Fal Oil Co Ltd & Anor v Petronas Trading Corporation SDN BHD [2004] EWCA Civ 822

Case Nos: A3/2003/2289 & 2299[2004] EWCA Civ 822
IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
[2003] EWHC 2225 (Comm)
ON APPEAL FROM THE QUEEN’S BENCH
DIVISION (COMMERCIAL COURT)
MORISON J.
Venue Royal Courts of JusticeDate Wednesday 07 th July 2004
Strand,
London, WC2A 2LL
Before
THE RIGHT HONOURABLE LORD JUSTICE JUDGETHE RIGHT HONOURABLE LORD JUSTICE BUXTONTHE RIGHT HONOURABLE LORD JUSTICE MANCE
Between
FAL OIL CO. LTD.(2) CREDIT AGRICOLE INDOSUEZ (SUISSE) SA Claimants/Respondents in 1 st appeal Appellants in 2 nd appeal- and -PETRONAS TRADING CORPORATION SDN BHD Defendant/Appellant in 1 st appeal Respondent in 2 nd appeal
(Transcript of the Handed Down Judgment ofSmith Bernal Wordwave Limited, 190 Fleet StreetLondon EC4A 2AGTel No: 020 7421 4040, Fax No: 020 7831 8838Official Shorthand Writers to the Court)
Mr. Richard Millett QC & Mr. Edmund King (instructed by Ince & Co.) for theClaimant /(1)Respondent & (2)AppellantMr. Stephen Males QC & Mr. Lawrence Akka (instructed by Holman Fenwick Willan) for Defendant/(1)Appellant & (2)Respondent
Judgment

Introduction

[1]This is an appeal and cross-appeal by permission of the single judge from a judgment of Morison J given on 10th October 2003. Morison J held after a five-day trial in July 2003 that the defendants, Petronas Trading Corporation Sdn Bhd (“Petronas”), had not established any defence to a claim for the price of fuel oil allegedly sold and delivered to them by the claimants, Fal Oil Co Limited (“Fal Oil”). Petronas appeal against that decision, while Fal Oil appeal against the judge’s decision in Petronas’s favour on a lesser claim relating to demurrage. Fal Oil’s claims were made under a contract of sale dated 29th November 2000. This provided for delivery of four cargoes of 80,000 metric tons (10% plus or minus in sellers’ option) of cracked fuel oil 700 CST of “ex-Yanbu Standard Quality” C and F one/two safe port(s) Singapore/Pasir Gudang range. To be of Yanbu standard quality, cracked fuel 700 CST must have no more than 1% water and sediment.[2]For Fal Oil’s operational convenience, a variation was agreed in respect of the first cargo and recorded in faxes of 13th and 14th February 2001, whereby the intended cargo would be lifted from Yanbu in the Centaur, but would be the subject of a ship-to-ship (“STS”) transfer onto the Devon off Port Sudan, and:
“1. B/L should be cut as per the quantity received by daughter vessel [i.e. the Devon] taking into account the vessel experience factor. 2. Quality of each sample taken from individual tanks of daughter to meet the contractual quality. ….. 5. Buyer accepts Saybolt as independent inspector for the STS operations (for the inspection of quantity and quality in relation to items 1 and 2 above), and agrees to share the inspection fee with the Seller. …..”
[3]The Centaur loaded cargo at Yanbu on 25th February 2001, a STS transfer onto the Devon took place off Port Sudan on 26th February 2001 and the Devon arrived on 17th March 2001 at her first nominated discharge port of Singapore, her second nominated discharge port being Pasir Gudang. The percentage of water and sediment in samples taken from the Yanbu shore tanks on 20th and 22nd February 2001 was 0.2%. This was consistent with the range of water and sediment which was subsequently ascertained to be present in samples taken from the Centaur’s tanks after loading at Yanbu. 0.2% of 85,000 tonnes is 170 tonnes. However, on discharge of the Devon in the Far East, her cargo was found to contain about 1,700 tonnes of water (mainly sea-water) - an excess over the reported Yanbu quantities of some 1500 tonnes. This discovery led to great delay in discharging and to a claim by Petronas to reject the whole cargo – hence Petronas’s refusal to pay its price. The issues before Morison J were(1) whether Petronas had established on the balance of probabilities that the cargo loaded onto the Devon at the STS contained more than the permitted contractual maximum of 1% water and sediment and(2) the nature of any contractual liability for demurrage by Petronas towards Fal Oil. (The order in terms also required the judge to determine Petronas’s liability for demurrage, but this would raise further issues which were not addressed before him or us.) The judge held that (a) Petronas had failed to establish any excess of water and sediment in the cargo loaded at the STS point, and (b) the demurrage provision in the sale contract between Fal Oil and Petronas operated as a contract of indemnity (that is in respect of liability, if any, on Fal Oil’s part under its charterparty), rather than as a free-standing provision. Petronas appeal in respect of (a), while Fal Oil appeal in respect of (b).[4]The water discovered on discharge of the Devon included some fresh water arising from leaking coils on the Devon. But the bulk of the additional 1500 tons consisted of sea-water. Its discovery, in the absence of any known casualty or identifiable incident, represented a most unusual event. It becomes even more unusual, once it appears that the total quantities of liquid loaded on the Centaur at Yanbu and on the Devon at the STS point closely correspond, after making appropriate adjustments for cargo and atmospheric conditions, with the total quantity of liquid (including the 1500 tonnes) ascertained to have been in and discharged from the Devon’s cargo-carrying tanks at Singapore and Pasir Gudang. On this point, the judge accepted the evidence of Petronas’s expert, Mr Severn, who had calculated the following figures. The quantity metered by the terminal at Yanbu and used in the bill of lading by the shippers (Exxonmobil Sales and Supply Corporation) became after adjustment 75,484 tonnes. Saybolt as surveyors acting for Fal Oil at Yanbu had in fact calculated the Centaur’s ship’s loaded quantity, using the vessel’s ullage tables, and had arrived at a total which led them to give notice to the terminal of an apparent discrepancy of some 230.49 tonnes (a relatively small quantity). But Mr Severn evidently attached no significance to this, taking as accurate the quantity given by the meters (which were recorded as having been proved by the surveyor during the loading). Mr Severn’s adjusted quantity for cargo loaded on the Devon based on Saybolt’s measurements at the STS point gave a figure of 75,953.988 tonnes, that is some 469.988 tonnes more than loaded at Yanbu. But Mr Severn’s calculations for discharge at Singapore and Pasir Gudang gave a total adjusted tonnage only 41.973 tonnes in excess of the adjusted Yanbu load figure. Mr Severn considered that the reliable measurements were those at Yanbu and on discharge in the Far East. The judge was on that basis faced with a situation where any ingress of water must have been matched by an equivalent loss or discharge of fuel oil. Fal Oil ran a positive case, to the effect that this was likely to have occurred as a result of some accident on board the Devon, in particular as a result of defects in her piping or valves drawing or allowing the entry of water during discharge and/or an interchange of oil and ballast connected with recent modification to her ballast piping and cargo systems. But the judge did not accept any such theory, which also failed to explain where 1500 tonnes of oil had gone, and Fal Oil do not on this appeal pursue it. The judge ruled out the possibility of 1500 tonnes of oil being discharged into the sea at any stage, considering that this would have involved a major international incident. He concluded that there must have been a deliberate substitution of water for oil, but that it was impossible to say whether this occurred on the Centaur or the Devon. On that basis, he held that Petronas had failed to prove that the 1500 tonnes of sea-water had been loaded as cargo onto the Devon at the STS point.

Scope of permission to appeal

[5]Permission to appeal was granted by the single

judge in terms which (as amended) stated:

“1. The Defendant be granted permission to appeal with respect to the Judge’s approach to the assessment of the inferences to be drawn in respect of three possibilities, namely theft on board the DEVON, failure to transfer the full cargo to the DEVON/theft on board the CENTAUR, or contamination at Yanbu. 2. The Claimants be granted permission to appeal in relation to the demurrage issue. 3. The permission granted at paragraph 1 above be subject to the condition that no primary findings of fact are challenged.”
[6]This formulation led initially to some argument about what constituted a “primary” fact. The word may be capable of different meanings. For example it may signify at one extreme a judge’s conclusions on a decisive factual element in a cause of action or at another extreme no more than a judge’s finding based on direct oral or documentary evidence, in the light of which other inferences of fact might be drawn. It cannot mean the former in this case, since the decisive factual element in Petronas’s set off and counterclaim is the water content of the liquid loaded on the Devon at the STS point, which is clearly in issue in the light of paragraph 1 of the order granting permission to appeal. Equally, it cannot, in the present context, mean the judge’s findings that “The measurements and sampling at Yanbu …. may be relied on” (paragraph 22(2)) and that “At that time [on loading at Yanbu] the water content was not more than 0.2%” (paragraph 26(2)) and that in consequence either the water must have entered after the STS point or the Centaur must have off-loaded 1500 tonnes or have concealed 1500 tonnes of oil on-board (e.g. perhaps, in a ballast tank) from Mr Pantouvakis (the Saybolt surveyor) during the STS transfer. This is so, even though the judge described the first finding (in paragraph 22(2)) as a “hard” fact about which he felt “confident”. If all or any of those findings were regarded as “primary findings of fact”, not open to review on this appeal, the permission to appeal granted “with respect to the Judge’s approach to the assessment of and the inferences to be drawn in respect of …. contamination at Yanbu” would be deprived of content. Further, many of the submissions directed at the limitations of the sampling after loading at Yanbu were and are equally applicable to the sampling undertaken at the STS point, so that either the permission to appeal “with respect to the Judge’s approach to the assessment of and the inferences to be drawn in respect of …. failure to transfer the full cargo to the DEVON” would also be deprived of effect, or we would be in the odd position that we could take into account submissions about the limitations of sampling in relation to the STS point, but not take them into account in relation to the original sampling at Yanbu. In the result therefore, and rightly in my view, we heard substantial submissions about the extent to which any firm conclusions could be drawn, in the light of the sampling undertaken and the evidence as a whole, as to the water content of the cargo as loaded both at Yanbu and at the STS point.

The first issue (the 1,500 tonnes of water)

[7]It is Petronas’s case on this appeal that(i) once the judge rejected Fal Oil’s positive case regarding contamination on the Devon, he was bound to decide in favour of Petronas;(ii) that the judge overlooked the fact that it was always Petronas’s case that “either the water was loaded with the cargo at Yanbu or there must at some stage have been a removal of 1500 mt of water from the system”, and that this oversight led him wrongly to decide the case on the basis that there must necessarily have been a deliberate substitution of water for cargo;(iii) that it was not open to him, having regard to the way the case was conducted by Fal Oil, to decide the case on the basis that there was a possibility of oil having been deliberately replaced by water on the Devon, since this was the judge’s own theory not advanced by either party;(iv) that he should have stood back and concentrated on the probable condition and content of the cargo loaded on the Devon, looking at all the evidence overall; and(v) that there were sufficient evidential indications that the additional 1500 tonnes were already admixed in the cargo by the STS point for Petronas to be entitled to succeed, however the matter should be approached.[8]Fal Oil support the judge’s reasoning and decision. They also maintain that an alternative logical approach to the resolution of this case would be to follow the cargo through from loading at Yanbu to the STS point, and to dismiss the claim if it appeared that nothing untoward occurred as between Yanbu and the STS point.[9]I do not accept Fal Oil’s alternative approach. The logic of the commercial relationship points in an opposite direction. Petronas as STS buyers had no direct concern with the position prior to the STS point, and no possible access to the cargo until after the STS loadpoint. Their problem here is that the presence of an additional 1500 tonnes of water was not identified at that point. But it was identified on discharge in the Far East. It is entirely logical for Petronas to seek to work back to the STS point from discharge in the Far East, by excluding any incident on the Devon. If Petronas can thereby show a prima facie case that the water was present (despite Saybolt’s failure to detect it) at the STS point, an evidential onus arises on Fal Oil to adduce evidence to the contrary. At that point, Fal Oil are able to rely on the prior history of the cargo, including its loading at Yanbu, the shortness of the voyage from Yanbu to Port Sudan and the unlikelihood of any water having been loaded at Yanbu or having been exchanged for oil there or between Yanbu and Port Sudan. Once all the evidence was before the judge on such points, it was his task to look at it overall, and to decide what conclusions he could draw from it and as to the nature and water content of the cargo loaded on the Devon at the STS point.[10]Moving to Petronas’s submissions, I take first the suggestion that, once the judge rejected Fal Oil’s positive case of defects on the Devon, he was bound to conclude that the water was present in the cargo transferred at the STS transfer. This submission mirrors, and suffers from a similar defect to, Fal Oil’s alternative case considered in the previous paragraph. Merely because Fal Oil have not made out their positive case as to the cause of contamination, it does not follow that Petronas have made out their case that the cargo transferred at the STS point was already contaminated. It is open to a judge to conclude simply that he does not know what the position is. Although a judge will only reach such a conclusion rarely and reluctantly, it is necessary to recognise and to identify such cases where they exist: see Rhesa Shipping Co. SA v. Edmunds (The Popi M) [1985] AC 948.[11]I turn therefore to the second criticism which I have identified, namely that the judge failed to consider the possibility of contamination at Yanbu, and that this led him to overlook the possibility of innocent contamination, rather than some deliberate switch. The quotation at paragraph 7(ii) above comes from paragraph 17 of Petronas’s closing written submissions to the judge. The submissions went on to ask where “the missing oil” went, and to give the answer in paragraph 21:
“It either remained on board CENTAUR or on board DEVON”
. However, in the course of 95 paragraphs of submissions there is only one further paragraph, numbered 68, which touches on the possibility of sea-water at Yanbu, in these terms:
“Some doubt may be cast on the validity of the inspections of CENTAUR at Yanbu. The entire operation was conducted in 30 minutes, which Mr Severn does not consider long enough to obtain a really coherent set of data. He does not believe that the surveyor would have had time to detect any free water had it been there.”
This is also directed to the presence of water in the cargo on the vessel, albeit immediately after loading, rather than to the loading of water as part of the cargo.[12]During closing oral submissions at trial, which were limited to one hour on each side, counsel for Petronas did not explicitly address any possibility that water had been loaded as part of the cargo. That does not mean that it was abandoned, but it was clearly not a possibility at the forefront of Petronas’s case by this stage. In his first report, Mr Severn had expressed the view that “the only possibilities appear to be during loading at Yanbu (onto Centaur), during the short voyage from Yanbu to Port Sudan or during the STS from Centaur to Devon” (paragraph 4.3). During cross-examination, Mr Severn confirmed that one could eliminate the ingress of seawater into the shore tanks at Yanbu. He repeated his doubt about the brevity and therefore quality of the post-loading sampling of the Centaur’s 12 tanks at Yanbu, and buttressed it with reference to apparent inconsistency between Saybolt’s ballast report reporting the Centaur as having 19,000 tonnes before loading (and 2,700 tonnes after loading) at Yanbu, and other Saybolt documents suggesting that she “arrived” for loading at Yanbu with 23,100 tonnes ballast. But he agreed that he was unable to make any (positive) statement that the Centaur’s cargo contained 1,700 tonnes of water when she sailed from Yanbu. The questioning and answers continued:
“Q. Can we also take it that there is no basis for saying that there had been any swap over of oil for water at that stage? A. I think that is a fair conclusion, yes. Q. Let us just look at paragraph 4 of your witness statement …. You say: “Accordingly, the only possibilities appear to be during loading at Yanbu (onto Centaur), during the short voyage from Yanbu to Port Sudan or during the STS from Centaur to Devon”
. Those are the only possibilities you have set out. On what we have been discussing so far, can we therefore discount the first of these, namely loading at Yanbu? A. Yes, I think I did that in my report anyway, where I was coming to the STS. Q. Right, let us look at the second and third possibilities, shall we? ….” “Accordingly, the only possibilities appear to be during loading at Yanbu (onto Centaur), during the short voyage from Yanbu to Port Sudan or during the STS from Centaur to Devon”. Q. Right, let us look at the second and third possibilities, shall we? ….”[13]Mr Males QC for Petronas submits that Mr Severn cannot have meant to abandon a possibility of contamination by water during loading at Yanbu expressed in Mr Severn’s own report to which Mr Severn referred in his answer. Mr Millett was prepared to accept that Mr Severn cannot have been intending to say that there was no possibility at all of such contamination at that stage, but was saying that it could be discounted. However, two matters are clear. First, no-one was by the end of the trial suggesting that there was any serious possibility that the terminal could or would have loaded out of the shore tanks a cargo including something approaching 1,700 tonnes of water. Second, no-one had advanced any positive evidence that such water entered the Centaur during her loading or any positive explanation, if it did so, for what happened to the equivalent quantity of oil metered by the terminal as having been loaded. In reality, any suggestion that the water came on board during the course of loading at Yanbu would appear to have raised the same conundrum as the parties and the judge faced when trying to explain or identify the time of any subsequent admixture of water with the cargo. This conundrum arose unavoidably, once Mr Severn demonstrated that the quantity of liquid remained unchanged from loading on the Centaur at Yanbu to discharge in the Far East. Further, having established this equivalence of liquid quantities throughout the period from loading at Yanbu to the discharge in the Far East, Petronas could not and did not seek to pursue any alternative hypothesis before us.[14]In relation to the judge’s conclusion that there must at some stage have been a deliberate substitution of water for oil, Mr Males emphasises the authorities in which courts have stressed that the more serious the conduct, particularly if it is criminal in nature, the weightier the evidence that will be required to prove it on the balance of probabilities. Accordingly, he submits, the judge should have looked first for any other possible explanation, rather than conclude that there had been a deliberate substitution of water for cargo by someone, even though he could not and did not claim to identify by whom. I for my part certainly see the potential attraction of an alternative explanation of the facts, if one could viably be suggested - for example that which would exist if it could be suggested that the cargo loaded on the Centaur at Yanbu consisted of no more than about 83,500 tonnes of oil (with a normal water/sediment content of up to about 200 tonnes) to which became added about 1500 tonnes of seawater, owing to some unknown defect in the pipes and/or sea valves of the Centaur. But such an explanation simply does not square with the terminal’s metered quantity of 85,000 tonnes, which was used as the basis of the bill of lading quantity by Exxonmobil (and presumably therefore billed to and paid for by Fal Oil – if the contrary were not so, then Fal Oil would have been obliged to disclose the relevant documents). Mr Severn never suggested any possibility of contamination of oil prior to the terminal’s meters, and he explicitly accepted the accuracy of the Yanbu terminal metered quantity. Petronas in their closing submissions summarised their position as follows (paragraph 14):
“Mr Severn’s submissions are accepted by Fal, and are beyond doubt. The analysis is detailed and careful. It is based upon a comparison between b/l figures and shore-side discharge figures but it also takes into account matters such as: (a) the terminal loading quantities at Yanbu (which were metered); ….” (a) the terminal loading quantities at Yanbu (which were metered); ….”
[15]Accordingly, while it is true that the judge did not direct himself specifically to loading at Yanbu as a possible cause of contamination, and I would accept that it remained in the arena at the end of the trial despite Mr Severn’s answer in cross-examination, I cannot see how it would have altered the fundamental parameters of the conundrum with which he had to wrestle. It may be also because this was generally understood at trial that loading at Yanbu received so little attention in the appellants’ final submissions.[16]This brings me to the third main criticism that I have identified as being made by the appellants, namely that it was not open to the judge, having regard to the way the case was conducted by Fal Oil, to decide the case on the basis that there was a possibility of oil having been deliberately replaced by water on the Devon, since this was the judge’s own theory not advanced by either party. It is true that deliberate substitution of water for oil by the Devon, or even a deliberate covering up by the Devon of some loss of oil by pumping in water, was never part of Fal Oil’s positive case. This is not a case where it was or could be suggested that any such substitution occurred because the carrying vessel stole some of the cargo for the purpose of using it as fuel oil. There was no motive for the crew to do this even if it had been feasible - both the Centaur and the Devon were on time charter to an associated company of Fal Oil, which would have been responsible for bunkers. But in any event the oil cargo loaded at Yanbu was too viscous to be used as fuel oil, at least without admixture with a lighter cargo which neither vessel had available. Substitution of 1,500 tonnes of cargo would therefore involve either concealment elsewhere on board or off-loading.[17]Mr Severn, in giving evidence, pointed out that the attention given to the Devon in the Far East would have ensured that, if 1,500 tonnes of oil had remained on board in any tank, including a ballast tank, that would have been spotted. He suggested that this would not necessarily have been the case if such a quantity remained on board the Centaur after the STS transfer. He pointed out that, although the Centaur left Yanbu with 2,700 tonnes ballast, she was recorded as arriving at the STS point with only 70 tonnes ballast.

The judge then asked him:

“Q. From your experience, let me be quite blunt, even if there be no evidence, if 1,500 tonnes goes walkabout, is it possible that it was put onto the Devon and pumped off into some other vessel or some other arrangement so that it gets rid of its 1,500 tonnes? A. If we are going to be blunt, yes, of course. There are two vessels involved. It would appear that one of those two vessels is somehow responsible, possibly covering up an incident involving leakage and/or loss of oil. Of course, I cannot eliminate either vessel as a possible candidate. I would qualify that by saying that the whole incident would need to have to be deliberate from the outset as opposed to an accidental contamination because the operations of each vessel was such, had Devon done the dirty deed, I believe an accidental leakage would have been detected. Q. Could it be done in such a way as to fool a reasonably competent cargo inspector …..? A. It is the case in my experience that I have been on board vessels where there has been a deliberate attempt to conceal cargo. Yes, that could be concealed from a routine cargo survey, of course. I do not believe it would escape the attention of the investigative surveyors who went on board in Singapore. Again, I have no evidence to suggest that is the case.”
[18]Mr Males submits that the judge’s question and Mr Severn’s answer relating to the possibility of substitution on the Devon, involving either off-loading of cargo during the voyage or its concealment during discharging, were isolated references, which went unnoticed and were not followed up in closing submissions. Mr Millett points out that Mr Akka, counsel for Petronas below, submitted in his closing submissions that, given the lack of evidence relating to the Centaur, “it is entirely possible that there could have been a transfer of oil and water (deliberate or otherwise) and that the oil could have remained on board her” (paragraph 88). But this relates to the Centaur, not the Devon. Mr Millett accepts that a possibility of deliberate substitution or a deliberate cover up on board the Devon was not “at the forefront” of his closing submissions. However, his written closing submissions (paragraph 6) did refer to the first question and the answer by Mr Severn quoted in the previous paragraph, but went on to acknowledge that Mr Severn had discounted the possibility that the oil could have been concealed from the inspectors who looked so thoroughly all over the Devon during her discharge. Neither Mr Severn nor Fal Oil’s closing submissions raised as a possibility off-loading by the Devon onto another vessel before arrival at Singapore. Mr Males submits that there was no suggestion by Fal Oil of this, and no examination of it, as a real possibility at the trial. Had there been, evidence from the master and crew of the Devon would have been material about her activities and movements, although Fal Oil, as voyage charterers from an associated company which had time-chartered that vessel, would have been better placed to adduce such information.[19]Mr Males submits that the judge was in these circumstances wrong, after concluding in paragraph 22 of his judgment that there must have been substitution of water for oil at some stage, to go on to say that “If it occurred on the DEVON, then the crew must have off-loaded 1500 tons of fuel to a third party en route to Singapore and topped up the cargo tanks with an equivalent amount of sea-water”. He should not only have discounted (as he evidently did) any suggestion of concealment of cargo on board at the discharge port(s), but should also have put aside any such thought of substitution involving discharge to another vessel en route. Before us, Mr Millett sought to maintain such a possibility on the basis that the Devon stopped for less than a day in the Indian Ocean. According to her radio report this was for the familiar reason that she needed to repair or replace a broken cylinder liner. Mr Millett suggested, for the first time on this appeal, that this might be an occasion when a discharge of 1,500 tonnes took place. There is nothing to commend this new and entirely uninvestigated suggestion. There is no reason to doubt the vessel’s explanation, and, if she was going to steal cargo, one might have thought it unlikely that she would announce any stoppage at all. Still more cogently, we were told that the vessel was at the time 300 miles off the south west coast of India, hardly a plausible place for a transhipment of cargo.[20]In my view, there is therefore force in the points made by Mr Males in this area, but it relates more to the question what if any weight or degree of likelihood may be ascribed to the possibility of deliberate substitution by the crew of the Devon than to whether it was open to the judge to consider such substitution as a possibility at all. The possibility of a deliberate substitution or cover up at some stage was one that was being canvassed generally. Petronas’s suggestion was that this could have occurred on the Centaur, despite the shortness of her voyage. The judge not surprisingly raised with Mr Severn the possibility that it might have occurred on the Devon’s longer voyage. Neither party could suggest any positive motive for such substitution on either vessel (other than the simple, but on the face of it more than a little implausible, motive of theft). Once the judge’s suggestion was made and answered, it was in the arena as a possibility, and was touched on, albeit briefly, in Fal Oil’s closing written submissions. The complaint thus becomes whether the judge gave appropriate weight to the possibilities before him, when concluding that Petronas had not made out their case.[21]That brings me to the criticisms that the judge should have stood back and concentrated on the probable condition and content of the cargo loaded on the Devon, looking at all the evidence overall, and that there were sufficient evidential indications that the additional 1,500 tonnes were already admixed in the cargo by the STS point for Petronas to be entitled to succeed, however the matter should be approached. Mr Males makes in this connection a series of submissions: i) The information about the cargo loaded at Yanbu and transferred at the STS point and the information about the Centaur is exiguous and much more questionable than the information about the cargo’s condition at the time of discharging in the Far East. Contamination during discharging in the Far East can be and was excluded as a serious possibility. ii) Even though the mechanism cannot be precisely identified, the evidence did not exclude either of two possibilities:(a) the loading with the cargo at Yanbu of excess water,(b) the loss from the cargo of 1,500 tonnes of oil and/or the accidental or deliberate substitution of an equivalent quantity of water on the Centaur - accompanied in the latter case (because of the absence of any suggestion or likelihood that such a quantity of oil could have been discharged unnoticed into the sea) by concealment of the lost oil in a ballast tank (probably the forepeak). iii) Substitution of water for oil during the Devon’s voyage, assuming that it was in the arena, was highly implausible. It would have involved an illicit operation on passage, which had never been positively suggested or investigated and which could not have occurred without widespread crew complicity. Any accidental loss of oil on the Devon could be and was excluded because this could not have occurred unnoticed. iv) Any deliberate substitution of water for oil at any point would have involved someone committing a serious criminal offence. But the court’s task was not to identify any such offence or the offender. It was to decide on the balance of probabilities whether the water was present in the cargo loaded at the STS point. In that connection, the court was entitled to take into account the extreme improbability of any loss of oil or substitution of water for oil after the STS point, and simply to conclude that the water was in probability present at the STS point, even if it remained a complete mystery how it got there.[22]In relation to point (i), Mr Males stressed certain further matters: (a). The only evidence of the water content of the cargo after loading at Yanbu derived from certificates of quality relating to the shore tanks on 20th and 22nd February 2001 (i.e. five and three days before loading), showing a 0.2% content, and composite samples taken from the Centaur after loading which, on analysis in Singapore on 15th-18th May 2001, showed the water content in individual tanks ranging between 0.15% and 0.25%. Mr Males reminded us of Mr Severn’s evidence that these samples had all been collected from some 11 tanks in 30 minutes and that this threw doubt on their likely accuracy. He further reminded us of the judge’s finding that:
“It is difficult to detect water in this type of fuel oil by carrying out running samples ….. Taking a sample in a ship’s cargo tank involves judgment and experience and a cargo inspector, acting competently may not detect water that is there. This type of fuel oil makes it difficult to detect water using standard industry approved equipment. ….” (paragraph 22(3)) I note in parenthesis that this finding is amply borne out by the history of the Devon in the Far East. On arrival at Singapore, the first survey of the Devon’s tanks failed to detect any significant water at all, using Mr Pantouvakis’s method, even though it is now effectively common ground and certainly clear that the cargo contained as much as 1,700 tonnes. The water had by then of course had ample time to sink to the bottom, where it is to be presumed that it was not detected. The problems of detection to which the judge was referring are associated with the difficulties of identifying both water mixed with oil and free water at the bottom of tanks. (b). Not only was there the unexplained discrepancy in the reported ballast on the Centaur’s arrival at Yanbu, there was also the odd difference in the figures reported for her ballast on sailing from Yanbu (2,700 tonnes) and on arrival at the STS point after the day’s voyage to Port Sudan (70 tonnes). Mr Millett’s response to this was that the Centaur could well have discharged ballast during her short voyage. He commented that the Devon discharged 4,500 tonnes of ballast during her (admittedly much longer) voyage to the Far East. Mr Males also pointed out that the Centaur’s drafts as reported by Mr Pantouvakis at the STS point were identical with those reported after loading at Yanbu. However, the judge accepted that Mr Pantouvakis simply relied on what the crew told him in this respect, there being a 2 metre swell, and did not identify any trim change that would have resulted from any discharge of ballast. (c). Mr Males also focussed on the surveys undertaken by Mr Pantouvakis at the STS point. Mr Pantouvakis attended on board the Centaur, measured the ullage figures and calculated the cargo quantity. His ullage report reported that he had found no free water. But Mr Males relies on (1) the judge’s finding quoted in (a) above about the difficulty of detecting water in this type of fuel oil and by carrying out running samples, and (2) the associated fact that Mr Pantouvakis did not have with him any bottom sampler, and therefore had no reliable means of detecting any free water in the bottom 30 cms of the vessel’s tanks. (d). Petronas suggested that Mr Pantouvakis might have overlooked a quantity of oil, remaining on board the Centaur, if it had been put in the forepeak ballast tank. The judge said that he thought this unlikely, although he could not rule it out as a possibility. He indicated that he had formed a favourable view of Mr Pantouvakis as a hard working surveyor, who had “seen all sorts of attempts to pull the wool over his eyes”
. He said that he would be surprised if Mr Pantouvakis did not look into the forepeak tank, or check the cargo sounding rod which would have satisfied him that her cargo tanks were empty. Mr Males challenges this on the ground that Mr Pantouvakis acknowledged in oral evidence that, contrary to the impression given by his witness statement, he had not actually checked the tanks himself, but was a witness to the crew doing so. Nevertheless, Mr Pantouvakis also said, both in his statement and orally (albeit in response to a very leading question by Mr Millett in re-examination), that there was no smell of oil and no visible trace of oil on the tape on sounding any ballast tank. The judge accepted this evidence as showing that the position was likely to have been as he found it, and it seems to me that it was open to him to do so. This is so, independently of the mild support which it also receives from the fact that the documents relating to the Centaur’s return to Yanbu after the STS transfer do not show the presence of any oil remaining on board her at that point. (e). There was no water in the Devon’s cargo tanks before she loaded. During her loading, Mr Pantouvakis took samples from the Devon’s manifold, and these were collected into canisters, one of which was retained by the Centaur, the other given to Fal Oil’s agent at Port Sudan. Both were subsequently lost. The judge declined to draw any adverse inference from this, and it has not been suggested that we should do so. On completion of the loading, Mr Pantouvakis took samples from the Devon’s tanks. Clause 2 of the variation fax dated 13th February 2001 required the independent surveyor to take or test the quality of each of the Devon’s individual tanks. But Fal Oil’s instructions to Saybolt, copied to Petronas, were simply for such a surveyor to conduct a quantity and quality survey “as usual”. Mr Pantouvakis took samples from each of the Devon’s tanks, but then composited them and split the resulting bulk into four parts. The judge found that this was in accordance with industry practice on a STS transfer. Again, however, Mr Pantouvakis did not have a bottom sampler. Further, the judge found that compositing samples (measuring out the samples from individual tanks in the right proportions) and then splitting them into four equally representative parts “on board a vessel in a swell cannot be easy and there is obvious scope for miscalculation and error”. On testing in Singapore two of these composite samples gave a reading of only 0.25% water (i.e. about 212.50 tonnes). (f). During the process of checking the Devon’s tanks, Mr Pantouvakis heard frequent “beep” sounds, indicating the presence of water at different levels, which led him to give notice to both vessels as follows (and subsequently led to the Devon giving notice of protest to the Centaur):
“In tk No 1C, 2P, 2S 5P traces of water In tk No 2C approximate 2cm of traces of water of approximate 13.7m3 In tk No 3C approximate 5cm of traces of water of approximate 55.0m3 In tk No 4C approximate 14cm of traces of water of approximate 96.2m3 In tk No 5C approximate 9cm of traces of water of approximate 58.4m3 In tk No 5S approximate 3cm of traces of water of approximate 4.8m3 In tk No sl P approximate 105 cm of traces of water or approximate 30.8m3”
In tk sl S approximate 2cm of traces of water of approximate 12.1m3 Because of the nature of the cargo (High Density C.9995) and because it was not possible to verify exactly the total amount of the water the final amount of water will be verified at Discharging Port when the free water will be settled.” In tk No 2C approximate 2cm of traces of water of approximate 13.7m3 In tk No 3C approximate 5cm of traces of water of approximate 55.0m3 In tk No 4C approximate 14cm of traces of water of approximate 96.2m3 In tk No 5C approximate 9cm of traces of water of approximate 58.4m3 In tk No 5S approximate 3cm of traces of water of approximate 4.8m3 In tk No sl P approximate 105 cm of traces of water or approximate 30.8m3” In tk sl S approximate 2cm of traces of water of approximate 12.1m3 Mr Pantouvakis’s preliminary estimation of quantities in the notice thus totalled some 272 tonnes, consisting according to his evidence of both free water and water and oil mixed. Petronas rely on Mr Pantouvakis’s final paragraph, which is, they suggest, consistent with a possibility of larger quantities. Fal Oil point out in response that Mr Pantouvakis said that “this quantity is the maximum of traces I found and because I was not sure that it is 100% of water, can be less or can be more. But most probably can be less, but I am not sure.” The judge seems to have regarded the water identified by Mr Pantouvakis as additional to the 0.2% measured at Yanbu. That assumption, which favoured Petronas, cannot be correct. 272 tonnes is only about 0.32%. Even on the judge’s basis, the total percentage (based on a water content of 272 plus 170 tonnes) would only be about 0.52%. The analysis of Mr Pantouvakis’s running samples revealed a water content of 0.3%. The judge pointed out that all these figures were well within specification and that the running samples did not show an “enormous quantity of water”. Petronas submits that this misses the point, which is that there had been (a) sufficient water on loading the Devon (as distinct from the Centaur) to cause Mr Pantouvakis to give the notice he did and (b) that the volume of such water as noted by Mr Pantouvakis had increased by an estimated 100 tonnes over that which would follow from the percentage of 0.2% noted in the tanks and on loading of the Centaur at Yanbu. These are fair points. But there is, on the other hand, nothing in Mr Pantouvakis’s notice and no positive evidence to suggest quantities beginning to approach those discovered in Singapore. Further, although the water content suggested by Mr Pantouvakis’s notice and samples (0.32%) is somewhat greater than the water content of the cargo as reportedly loaded at Yanbu (0.2%), the evidence does not go so far as to establish that such a difference is itself a pointer to some untoward incident or influx of water occurring on the Centaur or during the STS transfer, rather than, for example, the result of the inherent inaccuracies in measuring water and sediment. There are other problems about understanding how as much as 1,500 tonnes of water could have been missed on the STS transfer. The in- and outlets for the cargo tanks of tankers such as the Centaur and Devon are at the tank bottoms. Mr Severn’s principal explanation as to how and why Mr Pantouvakis could have missed the transfer of as much as 1,500 tonnes of extra water at the STS point was that this water must have come on board at a late stage in the transfer, so as to find itself in the bottom of the Devon, out of reach of detection by Mr Pantouvakis in the absence of any bottom sampler. But this involves the difficulty that, if the water came on board at any time after loading at Yanbu, it would have been at the bottom of the Centaur’s cargo tanks (where the outlet/inlet pipes are) and would have been discharged first onto the Devon, and so (on Mr Severn’s own evidence) have been likely to be detected even by the sampling which Mr Pantouvakis undertook on the Devon.

In tk sl S approximate 2cm of traces of water of approximate 12.1m3

[23]Standing back from the detail, I agree that the critical question, to which the judge had to direct his mind, was whether the cargo transferred to the Devon at the STS point contained some 1,700 tonnes of water. The judge thought that the answer to that question depended upon whether Petronas could show that the presence of this water was more probably due to deliberate substitution of water for oil on the Centaur as opposed to the Devon. I would agree with Mr Males that the judge went too far in considering that he could close off all other options and make the ultimate issue depend on so stark a choice. This is particularly so, when(a) contamination during loading on the Centaur remained in the arena and(b) deliberate substitution of water for oil on the Devon was never positively suggested by Fal Oil and was touched only marginally in evidence.[24]However, it is still necessary to consider whether Petronas have established as a matter of probability that the water was present and was loaded on the Devon during the STS transfer. The suggestion that the water may have come on board during the loading of the Centaur at Yanbu faces the difficulty that the mechanism is entirely unexplained and on its face implausible, and that (perhaps for this reason) the possibility received no real attention or emphasis at trial. The possibility of loss of oil to sea from either the Centaur or the Devon is at best very remote, so much so that neither party suggested it and the judge disregarded it. The possibility of deliberate substitution (on either vessel) is essentially speculative. The judge examined the evidence about the Centaur and the STS transfer and found nothing to enable him to conclude that the water was probably on board the Centaur or transferred to the Devon. Although there are some oddities about the Centaur’s ballast, they are neither inherently suspicious nor capable of explaining the appearance of the water in lieu of oil. Although the judge could not rule it out as a possibility, he thought it unlikely that oil could have been concealed from Mr Pantouvakis on the Centaur, and one must again ask whether her owners and crew would have chanced so risky an exercise and why. It is true that concealment on the Devon in Singapore is even less likely, indeed it can be positively ruled out in view of the exhaustive surveys there undertaken. But merely because some scenarios can be completely excluded does not mean that any possibility that remains must “probably” be what happened. That would be to accept the fallacy which was exposed in The Popi M. The high point of Petronas’s case was and is Mr Pantouvakis’s conclusions and notice of protest regarding water at the STS point, which Mr Males described as “interesting”. The evidence is capable of arousing the suspicion that Mr Pantouvakis, using inadequate, although standard methods of sampling, identified a small part of the huge volume of water which was later discovered in the Far East. But suspicion is not enough. It must be possible to say as a matter of probability that the 1,500 tonnes was already present. But from where it could have come, and how, one cannot begin to understand, bearing in mind the unlikelihood of its having been either loaded with the cargo in Yanbu or substituted for oil during the 24 hours voyage since Yanbu. On the other hand, of course, the water got into the cargo, and oil apparently loaded went apparently missing, at some point before the Devon discharged in the Far East. At the end of the day, although by somewhat different reasoning, I find myself driven to the same conclusion as the judge, that Petronas have simply failed to prove that the water was loaded onto the Devon by or at the STS point. The case is a mystery. Accepting Mr Severn’s figures, as we must, one suspects that there must have been some incident or event at some stage about which someone must know, but what it was and when and how it could account for the apparent disappearance of 1,500 tonnes of oil and the appearance of an additional 1,500 tonnes of water is impossible to say, at least as a matter of probability. What is however critical is that it is impossible to say as a matter of probability that the cargo loaded on the Devon at the STS point was some 1,500 tonnes short of oil. I would therefore uphold the judge’s decision in Fal Oil’s favour on the first issue before him.

The second issue (the nature of liability for demurrage)

[25]The sale contract dated 29th November 2000 contained the following provisions:
“10.LAYTIME: LAYTIME ALLOWED SHALL BE A TOTAL OF 36 HOURS SHINC TO COMMENCE 6 HRS AFTER NOR IS TENDERED OR UPON BERTHING WHICHEVER IS THE EARLIER AND TIME SHALL CEASE TO COUNT AT DISCONNECTION OF HOSES. 11.DEMURRAGE: AS PER CHARTERPARTY PER DAY PRO RATA. ….. 15. OTHER TERMS AND CONDITIONS: WHERE NOT IN CONFLICT WITH FOREGOING, INCOTERMS 2000 WITH LATEST AMENDMENTS FOR CNF SALES TO APPLY.”
Incoterms for cost and freight sales provide:
“A3 Contracts of carriage and insurance a) Contract of carriage The seller must contract on usual terms at his own expense for the carriage of the goods to the named port of destination by the usual route in a seagoing vessel …. of the type usually used for the transport of goods of the contract description. b) Contract of insurance No obligation.”
No obligation.”[26]There is no indication that Petronas knew anything about Fal Oil’s chartering arrangements or intentions when the sale contract was entered into. The sale contract contemplates that there would be a relevant voyage charter, on usual terms. As at the time that the sale contract was made, it must have been likely that (as was in fact the case) no such voyage charter had yet been entered into. Neither party could therefore know its precise terms regarding laytime or demurrage, save that, under Incoterms, they must be “usual”. In the event, Fal Oil chartered in the Devon from an associated company, Fal Shipping Co. Ltd (which was it appears time-charterer under some unknown charterparty), for the first voyage under a voyage charter on the ASBATANKVOY form dated 22nd February 2001. The voyage charter provided:
“PART I …. B. Laydays: 25-26 FEBRUARY 2001 Commencing 0600 HRS Cancelling 1600 HRS …... H. Total laytime in Running Hours 72 HOURS SHINC I. Demurrage per day USD 18,000 P.D.P.R ….. PART II 5. Laytime shall not commence before the date stipulated in Part I, except with the Charterer’s sanction. …. 6. NOTICE OF READINESS. Upon arrival at customary anchorage at each port of loading or discharge, the Master or his agent shall give the Charterer or his agent notice ….. that the vessel is ready to load or discharge cargo, berth or no berth, and laytime, as hereinafter provided, shall commence upon the expiration of six (6) hours after receipt of such notice, or upon the vessel’s arrival in berth .…. whichever first occurs. However, where delay is caused to vessel getting into berth after giving notice of readiness for any reason over which Charterer has no control, such delay shall not count as used laytime. 7. HOURS FOR LOADING AND DISCHARGING. The number of running hours specified as laytime in Part I shall be permitted the Charterer as laytime for loading and discharging cargo, but any delay due to the Vessel’s condition or breakdown or inability of the Vessel’s facilities to load or discharge cargo within the time allowed shall not count as used laytime. ….. Time consumed by the vessel in moving from loading or discharge port anchorage to her loading or discharge berth, discharging ballast water or slops, will not count as used laytime. 8. DEMURRAGE Charterers shall pay demurrage per running hour and pro rate for a part thereof at the rate specified in Part I for all time that loading and discharging and used laytime as elsewhere herein provided exceeds the allowed laytime elsewhere herein provided. If, however, demurrage shall be incurred as ports of loading and/or discharge by reason of fire, explosion, storm or by a strike, lockout, stoppage or restraint of labour or by breakdown of machinery or equipment in or about the plant of the Charterer, supplier, shipper or consignee of the cargo, the rate of demurrage shall be reduced one-half of the amount stated in Part I per running hour or pro rata for part of an hour for demurrage so incurred.”
Commencing 0600 HRS Cancelling 1600 HRS H. Total laytime in Running Hours 72 HOURS SHINC I. Demurrage per day USD 18,000 P.D.P.R

PART II

[27]The issue presented to the judge and now to us is whether the demurrage provisions in the sale contract operate by way of indemnity or give rise to independent obligations. It is not easy at this stage to know the practical implications of this issue. It seems likely that most if not all of the delay in discharging the cargo was due to the discovery that it contained some 1,700 tonnes of water. If that was present in the cargo loaded onto the Devon at the STS transfer point, then Petronas would appear to have a good defence to any claim for demurrage in respect of the resulting delay. If it was not, then the presence of the water must have been the Devon’s fault. Petronas’s case on that factual hypotheses is that, if the sale contract demurrage provisions operate as an indemnity, Fal Oil can have no liability to the Devon against which Fal Oil can require any indemnity. But before us Mr Males put the matter more widely, suggesting that, even if the sale contract demurrage provisions operated as independent obligations, Petronas would still have a defence to any demurrage claim by Fal Oil, since, he submitted, fault of the carrying vessel delaying discharge was a general implied exception to the running of laytime and the obligation to pay demurrage. (Fal Oil would then have a similar defence vis-à-vis Fal Shipping Co. Ltd. under the, on this basis, independent charter.) If this be right, then the nature of the present sale contract demurrage provisions may be academic. Nonetheless, the preliminary issue requires us to determine their nature.[28]We were referred to authority going back to Suzuki & Co. v. Companhia. Mercantile Internacional (1921) 9 Ll.L.R. 171 (CA) on appeal from Shearman J (1921) 8 Ll.L.R. 174. The judgments in that case in fact distinguished the Privy Council case of Houlder Bros. v. The Commissioners of Public Works [1908] AC 276, which it is also useful to mention. In Houlder Bros. Lord Atkinson, giving the Privy Council’s advice, referred to Lord Blackburn’s judgment in Ireland v. Livingston LR 5 HL 395, 407, according to which, depending on the nature of the particular contract, the seller may either act as agent for the buyer in arranging carriage (e.g. taking a commission) or may commit himself to a fixed price (in which case the seller will contract with the carrier for his own account). The contract in Houlder Bros. fell into the second category, which is (now at any rate) the familiar position. The present sale contract clearly also falls into the second category. Addressing laytime and demurrage provisions in such a sale contract, Lord Atkinson said at p.291:
“There is, however, no rule of law that the vendor in a c.i.f. contract may not secure for himself a profit under a demurrage clause contained in it. Neither is there any indisputable presumption of law that the parties to such a contract did not intend that he should receive such a profit. To use the words of Lord Blackburn in Calcutta and Burmah Steam Navigation Co. v. de Mattos (32) LJ (QB) 322, 328, in such contracts “there is no rule of law …. preventing the parties making any bargains they please.”
[29]The argument that the laytime and demurrage provisions in Houlder Bros. were to be read as taking effect by way of indemnity was on any view extremely weak. As Bankes LJ pointed out in Suzuki, the sale contract in Houlder Bros. made no express reference at all to the charterparty. It provided for discharge immediately on arrival of the carrying vessels at their discharge ports at specified rates of tons per day (120 tons for “sailers” and 250 tons for steamers), failing which demurrage was to be paid at respectively 4d or 6d per vessel’s net registered ton per day. The code was on its face independent (and the buyers had also accepted as much in relation to prior contracts). But it is of interest to note Lord Atkinson’s further observation at pp.289-290 that: “It is, moreover, difficult to see how the principle laid down would work in practice”, in which connection he took as an example the situation of a charter of a sailing vessel providing for discharge at only 100 tons per day or for demurrage at only 3d per registered ton per day.[30]In Suzuki the terms of the contract dated 26th May 1919 appear most clearly from the first instance report:
“Shipment may be made in names other than that of Messrs. Suzuki & Co…..The rice to be discharged at Lisbon and/or Oporto at buyers’ option. Discharge to be at the rate of 500 tons per weather working day, Sundays and Government holidays excepted, time to count 24 hours after arrival of steamer at first port of discharge to be given before steamer’s arrival at Suez. Demurrage as per charter-party or freight agreement. The time taken by the steamer to shift from the first port to the second port of discharge not to count.”
The charter dated 1st June 1919 contained similar laytime provisions regarding discharge, and specified:
“Demurrage, if any, over and above the said laying days at £400 per day (or pro rata for part thereof)”
. In a separate clause it provided:
“15. Charterers’ responsibility to cease on steamer being loaded, provided the cargo is worth the freight, captain and owners having an absolute lien on the cargo for all freight, dead freight and demurrage which they are hereby bound to exercise.”
Shearman J appears to have viewed the relationship with regard to carriage as within the first of Lord Blackburn’s categories, saying at p.176:
“The sellers, acting, as it appears to me, as agents for the buyers, entered into a charterparty with the owners of the ship. The buyers agreed to pay demurrage. The point taken by the sellers is that, notwithstanding the fact that there is a liability on the buyers to pay the shipowner, I ought, under the terms of the contract, to follow the decision in the Houlder case. I think that the decision in that case is sound, but the real question I have to decide is what is the kind of contract between these parties. It is not necessary to discuss the Houlder case, but I am of the opinion that the true meaning of the contract is that the parties will pay to the shipowner the demurrage due, and the only right of action which the charterers could have would be in respect of damages which, under the charterparty they could prove they had suffered by the defendants’ breach of contract. …..”
[31]The Court of Appeal did not expressly refer to or endorse Shearman J’s view of the relationship between the sellers and buyers, but said that the only question that arose was “one of construction as to whether the contract between the buyers and sellers, which refers to the question of demurrage, is a contract of indemnity or whether it is an obligation to pay irrespective altogether of what the sellers’ position or responsibilities may be, if in fact the vessel is kept on demurrage”. Bankes LJ, giving the judgment, went on:
“Of course, if in fact it is a contract of the latter description, it really provides for an addition to the price of the rice depending upon whether the ship is or is not delayed in its discharge; and one does not quite see why the parties should enter into such a contract. But it seems to me that the language, from the nature of things, points to a contract of indemnity rather than to a contract of the nature contended for by the appellants …..”
He distinguished Houlder Bros. as concerning a contract “of a very special nature”, where there had also been a prior course of dealing. “Of course, if in fact it is a contract of the latter description, it really provides for an addition to the price of the rice depending upon whether the ship is or is not delayed in its discharge; and one does not quite see why the parties should enter into such a contract. But it seems to me that the language, from the nature of things, points to a contract of indemnity rather than to a contract of the nature contended for by the appellants …..”[32]The reluctance of the Court of Appeal to decide in favour of the sellers/charterers in Suzuki is easy to understand. The idea of a CIF seller, who as charterer has the benefit of a cesser clause, recovering demurrage for which he has no possible exposure to the shipowners, is unappealing. The general words “Demurrage as per charter-party or freight agreement” lent themselves to a construction whereby liability for demurrage under the sale contract followed (and so “indemnified” the seller against) whatever was the liability (if any) of the sellers for demurrage under their charter. However, nothing in the Court’s brief reasoning expressly limits its scope to circumstances as stark as those where a cesser clause means that a seller claiming demurrage has itself no conceivable liability for demurrage, and it is difficult to see that there could be any basis for so construing wording as sparse as “demurrage as per charterparty or freight agreement”.[33]What is, for whatever reason, absent in the reasoning in Suzuki is any acknowledgement of the theoretical basis on which, and the good reasons why, parties may construct an independent scheme regarding demurrage. It is now well-established that it is a breach of contract to exceed the specified laytime, and demurrage is the agreed damages to be paid for delay in loading or discharging arising from any breach: see the speech of Lord Guest in Union of India v. Compania Naviera Aelous SA[1964] AC 868, 899, adopted by Lord Diplock in Dias Co. Nav. SA v. Louis Dreyfus Corpn. [1978] 1 WLR 261, 263 (quoted recently in Kronos Worldwide Ltd. v. Sempra Oil Trading SARL [2004] 1 Ll.R. 260, 264). In the context of a C&F sale contract, where the seller is not the shipowner, the underlying rationale of the inclusion of any laytime and demurrage provisions (whether on an independent or indemnity basis) is that the seller will have to arrange carriage on terms which may expose the seller to liability for demurrage to a shipowner or other third party. Parties may, however, find it simpler and more acceptable to agree and operate an independent scheme which means that, in the event that delay occurs, they will know precisely where they stand, rather than to contract on a basis which makes their rights inter se depend upon the rights and liabilities of one of them, and possible disputes under some actual or future contract with a third party.[34]If (as here) the rate of demurrage depends on whatever charterparty the seller makes, that represents only a small qualification on the parties’ initial certainty (itself mitigated by the fact that the rate must be “usual”). Once the charter is made, the rate is defined and the two contracts can operate independently. The reason for adopting a charterparty rate into an otherwise independent sale contract scheme (rather than making a best guesstimate of the future demurrage rate(s) to which the seller is likely to be exposed) is obvious. It is to ensure that the sale contract demurrage provisions approximate sensibly to those ultimately agreed. Future demurrage rates are inherently uncertain (especially in the case of a sale contract like the present for four deliveries over a significant future period). But making the seller’s and buyer’s rights and liabilities relating to demurrage depend not just upon the actual charter rate, but upon resolution of the position regarding demurrage as between the seller and a third party under the charter and/or other intermediate sale contract(s), introduces a quite different potential for uncertainty and dispute. As to the position which may (on the authority of Mallozzi v. Carapelli [1975] 1 Ll.R. 229 (Kerr J); affirmed [1976] 1 Ll.R. 407 to which I refer below) arise if a sale contract refers to a voyage charterparty rate and the seller never enters into such a charter (but for example uses a time-chartered vessel or his own vessel), that situation can hardly have been contemplated at the time of the sale contract; and, since it prevents any demurrage at all arising between seller and buyer (whether the sale contract demurrage provisions are viewed as operating independently or by way of indemnity), it represents on any view something of a windfall escape for the buyer (rather than a windfall bonus for the seller).[35]Further, although Bankes LJ in Suzuki reinforced his conclusion that the sale contract provisions operated by way of indemnity by reference to the “windfall” nature of an undue addition to the price which could otherwise result, the general law of penalties offers potential protection against sale contract provisions for demurrage which lack a reasonable basis as a genuine pre-estimate of the recovering party’s exposure. So, if a charter containing a cesser clause had already been agreed, or was likely to be arranged, arguments based on the law of penalties could clearly arise for consideration. In Suzuki one notes that the charter was made only six days after the sale contract, but the reports contain nothing to suggest either that the sale contract was made with the particular terms of the subsequent charter in mind or indeed that any argument based on the law of penalties was raised or considered. Bankes LJ’s statement that, if the sale contract provisions operated independently of any liability for demurrage, then they really provided for an addition to the price, might, possibly, be understood as indicating that there was no question of such provisions being a genuine pre-estimate of any loss that the seller was likely to suffer by way of demurrage under any charterparty. On the other hand, it is of the nature of a liquidated damages provision that it may still be a genuine pre-estimate and enforceable, even though, in the particular circumstances which happen to materialise, it would not be possible to show any actual loss.[36]I turn to subsequent authority, in which there is some recognition of the considerations mentioned in the last three paragraphs. Petronas rely on Mallozzi v. Carapelli SpA. The sale contract provisions there provided for laytime, by reference to a rate of discharge per day, and for “Demurrage/half despatch on loading at the rates indicated in the charter-party for buyer’s account”. However, the sellers chartered a carrying vessel on a time-chartered basis. Kerr J considered both Houlder Bros. and Suzuki and concluded that the sale contract before him was “virtually indistinguishable from the Suzuki case because of the vital feature that liability to demurrage is imposed by reference to a charter-party and not directly under the contract itself” (p.246). 43. He also said:
“On this basis the present contracts fall into the class whereby the buyers are only made liable for demurrage on an indemnity basis if the sellers are themselves liable for demurrage under a charter-party.”
He regarded Suzuki as recognising “that in c.i.f. contracts a provision of this kind can and should be construed as being in the nature of an indemnity”. He also attached some importance to the words “for buyers account” as connoting an indemnity.[37]In the Court of Appeal Mallozzi was decided on the much narrower ground that the sale contract only provided for demurrage at a charterparty rate, and there was no charterparty from which any rate could be derived. The Court, particularly the commercial judges on it (Megaw and Roskill LJJ), went out of their way to emphasise that they had not heard submissions in support of the proposition that the relevant clause was an indemnity clause, and that the Court of Appeal’s conclusion did not in any way rest on that basis or on the authority of Suzuki. The actual decision in Mallozzi does not therefore lend Petronas any support.[38]In R. Pagnan & Fratelli v. Finagrain Co. Com. Agricole et Financiere SA (The Adolf Leonhardt) [1986] 2 Ll.R. 395, the FOB sale contract between Pagnan as seller and Finagrain as buyer dated 17th January 1978 provided:
“Special conditions …. Time to count as per Centrocon charter party WIBON, WIPON, WIFPON. Demurrage/Despatch as per C/P. ……”
Finagrain sub-sold the goods under a contract dated 1st February 1978 with Exportkhleb, which was accordingly responsible for chartering in a carrying vessel to uplift the goods. Staughton J, agreeing with the GAFTA Board, considered (obiter) that the sale contract imposed an independent liability for demurrage on Pagnan, rather than a liability to indemnify Finagrain for any demurrage liability which Finagrain might have to its sub-buyer (Exportkhleb). He said:
“Issue (3) is whether the sellers have an independent obligation to pay demurrage to the buyers, or whether they are only obliged to indemnify the buyers against liability to V/O Exportkhleb. ….. My answer would be that the sellers have an independent obligation, as the Board of Appeal held. I do not find it surprising that a buyer should contract to receive demurrage at a different rate, or on different conditions, than those governing his liability to pay a shipowner or a sub-buyer. Normally one might perhaps expect the terms to be the same, as I said in Eurico S.p.A. v Phillip, [1986] 2 Lloyd’s Rep. 387, immediately before the argument in this case; but they may be different. What persuades me that an independent obligation was intended here is the reference in the sale contract to the Centrocon charter-party, scilicet in its printed form. Whatever terms might be agreed between the buyers and a shipowner, or their sub-buyers, it was all Lombard Street to a china orange that they would not be precisely the printed terms of the Centrocon form. The buyers had not, when they contracted with the sellers, concluded their sub-sale, at any rate in point of form; it makes good sense that they should bargain for an independent obligation in the terms of the printed form, if only as an approximation to what they might agree with their sub-buyers.”
The underlying thinking here is that it is not surprising to find a party, who may become liable to demurrage (either directly under a charterparty or indirectly by reason of a sub-contract), stipulating not for an indemnity, but for provisions of independent operation which approximate to (or represent a “genuine pre-estimate” of) what he anticipates is likely to be his own liability. That, as I have explained, fits with the concept of demurrage as liquidated damages: see paragraphs 33-35 above.[39]Then there is Ets. Soules et Cie v. Intertradex SA [1991] 1 Ll.R. 378. The sale contract contained both a laytime provision and an expressly stated rate of demurrage (US$3,000 per day pro rata with half despatch). Hobhouse J pointed out that there was no cross-reference to any charter, and so no question of the demurrage provision being drafted on some basis of indemnity. It appears that the buyers had conceded this. On appeal, there was an attempt to withdraw the concession, but Neill LJ expressed the view that the concession was rightly made (p.388) while Staughton LJ, without expressing any final opinion, saw good grounds for treating it as correct (p.385), mentioning considerations similar to those which he had identified in The Adolf Leonhardt.[40]In Gill & Duffus SA v. Rionda Futures Ltd. [1994] 2 Ll.R. 67, the sale contract contained detailed provisions regarding laytime and notice of readiness and went on:
“Despatch and demurrage at discharge to be for buyer’s account. Demurrage as per C/P half despatch. ….. Demurrage to be settled as incurred by buyers every 15 days.”
Clarke J was influenced by the detailed provisions regarding laytime and notice of readiness, and concluded that in that context the expression “demurrage as per C/P” meant no more than that the rate of demurrage in the relevant charter-party should be the rate of demurrage for the purpose of he contract of sale. After referring to Mallozzi, he said that, “whatever inference might be drawn from the use of the words “for buyer’s account” if they stood alone, here they do not. In this contract, …. there are both detailed provisions for the commencement and calculation of laytime and an express provision that demurrage was to be “settled as incurred by buyers every 15 days”. In these circumstances the parties cannot in my judgment have intended that the limit of the buyers’ obligations was whatever was paid by the buyers.” Thus in Clarke J’s view the terms of this particular sale contract provided, as a matter of construction, for an independent demurrage liability, not an indemnity.[41]The last case which it is necessary to mention in any detail is OK Petroleum AB v. Vitol Energy SA [1995] 2 Ll.R. 160, where two sale contracts each provided for laytime consisting of a specified number of hours SHINC, and (in the one case) for “Demurrage As per Charter Party” and (in the other) for “Demurrage. As per Charter Party rate, terms, and conditions pro rata for part cargo”. The two charters (entered into in each case after the relevant sale contract) incorporated standard terms, which included a time bar provision excluding charterers’ liability for demurrage unless a claim was notified within 90 days of discharge. Colman J ultimately concluded that the incorporation of demurrage as per charter party or as per charter party terms and conditions could not in any event embrace a collateral provision like a time bar. But in the course of his analysis he endorsed counsel’s concession that the demurrage provisions in the sale contracts should not be read as indemnity provisions. He gave as his reason the inconsistency between the sale contract and charterparty regimes, the latter aggregating load and discharge port laytime and the former allowing half the total charterparty laytime for discharging. (In that respect the case mirrors the present sale contract and charter.) But this reasoning seems to me vulnerable to the criticism that it defines the nature of the sale contract demurrage provision by reference to the nature of whatever charter the sellers made subsequent to entering into the sale contract. It is one thing to point out, as Staughton J did in The Adolf Leonhardt, that, since there could be no certainty what terms a future charter might contain, the parties might well by the sale contract have decided upon an independent scheme which involved a genuine pre-estimate of or approximation to the sort of charterparty terms regarding laytime and demurrage which they believed it likely would be available. It is another to define the nature of particular sale contract provisions by reference to whatever charter may actually happen to be made in future.[42]This examination of the authorities leads to conclusions, which I can summarise as follows: i) Provisions in a sale contract regarding laytime and demurrage should be approached without any pre-conceptions or presumption as to their likely nature. That follows from general principle, as well as from the passage quoted above from Houlder Bros. ii) The scope and effect of such provisions is a question of construction: see Suzuki. iii) The underlying rationale of any sale contract demurrage provision is that the receiving party may suffer loss under a charter or other third party contract. However, this is consistent with the provision operating either by way of indemnity or independently. An independent provision can, subject to the law on penalties, be justified as a genuine pre-estimate of the receiving party’s exposure. iv) Although the authorities distinguish generally between(a) provisions operating as an indemnity and(b) independent provisions, the precise nature and effect of any demurrage provision depends upon the context and wording of the particular provisions, including the scope of any reference to or incorporation of the demurrage provisions of any charterparty or other third party contract. v) In the absence of any cross-reference in the sale contract provisions to a charterparty or other contract under which demurrage liability may arise, the natural inference is that the sale contract fall within category (b): cf Houlder Bros and also the dicta in Ets, Soules. But it is of course conceptually possible to have an implied as well as an express cross-reference of this nature. vi) In cases where there is some form of cross-reference to a charterparty or other third party contract under which demurrage liability may arise, the nature, purpose and effect of the cross-reference become critical. There are two broad situations, corresponding with categories (a) and (b) mentioned in conclusion (iv) above. In the first, the sale contract creates a liability for demurrage by way of “indemnity”, that is to pay only if and so far as such a liability exists under the charter or other third party contract (although Suzuki is the only clear reported example in the authorities). It would no doubt also be conceptually possible for sale contract provisions to operate by way of “indemnity”, but subject to the additional qualification or precondition that any liability for demurrage can and should only arise so far as consistent with other sale contract terms (e.g. as to the length of permissible laytime). But such a construction is likely to lead to the practical problems identified in Houlder Bros. and the authorities provide no positive example of it. The second situation (exemplified by a number of authorities) is one where the sale contract provisions simply refer to or incorporate provisions of a charterparty or other third party contract (or at least one of such provisions, e.g. as to the rate of demurrage) in an otherwise independent sale contract scheme. The extent of any such reference or incorporation is then itself of course a matter of construction. vii) Thus, for example (although it is unnecessary to express a view on the correctness or otherwise of the actual construction put on any previous contract differently worded to the present), in Suzuki the words “demurrage as per charterparty or freight agreement” were interpreted as meaning that the case fell within category (a). In contrast, in Gill & Duffus Clarke J considered that the particular provisions for demurrage there in view brought the sale contract within category (b). It is also unnecessary to comment on Staughton J’s obiter view in The Adolf Leonhardt that the obligation “Demurrage/Despatch as per C/P” in the particular contract there in issue was also to be construed as independent. However, the existence in a sale contract of its own laytime code is clearly a relevant factor (even though it was also present in Suzuki), and I find useful Staughton J’s general explanation as to why it may be appropriate to treat sale contract laytime and demurrage provisions as an independent code.[43]Applying these principles to the present case, I conclude that the present sale contract provisions constitute an independent code, falling clearly within category (b). First, the sale contract was made independently of, and without knowledge of the terms of, any charterparty. Since the sale contract covered four shipments, there might well have been four very different charterparties. The sale contract contained a specific laytime code (clause 10), which would not necessarily coincide with whatever charterparty had been or might in future be made. The two did not coincide in the case of the first shipment with which we are concerned, since laytime was under the charterparty reversible and so allowed a total of 72 hours for loading (with which Petronas were not concerned at all) and discharging.[44]Second, as soon as one has a situation where the laytime provisions may not coincide, problems (identified in Houlder Bros.) arise about treating sale contract demurrage provisions as operating by way of indemnity in respect of charterparty liability. In effect, in order for the present sale contract to fall within category (a), the sale contract laytime would have to be read, not as a period after which demurrage would become payable, but as a minimum discharging period which must elapse before demurrage might become payable; it would then only actually become payable, provided that demurrage was at the same time payable under the charterparty. On this basis, in the present case, demurrage would only be payable if and when the whole reversible laytime of 72 hours had been used up under the Fal Oil charterparty. Thus, if Fal Oil loaded the Devon in, say, 24 hours, Petronas would in effect have laytime of 48 (not 36) hours available before they could have to pay demurrage in respect of slow discharging. On the other hand, if the charterparty entered into had only allowed 48 hours reversible laytime, and Fal Oil had used up 24 hours laytime at the load port, Fal Oil would become liable to Fal Shipping under the charter for demurrage after a further 24 hours of discharging, but Petronas could not become liable for demurrage to Fal Oil until after 36 hours discharging. The sale contract cannot be construed by reference to the charterparty that was actually made. But the existence in the sale contract of its own laytime code, likely to create such problems in the event that the demurrage provisions are construed as operating by way of indemnity, is a relevant factor: see paragraph 42(vii) above.[45]Thirdly and most importantly, the present sale contract demurrage clause (clause 11) in my view clearly incorporates a rate, and no more. In that regard, the present case is on any view readily distinguishable from Suzuki, where demurrage was provided to be “as per charter-party or freight agreement” simpliciter. The reference to a rate takes one to clause I in Part I of the present charterparty, providing for $18,000 per day and pro rata. It may also take one to the more detailed provisions of clause 8 in Part II, which I have set out in paragraph 26 above, although this was not the subject of specific submissions before us. If so, the references in clause 8 to loading would have to be ignored, but otherwise there would seem to be no problem in incorporating into or applying under the sale contract the provision that demurrage should be at half rate under the conditions specified in clause 8 of Part II.[46]Fourthly, once it is concluded that the express words of the laytime and demurrage provisions do no more than refer to the charterparty rate, their natural reading and effect is as an independent obligation. So read, they have an understandable and acceptable rationale as a code containing an agreed approximation or pre-estimate of the loss which the sellers, Fal Oil, would be likely to suffer in the event of delay in discharging. There is no need to force them into category (a). We have not heard or been concerned with any suggestion that the present sale contract provisions were not, as and when agreed, a genuine pre-estimate of the seller’s likely exposure: see paragraphs 33-35 above.[47]I would therefore allow the appeal in respect of the second issue relating to the nature of the demurrage liability under the present sale contract, and hear counsel on the appropriate form of declaration to give effect to this conclusion.

Conclusion

[48]It follows that I would dismiss Petronas’s appeal on the first issue and allow Fal Oil’s appeal on the second issue.[49]I gratefully adopt the account of the history and documentation that is set out by Mance LJ. I venture to add some few words of my own. The “contamination” appeal Methodology

The “contamination” appeal

[50]We were pressed with various observations in this court, made in different cases addressing different difficulties that had arisen at the trials from which those appeals were brought, which it was alleged the judge had failed to respect, and which should have driven him to a different conclusion. I am afraid that I did not find this part of the argument at all helpful. An experienced commercial judge scarcely needs telling that, faced with two conflicting accounts of an incident, he should analyse the evidence to try to see which account is the more persuasive (Cf Cooper v Floor Cleaning [2003] EWCA Civ 1649[15]); or that, having reached the end of a difficult and enigmatic case, he should “stand back” and see if a further review enables him to make progress (as, it was suggested, did Phillips J in The Theodegmon [1990] 1 Lloyds Rep 52, and Clarke J in The Kapitan Sakharov [2000] 2 Lloyds Rep 255). Nor is he to be criticised if he takes these nostrums as read, without enunciating them. He is to be criticised if it is clear, or tolerably clear, that if he had taken either of those steps he would or should have reached a different result. But that is not this case. As my Lord has demonstrated, the judge did analyse the evidence, in a manner very far different from that of the trial judge in Cooper v Floor Cleaning, and reached the conclusion, that was open to him, that it yielded no clear answer; and any overall review at the end of the case would not have compelled the judge to revisit earlier conclusions, most notably about events at Yanbu, because his view of those events was based on cogent evidence that he would not have been justified in rejecting simply in order to find a more convenient answer to the puzzle that the case presented.

Misloading at Yanbu?

[51]I start with this issue, because it occupied a very large part of the appellant’s submissions before us. If the judge could have been persuaded that the missing oil never went on board, preferably by error rather than by fraud, then that would exempt him from further embarrassing enquiry into the respective activities of the Centaur and the Devon, and provide an easy answer to the question of whose responsibility it was that the cargo was insufficient. But the judge was precluded from that conclusion by the evidence that he received; or, if that is to put it too high, in rejecting that conclusion he was at least well within his proper area of judgement, and well outside any area in which this court could interfere. The much-controverted statement of Mr Severn that the possibility of error on loading at Yanbu had to be discounted, set out by my Lord in his §12, was not a casual or unthinking observation, but accepted by him to be a necessary conclusion from the loading records at Yanbu and the analysis of water content on loading. It would have been a remarkably strong step for the judge to reject that evidence in order to reach a conclusion favourable to Mr Severn’s clients.[52]I therefore think that the judge was right to conclude that error at Yanbu was excluded; to analyse the case thereafter on the basis that the right cargo had been loaded on to Centaur; and not to revert to events at Yanbu just because analysis of subsequent events faced him with serious difficulties. I would reach that conclusion in any event, on the basis of the normal approach of this court to findings of fact. I do not therefore find it necessary to pursue the debate as to whether the judge’s conclusion on the Yanbu issue were findings of “primary” fact, and therefore protected in this court in any event by the terms of the grant of permission to appeal. I would, however, emphasise my respectful agreement with my Lord’s conclusion, in his §6, that it was in any event right for the court to hear the substantive case on the points that might be affected by the scope of the grant of permission.

Burden of proof

[53]Petronas resist payment for the cargo on grounds of its insufficiency; and in addition, so we understood, claims the cost of putting right the contamination. Petronas must therefore establish that the cargo was contaminated when it passed the Devon’s manifold. Petronas seek to do that by demonstrating, as was the case, that the cargo was contaminated on first inspection by them, and that there was no evidence of any contaminating incident between passing on to the Devon and that inspection. That approach is reinforced by the only positive theory of contamination advanced by Fal Oil, leakage of seawater into the Devon, having been decisively rejected.[54]If there were no other evidence about the history of the cargo, then it might be open to a court to assume (though I doubt whether it would be compelled to do so) that, in the absence of any evidence of contamination between going on to Devon and Petronas’s inspection, the cargo was contaminated when it reached Devon. But there is other evidence about the history of the cargo. First, there are the judge’s findings about the position at Yanbu. Second, the judge investigated the evidence as to the quantity and condition of the cargo on Centaur at the STS and concluded, largely on the basis of the evidence of Mr Pantouvakis, that it was unlikely either that the cargo loaded on to Devon was already contaminated, or that the missing 1500 mt of oil had been successfully concealed on Centaur. These were both conclusions that were plainly open to the judge, based as they were on the evidence of a witness whom he had seen and heard. The judge was entitled to conclude that, just as there was nothing that demonstrated interference with the cargo when on the Devon, so there was nothing that demonstrated interference with the cargo when on the Centaur: bearing in mind, in the latter case, that the cargo had been shown to have been intact when loaded on to Centaur.[55]Accordingly, even if the judge had adopted the approach indicated at the beginning of §54 above, he could not have found that Petronas had discharged the burden of proof. It might well have been enough for the judge to have stopped there. He did, however, do what the guidance referred to in §49 above requires, and examined whether there was any other approach to the evidence that would yield a different conclusion. Since the evidence excluded accident, the judge was driven to conclude that the only other explanation was misconduct either on Centaur or on Devon: see § 22(4) of the judgment. Since there was no evidence of such misconduct on either vessel, and no material from which the judge could conclude without impermissible speculation that one vessel was more likely to have been guilty than the other, the analysis in §27 of the judgment was all that was available. The judge’s own theory?[56]It was suggested to us that, by even contemplating misconduct on the Devon as a possibility, the judge had gratuitously introduced into the case a theory for which there was no evidence, and which had never been addressed by either of the parties. I do not think that that criticism is justified. Misconduct on the Devon entered the case as a possibility, but no more than that, by the logical process reviewed in §54 above. Just as there was no evidence positively to support the theory, so there was no evidence that excluded it. Faced with the unusual circumstances of this case, the judge would have been wrong, once misconduct on the Devon had presented itself as a logical possibility, to have excluded that possibility entirely just because of the history of the case, and thus to have been forced back on to misconduct on the Centaur, in respect of which there was equally no positive evidence. If the judge had gone further than the modest steps that he took, and made positive findings that the cargo had been stolen on the Devon, then he would indeed have nodded; but that is what he did not do.[57]I would therefore dismiss Petronas’s appeal on the first issue for those brief reasons, as well as for the much fuller reasons explained by my Lord.

The demurrage appeal

[58]On this part of the case I have the misfortune to take a view that departs from that which commends itself to Mance LJ. I need hardly say that I reach that conclusion only with the greatest diffidence.[59]I would venture to make three preliminary observations. First, the review of the authorities set out by my Lord in his §§ 28-41 indicates that they yield no clear solution to the present issue. Second, therefore, it is necessary to investigate the reasons for and commercial justification of a provision for demurrage in a C&F contract. Third, where contracts are made by the exchange of formalised documents, such as the sale contract and charterparty in the present case, enquiry as to the effect of and reasons for the provisions has to be directed not so much to the intentions of the immediately contracting parties as to the reasonable commercial purpose of the use of those provisions in contracts of this type.[60]The first issue, therefore, is why it should be thought necessary or appropriate for there to be any demurrage clause at all in the sale contract. The price is a C&F, delivered, price, the cost of delivery therefore resting with the seller. There are no provisions as to time of delivery, or as to delay in delivery, until the vessel arrives at the port and gives notice of readiness: as to the time and reasonableness of which there are no express obligations between buyer and seller. In that context and against that background, the sudden intervention of delay provisions at one stage of the contract appears to be anomalous. The reason for such provisions is as explained by my Lord in §33 above: the seller will have to arrange carriage, and he may, and very likely will, do so by chartering vessels on terms that will expose him to a potential liability to pay demurrage to the owner.[61]That rationale strongly disposes me to think that the demurrage provisions in the sale contract should be viewed as being in the nature of an indemnity. They are included, not because of any inherent characteristic of the C&F contract itself, but because the seller, in performing that contract, may suffer collateral and as yet unquantified expense against which it is thought reasonable that the buyer should protect him. That is the very essence of an indemnity. Of course it is the case that, as Lord Atkinson said in Houlder Bros in the passage quoted in §28 above, once launched upon a demurrage clause the parties are free as a matter of law to make any bargain that they please securing to the seller, or withholding from him, a profit under that clause. But why should it be assumed, or even be likely, that the parties would have used that freedom to formulate provisions, included in the contract in order to protect the seller against demurrage liability, in terms that far exceed that objective, by conferring on the seller the possibility of a windfall profit if he does not incur demurrage liability at all?[62]My Lord points out, in his §33, that at the time of entering into the sale contract the parties will probably not know the terms of the charterparty, or even whether transportation is to be secured by charterparty at all; as opposed, for instance, by use of the seller’s own shipping. They may therefore prefer to operate a scheme that lets them know on contracting precisely where they stand in the event of delay, rather than contract on a basis that depends on the seller’s future liabilities. However, with deference, I would venture to think that the uncertainty at the time of contracting as to the terms, or even the existence, of the seller’s future charterparty makes it more, rather than less, unlikely that the buyer is to be taken to accept an independent, open-ended, liability in respect of the seller’s demurrage. Why should the buyer accept such an obligation, when he has no means of knowing whether, and if so in what terms and amount, there is going to be a cost to the seller at all?[63]Both of these questions were in my respectful view posed by Bankes LJ, speaking for a very strong commercial Court of Appeal, in Suzuki, in the terms set out in §31 above: questions to which in my equally respectful view subsequent cases have given no answer. Nor, with deference, do I think that any illumination is to be found from the law of penalties. It seems very unlikely that the issue of whether buyers generally are to be assumed to have agreed to a clause obliging them to pay damages for demurrage whether or not the seller had suffered actual loss can reasonably be influenced by the availability to those buyers of the sparse comfort that the clause might not be enforceable in cases where its circumstances or outturn can be characterised as oppressive: see Chitty on Contracts (29th edition) § 26-109.[64]Where there is no cross-reference at all in the sale contract to the or any charterparty, then the argument that the sale provisions are indeed intended to be independent is much stronger: see my Lord’s § 42(v). But once there is explicit reference in the sale contract’s provisions for payment of demurrage damages to the provisions in the charterparty that address the same category of loss, then the connexion between the two is much harder to resist. In our case, only the rate of demurrage and not the laytime provisions track the charterparty. But I would make two observations. First, if the provisions are intended to be independent, there is no good reason to refer to the charterparty rate at all, as opposed to a rate specific to the sale contract. Indeed, as was demonstrated in Mallozzi, such reference may be a positive barrier to enforcement of the provision in the sale contract. Second, on the facts of the present case there is no insuperable difficulty, I would say no difficulty at all, in applying provisions that provide for different laytimes. The sale contract permits 36 hours laytime. That is a floor, or ceiling, to Petronas’ liability. They then have to cover such liability as the seller incurs under the charterparty by reason of being delayed over the 36 hours. The seller cannot complain if he is therefore handicapped by different laytime provisions in his charterparty, because he enters into those charterparties knowing the terms of his sale contract and the implications of its laytime provisions.[65]I would therefore uphold the conclusion of the judge, shortly stated though it was. I respectfully agree with him in his §30 that that conclusion reflects the commercial reality of this transaction.[66]I agree with the judgments of Mance and Buxton LJJs upholding Morison J’s decision in favour of Fal Oil on the first issue. I cannot usefully add anything. In view of their disagreement on the demurrage issue, notwithstanding a degree of hesitation about expressing views which may be entirely academic, I cannot avoid explaining, if briefly, and without repeating the material closely analysed in the previous judgments, why I agree with Mance LJ.[67]We are primarily, but not exclusively, concerned with the contract of sale dated 29 November 2000. This made express provision for laytime and demurrage. “10. LAYTIME: Laytime allowed shall be a total of 36 hours SHINC to commence 6 hrs after Nor is tendered or upon berthing, whichever is earlier and time shall cease to count at disconnection of hoses. 11. Demurrage: as per charterparty per day pro rata.”[68]To my untutored eyes, the eight words which comprise the whole of condition 11 appear simple enough. The condition defines the rate on which it is agreed that demurrage will be calculated and payable if, but only if, liability to pay demurrage arises under the contract of sale. It does not create any further or additional obligation. It does not claim to be, nor expressly state that it is to be, treated or deemed to be a provision relevant to any obligation which may arise under the equivalent clause in the charterparty. And just as the language does not purport to create any such obligation, I can see nothing in principle, nor in the authorities, which compels me to the conclusion that it does.[69]Condition 11 therefore is concerned with rates or methods of calculating demurrage. The rate is fixed by reference to the demurrage agreed (or in this case, as we now know, to be later agreed) in the charterparty. The parties agreed that, if liability for demurrage arose between them, it would be paid at the same rate as the rate for demurrage agreed in the charterparty. As a matter of fact, that provides a link between the contract of sale and the charterparty, but does so without undermining the contract of sale as an independent, freestanding contractual arrangement of its own. When the charterparty was eventually concluded, the daily demurrage rate was fixed at US$ 18,000 pro rata.[70]At the time when the contract of sale was agreed, the charterparty rate was unknown. That does not present an insuperable difficulty. The parties appreciated that a charterparty would inevitably come into existence, and there was nothing to prevent them agreeing the rate of any potential liability for demurrage as between themselves by reference to it. Seen through the eyes of both parties, I do not discern any commercial naivety in using a separate, independent but linked charterparty as an agreed reference point by which the rate of demurrage in this contract of sale should be fixed. If improper advantage were taken of the condition it would be capable of remedy. However, it is not here suggested that the charterparty was blemished in this, or any other way, or that the agreed rate might be regarded as a penalty clause, or indeed that resort may be needed to the Inco Term 2000 to which reference is made in condition 15 of the contract of sale.[71]For these reasons, which mirror paragraphs 45 and 46 of Mance LJ’s judgment, the contract with which we are concerned was an independent, sensible, prearranged method of anticipating and calculating the loss recoverable for demurrage in the contract of sale: no more, and no less. Accordingly, I too would allow the appeal on the demurrage issue. "1 The first appeal shall be and is hereby dismissed. 2 The second appeal shall be and is hereby allowed." Paragraph 3 will be as per the draft but deleting the last sentence to which Petronas object and replacing it by the words to the effect that "the demurrage claim will be remitted to the Commercial Court for it to hear and determine further argument on any outstanding issues, including the issue whether any and, if so, what further amendment of Petronas's defence is required or should be permitted in relation to Fal Oil's demurrage claim". Paragraph 4: an order will be made in the alternative form sought by Petronas. Paragraph 5: an order will again be made in the alternative form sought by Petronas with the deletion of the words "to the Court or" (which this Order makes otiose). In other words, the effect of those two paragraphs of the order will be that there will be a limited condition of stay. The only addition ought, we think, to be that, in relation to the order in paragraph 4, there should be interpolated after the words "in relation to the First Appeal" the words "(any such application for permission to appeal to the House of Lords to be prosecuted with reasonable despatch)". Paragraph 6: Petronas's application for permission to appeal to the House of Lords shall be refused. In other words, an order is made in the form sought by Fal Oil. Paragraph 7: the order will neither be in the form sought by Fal Oil, nor in the form sought by Petronas. Instead it will be ordered that "Each side shall bear its own costs of dealing with and corresponding in relation to the scope of appeal prior to the hearing of the appeal (save for costs already covered by Lord Justice Thomas' order dated 19 January 2003) such costs specifically to include those of and in relation to the subject matter of Ince & Co's letter dated 12 March 2004 to the Civil Appeals Office and the correspondence enclosed therewith." Paragraph 8 - a new paragraph:
"Subject to paragraph 7, Petronas shall pay Fal Oil's costs of appeal on both issues. There will be no variation of the costs order below."
A new point has been made relating to judgment rate interest. I am not confident that that has been addressed by all members of the court. I may be wrong in thinking it has been. At the moment the order will be for judgment rate interest. I shall confirm with the other members of the court that that was also their intention, and in the meanwhile the order should not be drawn up. I am not confident that it has been considered but it may well have been and I think it better therefore to make that reservation. I say that partly because just before coming into court I have been given yet another e:mail skeleton argument making submissions, and I am not confident that that, amongst other things, has been seen by the other members of the court. There were one or two typographical amendments to the draft judgment of Lord Justice Judge which he has asked me to mention in court. In paragraph 68, third line, the first and second comma should be removed; in other words, the comma after the word "calculated" should be deleted and the comma after the word "payable" should also be deleted. The third and fourth commas in that line remain. In paragraph 69 in the seventh line there should be a comma after the word "independent". In paragraph 70 in the sixth line there should be an additional comma after the word "separate". (Court adjourned) NOTE: Having further consulted Lords Justice Judge and Buxton, I confirm that no order will be made varying the usual judgment rate of interest.

Order

Appeal Allowed. (Order does not form part of the approved judgment)