act responsibly with regard to their own safety. It was not reasonably foreseeable that plaintiffs would close their eyes and ears with regard to what was happening around them, nor that they would act recklessly with regard to the risk of passing trains.

[47] The headlight of a train, he submitted, is an important signal of the train’s approach. In this regard he referred to the following passage in South African Railways and Harbours v Orford 1963 (1) SA 672 (A) at 677F-H:

“It is a matter of common knowledge - and is, indeed, established by the evidence in this case - that in practice locomotives travelling at night employ powerful headlights. The road-travelling public is familiar with that practice, and anybody proposing to traverse a level-crossing at night will naturally expect to be able immediately to discern the presence of an approaching train by the powerful beam of its headlight. I agree with the observation of NESER, J., in Matcheke v S.A.R. and Public Utility Corporation, Ltd., 1948 (1) SA 295 (T) at p. 307, that

'the railway service have accustomed the public to expect the normal headlights on trains'.

The locomotive's headlight is thus, in effect, an established method of giving warning of a train's approach at night time which travellers by road have reasonably come to expect.”

[48] Mr le Grange referred to various decisions in which allowance was made for perception and reaction time on the part of a driver of a motor vehicle. In Coetzee v Shield Insurance Co Ltd 1980 (4) SA 621 (C) at 626F, for example, allowance was made “for a reaction time of say 3/4 second”. In Rodrigues v SA Mutual and General Insurance Co Ltd 1981 (2) SA 274 (A) at 279F reference was made to the evidence of an expert “that perception and reaction time combined are ordinarily accepted at 11/2 seconds”. In Masureik and Another (t/a Lotus Corporation) v Welkom Municipality and Another 1995 (4) SA 745 (O) at 767J the court accepted expert evidence to the effect “that a reasonable pilot has reaction time which is at least two seconds”.

[49] In this case, Mr le Grange submitted, it would be reasonable to allow for a total human reaction time of three seconds on the part of the driver. That would include the time taken by him for the observation of plaintiffs, the application of the siren, the observation of the fact that plaintiffs were not reacting to the siren, the decision to apply the brakes and the human reaction thereafter in applying the brakes. Had the driver decided to brake immediately, he submitted, his reaction time would have been at least one second.

[50] Mr le Grange argued that it was not unreasonable for Human to have sounded the siren first as a warning to plaintiffs. He pointed out that in many decided cases the courts have emphasised the sounding of a siren or hooter as the first requirement for a train driver that is approaching a crossing. In Geldenhuys v South African Railways and Harbours 1964 (2) SA 230 (C), for example, the following was said, at 232D/E-H:

“At this stage it will be convenient to set out the duties of persons in charge of trains, according to the principles that have been accepted in a number of decided cases. In Walker v Rhodesian Railways, Ltd., 1937 S.R. 62 at p. 73, HUDSON, J., sets out the said duties as follows:

(1) It is the duty of the Railway Administration, when intending to exercise its preferent right at open level crossings at night, to advertise the approach of a train by an unmistakable appeal both to the sight and hearing of travellers on the road.

(2) This duty, in respect of the appeal to sight, can only be discharged at night by displaying on the forefront of the train a light or lights sufficiently distinguishing or powerful to give adequate and unmistakable warning of its approach.

(3) The driver's first duty, whether by day or night, is to see that the line ahead of him is clear.

(4) On approaching a crossing he must keep his eye on the crossing and its immediate neighbourhood.

(5) Having satisfied himself that the crossing is clear or likely to be clear, he must then look to the public road to his right and left to see whether there are any vehicles approaching the crossing.

(6) In an emergency he must be in a position to give a special warning and apply his brakes if necessary.

These principles are quoted with approval by NESER, J., in Matcheke v South African Railways & Harbours and Public Utility Corporation Ltd., 1948 (1) SA 295 (T) at p. 301, and I respectfully accept them.”

[51] Mr le Grange submitted that plaintiffs were also negligent. If liability on the part of any of the defendants is established, he argued, plaintiffs’ claims would be subject to apportionment by reason of their own negligence.

Submissions on behalf of the third party

[52] Mr J C Marais appeared on behalf of Kuffs. On the questions of negligence he adopted the submissions advanced on behalf of defendants. He referred in particular to the following statement (in regard to level crossings) by Solomon J in Worthington and Others v C.S.A.R. 1905 TH 149, cited with approval by Ramsbottom J in Pretoria City Council v South African Railways and Harbours 1957 (4) SA 333 (T) at 336H-337A:

“It is the duty of the traveller to look out for and wait for the train. At the same time a condition is attached to the preference which the railway has, and that is that the train ought to give due warning of its approach when it is nearing a level-crossing of this nature, so that persons might stop and allow the train to pass. The train is bound, in my opinion, to give due and timely warning of its approach, and also not to be travelling at such an excessive rate of speed that the warning it might give should be of no avail. What is an excessive speed and what is due warning must entirely depend on the special circumstances of each case. Where there are obstructions to prevent persons travelling along the road from seeing an approaching train, or where there are any other circumstances which would make it difficult to ascertain that a train is approaching, then, of course, better warning would have to be given, and the train would have to travel at a slower speed.”

[53] Mr Marais also emphasised the fact that Human found himself in a situation of emergency. His conduct should therefore not be judged too finely. He referred to the recapitulation, in Young v Workmen's Compensation Commissioner and Another 1998 (3) SA 1085 (T), at 1095F-1096C, of earlier judicial statements on this topic. In South African Railways v Symington 1935 AD 37 at 45, for example, Wessels CJ said the following:

“Where men have to make up their minds how to act in a second or in a fraction of a second, one may think this course the better whilst another may prefer that. It is undoubtedly the duty of every person to avoid an accident, but if he acts reasonably, even if by a justifiable error of judgment he does not choose the very best course to avoid the accident as events afterwards show, then he is not on that account to be held liable for culpa.”

[54] Following the amendment of the third party’s plea Mr Marais also argued that second defendant was negligent in providing a train to Metrorail that had inefficient warning and braking systems.

Metrorail’s own negligence

[55] I proceed to consider the three main grounds of negligence relied upon by Mr Budlender on behalf of plaintiffs. I shall in regard to each ground also consider the related questions of wrongfulness and causation. The allegations of negligence against Metrorail are twofold. The first is that Metrorail was negligent in failing to warn Kuffs and plaintiffs that the unscheduled train in question was about to move from Cape Town to Woodstock. The second ground is that Metrorail should have ensured that the security guards receive proper training in cable patrol.

[56] I deal first with the alleged failure to provide plaintiffs with proper training in cable patrol. It is trite law that questions of wrongfulness and negligence must be determined with reference to all the particular facts of each case. Two factors complicate the factual enquiry in this regard. The first concerns the role and position of Kuffs as an important intermediary between metrorail and plaintiffs. The second is the vagueness of the evidence concerning the agreements or understandings between Metrorail and Kuffs in regard to the cable patrol. It seems clear that the cable patrol did not form part of the original agreement between them. It started a few months later. One must assume that the parties’ agreement contained at least a tacit term in connection with the questions of the training of the security guards. The witnesses that gave evidence in this regard (Gounder and Apollis) rather described the de facto situation that existed at the time of the incident. Not one of them provided first hand information of the terms agreed upon between Metrorail and Kuffs.

[57] The stance taken by Metrorail, through its witnesses and its counsel in argument, was that Kuffs was at all times fully informed of the factual situation and that it was left to them to provide such training as was required. The training in question, namely an instruction to watch for trains when one crosses a railway line, would after all have been a fairly elementary element of training.

[58] In the circumstances I am not persuaded that Metrorail’s stance can be described as unreasonable. In short, in my view the alleged duty of care on the part of Metrorail to individual security guards to ensure that they were properly trained for cable patrol, was not established in these proceedings.

[59] The next question is whether Metrorail was negligent in failing to issue a specific warning to Kuffs that an unscheduled train was about to move through the area. Metrorail would have been in possession of precise information in regard to the movement of each and every train. Kuffs was not. Unlike the question of training, Metrorail could not have assumed that Kuffs would issue the suggested warnings to individual security guards because Kuffs was dependant upon it for the relevant information.

[60] Generally speaking it might be said that the less frequent these unscheduled train journeys were, the greater the need for specific warnings. The evidence in regard to the frequency of these trains is unfortunately vague. A number of witnesses testified on behalf of Metrorail to the fact that there were such trains. Metrorail, one must assume, is in possession of precise data relating to this question as the movement of all the trains were at all times controlled from a central control room. Yet no such evidence was placed before the court. In my view Metrorail did not prove that the frequency of such trains was such that all security guards must have been aware of such movements.

[61] In my view Mr Budlender raised valid arguments in support of a finding that Metrorail’s failure was wrongful. The dangers created by these trains were significant and they were clearly foreseeable. The warnings could have been issued with little difficulty and at hardly any cost. It is a fair inference, furthermore, that had such a warning been issued in the present case, the accident would probably have been avoided.

[62] I find therefore that Metrorail was negligent in failing to issue a specific warning to Kuffs on the night in question that a train was about to pass through the area where its guards were performing cable patrol. That failure was causally related to the accident that occurred.

Causal negligence on the part of Human

[63] Plaintiffs’ third ground of negligence is that the train driver was negligent in that he took the wrong option by sounding the siren first and observing plaintiffs’ reaction before braking. This argument was explored in the cross-examination of Carver. Mr Budlender’s contention was that it had been shown on a balance of probabilities that the accident could have been avoided if Human had applied the brakes immediately after he saw plaintiffs.

[64] The reasoning underlying Mr Budlender’s contention appears to me to be sound. If it is accepted that the driver wasted valuable time by first sounding his siren and observing plaintiffs’ reaction then it follows that the train continued travelling towards plaintiffs during that period. Mr Budlender submitted that a period of at least three seconds was wasted in this manner. Assuming that the train was travelling at 40 kph at that stage then it follows that the train could have been brought to a halt some four metres before the point of collision with first plaintiff.

[65] The alternative leg of the contention is somewhat more involved. It is based on the more conservative assumption that the driver’s total human reaction time was only three seconds and not four seconds. Had the driver braked immediately after observing plaintiffs, his reaction time would have been only one second. This means that he wasted two seconds. In that period, at 40 kph, the train would have travelled some 22 metres. It would therefore have come to a halt at Point 37. First plaintiff was struck at Point 44. This means that the train would have come to a halt some 7 metres beyond the point where first plaintiff was struck. The argument on behalf of plaintiffs, however, is that in such event the train would have been travelling at a much slower speed at Point 44 than it actually did. Mr Carver estimated that the train, in terms of the 40 kph scenario, was travelling at a speed of about 29 kph at Point 44. That translates into 8 metres per second. Had the train come to a standstill at Point 37, its speed at Point 44 would have been significantly slower. According to my estimate it would have been about 14 kph. (At a speed of 15 kph, it may be noted from the table of stopping distances, the train would have come to a standstill in about 8 metres.) A speed of 14 kph is equivalent to about 4 metres per second. Plaintiffs were probably alerted to the approach of the train by the sound of the siren or the headlights shining upon them. In the case of the hypothetical slower train this would probably have happened when the train was at the same distance away from them. The effect of this line of reasoning is therefore that plaintiffs would have had an additional one second available in which to evade the slower train. The evidence was that each was probably struck a glancing blow. It is reasonable to assume therefore that each plaintiff would have been able to move an additional distance of about one metre in that one second. That would have been sufficient for them to avoid the accident.

[66] Various answers to this line of reasoning were suggested in argument by counsel for defendants and Kuffs. Thus it was submitted that Human was confronted with an emergency situation and that he should not be criticised with the benefit of hindsight.

[67] I agree that Human found himself in a situation where an immediate response was called for. In judging his conduct there are, however, a number of considerations that cannot be ignored. The first is that Human was a professional train driver with vast experience. On each and every suburban train journey undertaken by him, he would have been constantly on the lookout for the exact kind of emergency that presented itself, namely a pedestrian on the railway line. He had no reason to be less attentive because this was an unscheduled train. Another important consideration, in my view, is that Human was not confronted by various alternative courses of action that required to be weighed up against each other. The choices open to him were few and simple and the immediate braking option carried no additional risk.

[68] In response to Mr Budlender’s argument that the total reaction time was probably longer than three seconds, Mr le Grange pointed out that Carver’s calculation of his 40 kph scenario was based on the assumption that plaintiffs were visible to the driver from the position where the train was when his reaction time commenced. It does not follow, he submitted, if a longer total reaction time is postulated, that plaintiffs would have been visible to the driver from the position where the train would have been at that earlier stage. Although Carver agreed that plaintiffs would have been visible from Point 136 (the position of the first palm tree) he did not concede that plaintiffs would have been visible from any point before that.

[69] It seems to me, however, that Mr le Grange’s argument overlooks an important aspect of Carver’s scenarios. The speed of 40 kph is not a fixed premiss. It is one of the assumptions made in order to arrive at a feasible scenario. The length of human reaction time, also, is not a fixed premiss. Various assumptions may be made in that regard. A particular scenario can only be regarded as feasible if it is consistent with the known and assumed facts. One of these facts is that plaintiff was struck by the train at Point 44 m. Another obvious fact is that the driver actually saw plaintiffs when his reaction time commenced. If any assumption in regard to the length of the human reaction time entails that plaintiffs were seen from a position where they were not visible, then the scenario incorporating that assumption is simply not a feasible scenario. It means that a slower speed than 40 kph would have to be postulated in order to convert such a scenario into a feasible scenario.

[70] I have, for illustrative purposes, considered the following scenario: The train travelled at a speed of 37,5 kph; the driver observed plaintiffs at Point 136 (the first palm tree) and the human reaction time was four seconds of which three seconds represented wasted reaction time. In estimating the stopping distance at 37,5 kph I have taken the average of the stopping distances at 35 kph and 40 kph respectively. At a speed of 37,5 kph the train travels 10,42 metres per second. On this basis I arrive at the following estimates:

Distance to stop 49,54 metres

Train reaction (3 sec) 31,26 metres

Human reaction (4 sec) 41,68 metres

Total stopping distance 122,48 metres

Deduct 3 sec wasted time 31,26 metres

Distance 91,22 metres

[71] In terms of this scenario the train would therefore have come to a standstill at a point 91 metres beyond Point 136, which is at Point 45, ie one metre short of the point where first plaintiff was struck. The scenario postulated by me thus appears to be a feasible and reasonable scenario. Applying it to the alternative leg of plaintiffs’ contention (see para [65] above), this scenario also leads to a conclusion that the accident could probably have been avoided.

[72] Counsel for defendants and Kuffs suggested in argument that Mr Budlender’s contention is based on ingenious calculations which may not reflect the realities of the situation. I do not agree. The calculations are based on the evidence of Carver and plaintiffs are not required to meet any higher standard of proof than a preponderance of probabilities.

[73] Counsel for defendants and Kuffs also submitted that the primary duty of a train driver, according to the case law, is to sound his siren or hooter. I have considered the various cases referred to in this regard, but I cannot agree with this interpretation. In all these cases the courts were applying general principles of negligence to the facts before them. In some cases the sounding of a siren may be the obvious thing to do. In others it may be the only thing that can reasonably be done. In the present case, however, I am of the view that the situation called for an immediate braking reaction.

[74] Reverting then to Human’s conduct, it follows that his initial reaction, namely to sound the siren, unfortunately proved to have been a time wasting and futile exercise. Plaintiffs, one must assume, did not hear the siren at that stage. It is significant, moreover, that Carver accepted that the prevailing circumstances were such that plaintiffs probably did not hear the siren. In the circumstances, given the distance from the train to where plaintiffs were, the windy conditions and the fact that the siren did not make much noise, the initial sounding of the siren was not even likely to act as a proper warning of the train’s approach. Human was aware of these circumstances but he nevertheless wasted a few seconds by sounding the siren first.

[75] I conclude therefore that Human was negligent and that his negligent conduct was causally related to the collision with both plaintiffs.

Negligence on the part of plaintiffs

[76] That brings me to the question of contributory negligence on the part of plaintiffs. A relevant factual issue is whether plaintiffs actually knew that there was a risk that an unscheduled train was about to move down the railway line at the time that they were performing cable patrol work. Mr Budlender submitted that second plaintiff’s testified that he did not know that an unsolicited train might at any time be sent down the line. First plaintiff did not testify but it was a reasonable inference, he submitted, that he also was not aware of that risk. Defendants, he pointed out, led no evidence to contradict second plaintiff’s statement. Mr le Grange, on the other hand, submitted, that there was ample evidence from defendants’ witnesses that the movement of unscheduled trains after 10:00 was not an unusual occurrence. Bidli had also on occasion observed a train during that period.

[77] Insofar as the question of plaintiffs’ actual knowledge is relevant to the question of contributory negligence the onus of proving such knowledge, rests on defendants. In view of the evidence adduced by defendants I have no difficulty in finding that the movement of unscheduled trains did occur after 10:00. An important factual issue however concerns the frequency of such trains. I have already commented on the fact that Metrorail did not place any precise evidence in this regard before the court. It is important to note, furthermore, that the cable patrol covered the area from Cape Town to Salt River. Some of the unscheduled trains, according to defendants’ evidence, were only moved to a shunting siding or to a different platform. The inference can not be drawn that plaintiffs must have observed such trains.

[78] The question of plaintiffs’ actual knowledge is, however, not decisive of the question whether they were negligent or not. Second plaintiff did not claim that he was told or assured by Metrorail or by any of his supervisors that there would actually not be any trains moving through the area whilst they were carrying out cable patrol. Second plaintiff appears to have assumed this fact on the strength of what he was told by a colleague. In my view he was not acting reasonably in acting on that assumption. Plaintiffs must have been aware of the risks involved if a train should proceed down the line whilst they were walking on the line. It was not necessary for them to walk on the actual railway line and there was nothing that precluded them from keeping a proper lookout. In my view they were negligent in failing to keep a proper lookout. It seems obvious that they would have seen and heard the train at an earlier stage, had they done so.

[79] I find therefore that plaintiffs were also causally negligent by walking on the railway line without keeping a proper lookout for trains.

The apportionment of plaintiffs’ damages

[80] As plaintiffs were also negligent any damages that they may recover in this action are subject to apportionment in terms of the provisions of section 1(1)(a) of the Apportionment of Damages Act 34 of 1956 (“section 1(1)(a) of the Act”). This provision reads as follows:

“Where any person suffers damage which is caused partly by his own fault and partly by the fault of any other person, a claim in respect of that damage shall not be defeated by reason of the fault of the claimant but the damages recoverable in respect thereof shall be reduced by the court to such extent as the court may deem just and equitable having regard to the degree in which the claimant was at fault in relation to the damage.”

[81] I have found that Metrorail and Human are both liable to pay damages to plaintiffs. In what manner are the provisions of section 1(1)(a) of the Act to be applied in a situation where there are two wrongdoers? As far as I am aware, this question has not yet been considered in any reported South African judgment. There are, however, English, Australian and Canadian cases that may serve as guidelines.

[82] Before considering the legal question, I need to point to three elements of the factual context in the present case that are relevant to the enquiry:

(i) The first point is that there does not appear to be any reason to distinguish between the respective positions of the two plaintiffs in regard to the measure of apportionment to be applied. The same considerations would apply to the claim of each of them.

(ii) The second point is that Metrorail is also liable to plaintiffs by reason of its vicarious liability as the employer of Human. In that capacity it is also subject to the provisions of section 1(1)(a) of the Act. See Becker v Kellerman 1971 (2) SA 172 (T) at 177. Metrorail’s vicarious liability is, however, not relevant to the question under consideration. Its liability in that regard is entirely co-incidental.

(iii) The third point to be noted is that Metrorail and Human are to be regarded as concurrent wrongdoers, ie persons whose independent or 'several' delictual acts (or omissions) combined to produce the same damage. They are not joint wrongdoers, ie persons who, acting in concert or in furtherance of a common design, jointly commited a delict. For this distinction, see Nedcor Bank Ltd t/a Nedbank v Lloyd-Gray Lithographers (Pty) Ltd 2000 (4) SA 915 (SCA) at 922D-F.

[83] The starting point of the enquiry must be the South African case law on the application of section 1(1)(a) of the Act to the ordinary situation where there is a single wrongdoer. There are three important judgments of the erstwhile Appellate Division on this topic. The first is that of Ogilvie Thompson JA in South British Insurance Co. Ltd., v Smit 1962 (3) SA 826 (AD). In the second judgment, Jones NO v Santam Bpk 1965 (2) SA 542 (A), the Appellate Division followed, and, in one respect clarified, what was said in South British Insurance Co. Ltd., v Smit. See the following passages, at 554G-555D:

“The provisions of sec. 1 (1) (a) of the Apportionment of Damages Act were particularly considered by this Court in South British Insurance Co. Ltd., v Smit, 1962 (3) SA 826 (AD). At p. 836 C OGILVIE THOMPSON, J.A., in summing up his conclusions after a consideration of the terms of the statute and the possible meanings to be attached thereto, finalised the views of the majority of the Court in these words:

'What the Court is required to do is to determine, having regard to the circumstances of the particular case, the respective degrees of negligence of the parties. In assessing 'the degree' in which the claimant was at fault in relation to the damage' the Court must determine in how far the claimant's acts or omissions, causally linked with the damage in issue, deviated from the norm of the bonus paterfamilias. In thus assessing the position, the Court will, as explained above, determine the respective degrees of negligence, as reflected by the acts and omissions of the parties, which have together combined to bring about the damage in issue.'

It is important to note the statement that it is 'the respective degrees of negligence' of the parties which has to be determined, not only the degree of any negligence on the part of the claimant. This was emphasised in a previous passage of the same judgment where, at p. 835 H, OGILVIE THOMPSON, J.A., said that, although the sub-section only refers to the claimant,

'it is, I think, plain from a consideration of the section as a whole that what the Court has to measure is the conduct of all parties whose fault caused the damage. Postulating a single defendant, the determination of the 'degree in which the claimant was at fault in relation to the damage' will also automatically determine the degree in which the defendant was at fault in relation to the damage'.

I concurred in this judgment of OGILVIE THOMPSON, J.A., in Smit's case, but on further consideration I have come to the conclusion that the last sentence of this quotation does not make clear my view as to how the respective degrees of fault of the different parties must be assessed. A determination of the degree of fault on the part of the claimant does not by itself

'automatically determine the degree in which the defendant was at fault in relation to the damage';

the Court must first also determine in how far the defendant's

'acts or omissions, causally linked with the damage in issue, deviated from the norm of the bonus paterfamilias'.

It is on the basis of comparison between the respective degrees of negligence of the two parties (or several parties if there be more than one claimant or defendant) that the Court can determine in how far the fault or negligence of each combined with the other to bring about the damage in issue.”

[84] For present purposes it is instructive to consider the judgment of Ogilvie Thompson JA in South British Insurance Co. Ltd v Smit, supra, more fully. A central question in that case was whether the relative degrees of blameworthiness of the parties is the sole criterion of apportionment or whether the causal significance of the acts of the parties must also be taken account. See the following passage, at 833F/G-833 in fine:

“Considerable divergence of view has manifested itself in the writings of learned authors concerning the correct interpretation of the concluding portion of para. 1 (a) of this section. According to one view, the causal significance of the acts of the parties is irrelevant, since 'the relative degrees of blameworthiness of the parties must be the sole criterion of apportionment' (McKerron, Law of Delict, 5th ed., p. 261). The opposing view - of which Mr. Boberg of Witwatersrand University would appear to be a staunch protagonist: see 76 S.A.L.J. 259 and Annual Survey of South African Law (1960) p. 160 - emphasises the words 'fault in relation to the damage' and maintains that the criterion is not the degree of blameworthiness alone, but the degree of blameworthiness in relation to its causal effect in producing the damage. Prof. Swanepoel, Tydskrif vir Hedendaagse Romeins-Hollandse Reg (1959) at p. 271, has expressed the opinion that, since the Court is dealing with two unlawful acts, all that is required is that the two unlawful acts be weighed against each other, and that 'die grondslag van verdeling van skade is dus die twee onregmatige dade.'”

In the course of his discussion of this question, Ogilvie Thompson JA referred to and distinguished the position in English law. See the following passages, at 834D/E-835B:

“Sec. 1 of Act 34 of 1956 is couched in very similar terms to those of sec. 1 of the Law Reform (Contributory Negligence) Act, 1945, of England. Indeed, para. 1 (a) of our Act is in virtually identical terms with that of sub-sec. 1 (1) of the English Act save for the concluding words of the latter, which provide that the damages recoverable shall be reduced to such extent as the Court thinks just and equitable 'having regard to the claimant's share in the responsibility for the damage'. These last cited words have apparently also given rise to some controversy, similar to that outlined above in relation to our own Act, regarding the criterion thus prescribed by the Legislature. Thus Pollock, Torts, 15th ed., p. 352, is of the view that 'responsibility' means 'causal responsibility'; while Glanville Williams, Joint Torts and Contributory Negligence, sec. 98, although alluding to such a view as having been judicially expressed in relation to identical words in the Tortfeasors Act of 1935, says that, in all cases under the 1945 Contributory Negligence Act, it has been assumed that 'apportionment is on the basis of fault or blame'. The learned author, however, goes on to cite a passage from the judgment of DENNING, L.J., in Davies v Swan Motor Co. (Swansea) Ltd., 1949 (1) A.E.R. 620 at p. 632, which, with the greatest respect, would not appear entirely to exclude the causation factor. The passage in question reads:

'While causation is the decisive factor in determining whether there should be a reduced amount payable to the plaintiff, nevertheless the amount of the reduction does not depend solely on the degree of causation. The amount of the reduction is such an amount as may be found by the court to be 'just and equitable', having regard to the claimant's 'share in the responsibility' for the damage. This involves a consideration, not only of the causative potency of a particular factor, but also of its blameworthiness. The fact of standing on the steps of the dustcart is just as potent a factor in causing damage, whether the person standing there be a servant acting negligently in the course of his employment or a boy in play or a youth doing it for a 'lark', but the degree of blameworthiness may be very different.'

The word 'responsibility', appearing in the English Act, may perhaps be said expressly to connote incorporation into the enquiry of the element of causation. As already indicated, our Act deliberately departed from the wording of the English Act reading 'having regard to the claimant's share in the responsibility for the damage'. Under our Statute the Court is enjoined to have 'regard to the degree in which the claimant was at fault in relation to the damage'. The critical question, accordingly, is: what does 'fault' ('skuld') here mean?”

In his discussion of this “critical question” Ogilvie Thompson JA then made the statements that were considered in the passages in Jones NO v Santam Bpk, supra, which I have quoted above.

[85] In the light of Jones NO v Santam Bpk, supra, read with South British Insurance Co. Ltd., v Smit, supra, the interpretation of section 1(1)(a) of the Act thus appeared, at that stage, to be fairly clear: The court had to compare the respective degrees of blameworthiness of the two parties. Unlike the position in England, the relative degrees of the causal significance of the parties’ acts would not play a direct role. Although it was made clear in Jones NO v Santam Bpk, supra, that the conduct of all parties had to be considered, the method of comparison to be followed in the case of two or more wrongdoers, was not dealt with.

[86] Then came the third important judgment, General Accident Versekeringsmaatskappy SA Bpk v Uijs NO [1993] ZASCA 58; 1993 (4) SA 228 (A). The plaintiff (he was represented by a curator ad litem) had been a passenger in a motor vehicle when it was involved in a collision in which he sustained serious head injuries. It was common cause between the parties that the driver of the vehicle had caused the accident in a grossly negligent fashion. The trial court found that the plaintiff had been causally negligent in that he had failed to wear his seatbelt and accordingly reduced his damages by one third. Van Heerden JA dismissed the appeal. He held that justice and equity demanded that allowance be made for the fact that the plaintiff had in no way contributed to the accident and that his fault was of a different kind to that of the driver. He said, inter alia, the following, at 234J-235E:

“Soos welbekend, bepaal art 1(1)(a) dat waar iemand skade ly wat deels aan sy eie skuld en deels aan die skuld van 'n ander persoon te wyte is, 'n vordering ten opsigte van die skade nie ten gevolge van die skuld van die eiser verydel word nie, maar dat die verhaalbare skade in so 'n mate verminder word as wat die hof, met inagneming van die mate van die eiser se skuld met betrekking tot die skade, regverdig en billik ag. Wat betref skade gely as gevolg van 'n botsing tussen twee voertuie, vind art 1(1)(a) normaalweg toepassing indien die botsing aan die nalatigheid van altwee bestuurders te wyte was en albei as gevolg daarvan skade gely het. In so 'n geval sou 'n bepaling van die graad van kousale nalatigheid van bestuurder A in baie gevalle - maar nie altyd nie (Jones NO v Santam Bpk 1965 (2) SA 542 (A) op 555) - ook uitsluitsel gee oor die skuldgraad van bestuurder B, en sou dit normaalweg billik wees om die skade van bestuurder A met die graad van sy nalatigheid te verminder. Moeiliker is die toepassing van die subartikel in 'n geval soos die onderhawige, waar slegs Stander skuld met betrekking tot die veroorsaking van die botsing dra. Dit het Van Huyssteen se skade tot gevolg gehad en Stander se afwyking van die norm van die bonus paterfamilias kan op naby 100% gestel word. Maar wat nou as Van Huyssteen se afwyking van die norm ook op bykans 100% gestel word? Moontlik sou dan gesê kan word dat hy en Stander gelyke skuld met betrekking tot die veroorsaking van sy skade het. Artikel 1(1)(a) bepaal egter nie dat 'n eiser se skade verminder moet word in verhouding tot sy skuld nie, maar wel tot die mate wat, met inagneming van die omvang van die eiser se skuld, regverdig en billik is. En in 'n geval soos die onderhawige verg regverdigheid en billikheid inagneming van die feit dat Van Huyssteen geensins tot die plaasvind van die botsing bygedra het nie, en dat sy skuld andersoortig as dié van Stander was.”

[87] It is interesting to note that Van Heerden JA did not in General Accident Versekeringsmaatskappy SA Bpk v Uijs NO, supra, refer to the earlier judgment in South British Insurance Co. Ltd., v Smit, supra, nor to English cases on the topic of the non-wearing of a seatbelt. He nevertheless had regard to considerations relative to the causative potency of the parties’ conduct but he did so in the course of applying the criteria of justice and fairness.

[88] I revert then to the question of the method of apportionment of damages between a plaintiff and two wrongdoers. This question has been considered in English, Australian and Canadian cases. An authoritative judgment in England is that of the House of Lords in Fitzgerald v Lane and another [1988] UKHL 5; [1988] 2 All ER 961 [HL]. The plaintiff was a pedestrian who had been injured in a road traffic accident involving two negligent motorists. The statute in question was still section 1 of the Law Reform (Contributory Negligence) Act, 1945, of England. (This provision, it was pointed out in South British Insurance Co. Ltd v Smit, supra, at 834D/E-835B, is in virtually identical terms with that of section 1(1)(a) of the South African Act, save for the concluding words which read “having regard to the claimant's share in the responsibility for the damage”.) The House of Lords held that the plaintiff’s conduct had to be contrasted with the totality of the defendants’ tortious conduct. As the plaintiff had been substantially the author of his own misfortune and as his share in the responsibility for his injuries was at least as great as that of the defendants jointly, the plaintiff was only entitled to judgment for 50% of his claim. In the course of his judgment Lord Ackner said, at 970e-f:

“While the plaintiff’s conduct has to be contrasted with that of the defendants in order to decide to what extent it is just and equitable to reduce the damages which would be awarded to him if the defendants were solely liable, it does not involve an assessment of the extent to which the fault of each of the defendants contributed to that damage. What is being contrasted is the plaintiff’s conduct on the one hand with the totality of the tortious conduct of the defendants on the other.”

[89] The same approach is followed in Australia with respect to statutory provisions that are identical to the English statute. Thus, in Donaldson v Canberra Tyre Service Pty Ltd & Anor [2004] ACTSC 26 (5 May 2004) Crispin J said the following:

“15. In a case involving joint tortfeasors there may be some debate as to the manner in which any reduction in damages on that ground should be determined. As the New South Wales Court of Appeal observed in Barisic v Devenport [1978] 2 NSWLR 111, there has been an almost universal practice both in the United Kingdom and Australia of regarding the plaintiff as one unit and the defendants, if they are concurrent tortfeasors, as another. The plaintiff's negligence is then compared with the aggregate degree of negligence or blameworthiness of the defendants. The extent to which the plaintiff's damages should be reduced is determined as a result of this single balancing exercise and judgment, for the sum so reduced is then entered against both defendants. The extent to which each should contribute to the amount of such judgment sum is determined in proceedings for contribution or indemnity between them.”

[90] The Supreme Court of Canada applied a similar approach in Ingles v. Tutkaluk Construction Ltd 2000 SCC 12 with respect to a statutory provision (described as section 3 of the Negligence Act, R.S.O. 1990, c. N.1.) that read as follows:

3.  In any action for damages that is founded upon the fault or negligence of the defendant if fault or negligence is found on the part of the plaintiff that contributed to the damages, the court shall apportion the damages in proportion to the degree of fault or negligence found against the parties respectively.”

 In para [55] of the judgment the court said the following:

“When there are two or more tortfeasors, and a plaintiff has also been found negligent, the proper approach to apportionment is to first reduce the extent of the recoverable damages in proportion with the plaintiff’s negligence, and then to apportion the remaining damages between the defendants, in accordance with their fault; see, for example, Fitzgerald v. Lane, [1988] UKHL 5; [1988] 2 All E.R. 961 (H.L.); Bow Valley v. Saint John Shipbuilding, supra; Colonial Coach Lines Ltd. v. Bennett, [1968] 1 O.R. 333.”

 

On the facts of that case (it concerned the carrying out of building work without a building permit) the plaintiff, an owner builder, was found to have been 6% liable, the one defendant (the City of Toronto) 14% and the other defendant (a building contractor) 80%. The plaintiff’s damages were reduced by 6%.  

[91] That brings me back to the question of the application of section 1(1)(a) of the South African Act to two concurrent wrongdoers. Subject to two qualifications it seems to me that the approach in the jurisdictions referred to above, namely to assess the responsibility of the claimant against the totality of the tortious conduct of the wrongdoers, may also be followed here. The first qualification, in the light of South British Insurance Co. Ltd., v Smit, supra, is that the causative potency of the conduct of each of the parties (as opposed to their fault) would not play an immediate role in the comparison. The second qualification, in the light of General Accident Versekeringsmaatskappy SA Bpk v Uijs NO, supra, is to recognise that the entire process remains subject to considerations of justice and equity. Such considerations, I may point out, may become particularly relevant when the number of the wrongdoers involved or the nature of the conduct in question, would, on a mechanistic application of the formula, give rise to inequitable results.

[92] The application of this approach to the facts of the present case leads me to the following three conclusions which can be stated shortly:

(i) The blameworthiness of each of the plaintiffs is substantially outweighed by the aggregate of the blameworthiness on the part of Metrorail and Human, the two of them being regarded as one unit for purposes of the comparison.

(ii) In the circumstances of this case there are in my view no specific considerations of equity or justice, other than the comparison in respect of blameworthiness itself, that need to be taken into account.

(iii) In all the circumstances a reduction by one third falls to be applied in determining the damages that would be recoverable from Metrorail and Human by each plaintiff.

Metrorail’s claim against Kuffs

[93] That brings me to Metrorail’s claim against Kuffs. Apart from the denial that defendants were negligent, Kuffs did not raise any defence to Metrorail’s claim in terms of the indemnity. Following the amendment of its plea, Mr Marais did argue that second defendant was negligent in failing to provide Metrorail with trains having more efficient warning and braking systems. It is not necessary for me, however, to consider the force of this argument. Plaintiffs did not seek to advance a claim against defendants on that basis. A finding that second defendant provided Metrorail with such trains would in any event not have assisted Metrorail in defending plaintiffs’ claims as it would have been equally negligent in using such trains. It would therefore not have afforded Kuffs with a defence to Metrorail’s claim against Kuffs in terms of the indemnity.

Costs

[94] Plaintiffs have been substantially successful against Metrorail and Human. They are entitled to their costs. I have not heard argument on the question of costs as between Metrorail and Kuffs. This must stand over for later determination.

Conclusion

[95] In the result, I make the following declaratory orders:

1. First defendant (Metrorail) and third defendant (Human) are jointly and severally liable to pay damages to plaintiffs.

2. The damages to be recovered by each of the plaintiffs are subject to a reduction by one third in terms of the provisions of section 1(1)(a) of the Act.

3. First and third defendants are jointly and severally liable for the costs incurred by plaintiffs to date.

(d) The third party (Kuffs) is obliged to indemnify first defendant (Metrorail) against plaintiffs’ claims.

(e) All questions of costs as between first defendant and the third party stand over for later determination.

--------------------------

A P BLIGNAULT

---