Coopers Payen Ltd & Anor v Southampton Container Terminal

LORD JUSTICE SCHIEMANN: Lord Justice Clarke will deliver the first judgment.

LORD JUSTICE CLARKE:

These proceedings arise out of damage to a Mori 600 ton press which occurred at Southampton Container Terminal on 31 July 1998. The damage occurred when the press toppled off a mafi flat trailer which was being towed by a tug at Southampton Container Terminal. The appellants are Coopers Payen Limited ("Coopers") and Sanwa Parking Industry Co Limited ("Sanwa"). The press had been sold by Sanwa to Coopers on C&F Southampton terms. At the time of the accident the press was in possession of the respondents, Southampton Container Terminal Limited ("SCT"). It is common ground that SCT was the bailee or sub-bailee of the press at the time of the accident. The appellant's case is that the damage was caused by the breach of SCT's obligation as bailee or sub-bailee to take reasonable care of the press. However on 7 August 2002 the trial judge, Her Honour Judge Faber, rejected the appellant's claim and held that SCT was not liable for the damage. She gave judgment for SCT and refused permission to appeal. Permission to appeal was, however, subsequently granted by Carnwath LJ.

The appeal potentially gives rise to two issues. The first is whether the judge was right to conclude that SCT was not liable to the appellants. The second was argued before the judge but not decided by her; it only arises if the appeal succeeds. It is whether, if SCT is liable to the appellants, it is entitled to limit its liability by relying upon the Hague Rules package limitation said to be contained in the relevant bill of lading. It was initially agreed by the parties that this court should determine that question if the appeal on the first issue was allowed. Tuckey LJ so directed. However since then the Court of Appeal has decided The Rafaela S [2003] EWCA Civ 556. The appellants say that the effect of that decision as applied to the facts of this case is that the relevant carriage was governed by the Hague-Visby Rules and not the Hague Rules, which would mean that there would be no relevant package limitation. The House of Lords has given permission to appeal from the decision in The Rafaela S. In those circumstances the parties agreed that it would be sensible to await the decision of the House of Lords before arguing the limitation point in this appeal if it should arise. Schiemann LJ agreed and so directed. It follows that we are at present concerned only with the issue of liability.

The appellants claim damages in the sum of £47,830.93. It is not in dispute that that is the correct measure of damages, subject to any right which SCT may have to limit its liability.

The press had been manufactured by Mori in Japan. It was carried by road on a low loader from Mori's factory to Kobe. At Kobe it was packed in a wooden case by specialist packers and various separate parts were packed in a second case. A symbol purporting to show the centre of gravity of the load was marked on the case. The symbol was incorrectly placed.

A decision was taken by freight forwarders engaged by Sanwa to unitise the case on to a flatrack for the carriage to Southampton. It was latched to the flatrack and covered by blue tarpaulin which was battened down. The tarpaulin completely concealed the centre of gravity symbol on the case. The second case containing the parts was packed in a separate container.

The two containers were shipped on board the Mosel from Kobe to Southampton, pursuant to a bill of lading issued by Hyundai dated 30 June 1998. The bill of lading named Sanwa as the shipper and Coopers as the consignee and notify party. It named the port of discharge as Southampton and the place of delivery as "Southampton CY". The gross weight was said to be 38,350 kg.

The load has been described in some of the argument as "the unitised load". During the loading of the unitised load it was noted that it was very heavy and that the centre of gravity was a little high. Hyundai Japan sent an email to Hyundai Southampton warning them to make sure that the load was handled carefully during discharge. The email was passed on to SCT. As I understand it, the email played some part in the trial but it has not been referred to as being directly relevant to the issues in this appeal.

The Mosel arrived at Southampton on the morning of 31 July 1998. It appears that the press was discharged around 1700 hours in the evening. It was discharged into the possession of SCT. The unitised load was over-height and therefore was to be taken to a storage area at the end of the quay to await collection, rather than being put in one of the usual container stacks adjacent to the berth.

There are four berths at the Southampton Container Terminal. From west to east they are berths 207, 206, 205 and 204. The Mosel was berthed at berth 206. Initially it seems that SCT intended to take the load to a storage area at the end of berth 204. However in the event it was decided to take it to a storage area at the western end of berth 207. The unitised load was to be moved from alongside the vessel to the storage area on a mafi flat. The mafi flat was to be towed by a tug.

When the load was discharged from the vessel on to the mafi flat the tug turning the flat was facing in the direction of berths 205 and 204. Thus in order to go to the storage area beyond berth 207 it was necessary to perform a U-turn on the quay. The load was loaded on to the mafi flat and secured with twist locks.

The driver, Mr West, pulled away, initially in the direction of berth 204. It appears that he proceeded some 30 yards along the quay and then began to turn to the left, starting his U-turn in order to head in the other direction towards berth 207. During the turn the unitised load toppled over and fell on its side. The mafi flat trailer toppled over with it but fortunately the connection between the trailer and the tug came free so that the tug righted itself.

There was an issue at the trial as to whether the load toppled to the left or to the right. SCT contended for a topple to the left and the appellants contended for a topple to the right. The judge held that it toppled to the right (that is on the outside of the turn) and that is now accepted.

The appellants' case was put in various ways before and at the trial, but they invite this court to say that the load fell because the driver of the tug turned too fast and/or too tightly. They further say that SCT failed to discharge the burden of proving that it exercised all reasonable care and skill as a bailee. It is common ground that so far as the appeal on liability is concerned SCT's position is the same whether it was a bailee or a sub-bailey. I shall therefore refer to it as a bailee. The judge acquitted SCT of negligence or breach of duty and held that it had discharged the burden of proof. The appellants say that her conclusions were wrong and should be reversed.

The correct approach of the Court of Appeal to appeals of this and other kinds has been considered in many cases. I have myself had occasion to consider it in Asscicurazioni Generali SpA v Arab Insurance Group [2003] 1 WLR 577, [2002] EWCA Civ 1642. I will not repeat here what I said there. It is sufficient to refer to only two points. First, I there expressed the view (in paragraph 13) that in the type of appeal in which the court is asked to reverse findings of fact based upon the credibility of the witnesses, the same approach should be adopted whether the appeal is by way of review or rehearing. I adhere to that view for the reasons there given.

The second point is that in such a case, that is where the court is asked to reverse a judge's findings of facts which depend upon his or her view of the credibility of the witnesses, it will only reverse those findings if satisfied that the judge was plainly wrong (see paragraph 12 and the cases there cited). See also the discussion at paragraph 14 and following in relation to the different types of problem faced by the Court of Appeal.

The judge heard oral evidence from four witnesses on behalf of SCT. They were Mr Hatchard, the shift manager; Mr Wilton, the shipside supervisor; Mr Strange, the shoreside supervisor; Mr West, the tug driver; and Mr Aspinell, the insurance and claims negotiator. She also had the witness statement of Mr Green, who was a loss adjustor who inspected the goods on 5 August 1998. The appellants called only one witness, namely Mr Perera, who was a marine surveyor instructed by their underwriters. He had inspected the goods on the day of the accident and also investigated the claim in Japan. The judge also read the witness statements of Mr Hamada, the manager of the overseas division of Sanwa and Mr Perry, Coopers' engineering manager.

The judge directed herself that it was common ground that the burden of proof lay upon SCT as bailee, which meant that it must prove that it exercised reasonable care of the press or that its failure to do so did not contribute to the damage. She added that SCT did not have to prove the actual cause of the damage, and referred to pages 49 and 782-785 of the second edition of Palmer on Bailment.

The judge identified four main allegations of breach of duty made by the appellants. They were the speed at which the tug was driven, the angle of the turn, that it was performing a U-turn and SCT's failure to act on the Hyundai email. If the container had been loaded on to a Mafi in lane 1 instead of lane 3 it would have allowed a larger diameter of turning circle, and if the driver had not driven them at more than 5km an hour the accident would not have happened.

The judge considered the evidence in some detail and set out her findings between paragraphs 51 and 62 of the judgment. They may be summarised as follows:

SCT does not challenge any of those findings, whereas the appellants challenge conclusion no 8 that the tug was proceeding between 3.6 and 4.8 kph and conclusion no 9 that the mafi was not travelling at about 9.8 kph. They also say that if (as the judge held) there was no evidence that the turn was uniform the judge should have held that SCT had failed to discharge the burden of showing that the turn was not carried out too tightly. The appellants further challenge conclusions nos 5 and 6. They submit that the evidence shows that the casualty was not caused by the loosening of the washers and thus of the bolts holding the press in its case and that the judge should have so held.

I shall briefly focus on that point first because it seems to me to be of some potential significance. In reaching her conclusion nos 8 and 9 that the mafi was proceeding at a slow walking speed of between about 2 and 3 mph or 3.6 and 4.8 kph the judge said this (in paragraph 60):

Mr West said in evidence that the tug remained in first gear during the turn. The judge set out the unchallenged evidence that in first gear the tug could attain a maximum speed of 4.8 kph loaded and 5 kph unloaded. The load was about 75 per cent of the full load; so if the tug remained in first gear it could not have achieved a speed of over 5 kph or anything like the 9.82 kph spoken of by the expert and relied upon by the appellants. The judge might have held that Mr West was a truthful witness and that she believed his evidence that the tug was in first gear throughout. However she did not so hold. On the contrary she observed that if his had been the only evidence she could well have rejected it. There is no doubt that Mr West was not a wholly reliable witness because he thought that the press had toppled off to the left on the inside of the turn and not to the right on the outside of the turn, as the judge held had in fact occurred. As I have indicated, although one of the main planks of SCT's claim at the trial was that the press toppled over to the left, it does not challenge that finding.

Mr Russell submits that once it was held that the press toppled off to the outside of the turn the only explanation was that identified by the joint expert, Mr Krabbendam, namely that the mafi must have been driven at some 9.82 kph, which was too fast.

The judge's conclusion that the tug was being driven at less than 5 kph was based upon the evidence of Mr Strange whom she accepted as both a credible and a reliable witness. Mr Russell accepts that Mr Strange was a credible witness but submits that he was not a reliable witness. He submits that he was simply mistaken as to the speed at which the mafi was going.

It is to my mind important in this connection that in considering Mr Strange's evidence the judge said this:

In my opinion she was right in that regard. Indeed it seems to me to have been an important part of the exercise upon which the court was engaged, namely to decide whether the damage was caused by a failure on the part of SCT to exercise reasonable care. Mr Buckingham correctly submits, (as the judge observed) that it is not necessary in order to avoid liability for breach of duty for a bailee to show what caused the loss or damage: see eg Palmer on Bailment and Bullen v The Swan Electric Engraving Company [1887] 23 TLR 258. He must simply show (as the judge correctly put it) either that he took reasonable care of the goods or that his failure to do so did not contribute to the damage.

While it is not necessary as a matter of law for the bailee to show what the cause of the damage was the identification of the cause may be a significant pointer as to whether or not the bailee exercised reasonable care. The cause of the loss is in most cases far from irrelevant.

This can I think be seen on the facts of this case. The press toppled over while the mafi was being towed in a U-turn round to the left. Mr Krabbendam identified the mechanism which caused the topple as being the effect of a resultant force comprised of the horizontal centrifugal force caused by the turn and the vertical force caused by the press when the vessel force goes over what the expert called the tipping line. So far as I can see on the evidence those must have been the forces which ultimately caused the topple. One of them was the centrifugal force caused by the actions of the tug. If the tug was driven at a safe, non-negligent speed of less than 5 kph, even assuming the smallest possible (albeit uniform) turning circle, the accident could not have happened unless there was or were some other cause or causes.

What then was the cause of the accident, assuming that Mr Strange's evidence that the speed of the tug did not exceed 5 kph is accepted? The answer to that question must, in my opinion, have a potentially significant effect on the answer to the question whether Mr Strange's evidence as to speed can be accepted. It was no doubt for that reason that the judge referred to the other causes relied on by the parties. It seems clear to me that, if the turn was carried out at a safe speed, since there is no evidence or suggestion of a collapse of the container or any defect in the road surface, the cause must have been related in some way to the press or the packing.

As to that, the judge correctly considered those questions and, with one possible exception, rejected the possibilities. Thus, as already indicated, in conclusion nos 1 and 3 she rejected other causes on the quay and, importantly, in conclusion no 4, in the light of the evidence of Mr Krabbendam (whose objectivity she said she had no reason to doubt), she held that the distribution of weight did not render the unitised press unsafe for normal container transit. It is thus clear that she held that the somewhat higher centre of gravity and the shape of the press, which was by no means uniform, was not a cause of the topple. The only exception is her findings as to the washers.

Mr Buckingham submits that as long as this remains a potential cause of the accident there is no basis upon which the judge's conclusion can properly be challenged. He also suggested in the course of his oral submissions that there were a number of other possibilities which might have contributed to internal failure of the bolts. They included the way in which the press was treated in Kobe, both while being transported by road and while being loaded on board the vessel. However none of those were suggested as serious possibilities to the judge in the course of the respondents' final submissions and they are not referred to in the respondents' skeleton argument. They do not to my mind merit any consideration.

In Mr Buckingham's final outline submissions at the trial he concluded thus:

However Mr Buckingham correctly recognises that the logic of the judge's conclusion is that it is more probable than not that the casualty occurred because the failure of the bolts was caused by the failure of the washers. That is because unlike, for example the case of Popi M this is a case in which the various possible causes were all considered. Mr Russell identified six possibilities in the course of his submissions. They were these:

Possibilities 1 and 3 were never serious candidates. Possibility 2 is no longer a live candidate, having regard to the conclusions of the judge which were based upon the evidence of Mr Krabbendam. Possibility 4 is also no longer a candidate, having regard to the conclusion of Mr Krabbendam and the findings of the judge. That leaves possibilities 5 and 6.

As I indicated earlier the judge, to my mind correctly, regarded the consideration of the possible causes of the casualty other than a breach of duty by SCT as relevant to whether Mr Strange's evidence should be accepted. As I read her judgment she said, in effect, that, although she was unable to find that any of the other causes relied upon was responsible for the accident, she found Mr Strange's evidence convincing. She did not say what conclusions she would have reached if she had concluded on the balance of probabilities that none of the other suggested causes was causative of the accident. In that event, short of deciding the case on the burden of proof, she would merely have held that Mr Strange's evidence was not sufficiently reliable to lead to the conclusion that the speed of the tug was less than 5 kph. When the judge said that she was unable to find that any of the other causes relied upon was responsible for the accident she must have meant that that was the case if those causes were viewed in isolation. Given that she rejected as possibilities all the other possible causes except 5, it must follow from her conclusion that failed washers and thus failed bolts was the probable cause. However I do not think that she expressly reached that conclusion because, if she had, she would have said so.

The question for the judge to decide was ultimately whether the press toppled over because of the breach of duty of SCT. This question had to be considered by reference to the evidence as a whole, including the eyewitness evidence and the expert evidence, all of which (like any evidence) must be considered against the probabilities.

Mr Russell's submissions depend to a significant extent upon the evidence of the joint expert. Mr Russell submits that the judge rejected his evidence in circumstances in which she should not have done, given that he was a joint expert. He relies upon this statement of Lord Woolf MR in Peet v Mid-Kent Healthcare Trust [2001] EWCA Civ 1703, [2002] 1 WLR 210 (at paragraph 28):

Mr Russell submits that it should be the rare case indeed in which it is appropriate for the court to disregard the evidence of a single joint expert, and such a case will be limited to circumstances where the witness has failed to comply with his over-riding duty to the court or has plainly erred. He further submits that where such evidence is disregarded the judge must give clear and cogent reasons for doing so. There is force in those submissions.

Mr Buckingham by contrast, summarised his relevant submissions in this regard as follows:

All depends upon the circumstances of the particular case. For example, the joint expert may be the only witness on a particular topic, as for instance where the facts on which he expresses an opinion are agreed. In such circumstances it is difficult to envisage a case in which it would be appropriate to decide this case on the basis that the expert's opinion was wrong. More often, however, the expert's opinion will be only part of the evidence in the case. For example, the assumptions upon which the expert gave his opinion may prove to be incorrect by the time the judge has heard all the evidence of fact. In that event the opinion of the expert may no longer be relevant, although it is to be hoped that all relevant assumptions of fact will be put to the expert because the court will or may otherwise be left without expert evidence on what may be a significant question in the case. However, at the end of the trial the duty of the court is to apply the burden of proof and to find the facts having regard to all the evidence in the case, which will or may include both evidence of fact and evidence of opinion which may interrelate.

In the instant case the judge did not disregard the evidence of the joint expert. On the contrary in some respects she accepted it. A judge should vary rarely disregard such evidence. He or she must evaluate it and reach appropriate conclusions with regard to it. Appropriate reasons for any conclusions reached should of course be given.

In this case, as I have already indicated, the parties appointed Mr Richard Krabbendam as a joint expert. They did so in accordance with directions given by the court on 4 April 2002. The instructions to the expert which, as I understand it, had been prepared jointly by the parties were contained in a letter of instruction dated 12 April 2002. That letter also set out certain assumptions upon which the expert was asked to base his opinion. The expert wrote a report dated 15 May and a further report dated 21 May 2002, which was the first day of trial.

In the period before the trial SCT's solicitors asked the expert to consider a number of further questions. I think they did so in the light of some feedback which they had received from the expert, including (it may be) his first report. His report of 21 May, which was not in fact delivered until after the second day of the trial, considered two of the further questions to which I have referred. That report (which I shall call "the report") was not very different from his first report.

In his report he set out his conclusions in some detail. Since, as I have indicated, Mr Krabbendam produced his first report only shortly before the trial and his second report after the trial had began and since he had not answered all the questions which SCT's solicitors wanted answered, it was agreed that he would give oral evidence and that both parties would have an opportunity to cross-examine him. Indeed in the event it appears that the judge, as it were, examined the witness in chief and both parties were given an opportunity to cross-examine the witness.

Mr Krabbendam's conclusions in his report included the following:

In paragraph 6.5 of the report Mr Krabbendam considered the first of the two further questions (to which I have referred) which considered the question whether the press was offset. His conclusion was that that made some, but little, difference to his overall conclusion, which is set out in paragraph 7 of his report in this way:

The expert does not refer to what I may call the washer theory because it had not been put to him in clear terms before making his final report, if indeed in any terms at all. That seems to me to have been most unfortunate. It is, in my opinion, of the utmost importance that if at all possible all relevant questions should be put to experts, including a joint expert, together with all relevant assumptions of fact long before the trial in order to afford the parties the best chance of settlement and, failing settlement, the fairest and most efficient trial.

However that may be, the expert was examined in chief by the judge and cross-examined by both counsel. He essentially adhered to his opinion which was that the only possible cause of the casualty was that the mafi was driven too fast and/or turned at too tight an angle.

Both he and Mr Perera were cross-examined with regard to the washer theory. However, neither was cross-examined quite on the basis of the way the judge put it (in paragraphs 57 and 58 of her judgment) which was as follows:

I note in passing that although it is true that there is no evidence as to the size of the washers, there is a photograph of one of the washers taken by or for Mr Perera which shows one of the bolts and one of the washers after the casualty. Mr Perera, as I understand it, saw two of the bolts and two of the washers but not the other two bolts and washers. No one suggested at the trial either in evidence or in argument that the bolts might have been loosened in part as a result of "the wobble when hanging from the crane and then at a normal turn to the left".

I am bound to say that in those circumstances I do not for my part think that the judge should have included those as possible contributory causes to a loosening of the washers. We were shown all the evidence relating to this theory. There was very little of it. SCT did not call an expert with regard to it. In my opinion the evidence does not support the conclusion of the judge unless by "a possibility" she simply meant theoretical scientific possibility. Mr Perera accepted in cross-examination that the press would have been exposed to lateral forces while at sea. However he refused to accept that the bolts (of which there were four, each of 20 mm) would not have stayed firm.

It was suggested to him that there might be rocking and movement as a result of the press being secured only at the bottom. In answer to such a suggestion he said (page 88):

He accepted that there would be forces acting on the bolts tending to pull them, at any rate when the vessel had gone on one side of a roll, and he was asked this:

Mr Krabbendam gave evidence to very much the same effect, although he had not of course himself seen the damaged bolts. He was cross-examined as to whether he had relevant expertise on this question. He said that he had and that evidence was not challenged. His evidence may perhaps be seen from his cross-examination at page 178:

It appears to me that in these circumstances the position is that there was really no evidence to support the washer theory as a realistic possibility for the cause of this casualty. It seems to me that that conclusion is significantly different from the conclusion which was reached by the judge, to which I have already referred. Indeed to my mind the appellants have shown on the balance of probabilities that this was not the cause of the casualty, at any rate if looked at in isolation.

Given the different conclusion which I would reach as to the relevance of possible loosening of the washers it is, in my opinion, open to this court to reach its own conclusions. This is not a case where the conclusion of the judge or the outcome of the case depended upon the credibility of the witnesses. As I have indicated, the judge's conclusion depended largely upon the evidence of Mr Strange which, it is fair to say, was consistent with that of other witnesses. However, her conclusion depended not on his credibility but upon the reliability of his evidence. In my judgment it is very important to judge the reliability of his evidence in the context of the evidence as a whole. It appears to me from the way in which the judge approached the matter that she rather considered his evidence and that of the other witnesses from SCT in isolation and not in the context of the convincing expert evidence of the single expert, or in the context of the correct conclusion in relation to the only other possibility, namely, the washer theory.

Mr Strange, as Mr Russell correctly put it, had a vivid recollection of getting out of the way of the collapsing press. He was giving evidence four years after the event as indeed were the other witnesses. He was not driving the tug. It appears that he was first on the shore side of the tug and at some stage thereafter moved from the shore side of the tug to the river side of the tug. The tug must have been gradually increasing speed. Since it started from a stationary position that would have taken some time. It must have been difficult for him to judge the speed of the tug, even if he had been particularly observant, especially since it was increasing speed.

I have considered in some detail the transcript of his evidence from which it is plain that he was concentrating on a straddle carrier which was further along the quay. While fully recognising that we are in a worse position than the judge because we have not seen Mr Strange give evidence, the transcript of his evidence strongly suggests that his focus was on the straddle carrier and that his key memory was of taking evasive action to get away from the toppling of the press.

Mr Buckingham submits with great force that the judge had the advantage of seeing Mr Strange give evidence and that we should not reach a conclusion different from her. I accept that we are in a disadvantageous position in relation to his particular evidence by comparison with her but it is of the utmost importance that his evidence should be put in the context of the case as a whole. I have reached the firm conclusion that, if his evidence is put in the context of the case as a whole, including in particular the expert evidence and the evidence relating to the only other possible cause, the only fair conclusion that could be reached is that this casualty was, as Mr Krabbendam thought, caused by this tug proceeding too fast and/or at too tight a turn.

Mr Buckingham has submitted that the judge chose the turning diameter of 15 m on the basis that it was the diameter most favourable to the appellants and that on the evidence as a whole a much greater turning circle is justified. There is undoubted force in those submissions, but these are small distances and not easy to judge. It may well be that the turning circle was a little greater than 15 m which would involve the speed being somewhat greater than 5 kph if it was indeed the speed and/or the tightness of the turn which caused the casualty. Those considerations do not outweigh the other considerations to which I have referred. I would hold that this casualty was caused by the tug proceeding too fast and/or at too tight a turn. I reach that conclusion as I have said on the hypothesis that the turn was uniform, which is a hypothesis favourable to SCT.

For these reasons I would allow the appeal.

MR JUSTICE LIGHTMAN: I agree with the judgment of my Lord Clarke LJ and will only add a few words of my own. The issue on this appeal is whether it was open to the trial judge to accept the evidence of the defendants' witness, Mr Strange, and to reject the evidence of Mr Krabbendam, the single joint expert appointed in the case, on the issue of the speed at which the trailer was travelling when it toppled over.

The defendants' mafi flat trailer, whilst loaded with the claimants' Mori 600 ton press and towed by a tug toppled over damaging the press. At the time the trailer was doing a U-turn. The claimants sued the defendants for negligence as bailees for the damage occasioned. It is common ground that in this action the onus is upon the defendants as bailees to establish that they were not negligent or that any negligence on their part was not causative of the damage.

On the issue of negligence the critical question at the trial was the speed at which the trailer was travelling when it toppled over. The defendants' witness, Mr Strange, gave evidence that it was travelling at a slow walking pace not exceeding 3.6 to 4.8 km and certainly was not exceeding 5 kph. If this evidence is accepted and correct, the defendants could show that the toppling over was not attributable to any negligence on their part. The expert, Mr Krabbendam, however, gave evidence that the only possible explanation for the trailer toppling over whilst doing a U-turn must have been that it was travelling at at least 9.82 kph. If this expert's evidence is correct, the claimants establish that the speed was excessive and that for this reason the defendants were negligent.

Where a single expert gives evidence on an issue of fact on which no direct evidence is called, for example as to valuation, then subject to the need to evaluate his evidence in the light of his answers in cross-examination his evidence is likely to prove compelling. Only in exceptional circumstances may the judge depart from it and then for a good reason which he must fully explain. But if his evidence is on an issue of fact on which direct evidence is given, for example the speed at which a vehicle was travelling at a particular time, the situations is somewhat different. If the evidence of a witness of fact on the issue is credible, the judge may be faced with what, if they stood alone, may be the compelling evidence of two witnesses in favour of two opposing and conflicting conclusions. There is no rule of law or practice in such a situation requiring the judge to favour or accept the evidence of the expert or the evidence of a witness of fact. The judge must consider whether he can reconcile the evidence of the expert witness with that of the witness of fact. If he cannot do so, he must consider whether there may be an explanation for the conflict of evidence or for a possible error by either witness, and in the light of all the circumstances make a considered choice which evidence to accept. The circumstances may be such as to require the judge to reach only one conclusion.

In my judgment in the circumstances of this case and on the evidence before her, the judge could only properly have chosen to accept the evidence of the expert. There was by the close of the trial no surviving alternative explanation for the trailer toppling over other than it was travelling at an excessive speed. No other explanation was put forward by the defendants' counsel in his closing submission and the judge held that none was shown. She said, when dealing with the evidence of Mr Strange (in paragraph 60):

As it seems to me the error of the judge was to view the evidence of Mr Strange in isolation, and after holding that he was a credible witness to accept his evidence and on this basis to give a judgment for the defendant. The fact that Mr Strange was a credible witness did not absolve the judge from the duty of evaluating his evidence critically and deciding the weight to be afforded to it in the context of the situation which prevailed, namely, that it was inconsistent with the clear view of the single expert and would leave the accident totally inexplicable. If the judge had undertaken this exercise, she would have seen that Mr Strange's estimate of speed, though honest, could not have been correct. She must have preferred the evidence of Mr Krabbendam as to the speed at which the trailer was travelling and held that the defendants had not established that the damage was not attributable to their negligence. In a word she would have held that the trailer was travelling at an excessive speed.

I would therefore also allow this appeal.

LORD JUSTICE SCHIEMANN: I agree with both judgments and the reasons given by my Lords. I only add that in this court we have enormous advantage over the trial judge. She has cleared out of the way an enormous amount of matter which was discussed in front of her and made a crucial finding as to where the machine fell. This has enabled all in front of this court to concentrate on what are indeed the issues as identified by my Lord, and has enabled us to reach the conclusion which we have reached.