owners of the MV 'Banglar Mookh' v Transnet Ltd (842/2011) [2012] ZASCA 57; 2012 (4) SA 300 (SCA); [2012] 3 All SA 632 (SCA) (30 March 2012)

THE SUPREME COURT OF APPEAL OF SOUTH AFRICA

JUDGMENT

REPORTABLE

CASE NO: 842/2011

In the matter between:

THE OWNERS OF THE

MV ‘BANGLAR MOOKH' ….................................................................APPELLANT

and

TRANSNET LTD …...........................................................................RESPONDENT

Neutral citation: The owners of the MV ‘Banglar Mookh’ v Transnet Ltd (842/11) [2012] ZASCA 57 (30 March 2012)

Coram: Farlam, Cachalia, Tshiqi, Wallis JJA et Plasket AJA

Heard: 21 February 2012

Delivered: 30 March 2012

Summary: Vessel colliding with harbour wall while entering harbour – alleged negligence of the pilot – approach to evidence – unsafe to rely unduly on demeanour instead of the inherent probabilities – expert evidence reconstructing the incident only reliable where the underlying facts on which it is based are established – negligence not shown – negligent failure to retain records – does not warrant striking out defence.

_____________________________________________________________

ORDER

______________________________________________________________

On appeal from: Western Cape High Court, Cape Town (Binns-Ward J, sitting as court of first instance):

‘The appeal is dismissed with costs.’

_____________________________________________________________ _

________________________________ ____________________________

FARLAM ET WALLIS JJA (CACHALIA, TSHIQI JJA ET PLASKET AJA CONCURRING)

Introduction

[1] The appellant in this matter, the owners of the Banglar Mookh, the Bangladesh Shipping Corporation, instituted an action in the Western Cape High Court, Cape Town, exercising its admiralty jurisdiction, against the respondent, Transnet Ltd, and the National Ports Authority of South Africa. They claimed payment of the

damages suffered on 5 September 2005 when their vessel, the MV ‘Banglar Mookh’, which was at the time being piloted by Mr Tadeusz Jan Grelecki, an employee of the respondent, collided with the knuckle at the ‘A’ berth at the entrance to Duncan Dock in the Cape Town harbour. (It was subsequently agreed between the parties that the respondent was the party which would be responsible if the appellant were to establish a basis for liability for the damages sustained as a result of the collision and the National Ports Authority of South Africa, which had been cited as second defendant, took no part in the proceedings and no relief was sought against it.)

[2] In its particulars of claim the appellant alleged that the cause of the collision was the gross negligence of Mr Grelecki (whom we shall call in what follows ‘the pilot’). When the appeal was called in this court the appellant was granted leave to amend its particulars of claim to allege recklessness.

[3] The appellant accordingly sought to prove that the collision between the appellant’s vessel and the knuckle had been caused by the recklessness or gross negligence of the pilot. It did this in an attempt to circumvent the exemption from liability enjoyed by the respondent in terms of item 10(7) of Schedule 1 to the Legal Succession to the South African Transport Services Act 9 of

1989, which reads as follows:

‘The Company [i.e., the respondent] and the pilot shall be exempt from liability for loss or damage caused by a negligent act or omission on the part of the pilot.’

[4] In two High Court judgments, Yung Chun Fishery Co Ltd v Transnet Limited t/a Portnet, an unreported judgment of the Western Cape High Court delivered on 1 September 2000 in case AC 30/97, and Owners of the MV Stella Tingas v MV Atlantica & another (Transnet Ltd t/a Portnet & another, Third Parties) 2002 (1) SA 647 (D), SCOSA A 46(D), it was held that the exemption does not apply if the pilot’s acts or omissions were grossly negligent or reckless. When the Stella Tingas case came before this court1 it assumed, without deciding, that ‘the exemption would not apply if the pilot were found to have been grossly negligent’ (see para 7 of the judgment at 480 B–C).

[5] The appellant relied on the two High Court decisions to which we have referred and submitted that the pilot in this case was reckless or grossly negligent and accordingly that the exemption did not apply.

[6] The case came before Binns-Ward J in the court a quo.2 Although the learned judge had, as he put it, ‘some reservations’ whether item 10(7) had been properly construed in the two cases mentioned earlier, the issue did not arise because he held that the appellant had not succeeded in proving that the pilot had been guilty of gross negligence. Having found that gross negligence on the part of the pilot had not been proved, he held that the exemption contained in item 10(7) applied and consequently dismissed the appellant’s action, but gave the appellant leave to appeal to this court against his judgment.

[7] There were, as will appear more fully later, two conflicting versions of the events which led to the collision, one in the evidence of Captain Shahidul Islam, the master of the vessel, the other in the evidence of the pilot. The judge rejected the pilot’s

version and accepted that of Captain Islam. He held that the pilot had been negligent but not grossly negligent, hence the dismissal of the action.

[8] Mr MacWilliam SC, who appeared for the appellant, submitted, as was to be expected, that the judge had correctly accepted Captain Islam’s version of the events leading up to the collision, but had erred in not holding that the pilot was reckless or grossly negligent. He contended that the onus to establish that the pilot had not been grossly negligent was on the respondent, with the result that the judge’s finding, ‘that he was not persuaded that it had been established that the pilot was grossly negligent’,

amounted to a finding of absolution. This he submitted meant that the principle that where a defendant fails to establish its defence,

judgment must be given in favour of the defendant, should have been applied: cf Arter v Burt 1922 AD 303 at 306. He also argued that the judge had erred in failing to uphold a contention advanced at the end of the trial that, because the respondent had, despite giving an undertaking to do so, failed to preserve the vessel tracking service (VTS) records (which would have provided an objective and reliable record of what had led up to the collision), the court should strike out the respondent’s defence and give judgment in favour of the appellant, effectively as if by default.

[9] Mr Wragge SC, who appeared for the respondent, submitted that the judge had correctly rejected the contention that the respondent’s defence should be struck out because of its failure to preserve the VTS records. He also submitted that the onus of proving that the exemption contained in item 10(7) did not apply was on the appellant: consequently the principle that absolution from the instance is not an appropriate order in a case where the onus is on the defendant does not apply.

[10] Mr Wragge devoted the main part of his argument, however, to the submission that the judge had erred in preferring Captain Islam’s version of the events to that of the pilot and that on the pilot’s version he had not been negligent at all, much less grossly negligent or reckless. He contended further that the judge had misdirected himself on a number of material points

and had adopted an incorrect approach to the resolution of the factual disputes before him. He argued further that in the circumstances this court is at large to decide the matter afresh on the record and that it should dismiss the appeal on the basis that the pilot had not been negligent.

The evidence

[11] In order to facilitate an understanding of the evidence the judge gave a helpful summary in his judgment of what he called ‘the physical interrelationship of some of the salient features around the harbour basin outside the entrance to Duncan Dock’3. This summary reads as follows:

‘What the parties referred to as “the basin” is defined on its seaward aspect by the breakwater on the north western side and by the North Wall, which is part of the seaward wall of the Ben Schoeman Dock, to the south east. The breakwater runs out from the land at an angle in a north easterly direction, while the North Wall runs outward from the seaward boundary of the Ben Schoeman Dock in a north westerly direction, pointing towards the end of the breakwater wall on the opposite side of the mouth of the basin. A vessel sailing in an easterly direction so as to pass the breakwater from the west, as did the Banglar Mookh, would ordinarily turn to starboard at an obtuse angle to cross the basin following the leading line into Duncan Dock. The North Wall and the entrance to the Ben Schoeman Dock would be on the vessel’s port side as it crossed the basin; and the North Spur on its starboard side. The part of the basin immediately outside the entrance to Duncan Dock is characterized by the North Spur, which is a wall running out in a north easterly direction from the seaward side of the A Berth wall of Duncan Dock and, on the southern aspect, by the South Spur, being a wall running out in a generally north westerly direction from the end of the quay that comprises the boundary between the southern edge of the Ben Schoeman Dock and what is known as the Eastern Mole of Duncan Dock. The walls of the North Spur and the South Spur define, in effect, an inner basin immediately outside the entrance to Duncan Dock. As mentioned, the entrance to Duncan Dock is between the knuckle of the A Berth wall and the knuckle of the Eastern Mole.’

[12] The judge also set out in his judgment measurements of the distance between the salient points furnished by the appellant’s expert witness, Captain McAllister, which were, with rare exceptions the same as or a little bit longer than those derived from the charts. The judge used these measurements because, he said, the longer distances favour the respondent. These measurements were as follows:

‘(a) from the end of the A Berth knuckle to the end of the breakwater 1071,5m

from the end of the breakwater to the end of the North Wall 722m

from the end of the breakwater to the end of the North Spur 851m

from the end of the North Wall to the end of the North Spur 509m

from the end of the North Spur to the end of the A Berth knuckle 230m

from the end of the North Wall to the end of the A Berth knuckle 676,3m

from the No. 4 buoy to the end of the A Berth knuckle 1180m.’

[13] As a further guide to the understanding of the evidence the judge also gave three examples, taken from a table produced in evidence, illustrating the distance a vessel will cover travelling at various speeds. The examples were:

‘at a constant speed of 5 knots a ship covers 154 m a minute; at 7 knots, 216m a minute and at 9 knots, 278m a minute.’

[14] The judge gave the following summary of the two conflicting versions of how the collision occurred:

‘On the plaintiff’s version, which is founded on the evidence of the master of the vessel, the pilot found himself obliged, during the crossing of the basin that lies inside the breakwater but outside the entrances to the Ben Schoeman and Duncan Docks, to order the execution of a turn hard to starboard because the vessel was approaching too close to one of the outer structures of the harbour, identified on the charts as the “North Wall”. According to the master, the effect of the turn hard to starboard was to then place the vessel on a course, within the relatively narrow confines of the basin, which required a subsequent corrective hard to port manoeuvre if the vessel was to avoid another hard structure, known as the “North Spur”, on the opposite side of the basin. Captain Islam’s evidence had it that while the turn hard to port resulted in a successful clearance by the vessel of the North Spur the vessel was, however, thereby put on the course that resulted in the glancing blow of the starboard bow against the A berth knuckle when the ship passed into the Duncan Dock.

…

The defendant’s version, established principally through the evidence of Pilot Grelecki, also had the vessel turning sharply to starboard when it entered the basin after passing the breakwater. On the defendant’s version, this occurred involuntarily, due to the effect of prevailing conditions, and was corrected by putting the vessel hard to port and back on the leading line through the entrance to Duncan Dock. Grelecki’s evidence is that because the vessel was to be berthed alongside the Eastern Mole (also known as “landing wall 1”), which would be to the portside as the vessel entered the Duncan Dock, he gave the helmsman orders to move the wheel gradually to port as the ship approached the entrance to the dock. According to Grelecki, he noticed, however, that the bow of the vessel instead started to veer to starboard. He shouted orders of ‘hard to port’ to correct this. He simultaneously rushed over to the wheel from the position at which he had been standing, on the port side of the bridge, only to find that the helmsman had swung it hard to starboard. Grelecki testified that he had then pushed the helmsman

aside and himself swung the wheel hard to port, but too late to avoid the glancing collision with the A-berth knuckle.’

[15] Although the VTS records were not retained and were thus not available at the trial, a record of radio transmissions between the pilot, the masters of the two tugboats involved and port control was available and a transcription was handed in at the trial. It was accepted by the parties that the times reflected on this record were not accurate. They were adjusted by Mr Kieron Cox, an expert who testified on behalf of the appellant. Mr Cox’s adjustments were predicated on the assumption that the collision occurred at precisely 11h20, an assumption which was not necessarily correct, although the collision did occur at approximately that time and the approximation was a close one. The adjusted times, though not precisely accurate, are, as the judge put it, ‘a true reflection of the relative times in abstract vis-à-vis each other’.

[16] The material portions of this transcript from the time when the pilot spoke to Mr Le Blond on the aft tug until the forward tug was finally fast about seven minutes after the collision, with the adjusted times inserted and, in brackets, the sound byte

length of some conversations, read as follows:

‘11:08 (01:06)

Aft Tug: Pilot Grelecki, Enseleni, good morning?

Grelecki: Good morning Enseleni and good morning Pierre [Le Blond] is the forward tug?

Aft Tug: Ah no, I will be on the stern.

Grelecki: Thank you very much. Right astern, right astern, Eastern mole 1, Eastern Mole 1, port side to.

Aft Tug: Righto, all received

Forward Tug: How far is the …Pinotage forward tug?

Grelecki: Good morning Pinotage, good morning Henk [Turkstra]. Centre Lead forward, centre lead forward, landing wall 1, port side to, please

Forward tug: I think I must check, but Port Control when they call on, they call say landing wall 1

Grelecki: Landing wall 1, landing wall 1

Forward tug: Aye, Landing wall 1

Grelecki: Landing wall 1, landing wall 1, port side to, Centre Lead forward, please

Forward tug: Aye, aye.

11:18 to 11:19:32 (01:31)

Grelecki: Forward tug are you fast?

Forward Tug: Our messenger line going up Pilot

[inaudible]

[Period of silence in the recording. No transmissions]4

Person: Harry Harry – copy

Grelecki: Forward tug, Pull back, Pull back – forward tug bow to port (28 seconds after start of communication)

Forward Tug: Messenger line is still going up Pilot

Grelecki: OK.

[Period of silence in the recording. No transmissions]

Grelecki: Pull, pull, pull, pull! Forward tug bow to port

Forward Tug: I haven’t got the line up yet pilot

Grelecki: Pull. Pull to port, pull! Make full to port (1:04 after start)

Person: Harry

11:20 (Point of collision) (00:20)

Grelecki: Bow full to port

Forward Tug: The line is only going up now Pilot (6 second for whole transmission)

11:21 (00:55)

Grelecki: Pull the bow to starboard now (Eh!) (an exclamation)

Forward tug: Guys gone. Ran away from the bow. My wire isn’t up yet Pilot. There’s nobody up there.

Grelecki: Nobody up there, what must I do? (19 seconds into transmission)

11:22 (00:47)

Grelecki: Are you fast? (17 seconds into transmission)

Forward tug: I’m not fast yet Pilot.

Grelecki: Not fast yet

Aft tug: I’m fast aft (25 seconds into transmission)

Grelecki: OK. Can you heave up the bow? Bow to starboard? Sorry stern to starboard

Aft tug: I know.

11:24 (00:27)

Grelecki: The problem was that I gave the command “hard to port” and the helmsman was keeping hard to starboard and I miss the point. Back to the tugs I presume?

Grelecki: The forward tug was not fast yet. They didn’t get the heaving line. I was alone.

Forward tug: Aye. It’s going up again. They all ran away and then it got looped behind the fender. OK it’s going up again.

Grelecki: OK

11:27 (00:17)

Grelecki: After tug stop please.

Aft tug: Stop aft.

11:27 (00:30)

Grelecki: Forward tug, forward tug stern to starboard please. The after tug, after tug, stern to starboard.

Aft tug: Stern to starboard

11:27 (00:32)

Forward tug: It’s up there, but we are waiting and they keep on telling us to wait. I don’t know what they’ve got, what’s happening up there.

Forward tug: OK, we are finally fast.’

(The transcript has been slightly amended after listening to the recording.)

[17] Two days after the collision Captain Islam completed a ‘casualty/accident report’ in terms of section 259 of the Merchant Shipping Act 57 of 1957, as amended. The report he completed reads as follows:

‘While under pilotage on entering the port of Cape Town, vsl came into contact with concrete wall section A Berth knuckle vessel damaged at starboard side shell plating in way of fore peak tank and no. 1 tween deck along length of approximately 18M. All damage above waterline.’

[18] On the same day he had an interview with Mr F Hartzenberg, the attending surveyor, as part of an investigation into the collision conducted by the South African Maritime Safety Authority (SAMSA). The surveyor recorded the following:

‘The vessel made its way into port at 10:54. A wind of force 6 prevailed at the time. The vessel was moving at 9 knots on entering the port. The harbour tugs were not connected to the vessel at this time. The vessel collided with the knuckle at A-Berth at 11:18.

All this information was verbally furnished to the undersigned by the master of the Banglar Mookh. This interview was held on 07 September 2005 at approximately 15:00. This event occurred on 05 September 2005 at 11:18. If this office had been advised or notified earlier a more valuable comment may have been possible.’

The result of the SAMSA investigation was that no further action was required.

[19] On the day of the collision written reports were made by the pilot, the master and chief engineer of the forward tug, which was trying to make fast on the bow of the vessel (Messrs Turkstra and Stein), and the master of the aft tug which was trying to make fast on the stern of the vessel (Mr Le Blond).

[20] Messrs Stein and Le Blond testified for the respondent, Mr Turkstra having died before the trial.

The trial court’s approach

[21] The judge summarised and discussed the evidence of Captain Islam and the pilot at some length. It would unduly protract this judgment were we simply to quote what he said in its entirety. The following is a synopsis that we trust does him justice. We start with Captain Islam whose ‘description of the vessel’s approach to the port and the collision was not marked by any noticeable confabulation and was not upset in cross-examination.’

[22] The judgment said the following about Captain Islam’s evidence. Captain Islam said that he met the pilot when he came on board and handed him a pilot card dealing with the vessel’s specifications and performance. That reflected the sea speeds of the vessel in knots when fully loaded as:

‘Full ahead 9.0

Half ahead 7.5

Slow ahead 6.5

Dead slow ahead 4.5’

He also said that he told the pilot that if the vessel was travelling at a speed higher than 3 to 4 knots it was impossible to alter the engine movements from ahead to astern without first stopping the engine. The pilot said that he was not given the pilot card, but that Captain Islam told him that the vessel’s slow ahead speed was 7.5 knots,5 which he did not accept. The judge discussed the evidence of the pilot in this regard and found it to be inconsistent and improbable. He concluded from this that his evidence6 that he took the vessel up the channel at 6 to 7 knots could not be accepted and said that he was left with the impression that the pilot was in no position to state the speed of the vessel in the approach channel with any degree of reliability.

[23] The judge continued with Captain Islam’s description of the vessel’s journey up the approach channel. He said in his evidence that the vessel proceeded up the channel at half ahead and that the wind from the east tended to drive the vessel to

the eastern side of the channel, which the pilot controlled by small changes in speed and the helm. Captain Islam said, on the basis of what is contained in the bridge log, that the vessel passed the breakwater at 11h16. It was unclear what was meant by this: whether it meant that the bow passed the breakwater or that the bridge came abreast of it or that the entire vessel had passed this point. The judge recognised that this would affect the calculations of the experts, but concluded that, as the vessel’s length was only 159 metres, it was unnecessary to resolve this. The captain said that the hard to starboard manoeuvre, mentioned in the judge’s summary of the conflicting versions, was occasioned by a need to avoid a collision on the portside of the vessel with the North Wall. He did not suggest that the vessel was proceeding too fast at this stage, but claimed to have become anxious about its speed shortly before the collision, when he says that he noticed it was reflected on the ship’s instruments as 9 knots. He was at all times aware of the danger of a collision with the harbour walls.

[24] There was a difference between the master and the pilot over the former’s involvement during the passage up the channel. Captain Islam said that he was with the pilot and engaged with him about the navigation of the vessel, while the pilot said he showed no interest and was in conversation with another crew member on the bridge. What is clear is that the master did not intervene at any stage, but claimed that when a collision was imminent it was too late for him to do so. The judge did not resolve this dispute but noted that, after the collision, the pilot ‘acted expeditiously and appropriately to avoid the stern of the vessel also coming into collision with the harbour structure’. Lastly the judge noted the master’s evidence that the helmsman and duty officer were punctilious in complying with the pilot’s orders. He specifically denied that the helmsman had put the helm to starboard contrary to the pilot’s order and that the pilot had intervened and taken the helm himself to remedy that by turning the vessel hard to port. He did however accept that the pilot had complained about the helmsman both before and after the collision.

[25] Turning to the pilot the judge summarised and discussed his evidence as follows. He started with the passage down the channel, which he said took place at a speed of about 6 to 7 knots. He maintained a steady course with minor movements of the wheel. When the vessel passed the breakwater it was in the middle of the channel. At that stage he gave the order ‘full ahead’ in order to counteract the swell effect at the end of the breakwater, which tends to push the vessel to starboard. He said that in his experience vessels of this type steered more easily at speed and that the purpose of this order was to improve the handling

of the vessel. Just past the breakwater he linked up with the two tugs, which ordinarily wait for incoming vessels at a point inside

the breakwater. The tugs could not be made fast while crossing the basin, which he said was due to the incompetence of the crew of the Banglar Mookh, but this did not concern him as in eight cases out of ten, with vessels this size, they only made fast inside the dock. The judge found this strange as it left unexplained why the tugs should then wait inside the breakwater and why they had attempted to make fast while the vessel was crossing the basin. He preferred the view of Captain Woodend that it was preferable for the tugs to make fast while crossing the basin.

[26] Once the vessel crossed the breakwater and entered the basin it was committed to entering the Duncan Dock.7 The pilot said that the vessel regained the leading line into the dock, having corrected for the veer to starboard caused by the swell at the end of the breakwater. It then proceeded smoothly across the basin towards the entrance to the dock, maintaining its

line by minor course changes of no more than five to ten degrees either to port or starboard. When approaching the entrance he gave an instruction for the vessel to commence a general and gradual turn to port in order to enter the dock and berth at the Eastern Mole on the port side of the dock entrance. He was disconcerted to realise that, notwithstanding his order, the bow was turning to starboard. He rushed to the wheel and saw that the helm was to starboard. He then grabbed the helm and turned hard to port. Whilst the vessel started to correct itself it was too slow to avoid a glancing collision with the knuckle of A berth. He then turned the helm hard to port and thereby brought the stern of the vessel round and prevented the stern from colliding with the knuckle of A berth.8

[27] At this point in the judgment the judge evaluated the evidence of the pilot in regard to the incident with the helmsman to which the pilot ascribed the collision. He started with his demeanour and said the following:

‘[41] I have to say that I perceived that Grelecki was noticeably discomfited in the witness box during his evidence in chief when describing the vessel’s approach across the basin to the point of collision. He became more so under cross-examination. I also found his description of events markedly vague. It is clear that the collision was a traumatic event in Mr Grelecki’s life. He showed every sign of still finding difficulty working through the experience nearly five years after the event.

[42] His professed inability, during his evidence in chief, to recall whether the vessel had been turned hard to starboard was perplexing and appeared to be inconsistent with the answers counsel expected to elicit. As a matter of inherent probability the detail of the cause of the incident would be deeply engrained in the witness’s mind. On Grelecki’s version it was the alleged putting of the wheel to starboard, instead of steering to port, that was the fundamental cause of the collision. If Grelecki had indeed seen the wheel swung hard to starboard, I consider it most improbable that he would have forgotten the fact. Grelecki’s written report to the port authority made on the day of the collision or the day thereafter, describes the wheel having been swung hard to starboard.’

[28] The judge found it hard to credit that, after 20 minutes in which the vessel had traversed the channel without any misunderstanding between the helmsman and the pilot, there would at this crucial point be a mistake by the helmsman. He recognised that in the period immediately after the collision this was what the pilot said on more than one occasion. However he discounted this because he thought it inconsistent with his evidence that the order to turn hard to starboard was given ‘only in reaction to the bow having already noticeably veered to starboard’ in other words ‘after the helmsman had already steered the vessel in the wrong direction. He concluded by saying:

‘All in all Mr Grelecki’s evidence in respect of the alleged error by the helmsman was vague and inconsistent. As a result it falls to be rejected as unsatisfactory and unconvincing.’

[29] The judge then dealt with the pilot’s evidence of the speed of the vessel across the basin. He found his answers on the information he had received from the master inconsistent and regarded his estimate of the speed of the vessel across the ground inconsistent with the information in the pilot card and that of Captain McAllister. He concluded that the vessel must have been travelling faster than 7.5 knots while crossing the basin and may have been going faster. He then criticised the pilot for not taking up the conning position where he could see various instruments that would have provided some assistance in keeping him informed of the vessel’s movements. Lastly he said that he found his evidence of the course that the vessel took while crossing the basin unconvincing. He did so on the basis of matters such as the pilot’s unwillingness to concede that the vessel must have crossed to the westward side of the approach line,9 and his difficulty in explaining certain manoeuvres undertaken by the vessel in the course of its approach. He thought that the description of the vessel’s position at the time of the ‘emergency caused by the helmsman’s aberration’

was incompatible with the distance the vessel ‘must have covered’ during the final two and a half minutes prior to the collision. Lastly he thought that the pilot’s insistence that the vessel was not on the side of the channel furthest from the breakwater as it entered the basin was inconsistent with his telling Captain Woodend that the vessel came down the channel steering a course to put the number 4 buoy10 ‘fine on the port bow’.

[30] The judge then briefly discussed and summarised the evidence of Mr Le Blond, (which he said ‘contributed nothing material to assist in the determination of liability in this case’) and Mr Stein (about which he said that he did not consider it necessary to say much). Whether this was a correct approach will be dealt with later in this judgment.

[31] He then proceeded to discuss the evidence of the two experts, Captain McAllister, who testified on behalf of the appellant, and Captain Woodend, who testified on behalf of the respondent. ‘The essence of Captain McAllister’s evidence’, said the judge ‘was that it is important that a pilot should not bring a vessel into port at excessive speed.’ He continued:

‘Captain McAllister pointed out that while proceeding at a relatively high speed might give rise to good steerage, it reduces the pilot’s ability to control the vessel within the dangers presented by the confines of a harbour. The pre-eminent duty of a pilot, so testified Captain McAllister, is to keep the vessel under full control and to manage its progress in a pro-active, rather than a re-active, manner.’

[32] The judge’s summary of Captain McAllister’s evidence is set out in para [58] of his judgment, which reads as follows:

‘On the basis of the prevailing weather conditions, the description provided by Captain Islam and the cross-checking control afforded by the voice recordings, Captain McAllister opined that the vessel had been brought up the easterly (seaward) side of the approach channel with the use of a combination of speed and steering to starboard to counter the easterly drift. In the witness’s opinion the high speed of approach, coupled with the positioning of the vessel to the eastern side of the approach channel as it arrived at the position at which a turn to starboard was required to line up with the leading lights of the approach into the Duncan Dock, resulted in a loss of control manifested in the vessel’s drift towards the North Wall on the eastern side of the basin, which necessitated reactive steps by the pilot in the form of an increase of speed to improve steerage and a hard turn to starboard. The limitations imposed by the physical confines of the basin required the last-mentioned manoeuvre to be followed by a hard turn to port to avoid the vessel coming into collision with the North Spur on the south western side of the basin.’

[33] During the course of his evidence Captain McAllister submitted a series of calculations that suggested that the average speed of the vessel from the time when the pilot boarded her to the moment of the collision was in excess of seven knots. Though the witness accepted that his calculations were not definitive he suggested that they provided a useful guide, which corroborated his opinion that the pilot had brought the vessel in at an excessive speed. He was not however willing to commit himself definitively to a particular speed as being a ‘safe’ speed to approach the port. The witness also expressed the view that the failure of the tugs to make fast and be in a position ‘to render timely assistance’ was due to the fact that the pilot had brought the vessel to the point of collision at an excessive speed.

[34] The judge concluded his summary of Captain McAllister’s evidence as follows:

‘Captain McAllister impressed as an articulate and self-confident witness, who succeeded in providing a rational and easily comprehensible

foundation for the opinions which he ventured. He candidly conceded that his approach was reconstructive in nature – that he had worked backwards from the given fact of the collision to determine why it had happened. In assessing the witness’s opinion I have been astute to caution myself against the danger of being led by it into judging the conduct of the pilot too stringently with the benefit of wisdom after the event.’

[35] The judge was less impressed by the evidence of the respondent’s expert, Captain Woodend. He listed what he called ‘a number of indications of a tendency by Captain Woodend to tailor his opinion to support [the pilot’s] evidence’. The judge also commented that he ‘seemed extremely reluctant, when pressed, to question the reliability of what he had been told by [the pilot]; even in the context of the difficulties posed for [the pilot’s] version by the objectively established considerations of time and distance’.

[36] The judge summed up his assessment of Captain Woodend’s evidence in the following sentence:

‘In my judgment the effect of Captain Woodend’s evidence was undermined by an a priori and generally inflexible presumption in favour of the factual correctness of [the pilot’s] version of events.’

What is important to note about this conclusion is that its validity as a criticism of Captain Woodend was entirely dependent upon the pilot’s version being rejected. If it should have been accepted then it is no criticism of Captain Woodend that he relied on it. Captain Woodend had ‘fairly conceded’, as the judge put it, ‘that his opinion would have been different in certain respects were it to have been premised on the acceptance of Captain Islam’s evidence’.

[37] The judge largely accepted the evidence of Captain Islam and Captain McAllister and rejected that of the pilot and Captain Woodend. For the reasons already canvassed above he rejected the pilot’s evidence that the helmsman created a situation of sudden emergency by disregarding his order to turn to port and instead turned to starboard. He also rejected his evidence concerning the vessel’s position in the approach channel as it passed the breakwater. For that reason he rejected the evidence of both the pilot and Captain Woodend regarding the swell effect on passing the breakwater creating a veer to starboard and accepted the evidence of Captain Islam and Captain McAllister that this was necessitated by the risk of collision with the North Wall and the fact that the effect of a near gale force wind and the swell was to set the vessel towards the eastern side of the channel. That he said set it on a collision course with the North Spur on the western side of the channel and, because of the speed at which the vessel was travelling, (which he assessed as being at least 7 knots), the distance between the various harbour structures was too little to slow the vessel’s approach or avoid a collision. He held that the order ‘hard to port’ was given in order to avoid a collision with the North Spur, but said that it would not affect matters if it was given to avoid a collision with the A berth knuckle and that the pilot was aware at least three minutes prior to the collision that there was a problem, as evidenced

by his conversations with the tug masters. His conclusion was that the pilot lost control over the vessel as a result of having

approached at an excessive speed. This he linked to the failure of the tugs to make fast before the collision and their resultant

inability to assist in preventing the collision.

Discussion

[38] As can be seen from this summary and the quoted extracts from his judgment set out above the judge was strongly influenced in the conclusions to which he came by (i) his impressions as to the demeanour in the witness box of Captain Islam and the pilot and (ii) the opinions of Captain McAllister and in particular his estimates as to the speed at which the vessel was travelling at various points of its approach to the point of collision from the time it passed the breakwater. Before we say anything further about his reliance on his demeanour findings and the appellant’s expert’s reconstruction of what happened, it is necessary to say something about items of evidence which the judge did not mention, either because he overlooked them or did not consider them to be important.

[39] The first item of evidence to which we refer is the fact that unlike the pilot, who shortly after the collision – less than five minutes according to the transcript – said over the radio that he had given the command ‘hard to port’ but that the helmsman was keeping hard to starboard, From the outset the pilot accordingly blamed the helmsman for the accident. Captain Islam in the reports he made two days after the incident did not say anything about the collision being caused by the pilot. In the statement he made in the casualty/accident report (quoted in para 17 above) the account he gave was under a printed heading ‘Brief account of cause of casualty/accident and any other relevant information …’ In a box on the page above the space for the account of the cause of the incident where information was sought as to ‘the locality of ship where casualty/accident occurred’ he had placed a tick above the word ‘accident’. When it was put to him in cross-examination that he had not stated that it was the fault of the pilot he said:

‘Ja no casualty happened, that’s why no – there’s no need here to description the reason. There was no casualty

(indistinct) on board the ship.’11

[40] When pressed further on the point, when it was put to him that he was asked to provide the cause of the accident and had not said it was the fault of the pilot, he said, ‘no this is not necessary that it should be put there.’ He proceeded, ‘This (indistinct) describe this the cause of the collide with the A-berth knuckle, this was the cause. This is a collision with the A-berth knuckle that was the collision of this accident – this is the cause of the accident.’ As Mr Wragge submitted this was a disingenuous answer.

[41] When the SAMSA report was put to Captain Islam he stated that he gave information to the surveyor but could not remember what he had said. This left unexplained his failure to attribute blame to the pilot in a statement made shortly after the collision to the functionary charged with investigating the collision. If the position had truly been that the vessel was travelling too fast and narrowly avoided colliding with both the North Wall and the North Spur it is remarkable that he did not think to mention that.

[42] In our view the answer Captain Islam gave regarding the casualty/accident report form and the fact that it appears that he

made no allegation to the SAMSA surveyor that the pilot was to blame for the collision were important facts which were of relevance in deciding whether his version should have been preferred to that of the pilot. However the judge did not mention them in his assessment of Captain Islam’s evidence.

[43] It will be recalled that the judge said that he did not consider it necessary to say much about the evidence of Mr Stein, the engineer of the forward tug, the Pinotage. There were in our view at least two aspects of Mr Stein’s evidence which were important and which the judge did not mention.

The first was his evidence that if the vessel had been doing 9 knots he and Captain Turkstra would have noticed it and would have informed the pilot that he was going too fast. As appears from the transcript no-one at any stage told the pilot that he was going too fast and Captain Islam’s evidence was that until a late stage of proceedings he was not concerned about the speed of the vessel. The second item of evidence to which we wish to refer in this regard is Mr Stein’s description of the difficulties experienced in the attempts to make the Pinotage fast to the Banglar Mookh. He said in this regard that after the crew of the Banglar Mookh dropped the leading line down the crew of the Pinotage made it fast to the messenger and ‘then it was very, very slow in going up and at times it was not moving at all and by the time the vessel went over to starboard it was too late to do anything, we had to get out of the way.’ This was consistent with Captain Le Blond’s observation and assessment of the quality of the crew on the stern of the vessel.

[44] Mr Stein never suggested, nor was it put to him, that the reason his tug could not be made fast to the appellant’s vessel was, as the judge suggested in para [84] of his judgement, that this was caused in part by the fact that ‘the speed and course taken by the vessel hindered rather than assisted the process.’ Had there been a problem in either tug making fast, occasioned either by the speed of the vessel or any unusual manoeuvres in the course of its passage, the probability is that this would have been reflected in the radio conversations between the pilot and the tug masters. Instead Captain Le Blond said that it was a ‘normal day at the office’ until the collision occurred and Mr Stein’s evidence was that the only peculiarity was the behaviour of the crew in assisting the Pinotage to make fast. This evidence was disregarded by the judge. So was Captain Le Blond’s evidence that when the Banglar Mookh came down the channel towards where the tugs were waiting there was nothing untoward because: ‘This is just a normal ship coming down the channel.’ He stressed that there was nothing unusual about its speed or its movements, which was inconsistent with Captain Islam’s evidence that it was on the easterly side of the channel and at risk of colliding with the North Wall.

[45] The judge referred in his summary of Captain Islam’s evidence to the fact that Captain Islam said that the pilot had complained about the helmsman in connection with the collision, both shortly before and after the collision. He recognised that the radio transcript confirms a complaint after the collision, but then commented that the nature of any complaints made before the collision was not explored. Of greater importance and not considered in the judgment was that the transcript clearly showed that on three occasions in the immediate aftermath of the collision the pilot said that the helmsman disregarded his commands and turned the helm to starboard and not to port. Apart from the passage already quoted 23 minutes after the collision he said to port control: ‘‘I told the wheelsman hard to port. We were heading nicely, he repeated hard to port, but he was keeping all the time hard to starboard, so I immediately stopped.’ Immediately after this he told the two tug masters that he had kept saying ‘hard to port, hard to port’ and then noticed that the helmsman was steering hard to starboard.’

[46] In our view this evidence went a considerable way to undermine any suggestion that the pilot’s version was a contrivance and this should have been taken into account before the conclusion was arrived at that the pilot’s version was to be rejected. The statements were made at a time when the pilot had not had an opportunity to fabricate a version or collude with the two tug

masters. If untrue the master, helmsman and other crew of the Banglar Mookh were available to refute them. In addition the pilot was not to know that the VTS records, which might reveal a different picture, would become unavailable. Contemporaneous statements of this character cannot simply be disregarded, but the judge did so without any consideration of the improbability of the pilot being able to invent this story on the spur of the moment. He rejected his evidence of this incident on three bases. First, he thought the pilot was ‘visibly discomfited’ in giving this evidence. Second, he placed great store on the pilot’s inability to remember whether the helmsman had placed the helm hard to starboard and an impression he formed that this was not what counsel expected. Third, he found the pilot’s description of this incident in the course of his evidence confusing.

[47] We will deal with the question of demeanour below. The judge’s emphasis on the pilot’s inability to remember that the helm was put over hard to starboard, was, as Mr Wragge correctly submitted, misplaced. The pilot was consistent in his evidence that contrary to his instruction the helmsman had put the wheel over to starboard. What he could not remember when in the witness box four and a half years after the incident was whether the helmsman had put the wheel over hard to starboard. In his conversation with the tug masters he had mentioned that the wheel was hard to starboard and this was accepted. His unwillingness at the trial to say definitely that this was what he observed redounds rather to his credit as a witness. The judge found it improbable that he would be unable to recall this detail. That suggests that it would be obvious visually. However, as the photographs show, the wheel was little larger that a conventional steering wheel in a motor car with six spokes protruding from the outer rim and no markings of the helm position. That could only be read off the ship’s instruments and in a situation of emergency there was little time to observe those. When that is borne in mind the perceived improbability disappears.

[48] We have already referred to the extent to which the judge relied on his demeanour findings in coming to his conclusion on the facts. This court has on a number of occasions in the past warned about the risks inherent in relying on demeanour: see in particular the judgment of Harms JA in Body Corporate of Dumbarton Oaks v Faiga [1998] ZASCA 101; 1999 (1) SA 975 (SCA) at 979 C–I, where some of the decisions on the point are referred to.12 What is always important is to decide the case in the light of what Harms JA called (at 979 I of the Dumbarton Oaks case) ‘the wider probabilities’. This required that Captain Islam’s evidence be subjected to the same close scrutiny as that of the pilot. Had that occurred no doubt the judge would have taken account of his repeated statements that he could not remember pertinent detail; his inconsistencies on certain aspects, such as between his description of the way his crew performed and that of the tug masters; his unfamiliarity with the layout and conditions at a port he was visiting for the first time and his lack of knowledge of the proper way to con a vessel into the Duncan Dock.

[49] In assessing the wider probabilities a most important factor was the failure of Captain Islam shortly after the incident to cast any blame on the pilot. The judge’s failure to have regard to this factor is a clear and in our view serious misdirection. His failure to give proper weight to the pilot’s complaints immediately after the collision that the helmsman had disregarded his orders and steered hard to starboard is likewise a serious misdirection. So too was his failure to give adequate weight to the handicaps of language and the elapse of time in assessing the pilot’s demeanour.

[50] We referred earlier to the fact that the judge had relied to a considerable extent on the expert opinion of Captain McAllister and in particular his reconstruction of the speed of the vessel at various stages. This court has recently had occasion to consider how reconstructions by experts, in particular in motor collision cases, should be approached: see Biddlecombe v Road Accident Fund [2011] ZASCA 225 (30 November 2011). In para 9 the court pointed out that in some cases expert evidence may provide ‘a definitive factual background against which to weigh the merits of the eyewitness accounts of what occurred.’ An example of this will be where

physical evidence, such as skid marks, location of debris, etc, is viewed in the light of established scientific data. But as is pointed out in para 10 of the judgment ‘(t)he expert tasked with reconstructing what occurred is often dependent for the

reconstruction not simply on the application of scientific principle to accurate data but on calculations based on imperfect human

observation. The fact that the reconstruction rests on a potentially imperfect foundation is the reason for caution in determining its evidential value’.

[51] We do not think that Captain McAllister’s reconstruction can be regarded as having been based on accurate data – on the contrary we think that it rested on ‘a potentially imperfect foundation’. A number of aspects are not clear. The deck and engine logs did not coincide and the assumption that the engine log could be taken as reliable was not based on any factual foundation. Accordingly there was no clarity on the engine speeds at different stages. To take but one example, the deck log said that the order ‘full ahead’ was given after passing the breakwater, whilst the engine log showed it as having been given before passing the breakwater. The difference between the two is one minute and that would materially affect the calculations. What was meant by ‘past the breakwater’? If that point was taken only once the vessel’s superstructure was past the end of the breakwater it reduced the distance to be covered to the point of collision by around 10 per cent and the speed by between one and two knots, from the 9 knots calculated by Captain McAllister to a little over seven knots, which no-one described as too fast. The position of the vessel at the various stages was not clear. What effect did the heading of the vessel have on the calculation? Captain McAllister agreed that it was not possible to assess the speed of the vessel as it passed the breakwater. Was the vessel on the easterly side of the channel as it passed the breakwater? Captain Islam said it was and the judge accepted this evidence and found that this was a result of an easterly set caused by the wind and swell. But in coming to this conclusion he ignored the unchallenged evidence of the respondent’s expert, Captain Woodend, who had extensive experience in piloting

vessels entering the port of Cape Town,13 that the effect of wind and tide at the point is to set the vessel to the west as described by the pilot and not to the east, which is what mariners unfamiliar with the port would expect. While the judge was correct in criticising his evidence because he was unwilling to reject the pilot’s version on certain issues, that criticism does not apply to his evidence on this point. Here he was in any event not testifying as an expert but on his experience as a pilot which was that there is no easterly set in the entrance to Table Bay if the wind and swell are coming from the west (as they were on the day of the collision). If there was no easterly set and the vessel was more or less on the leading line, subject only to minor course corrections as described by the pilot, then it was also travelling significantly slower than Captain McAllister’s calculations suggested. The fact that the pilot came down the channel, well before reaching the breakwater, with the number four buoy ‘fine on the port bow’ (ie at an angle of up to 45 degrees from the port bow looking ahead), lends no support to Captain Islam’s evidence that the vessel was on the easterly side of the channel.14

[52] One last aspect of the judge’s conclusions must be addressed. He found (and counsel supported this in argument) that the order by the pilot to go hard to port was an endeavour to avoid a collision with the North Spur and not an endeavour to avoid colliding with the knuckle of A berth. An examination of the chart of the entrance to Duncan Dock demonstrates that this is highly improbable. These two points are only 230 metres apart. They are so situated in relation to one another that a vessel the size of the Banglar Mookh (159 metres long) that successfully took evasive action to miss the North Spur by going hard to port, would then be on a heading that would take its bows clear of the knuckle of A berth. Once its stern cleared the North Spur its bow would be only about 70 metres from the entrance to the dock and heading across the face of the entrance. It would be more likely to collide port side on with the entrance to the Duncan Dock adjacent to Pier 1 than with the knuckle of A berth. Yet the collision was with the latter and involved a glancing blow with the starboard bow of the vessel. That makes it probable that the action of going hard to port was directed, as the pilot claimed, at avoiding the drift down on to the knuckle that started before passing the North Spur, as a result of the vessel going to starboard from a position near the leading line. That in turn is consistent with what both tug masters said in their reports after the collision.15 Far from that being irrelevant as the judge suggested it was strongly supportive of the pilot’s evidence. In all the circumstances we are satisfied that the judge erred in holding that Captain Islam’s version was to be preferred to that of the pilot.

[53] As we have endeavoured to indicate the judge misdirected himself in a number of respects in his approach to the evidence, with the result that this court is at large and obliged to decide the matter afresh on the record. In our view if the evidence is approached correctly, without misdirection, it is clear that the pilot’s version, despite his weaknesses in giving evidence, was supported by most of the probabilities and should not have been rejected. Accordingly, unless the point taken by the appellant as a result of the respondent’s failure to retain the VTS data and records is a good one, the appeal must be dismissed on the simple ground that no negligence on the part of the pilot was proved.

The unfair trial point

[54] We do not think that Mr MacWilliam’s contention that the respondent’s defence should have been struck out, because it breached the undertaking to preserve the VTS records for production at the appropriate time should litigation ensue, should be upheld.

[55] In response to a notice by the appellant in terms of Rule 35(3), (6) and (10), in which the appellant sought production, inter alia, of the VTS records, the respondent filed an affidavit dated 9 December 2009 and deposed to by Ms Lerato Maboea, the legal manager for the National Ports Authority, Cape Town, in which she dealt with the VTS records as follows:

‘Regrettably, the recordings in question (to the extent that they existed) were lost when the Port of Cape Town upgraded and replaced its vessel tracking system (“VTS”) in the first quarter of 2006. In any event, I am advised that the data recording system previously in operation had malfunctioned which would have prevented any copy of the data being made and stored. In addition, I am further advised that the hard drive of the data recording system previously in operation would override and update itself every 3 to 5 days.’

[56] The appellant did not seek to cross-examine Ms Maboea on the contents of her affidavit, nor did it apply at the outset of the trial to have the respondent’s defence struck out. Instead it participated fully in the trial and only raised the contention presently under discussion in its argument at the end. At no stage did it put the respondent on notice that it proposed to contend that Ms Maboea’s affidavit should not be accepted or that the records had been deliberately destroyed by the respondent or the port authority. In the circumstances we think that this aspect of the case must be approached on the basis that what she said was correct and that the failure to preserve the records was inadvertent or accidental.

[57] Mr MacWilliam’s main argument was based on the contention that the court should follow the decision of the English Court of Appeal in Arrow Nominees Inc v Blackledge [2000] EWCA Civ 200 (22 June 2000)[2000] EWCA Civ 200; ; [2000] 2 BCLC 167 CA[2000] EWCA Civ 200; ; [2001] BCC 591 (CA). In that case the Court of Appeal, reversing the decision of the court a quo, held that the judge should have struck out a petition for relief against unfair conduct by the majority shareholder of Arrow Nominees

Inc and two of its directors. The ground for doing so was that the petitioner, through its representative, had forged documents in the course of discovery thereby preventing a fair trial of the petition.

[58] In para 54 of his judgment Chadwick LJ (with whom Ward LJ, who also gave a separate concurring judgment, and Roch LJ agreed) adopted an observation of Millet J in Logicrose Ltd v Southend United Football Club Ltd (1998) Times, 5 March, that:

‘… the object of the rules as to discovery is to secure the fair trial of the action in accordance with the due process of the Court; and that, accordingly, a party is not to be deprived of his right to a proper trial as a penalty for disobedience of those rules - even if such disobedience amounts to contempt for or defiance of the court - if that object is ultimately secured, by (for example) the late production of a document which has been withheld.’

Chadwick LJ then went on:

‘But where a litigant's conduct puts the fairness of the trial in jeopardy, where it is such that any judgment in favour of the litigant would have to be regarded as unsafe, or where it amounts to such an abuse of the process of the court as to render further

proceedings unsatisfactory and to prevent the court from doing justice, the court is entitled - indeed, I would hold bound - to

refuse to allow that litigant to take further part in the proceedings and (where appropriate) to determine the proceedings against him. The reason, as it seems to me, is that it is no part of the court's function to proceed to trial if to do so would give rise to a substantial risk of injustice. The function of the court is to do justice between the parties; not to allow its process to be used as a means of achieving injustice. A litigant who has demonstrated that he is determined to pursue proceedings with the object of preventing a fair trial has forfeited his right to take part in a trial. His object is inimical to the process which he purports to invoke.

Further, in this context, a fair trial is a trial which is conducted without an undue expenditure of time and money; and with a proper regard to the demands of other litigants upon the finite resources of the court. The court does not do justice to the other parties to the proceedings in question if it allows its process to be abused so that the real point in issue becomes subordinated to an investigation into the effect which the admittedly fraudulent conduct of one party in connection with process of litigation has had on the fairness of the trial itself. That, as it seems to me is what happened in the present case. The trial was “hijacked” by the need to investigate which documents were false and what documents had been destroyed.’

[59] The Arrow Nominees case was subsequently considered by the Court of Appeal in two decisions, both reported in [2010] 1 All ER, viz. Shah v Ul-Haq [2009] EWCA Civ 542; [2010] 1 All ER 73 (CA) and Zahoor & others v Masood & others [2009] EWCA Civ 650; [2010] 1 All ER 888 (CA). The Shah case concerned a motorist, involved in an accident and entitled to recover damages for his injuries, conspiring with a third party to bring a fraudulent claim against the defendant on the basis that the third party was a passenger in the car at the time of the accident, which she was not. In holding that this conduct did not deprive him of his right to recover his own damages Smith LJ said (para 28):

‘Everything that was said in the Arrow Nominees case related to the situation which arose in the course of the trial, once it had become apparent that the petitioner’s dishonesty was such that a fair trial had become impossible.’

Similar views were expressed in the Zahoor case. There both parties in complex civil litigation were guilty of forgery and fraud in the presentation of their respective cases. This emerged in the course of a twenty day trial. It was then argued, as it has been here, that the claim should have been dismissed on the grounds of the claimant’s misconduct, but the trial judge declined to do so. On appeal Mummery LJ said, in giving the judgment of the court:

‘We accept that, in theory, it would have been open to the judge, even at the conclusion of the hearing, to find that Mr Masood had forged documents and given fraudulent evidence, to hold that he had thereby forfeited the right to have the claims determined and to refuse to adjudicate upon them. We say "in theory" because it must be a very rare case where, at the end of a trial, it would be appropriate for a judge to strike out a case rather than dismiss it in a judgment on the merits in the usual way.

One of the objects to be achieved by striking out a claim is to stop the proceedings and prevent the further waste of precious resources on proceedings which the claimant has forfeited the right to have determined. Once the proceedings have run their course, it is too late to further that important objective. Once that stage has been achieved, it is difficult see what purpose is served by the judge striking out the claim (with reasons) rather than making findings and determining the issues in the usual way … In a complex case (such as the present) which requires a good deal of evidence before the fraud can be established to the requisite standard of proof, it may be difficult to avoid a full trial.’

[60] Four points emerge from these cases. First, the power is only exercised in the case of fraud or dishonesty. Second, none of them go so far as to say that the power to strike out on these grounds is available against a defendant, thereby affording the plaintiff a victory by default, although it is possible, without the need to decide whether it is permissible, to conceive of an extreme case where that might be done. Third, only in an extreme case will it be exercised when the trial has run its course. Fourth, it is only if a fair trial was prevented that, as Mr MacWilliam correctly conceded, the point can be taken. Therefore, if the court concludes that the absence of the VTS records did not prevent a fair trial, the point must fail.

[61] In our view it cannot be said that the trial was unfair. The appellant was able to lead Captain Islam and its expert Captain

McAllister. The radio transcripts were available as were the ship’s logs. The pilot gave evidence and was cross-examined, as did the master of the aft tug and the chief engineer of the forward tug. Apart from this a large amount of other relevant data was available including the reports made by the master, the pilot, and those on the tugs, as well as a detailed hydrographic chart of the locality where the collision occurred. The missing records might have added greater certainty to the underlying facts on which the experts based their evidence, but they would not necessarily have shown that the plaintiff should have succeeded. In addition the loss of the records was at most due to negligence and not due to any dishonesty or reprehensible conduct on the part of the defendant. In this situation, the position is no different from that in any case where a document is lost or an important witness dies or disappears without any means of recovering their evidence. The parties must then make do with what is left to advance their respective cases. The absence of the evidence does not make the trial unfair.

[62] For these reasons it is clear that the judge correctly dismissed this point.

Conclusion

[63] In the circumstances we are satisfied that the appeal must be dismissed with costs. The following order is made:

The appeal is dismissed with costs.

I G FARLAM

JUDGE OF APPEAL

M J D WALLIS

APPEARANCES

APPELLANT: R W F MACWILLIAM SC

Instructed by Edward Nathan Sonnenbergs,

Cape Town;

Webbers Attorneys, Bloemfontein.

RESPONDENT: M WRAGGE SC

Instructed by Webber Wentzel Attorneys,

Cape Town

McIntyre & Van der Post, Bloemfontein.

1Stella Tingas, MV: Transnet Ltd t/a Portnet v Owners of the MV Stella Tingas 2003 (2) SA 473 (A); SCOSA A 59 (SCA).

2The Owners of the MV ‘Banglar Mookh’ v Transnet Ltd [2010] ZAWCHC 485 (12 October 2010).

3A reduced copy of a chart depicting the area concerned, which was handed in at the trial, is reproduced in the annexure to this

judgment. The A Berth knuckle is marked F. G. on the chart and is adjacent to A cargo shed.

4The transcript might otherwise suggest that this is a continuous conversation and convey a rising concern on the part of the pilot.

That would be unfair to him as after each order and the response that the line is not yet up, there is an interval suggesting that he was expecting the line to go up before he repeated the order. His tone of voice remains consistent throughout.

5This was an error on his part. The pilot said that in his experience with this type of vessel he suspected that the slow ahead speed would be 4½ knots but that the master told him that it was between 6 and 7 knots.

6Characterised by the judge as an assertion.

7Both experts agreed that this was so.

8In effect what he described was the vessel pivoting around the point of the knuckle of A berth. The bow collided with the knuckle but as the vessel then turned ‘into’ the point of collision by turning hard to port the stern moved away from the point of collision and further damage was avoided.

9Presumably he meant the leading line being the central line in the channel that is used by pilots as a guide for vessels to follow when entering the Duncan Dock.

10The buoy on the opposite side of the channel to the breakwater.

11It is clear that Captain Islam, like the pilot, was not speaking in his home language and that explains the slight incoherence of his answers. He was giving as his explanation for not blaming the pilot that this was an accident and did not amount to a casualty, which can have a technical meaning in maritime parlance. However that does not explain why he did not say that the pilot was responsible for the accident.

12See further H C Nicholas, ‘Credibility of Witnesses’ (1985) 102 SALJ 32 at 36 – 37, M M Corbett, ‘Writing a Judgment’ (1998) 115 SALJ 116 at 124 and Lord Bingham of Cornhill ‘The Judicial Determination of Factual Issues’ Current Legal Problems, Vol 58, 1 – 29, reprinted in his book The Business of Judging at 8 – 11, esp at 9 where he said, ‘the current tendency is (I think) on the whole to distrust the demeanour of a witness as a reliable guide to his honesty.’

13He was not only an experienced pilot but was formerly the port captain in Cape Town. Neither Captain McAllister nor Captain Islam had similar experience of local conditions as a pilot. Captain McAllister had been the master on board container vessels that

docked in the Ben Schoeman Dock not Duncan Dock.

14On any basis when the vessel came close to passing and passed the buoy it must have been broad on the port beam. When the transition from ‘fine’ to ‘broad’ occurred would depend on the vessel’s position in the channel.

15Captain Turkstra said ‘the ship started to veer to starboard’ and Captain Le Blond said ’it took a sheer to starboard’. It is significant that both mentioned this in their reports without either of them indicating that there had been sudden or unusual movements by the vessel prior to this point.