England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE QUEEN'S BENCH DIVISION (COMMERCIAL COURT)
The Honourable Mr Justice Cresswell
[2005] EWHC 2122 (Comm)

		Royal Courts of Justice Strand, London, WC2A 2LL
		28th November 2006

Lord Justice Rix :

Index	Para
1. Introduction	1-19
2. The essence of ITS's case	20-22
3. The essence of AIC's case	23-27
4. The dispute	28
5. The effect of the certificate	29-44
6. The parties	45-48
7. The inspection contract	49-52
8. Events at the loading port	53-60
9. Events at the discharge port, and their consequences	61-68
10. The Cooper retest (and further events of 16 April)	69-82
11. 17 April: Mr Rackham briefs Mr Lucas. What Mr Lucas knew	83-95
12. 17 April: the telephone conversation. What Mr Lucas said	96-122
13. The follow up correspondence	123-129
14. The premium parcel RVP	130-131
15. Other witnesses of the telephone conversation	132-145
16. After the telephone conversation	146-153
17. AIC's Swiss and Galaxy's English litigation	154-163
18. The expert evidence	164-168
19. International and ITS internal guidelines	169-186
20. ITS's practice and duties with regard to sample retention	187-192
21. Mr Lucas's evidence	193-205
22. The judge's findings about Mr Lucas	206-215
23. A third element	216-218
24. Mr Chalmers' evidence	219-245
25. The judge's findings as to Mr Chalmers	246-249
26. The issues considered by the judge	250
27. Deceit: the law	251-260
28. Deceit: the representation	261-280
29. Deceit: dishonesty	281-293
30. A third element; and the respondent's notice	294-300
31. Deliberate concealment	301-314
32. Section 32(1)(b)	315-325
33. The four tests	326-327
34. (1) Did ITS owe a duty to disclose the relevant information?	328-351
35. (4) Was the information relevant to AIC's right of action?	352-362
36. (2) and (3): consciousness of duty and deliberate decision to conceal	363-366
37. Section 32(2): a fallback case	367-369
38. Continuing duties	370-374
39. Causation, remoteness and loss	375
40. Conclusion	376-377
41. Annex A: The transcript of the 17 April Telephone conversation	A

1. Introduction

1. On 2 April 1996 Inchcape Testing Services (UK) Limited ("ITS") issued its certificate of quality in respect of a cargo of regular gasoline grade M2 due to be loaded on the Kriti Palm for carriage to New York (the "certificate"). The sample analysed was described in the certificate as that of a "Shore Composite Blend". It was a composite sample derived that day, before loading of the vessel, from four separate shore tanks at the refinery of Mobil at Coryton in England. The certificate described a large number of properties of the gasoline, the test methods applied, and the results. The property with which this case is concerned was the gasoline's Reid Vapour Pressure ("RVP"). The Colonial Pipeline Specification ("CPS") on which the gasoline had been sold and for which ITS had been instructed to provide a certificate of quality required that RVP should be maximum 9 psi as found by test method ASTM D5191. Unfortunately, in error ITS tested the RVP by an alternative test method ASTM D323. The certificate stated that the RVP as tested by D323 produced a result of 8.22. Both that and all the other results listed on the certificate were within the nominal result requirements of the specification. The certificate stated, at its foot, "FUEL MEETS SPECIFICATION".
2. The certificate did not in terms refer to CPS as the required specification. However, the reference to test method D323 showed that the required specification had not in fact been complied with.
3. On the same day ITS faxed the certificate to AIC Limited, in Bermuda, ("AIC") with a copy to Mobil. AIC and Mobil had jointly instructed ITS.
4. AIC had bought the gasoline from Mobil (Mobil Sales and Supply Corporation), FOB Coryton Mobil Refinery Installation, in March 1996 as part of a total contract purchase of 38,000 tonnes composed of two parcels, one of regular and the other of premium (grade R2) unleaded gasoline.
5. The "Inspection" clause of the sale contract provided:

"Quality and quantity at loadport: as determined by mutually acceptable independent inspectors, appointed by sellers, results to be final and binding for both parties save fraud and manifest error. Costs to be shared 50/50 buyer/seller."

6. Apart from the incorporation of the CPS, RVP of 9 psi max was expressly guaranteed. English law was chosen as the proper law.
7. Also on 2 April 1996, AIC sold the cargo on the Kriti Palm ex ship basis New York harbour. The on-sale was to Galaxy Energy (USA) Inc ("Galaxy"). The on-sale contract's quality clause relating to the regular gasoline parcel stated "meeting statutory baseline" ie the CPS which had statutory force in the United States "with the following guarantees…RVP 9.0 psi". The determination of quality clause stated: "As ascertained at loadport and confirmed by Calebrett". Caleb Brett is an internationally famous name in the world of testing houses, and it resided within the Inchcape group. (The ultimate holding company is now known as Intertek Group plc.) ITS's associate company in the US is ITS (or Caleb Brett) USA ("ITS USA"), which was appointed discharge port inspector under AIC's on-sale. The on-sale determination of quality clause was not expressly stated to be final and binding. The determination of quantity clause, however, was stated to be binding: quantity was to be ascertained at discharge port by Caleb Brett (viz ITS USA) "whose findings are binding". The contract's payments clause stated that "Payment to be made against telexed invoice, telexed inspection report of out-turn quantity and quality…". No mention was there made of any load port certificate of quality. The "general terms" clause incorporated the Inco terms of 1990 with latest amendments for "ex-ship duty-paid". There was no express choice of law clause.
8. The Kriti Palm completed loading at Coryton on 3 April 1996 and sailed for New York. It sailed with sealed samples from the shore tanks from which the cargo had been loaded: in particular with a sealed composite blend sample covering the four tanks from which the regular gasoline parcel had been loaded. Equivalent sealed samples were retained by ITS at its headquarters in West Thurrock.
9. On 14 April 1996 the Kriti Palm arrived at New York, and began discharging on the next day. On that day, 15 April, Galaxy faxed AIC to report that discharge tests gave an RVP for the regular gasoline parcel of 9.11 to 9.56. Those results were obtained from an analysis of samples of cargo taken from the vessel's tanks. Galaxy complained that AIC was refusing to release loadport samples for testing. Galaxy said that it would refuse to discharge further until such retesting had proved the parcel to be on specification. AIC's initial refusal to allow access to the loadport samples appears not long to have been persisted in, because on the same day ITS USA issued an analysis report on the regular gasoline parcel stating that its RVP as tested by D5191 was 9.04 and that analysis was carried out on the sealed sample for the composite blend ex shore tanks at the loading port. The seal number was recorded in the report. There was also an ITS USA D5191 test on a loadport sample ex the vessel's tanks after loading, which recorded an RVP of 9.3.
10. These results led to disputes between AIC and Mobil, AIC and Galaxy, and AIC and ITS. The discharge of the vessel was delayed, and Galaxy only took delivery of the regular gasoline after 5 July 1996. By then, still further analyses had been performed at Galaxy's request by another well known testing house, SGS, on further samples obtained from the vessel's tanks at New York, and these provided RVP results of 9.3 to 9.69.
11. In the meantime, on 16 April 1996 Mr Chris Rackham, international coordinator at ITS, was fielding urgent messages from AIC. It soon emerged that ITS had tested by the wrong method for the purposes of its certificate. He initiated a retest by the right method, D5191, on surviving residues of the testing (and therefore unsealed) samples, taken from the four individual shore tanks, which had been tested at the time of the blending of the original cargo. These residues were by then some two weeks old. The retest was not carried out on a composite blend from the four shore tanks. The results therefore had to be subject to a further calculation to provide an overall result. This calculation was performed by Mr Rackham. The calculated result was 9.33. The tests were carried out by Mr Michael Cooper, an ITS chemist, and have therefore become known in this litigation as the "Cooper retest".
12. On 17 April 1996 an important telephone conversation took place between Mr Tom Whitaker of AIC and Mr Nigel Lucas of ITS. Mr Whitaker was a trader at AIC. Mr Lucas was ITS's general manager. AIC had already intimated a claim, and the failure of the conversation led to the appointment of lawyers on both sides. The majority of that conversation was recorded by Mr Lucas, and a transcript of it survives. Mr Rackham was present with Mr Lucas and heard the conversation, but did not participate in it. There was a dispute at trial as to whether Mr Lucas then knew of the Cooper retest or its results. Mr Rackham certainly knew of them. The judge found that Mr Lucas did as well. In that conversation it was common ground between Mr Whitaker and Mr Lucas that for the purpose of its certificate ITS had tested the RVP by the wrong method. Mr Lucas also told Mr Whitaker that he did not know whether the D323 test had been inaccurate or not. However, at what has turned out to be a critical point in the conversation, there was the following exchange:

"TW: Well I have a quality certificate from you that says it is on specification.

NL: We will be standing by that certificate."

13. The judge, Cresswell J, found that in that remark, and in the conversation as a whole, Mr Lucas had been guilty of deceiving Mr Whitaker, rendering ITS liable to AIC in the tort of deceit. He therefore found that ITS was liable for all the financial consequences of AIC's reliance on the certificate in an ongoing dispute and ultimately unsuccessful litigation with Galaxy. He found that implicit in what Mr Lucas said was a representation that the certificate was good and valid, or good and reliable, and that, by reason of his knowledge of the Cooper retest results, Mr Lucas was reckless as to the truth of that implied representation, which was "probably wrong". He said (at para 284 of his judgment):

"The representation that the certificate (which stated 'Fuel meets Specification') was and remained a good and valid certificate, was a false representation made by Mr Lucas who was reckless as to its truth."

14. In this appeal, ITS, defendant at trial and now appellant, seeks to discharge what it accepts as the difficult burden of showing that the judge was wrong, indeed of showing, as Ward LJ said in Assicurazioni Generali SpA v. Arab Insurance Group [2002] EWCA Civ 1642, [2003] 1 WLR 577 at para 197, that the judge was "plainly" wrong in that finding of deceit. That is the first main issue on appeal – "deceit".
15. The second main issue is this. AIC commenced these proceedings on 17 May 2002, one month more than six years after the telephone conversation of 17 April 1996, and six weeks more than six years after the issue of the certificate. AIC explains the delay by reference to legal advice it received from its Swiss lawyers to the effect that the commencement of such proceedings against ITS would prejudice its defence of Galaxy's counterclaim in Switzerland in proceedings brought there by AIC. The effect, however, of the running of more than six years is that, subject to an argument concerning continuing duties, AIC is time barred in respect of any of its other causes of action for breach of contract or duty, apart from deceit, unless it can bring itself within section 32 of the Limitation Act 1980 and in particular section 32(1)(b) which prevents time running in the case of the deliberate concealment of "any fact relevant to the plaintiff's right of action". In this respect AIC has complained that ITS has deliberately concealed the Cooper retest and its results until their disclosure in these proceedings in June 2004. There are a number of sub-issues implicit in this point, but in general it may be described as "deliberate concealment".
16. The cause of action in deceit was also prima facie time barred after six years. However, no time bar was relied upon at trial in connection with deceit, for reasons which are not transparent (but may be connected with the fact that section 32(1)(a) extends the running of time where "the action is based on fraud…until the plaintiff has discovered the fraud"). In any event, the judge did not deal with fraud on the basis that it was prima facie time barred – even though at para 12 of his judgment he recorded that ITS said that all claims (other than a claim for contribution from ITS to AIC's liability to Galaxy) were time barred. Indeed, we were told on this appeal that there is no time bar in fraud, which is not correct.
17. The judge, having found deceit, said that it was not strictly necessary to consider section 32(1)(b), and dealt with it relatively lightly towards the end of his judgment (at paras 293/332). He concluded that there had been deliberate concealment of the Cooper retest results, which he described as "a key piece of evidence, highly relevant to the conclusion that the Vapour Pressure results by test D323 stated in the ITS certificates of quality for the regular cargo were probably wrong" (at para 328). The judge found that the deliberate concealment had been carried out both by Mr Lucas and by his colleague Mr John Chalmers. In 1996 Mr Chalmers was the ITS claims manager. The judge found that both Mr Lucas and Mr Chalmers knew about the Cooper retest at latest shortly after the conversation of 17 April 1996. It is not, however, immediately clear which of many pleaded causes of action the judge had in mind in coming to this (obiter) conclusion about deliberate concealment, but possibly all.
18. It is common ground, that were it to be necessary, that is to say in the absence of deceit, to come to a firm conclusion as to alternative causes of action and section 32 (or continuing duties), and if ITS remained liable on that basis to AIC in respect of such other causes of action, there would have to be a remission back to the commercial court for it to determine consequential issues of causation and quantum.
19. In sum, for the purposes of this introduction: there are two main issues, those of deceit and of deliberate concealment. Both depend on the unavailability to AIC until June 2004 of information about the Cooper retest and results. The finding of the judge at trial was that the original D323 results were probably wrong; that Mr Lucas was reckless as to the truth of his implied representation in his telephone conversation with Mr Whitaker on 17 April 1996 and was therefore guilty of deceit; and that both he and Mr Chalmers thereafter deliberately concealed the Cooper retest and its results from AIC in respect of unidentified causes of action. ITS challenge both these conclusions on appeal.

2. The essence of ITS's case

20. It may be helpful at the outset of this judgment to attempt to express as briefly as possible what the essence of the parties' respective cases on this appeal is.
21. On behalf of ITS, Mr Jonathan Gaisman QC, who was not ITS's counsel at trial, put the matter thus. There was no deceit. Mr Lucas, who had been away until the morning of 17 April 1996, had only a hurried briefing from Mr Rackham before his telephone conversation with Mr Whitaker. He was not told about the Cooper retest. Even if he was, he did not take in what he was told nor was he alive to its ramifications. And even if he had known about the Cooper retest, his honest judgment would have been that he was entitled, in a situation where the conversation was the prelude to the imminent appointment of lawyers by both parties to deal with a claim which AIC had already intimated, to reveal nothing. In fact, his dominant concern during the conversation was that Mr Whitaker was asking him to do something which he felt he could not do, which was to amend or supplement or replace the certificate. It was in that context that he said that ITS would stand by the certificate, meaning that it was a historical record which could not be tampered with. The judge, in finding deceit, erred in his interpretation of the conversation and in extracting from it an implied representation which accorded neither with what Mr Lucas actually said, nor with what he intended to say nor with what Mr Whitaker understood him to say. The judge failed to address the immediately following correspondence between the parties, which was a better guide to what was then intended and understood. In any event, he did not find that Mr Lucas knew that what he said was false, only that he was reckless as to the truth, but he applied a concept of recklessness which was more akin to negligence than the necessary dishonesty. Moreover, his reasoning was inadequate, and he paid scant if any attention to the need to address and to rationalise the rival factual cases, or to get into Mr Lucas's mind. His finding of deceit could not stand. It was plainly wrong.
22. As to deliberate concealment, Mr Gaisman said there was none. Such a concept required an unconscionable flouting of a duty to disclose, none of which existed. There was no unconscionableness, because the Cooper retest was not regarded as important, having been performed on unsealed residues of a testing sample, rather than on an official sealed and formally retained sample. There were better guides in the sealed loadport composite sample tested at New York, or in the sealed samples retained by ITS but untested by AIC. As for a duty to disclose, there was none. ITS's duty was to carry out its instructions properly and carefully. It had not carried out its instructions, because it had used the wrong test. On the judge's findings at trial, that even its D323 findings were probably wrong, it was open to conclude that there was additional negligence in the performance of the (wrong) test badly. All this was a breach of contract, but there was no additional duty of disclosure. As for those breaches of contract (or equivalent breaches of duty), AIC knew all that it needed to know, or could with reasonable diligence have learned all that it needed to know, in order to plead its right of action, back in April 1996. Therefore it was time barred, and section 32 did not avail it.

3. The essence of AIC's case

23. On behalf of AIC, Mr Nicholas Hamblen QC relied for the essence of its case on the judge's findings of deceit and deliberate concealment.
24. As to deceit, Mr Hamblen stressed that the judge had found that Mr Lucas (and Mr Chalmers) knew, understood and appreciated, at the relevant time, that the Cooper retest showed that the original D323 test had significantly understated the RVP of the regular gasoline and that the D323 certificate test result was therefore probably wrong. Thus the cargo, if it had been originally tested by the right D5191 test, would have been shown to be off-specification. The judge described the Cooper retest and its results as "key". Moreover, he regarded the retest as having been done clandestinely, with the purpose of using its results against AIC if they had turned out favourably; but as it was, they were deliberately kept from Mr Whitaker. In coming to his conclusion of deceit, the judge had been unimpressed by both ITS's principal witnesses, Mr Lucas and Mr Chalmers, whose credibility he regarded as suspect. If, as the judge concluded, Mr Lucas and Mr Chalmers were not telling the truth about the Cooper retest results and their respective knowledge of and understanding of their importance, then the implications were clear: they were lying because they appreciated that these were matters which AIC should have been told and were seeking to explain away their deliberate failure to do so. In such a case, where the principal witnesses had been disbelieved, it was particularly hard to reverse a finding of fraud made at trial.
25. As for deliberate concealment, that also followed from the judge's findings as to the knowledge and understanding of Mr Lucas and Mr Chalmers. Mr Lucas himself acknowledged, in his evidence, that AIC should have been told about the Cooper retest results. It also followed from applicable international and internal ITS guidelines, as the judge found. Moreover, the Cooper retest was performed on samples which were held to AIC's (and Mobil's) order, or were even their property, and was therefore unauthorised: these were additional or cumulative reasons why ITS was under a duty to disclose both the fact of the retest and the nature of the results.
26. In certain respects, moreover, AIC sought, by its respondent's notice, to improve on the findings and holdings of the judge. In deceit, Mr Hamblen submitted that even if Mr Lucas's mind had not been dishonest during the telephone conversation, he knew enough shortly thereafter to have become guilty of deceit when he failed, in breach of a duty to speak, to correct the way in which he had left matters during that conversation. Moreover, Mr Chalmers, as claims manager, was in the same position, and was also guilty of deceit in failing to correct the outcome of Mr Lucas's conversation. These submissions went beyond the judge's findings, but Mr Hamblen submitted were inherent as a matter of law in such findings as he had made about those witnesses' knowledge and understanding.
27. There were also further submissions relating to continuing duties and to section 32(2) of the Limitation Act.

4. The dispute

28. Thus, in a nutshell, the dispute was this. ITS had undoubtedly erred, and broken its contract, in failing to provide a contractual certificate relating to the testing of the cargo in accordance with its required specification. However, that cause of action had expired six years after 2 April 1996. The question was whether any cause of action survived the expiry of that limitation period. One answer was in deceit, premised not on a fraudulent certificate, but on a dishonest representation that the certificate was still effective when it was known to ITS, but in breach of contract and of duty not to AIC, that a retest demonstrated its inadequacy. The other answer, somewhat different but bound up in the first, was in the statutory concept of deliberate concealment which enabled some cause of action to survive limitation. A far distant third answer is said to be found in the nature of continuing duties.

5. The effect of the certificate

29. Halfway through his judgment, at a fulcrum of it, immediately after finding the facts and stating the agreed issues at trial, and before turning to answer those issues, the judge wrote a section headed "6. Mistake and departure from instructions contrasted" (at paras 166/174). He there considered the effect of a decision of this court in 2001 in Veba Oil Supply and Trading GmbH v. Petrotrade Inc [2001] EWCA Civ 1832, [2002] 1 Lloyd's Rep 295.
30. Veba Oil involved a dispute between traders, not between a trader and an inspection company. The defendant there had sold a cargo of gasoil on the basis of a certificate which, as here, was to be "final and binding for both parties save fraud or manifest error". Again as here, the inspection company (Caleb Brett) had erred by testing a property (density) by the wrong test, by D4052 rather than D1298. It was common ground at trial that the use of the wrong test was entirely irrelevant in practical terms. Thus Simon Brown LJ recorded (at para 7):

"Before coming to the arguments I should record certain further matters of agreement. First, that test method D4052 is more modern and accurate than D1298, having a margin of error of .0001 per cent. as opposed to 0.0007 per cent. Second, that had the inspectors used method D1298, they would still inevitably have found the density test satisfied in respect of the actual samples tested."

31. Nevertheless, at trial Morison J [2001] 2 Lloyd's Rep 731 and on appeal this court both held that Caleb Brett's error in using the wrong (albeit superior) test rendered Caleb Brett's certificate uncontractual, invalid and ineffective: the defendant seller was therefore potentially liable to its buyer for the consequences of the ultimate receiver's complaint, under a sub-sale from the buyer, that the gasoil was found at discharge to be out of specification as to density. That litigation was at the stage of an attempt by the seller to obtain summary judgment against the buyer on the basis of reliance on Caleb Brett's certificate. That defence failed completely and decisively.
32. The dispute before Morison J and this court was as to whether the certificate was effective or not. If it was, the seller had a complete defence. The buyer said that, even though use of the wrong test had made no practical difference, the certificate was ineffective. The seller's argument, however, was that the certificate remained valid and effective, because the buyer had to show that the error in using the wrong test had been material. That argument failed.
33. Thus Simon Brown LJ pointed out that an error in carrying out instructions is to be distinguished from a departure from instructions. In the former case, the error was only relevant if it was material in the sense of actually affecting the ultimate result, and the question would then be whether the error was "manifest" for the purposes of the clause. In the case of a departure from instructions, however, such as employing the wrong test method, it was irrelevant that the practical effect was immaterial. Only an immaterial departure from instructions, such as one that could not even potentially affect the scientific or commercial process, would be irrelevant. That certainly did not apply to the use of the wrong test method, even where that could not affect the result. As Simon Brown LJ said at para 26(vi):

"Once a material departure from instructions is established, the Court is not concerned with its effect on the result. The position is accurately stated in par. 98 of Mr. Justice Lloyd's judgment in Shell U.K. v. Enterprise Oil: the determination in those circumstances is simply not binding on the parties. Given that a material departure vitiates the determination whether or not it affects the result, it could hardly be the effect on the result which determines the materiality of the departure in the first place. Rather I would hold any departure to be material unless it can truly be characterized as trivial or de minimis in the sense of being obvious that it could make no possible difference to either party."

34. Tuckey LJ agreed (at paras 36/40). Dyson LJ agreed with the result, viz that the certificate even in that case was not binding, but disagreed about the test of immateriality of departure, preferring as his test (at para 46) –

"whether the parties would reasonably have regarded the departure as sufficient to invalidate the determination".

35. As to that, Dyson LJ said (at para 49):

"Turning to the present case, I am in no doubt that the fact that the use of the wrong method cannot have affected the "result" does not save the determination. As Mr. Goldstone has pointed out, it cannot be assumed that the choice of a particular method of testing is a consequence solely of the parties' desire to achieve an accurate result. There may be different reasons which have been dictated by the terms of other related contracts and/or related letters of credit. The possibility that there may be such other reasons is by no means far-fetched in the context of a commercial contract which is likely to be one of a chain. These are not matters about which the Court can or should speculate. The starting point is that if the parties have agreed that a determination using method A is to be binding, then a determination using method B will not be binding because the parties have not agreed that it will be. It follows that the determination is not binding and the appeal must be dismissed."

36. Thus Veba Oil was in all respects a parallel case to the present and demonstrates that the certificate here was invalid and not binding on AIC. Indeed, this is a much stronger case for that conclusion than Veba Oil itself, for in the present case it was never common ground that use of the wrong test was obviously immaterial. In the first place, and critically, conformity with CPS was necessary under US law for importation of the cargo into the US as regular gasoline of the appropriate grade. Thus, the cargo could not be imported as such, without either being blended anew so as to render its RVP to conform to CPS or being downgraded from regular gasoline to a blend stock. Moreover, although at trial, after hearing expert witnesses and choosing between them, Cresswell J found that the two tests, D323 and D5191, gave "comparable" results, that was a conclusion reached by the judge only after trial. Indeed, at the time of the conversation on 17 April 1996, it was believed that D323 would tend to give a lower, or at any rate a different, RVP reading vis a vis D5191. Thus the following exchange occurred between Mr Whitaker and Mr Lucas:

"TW: …because I believe this other [method ie D5191] will give a higher RVP reading.

NL: Ok well I don't know whether that is true or not, I don't think there is any correlation between the two methods…"

37. Moreover, AIC pleaded, until an amendment shortly before trial which deleted the following citation from para 22 of its points of claim, that –

"The use of ASTM D323 resulted in a lower reading than that which would have been shown by the correct test and gave the impression that the cargo met the contractual specification."

That plea was presumably premised at least in part on expert advice tendered to AIC.

38. At the appeal, it was acknowledged on both sides that Veba Oil demonstrated that the certificate in this case was in truth invalid. Although Mr Hamblen at one point submitted that this was new law in 2001, he was unable to maintain this position. Simon Brown LJ's judgment shows that the same doctrine had already been inherent in this court's judgments in Jones v. Sherwood Services Ltd [1992] 1 WLR 277 and in Lloyd J's judgment in Shell UK Ltd v. Enterprise Oil plc [1999] 2 Lloyd's Rep 456. It was the answer which Morison J gave at first instance in Veba Oil itself. The first of that trilogy, a decision of this court, had already been decided some years before the events of the present case.
39. An oddity of the present case, however, is that at trial neither of the parties nor the judge, despite dealing with Veba Oil, appears to have focused at any rate consistently on the invalidity of the certificate. It is not clear why that is so. It may have suited AIC to discount the certificate's invalidity for the purposes of arguments concerning causation. It may have suited ITS to do the same in the context of accusations about its failure to withdraw its certificate. Even so, Veba Oil was discussed in ITS's closing written submissions. It was there submitted that the law was still uncertain in 1996 and was only clarified in Veba Oil. It was therefore said that Mobil "(who are no slouches in this area of the law)" could be understood in the attitude it took in 1996 that it could continue to rely on the certificate even after the use of the wrong test was clear. Mr David Mildon QC, who was leading counsel at trial for ITS, then submitted that "That approach was correct on the then state of the law because the burden would have been on AIC to show that the use of the wrong method had materially affected the result." In my judgment, that analysis was wrong for 1996, but in any event it was definitively wrong, even on Mr Mildon's approach, at the time of trial.
40. What in these circumstances did the judge derive from Veba Oil? Having quoted from the judgments in it, he said this (at para 174):

"The following points should be noted. It would not have been easy for a client such as AIC inexperienced in the purchase and sale of gasoline (see issue 1) to determine whether a departure from instructions was trivial or de minimis. I refer under issue 6 below to test methods which are technically equivalent and would be expected to give results that are not significantly different. If the tests ASTM D323 and ASTM D5191 had been performed correctly the results should be comparable. Once it was clear that ITS had departed from instructions as to the test method (and used D323 instead of D5191) it was perfectly understandable commercially that AIC (with its very limited experience) would look to ITS (as an independent inspection company) for an answer one way or the other as to whether ITS was standing by the certificate which said "Fuel meets specification", and act accordingly vis-à-vis Mobil or Galaxy."

41. On this appeal, Mr Hamblen on behalf of AIC has not sought to rely on the judge's finding about AIC's inexperience. It is not clear how the judge himself deployed it later in his judgment. In any event, the status of the certificate in the circumstances was ultimately a question of law – and AIC appointed (US) lawyers almost immediately. During the argument of the appeal, the court made it clear that it regarded the question of AIC's inexperience as irrelevant, and Mr Hamblen did not seek to dissuade us otherwise. In any event, I do not think the judge was right to regard ITS, whether AIC was to be regarded as inexperienced or not, as being placed in the position of an adviser. ITS was an independent testing, inspecting and certifying house, but not a professional or commercial adviser, and least of all to one only out of a number of traders who might be affected by a certificate and because of that trader's absence of experience in need of advice.
42. It is to be observed moreover that the judge did not say expressly that one of the points to be noted about Veba Oil was its teaching that the certificate was in law – and thus in fact too – neither valid nor binding. This may or may not matter where questions of honesty or dishonesty are concerned, or even, if the point is not squarely taken, where causation is concerned: but it is very likely to be relevant to the question of what duties may or may not have existed in law as a result of the failure of a valid certificate. The judge did not subsequently refer to Veba Oil, although, as will appear below, he repeatedly returned in his questions to Mr Lucas and Mr Chalmers to the problem of a certificate drawn up on the basis of the use of the wrong test.
43. Thus Veba Oil is somewhat like Banquo's presence at the feast. I am therefore concerned that Veba Oil figures, but figures so inconsequentially, in the judgment below. I am also concerned to note in the passage cited above that, although the judge was prepared to assess AIC's reactions according to its inexperience at the time, he was nevertheless also prepared to assess ITS's reactions on the basis that the two tests, D323 and D5191, were known to be technically equivalent and to produce comparable results. One significant question will be whether the judge gave reasons for finding that this was known at the time.
44. Two further points may be made about Veba Oil. One is that, although it is unnecessary to find a further reason for the invalidity of the certificate, I would have thought it reasonably obvious that the statement on the certificate that D323 had been the RVP test method employed rendered this also a case of "manifest error". Secondly, it is relevant to refer to what Simon Brown LJ said by way of background to the specific problem in that case:

"15. Before turning to the authorities most closely in point, it is convenient first to recognize two principles which inevitably touch on the issue. The first, and that on which Mr. Nolan understandably places reliance, is to be found in Lord Justice Cairns' judgment in Toepfer v. Continental Grain Co., [1974] 1 Lloyd's Rep. 11 at p. 14:

When parties enter into a contract on terms that the certificate of some independent person is to be binding as between them, it is important that the Court should not lightly relieve one of them from being bound by a certificate which was honestly obtained and not vitiated by fraud or fundamental mistake on the part of the certifier. When, for instance, as in this case, the certificate called for by the contract is one relating to the quality of the goods sold, the business purpose is to avoid disputes about quality, and that purpose is defeated unless it is made difficult for a party to go behind a valid certificate.

16. The second, clearly countervailing, principle is surely this: inspectors should be astute to comply with their instructions and, if they depart from them, there should not then be much scope for dispute and litigation as to whether their determination is nevertheless binding. In short, the interests of finality cut both ways although, of course, one bears in mind that if a determination is set aside the underlying dispute is left unresolved."

6. The parties

45. I have already introduced the claimant, respondent to this appeal, AIC (and its Mr Whitaker); the defendant and appellant ITS (and its managers, Messrs Rackham, Lucas and Chalmers, as well as its chemist, Mr Cooper; and its associate ITS USA); the first seller of the gasoline, Mobil, and its ultimate receiver at New York, Galaxy.
46. It will be necessary to say something further about ITS and its set up, about the main witnesses, as well as about other persons who have parts to play in these events. For the present it suffices to say that there are three ITS locations in England which should be distinguished. The first is the laboratory within Mobil's Coryton facility where Mr Mailey performed his original D323 tests. The second is ITS's offices at West Thurrock at Caleb Brett House: this was where the ITS Inspection Division, also known as "Gray's", was based, and where the certificates had been issued. The third was a separate laboratory, also at West Thurrock, where Mr Cooper worked.
47. By the time of trial Mr Lucas had worked in the inspection industry for 25 years. In 1996 he was ITS's general manager and by the time of trial vice president, corporate compliance, for the Intertek Group. He was also then a council member of the International Federation of Inspection Agencies ("IFIA"), and had been chairman of a body responsible for producing professional standards for the inspection industry.
48. Mr Chalmers had by 2005 been in the industry for 22 years. In 1996 he was ITS's claims manager (and has since become global claims manager for Caleb Brett). He had been professional standards director for the Intertek Group, and had since 1999 been chairman of the legal committee of the IFIA.

7. The inspection contract

49. As of 1 January 1996, ITS and Mobil had entered into a contract for the provision by ITS of cargo inspection services. Although this contract was only between ITS and Mobil, it was pleaded by AIC and common ground at trial that, for the purposes of the joint instructions which Mobil and AIC had issued to ITS in respect of the Kriti Palm cargo, the contract bound ITS and AIC in addition. The terms of that contract therefore are at the basis of the relationship between those parties. The only term in that contract to which our (or the judge's attention) was drawn is in its Schedule A, which is introduced in clause 1 of the contract as describing the "Work" to be performed by ITS. The important clause in Schedule A is:

"Contractor [ie ITS] shall hold retained samples for ninety days (unless instructed otherwise by Company) [ie Mobil, and thus Mobil/AIC]."

50. There was another clause in Schedule A which related to "Quality and Performance Standards" and provided that certain consequences should follow if ITS received a less than "good" rating in any inspector audits performed by ITS, but that clause was not relied upon.
51. At paras 178ff the judge posed the agreed issue: "6. Was ITS retained by Mobil and AIC in any kind of advisory capacity? What was the proper scope of the contractual duties owed by ITS to AIC?" It is not clear how the judge answered the first of those questions, but he stated his answer to the second of them in the following terms:

"192. Thus in my opinion ITS' duty to take reasonable care to ensure that any certificate it issued was accurate as to those matters on which it was instructed to report, included the following implied obligations to both Mobil and AIC:

i) to determine whether Mobil had performed its contract with AIC in the relevant respects, applying the test methods ITS was instructed to apply.

ii) to exercise independent and impartial judgment and to act as an independent inspection company at all material times.

iii) to report the result of tests independently, accurately, clearly, unambiguously and objectively.

iv) to include in any certificate all information relevant to the validity and application of the test results and all information required by the test method and procedure used;

v) to make it clear whether the results reported referred to tests carried out on a single item, or on a batch of items, including where relevant details of any sampling carried out.

vi) to include in any certificate: - any departures from standard condition; reference to the test method and procedure used; any standard or other specification relevant to the test method or procedure or deviations, additions to or exclusions from the specification concerned.

vii) to issue material amendments to any certificate in the form of a further document by way of Supplement to the certificate, with a statement to the effect that the same should be passed onto any person to whom the original certificate had been provided.

viii) where a complaint or any other circumstance raised doubt concerning the quality of the tests, to ensure that the relevant work/tests were promptly audited/reviewed. Where the audit/review findings cast doubt on the correctness or validity of the test results such as to necessitate a Supplement to the certificate, to write to Mobil and AIC immediately enclosing the Supplement, with a statement to the effect that the Supplement to the certificate should be passed onto any person to whom the original certificate had been provided."

52. In drawing out these implied obligations the judge had primary regard to international and ITS internal guidelines, to which I will refer below. Suffice it to say at present that there is a dispute as to whether the judge had regard to the applicable guidelines. ITS says that he did not, and has erred accordingly. To some extent, the judge supported his implied obligations by reference to admissions made by ITS's principal witnesses, to which I will also refer below. Mr Hamblen accepts that the judge's implied obligations are at the cutting-edge of the law, in that no other authority has stated an inspecting house's duties in these detailed terms before. The judge himself recognised that his treatment of such duties rendered an appeal necessary. The first paragraph of his judgment stated that "This case raises important issues about the duties owed by inspection companies in domestic and international trade." When asked by ITS to grant permission to appeal to this court, he refused to do so on the ground that there was a realistic prospect of appeal, and only did so on the alternative ground that there is some other compelling reason why the appeal should be heard, adding "The judgment contains an analysis of the duties of inspection agencies. This is of wide international importance." Moreover, the judge did not derive his implied obligations by the traditional method of asking whether they were necessary. I will content myself for the moment with saying that the judge's implied obligations read rather like legislation. An obligation to take reasonable care in the performance of services may, I suppose, be broken in all kinds of ways: but the duty is at any rate easily stated. Mr Gaisman on behalf of ITS did not agree with the judge's restatement, and in particular challenged obligations (vii) and (viii). It will be necessary to consider the extent of ITS's duties below.

8. Events at the loading port

53. The Kriti Palm arrived at Coryton to load on 27 March 1996. No Mobil product was ready to be loaded, nor could it be loaded until ITS had certified that it complied with specification. On 30 March, ITS issued its first certificate, for one of the four shore tanks nominated for loading, namely tank 75 x 1. This individual shore tank certificate stated that RVP tested by D323 measured at 8.91/8.98 (upper and lower tank level readings). The vessel commenced loading.
54. On 2 April ITS produced further individual shore tank certificates, as well as its composite certificate referred to in the opening paragraph of this judgment. An individual shore tank certificate, again for shore tank 75 x 1 but in respect of a second parcel, stated the D323 RVP results at 7.69/7.60. Another individual shore tank certificate, for shore tank 100 x 1, stated the D323 RVP at 7.83/7.81. A fourth individual shore tank certificate, for shore tank 61 x 4, stated the D323 RVP test results at 8.85/9.01. It is agreed that the average of these last figures is 8.93 and thus would have been within specification if performed by the right test. Nevertheless, because the gasoline from that shore tank failed the tests for MON and Olefins, that certificate stated "Fuel does not meet specification". As for the overall blend certificate, the tests in respect of that were performed on a weighted lab blend from the other four samples. This produced a calculated overall RVP result of 8.22 and the other tests also fell within specification, possibly after some reblending. AIC and Mobil received all five certificates (the "certificates").
55. On the same day AIC faxed the composite blend certificate to brokers of its on-sale to Galaxy. It was not at that stage passed on to Galaxy, and the individual shore tank certificates were never passed on to Galaxy: but certain "oral indications" were given by the brokers to Galaxy over the telephone.
56. On 3 April the vessel completed loading and sailed for New York. Her master gave ITS a receipt for four 5 litre sealed samples, representing before loading shore and after loading ship's blends of the regular and premium gasoline parcels respectively. On the same day ITS drew up a "Sample List" of formal samples for retention for 90 days. The list referred to the four samples sent with the vessel together with their seal numbers, but also listed 29 other samples (which were retained by ITS in England), of which there were 4x2.5 litre samples of the shore tank blend, before loading, of the regular gasoline. No seal numbers are given, so that these samples may not have been sealed at that time. A manuscript internal draft of that sample list also exists.
57. Also found in manuscript internal draft dated 3 April (but only internal draft) is an additional sample list of the RVP testing samples on which ITS had performed its RVP tests: 4 500ml samples drawn before loading from the shore tanks from which the four parcels of regular gasoline derived. These were not sealed and of course had been already broached. The draft nevertheless said, in the style of the formal lists, that "Retained samples are intended to be held for a period of 90 days".
58. ITS's formal report to AIC and Mobil was dated 10 April 1996. The report commented that there had been a lack of available cargo on the vessel's berthing and that loading operations were protracted due to the unavailability of cargo. It continued:

"Analysis was performed as [sc at] shore tanks as soon as the tanks were available. The final tank for the regular parcel was to be outside the required specifications, however after re-testing with a columetric composite of all four tanks, the results were found to be acceptable."

59. Among the documents attached to the report were the certificates and the "Sample List" referred to above (at para 56) relating to the retained samples both kept by ITS and despatched to the vessel. This formal report was addressed to AIC c/o Banque Paribas in Paris.
60. There is a dispute between AIC and ITS about the status of the various samples. ITS submits that a contractual "retained sample" is a term of art and refers to those samples formally preserved (for 90 days) and listed on the formal Sample List. ITS would accept in respect of these samples, which were intended to be consulted, with the authority of the clients, should any dispute arise, that ITS would not be entitled to use them unilaterally for its own purposes. However, ITS says that that did not apply to the testing samples, which could have been thrown away and in any event were no longer fit for official use. AIC on the other hand submits that all the samples brought into existence, both the unused retained samples and the testing samples used by ITS for the purpose of its tests were contractually or at any rate professionally to be retained for 90 days, were the property of AIC and Mobil, and could not be used again without their authority. AIC accepts that the results obtained by retesting the already used testing samples could not be made certifiable, but nevertheless regards them as potentially valuable evidence. That was disputed by ITS's principal witnesses, albeit, as will appear below, its witnesses were not entirely unanimous on the subject.

9. Events at the discharge port, and their consequences

61. The Kriti Palm arrived at New York, in fact berthing across the Hudson River in Linden, New Jersey, on 14 April. On the following day the tests referred to at paras 9/10 above were performed, showing the cargo to be off-specification in respect of RVP when tested by D5191. The load port shore tank and ship's tank blend samples, which had travelled with the vessel, and post-voyage samples obtained at the discharge port from the ship's tanks, all showed the regular gasoline to have an RVP in excess of 9, to a greater or lesser extent. The important pre-loading shore tank composite blend sample which had travelled with the vessel, when tested by ITS USA, produced a result of 9.04.
62. This caused Galaxy to order the cessation of all discharge of the vessel, which up to that moment had only been discharging the premium parcel.
63. By 16 April the problem had come into the hands of Mr Rackham at ITS. He was its south east area manager. He kept a manuscript log book of telephone communications, which the judge described as careful and helpful. The first relevant entry is at 0843 on 16 April (a call from Mr Giovanni Sampino, by the time of trial AIC's senior trader, but then more of an operations man). Mr Rackham's internal email to Mr Chalmers at 1010 on 16 April (headed: "Possible Claim for Grays Inspection") takes up the story from the moment he first learned of the difficulties from AIC at 2200 on the previous night:

"Problem: ITS as Disport Inspectors declare high RVP and Octanes. Discharge is suspended awaiting further testing…

CR [Mr Rackham] called Linden and advised Jules Balogh (Lab Manager [at ITS USA]) of loading problems with quality…ITS Linden to re-sample and re-test multilayer samples all tanks…

A.M. 16/04/96

CR received call from Giovanni Sampino (GS) of AIC asking for list of retain samples and individual shore tank results prior to load. CR contacted Mobil and Mobil declined to issue individual shore tank results as cargo sold FOB and final document was bench blend as representative of cargo loaded. This passed to AIC."

64. Pausing there, I comment that AIC had already been sent by ITS, attached to the latter's formal report dated 10 April 1996, a full set of certificates, not only for the composite blend, but for the individual shore tanks as well. They were resent directly to Mr Whitaker, by Mr Rackham at Mr Lucas's request, on 17 April.
65. Mr Rackham's email continued:

"0845hr. GS called CR and stated that vessels retained loadport comp samples had been tested with Saybolt [another inspection house] as independent witness and results concur with load port C of Q. However not possible to perform RVP test as all samples comped in tin cans. ALSO ITS Grays loadport C of Q states RVP by ASTM D 323 whilst Colonial Pipeline Spec (stipulated load port spec) states RVP to be by ASTM D 5191.

RVP was tested by ITS lab Tech at Coryton to ASTM D 323 as an oversight and error not picked up at reporting stage by Inspection office.

Very early days yet but AIC state that vessel is held on demurrage and also if RVP method is found to be erroneously used they will hold us for total quality failure of cargo."

66. At this stage, therefore, the wrong test error had become known to the parties, there was a report of high RVP (and octanes) levels being found at the discharge port, but otherwise this report is rather confusing to the present reader and may indeed have involved an element of confusion. Thus "results concur with load port C of Q" does not appear to be correct: the whole problem was that they did not. This may possibly have been a reference to the premium gasoline parcel. Moreover, "not possible to perform RVP test" is also problematic, since high RVP had been found on the retained shore and ship's loading port samples: and this is Mr Sampino reporting to Mr Rackham. The possible meaning is that it was not possible to perform a reliable RVP test on the samples retained on board ship, owing to the nature of the samples and their containers. The judge does not comment on this email, neither when setting it out (at para 35), nor when dealing with Mr Sampino's and Mr Rackham's evidence.
67. At 0915 Mr Rackham had spoken to Mr Paul Mailey about the RVP problem. Mr Mailey was the chemist who had performed the original load port tests at Mobil's Coryton facility. The note reads "393 [sic] – RVP. 5191 Grabner". This may, in a compressed form represent Mr Rackham querying with Mr Mailey the "wrong test" information passed to him by Mr Sampino, and Mr Rackham learning of the reason for the error from Mr Mailey. Apparently it stemmed from the fact that RVP or Reid Vapour Pressure suggests the "Reid" method, which in essence involves a "wet" procedure, the D323 test. However, D5191, which requires the Grabner machine, is in essence a "dry" procedure or its equivalent, and results, with the help of a calculation using a correlation equation, in a dry vapour pressure equivalent or DVPE. I put the matter rather simplistically and unscientifically, but I hope sufficiently. It may be during this telephone call to Mr Mailey that Mr Rackham asked Mr Mailey to investigate whether the testing samples used by Mr Mailey still survived: as a result of learning that they did, Mr Rackham had them delivered to Mr Cooper and asked him to retest them by the D5191 method (see below).
68. At 1030 and 1055 Mr Rackham logged two further calls from AIC's Mr Sampino and Mr Whitaker, but without any further relevant note of their contents.

10. The Cooper retest (and further events of 16 April)

69. At some time that morning Mr Rackham, on his own initiative, commissioned the Cooper retest. No mention is made of it in his 1010 email, but there is an entry in his log at 1100: "LD retest regular unleaded. RVP ceta 8.67. ASTM D5191 Grabner 8.57". This is understood to be an initial report from Mr Cooper on the result of his retest using the Grabner machine which is necessary to the D5191 test method. Mr Cooper worked in ITS's Light Distillate (hence "LD") West Thurrock laboratory, where he was a senior chemist. This was a separate laboratory from that at Coryton, within the Mobil facility, where the original tests had been performed by Mr Mailey.
70. A manuscript note, prepared by Mr Mailey, faxed at 1128 on 16 April and addressed to Mr Rackham, is headed "Re Retain Samples ex 'Kriti Palm'". It describes 2 x 500ml RVP samples for each of the four shore tank parcels (and a 1 litre composite blend sample). The reference to two samples per shore tank is a reference to upper and lower levels within the tank. It would seem that these are the original testing samples used by Mr Mailey, and located with his help. His note states: "Samples will be retained until further notice". Mr Hamblen submits that this is evidence that these testing samples had been formally retained as "retained samples". So far as the document is concerned, I am not sure that it demonstrates much more than that as of 16 April Mr Rackham had asked Mr Mailey to locate the testing samples and retain them until further notice. It would seem that they had already by then been delivered to Mr Cooper for his retest.
71. When he had completed his retest, Mr Cooper prepared a manuscript note of his findings, dated 16 April. It is not known exactly at what time that day his retest was completed or his note was compiled. His note stated that all sample bottles were 500 ml containers 50/75% full. The findings were that RVP on the individual samples varied from 8.47 to 9.46. Two shore tanks gave results below 9 and two above 9. However, his note did not provide a composite calculation. His witness statement said that the samples were noted to be broached and therefore not valid for vapour pressure testing. He repeated this in his oral evidence, when he said that, being broached, they were not samples which would normally be analysed at all for vapour pressure and "should not really be analysed". That is something which he said he had mentioned to Mr Rackham at the time. However, as recorded in the judgment, he also said that if the broached samples remained intact and in good order one would generally expect vapour pressure to go down – because of the risk of light ends being lost. In this connection he also said that it would depend on how the samples were stored, and on the risk, with broached samples, of contamination. The judge found that both D323 and D5191 tests state "do not perform tests on broached samples" (at para 261).
72. At 1217 on 16 April Mr Rackham sent another internal email, addressed primarily to ITS USA, saying (inter alia) that "We are very sure at this end that overall vessel will show RVP under 9.0." This, as it turned out, optimistic and soon to be disappointed expectation, was presumably based on the early result so far gleaned from Mr Cooper from only one of the individual shore tank samples. Thus at 1409 Mr Rackham said in a further email to ITS at Rotterdam (where help had been offered) that "present indications are that situation is resolved".
73. At 1333 Mr Rackham's log contained an important entry:

"JC re Kriti Palm. Harass U.K. Need low RVP."

74. JC is a reference to Mr Chalmers, who, as ITS's claims manager was senior to Mr Rackham. Mr Chalmers was out of the office, away on a course. At 1300 the log had recorded: "JC ? K.Thurlow [Mr Chalmer's secretary] + Kriti Palm". The probable explanation of this note is that Mr Rackham had telephoned Mr Chalmers, had got on to his secretary in his absence, and had asked for her help in speaking to Mr Chalmers about the Kriti Palm. In these circumstances the call at 1333 probably came from Mr Chalmers, ringing in to Mr Rackham during a lunch break. Mr Chalmers had his laptop with him, which would have enabled him at some time to read Mr Rackham's 1010 email. So much is clear. However, the note at 1333 is disputed. AIC's case, which was accepted by the judge, was that this recorded Mr Chalmers telling Mr Rackham that he should harass Mr Cooper to come up with the results of his retest, so that a low RVP, ie under 9, could be established in order to resolve the issue. ITS's case, on the other hand, remains that this is not a report of what Mr Chalmers said to Mr Rackham, but of what Mr Rackham explained to Mr Chalmers, namely that AIC was harassing ITS UK (whereas Mr Cooper had been referred to in Mr Rackham's log as "LD") because it, AIC, needed a low RVP (vis-à-vis Galaxy).
75. The judge found: (1) that Mr Rackham accepted in cross-examination that the note suggested that Mr Chalmers was anxious to get the results (paras 110 and 275); (2) that this was to be contrasted, to the detriment of Mr Chalmers, with his own evidence that the Cooper retest was a "botched up job on a spent sample" and that he would probably have "blown up all over the telephone" when he became aware of it (para 110); (3) that the 1333 note demonstrated the importance and significance attached by ITS to the results of the Cooper retest (para 138); and (4) that Mr Rackham accepted when giving evidence that if the results of the Cooper retest had come out the way ITS hoped, the results would probably have been communicated to AIC (para 259).
76. Thus the judge viewed this note as a critical piece of contemporaneous documentary evidence which led him to reject witness evidence from Mr Cooper (to which I have referred above), Mr Rackham, Mr Chalmers and Mr Lucas (to which I will refer below) that they regarded the Cooper retest at the time as invalid and unreliable.
77. At some time on 16 April, but I do not think that it can be timed precisely – save that it was after Mr Rackham's email to Rotterdam at 1409 (see para 72 above), Mr Rackham received Mr Cooper's test results and tabulated them in a manuscript document of his own. He calculated what the judge described as "some kind of" average for the regular parcel as a whole (the nature of the calculation is not apparent from his document) and derived a figure of 9.16. He also "applied an EPA seasonal correction factor" (which has not been explained to us), deriving individual shore tank sample figures not found in Mr Cooper's document. He then averaged those figures, and calculated an average of 9.33. This 9.33 figure is the one that Mr Rackham reported to his senior colleagues (see his 17 April email below) and has been used as the critical figure for the purposes of argument in connection with the claim in deceit. Ironically, Mr Rackham's calculations were inaccurate. We have been told that the correct average figure to be derived from Mr Cooper's results, accepted at trial, was 9.1, a figure with which Mr Lucas was at times cross-examined. At the foot of Mr Rackham's document are to be found the words "Repeatability" and "Reproducibility", but without further explanation.
78. At 1515 Mr Rackham received a call from ITS USA. It is not mentioned in the judge's chronology. ITS reported that Saybolt had been appointed by AIC and SGS (yet another inspection house) by Galaxy. It also reported the results of its own tests (of 15 April) on the retained samples carried on the vessel: viz 9.3 for the composite blend ex ship's tanks after loading and 9.04 for the important composite blend ex shore tanks before loading (already referred to at para 61 above). Those figures are set out in Mr Rackham's log, together with the seal number in respect of the latter sample (0060969).
79. At 1615 another call is logged from Mr Sampino. Between that time and the next time at 1720, Mr Chalmers is mentioned twice. The first time is just the line "John Chalmers" and nothing further. The second time is "See John Chalmers" and nothing more.
80. Following the early entry at 1100, there is no other entry on Mr Rackham's log as to the Cooper retest or its results.
81. At 1741 on 16 April AIC telexed Mobil with copy to ITS, to hold them "fully responsible" in all costs and damages incurred by reason of the off-specification cargo, and concluding "This serves as a official notice".
82. There was also an exchange between AIC and Galaxy on that day. Galaxy made a proposal to resolve the problems caused by findings that both the premium gasoline was off-specification on its octane value and the regular gasoline on its RVP and octane values. Galaxy suggested a discount of 50 points per barrel on the premium and 150 (for the RVP) plus 60 (for the octane) points per barrel on the regular, which has been calculated to amount to a total discount of some $241,000. However, AIC, in a faxed letter signed by Mr Sampino, repudiated any liability, saying, somewhat strangely, that "the load port samples have been tested and the results show that the product is fully within spec, thus proving that the disport samples are not representative of the actual cargo". Galaxy replied to say that it was concerned to minimise costs and damages.

11. 17 April: Mr Rackham briefs Mr Lucas. What Mr Lucas knew.

83. 17 April began early for Mr Rackham, for his email to Mr Chalmers (and to Mr Mark Loughead, who was ITS regional manager for Europe, Africa and the Middle East) was timed at 0640. (There is a still earlier timed version sent at 0551. This has "VNL", ie Mr Lucas, written in Mr Chalmers' manuscript, presumably as someone being copied in to the email at some later stage, see below at para 220.)
84. The email stated:

"John, subsequent to our telecons.

Mark, subsequent to e mail from M Stokes.

Current situation is that MV Kriti Star loaded two grades Prem/Reg ums at Coryton a/c Mobil/AIC. AIC purchased FOB on Load C of Q.

Quality by ITS at Coryton to CPC spec grades R2/M2.

On arrival New York AIC had sold to Galaxy and ITS appointed for Q and Q [quantity and quality].

Regular pcl at disport found off on RVP. Spec max 9.0 load C of Q states 8.22.

Checking with Mobil Lab tech it appears RVP done by ASTM D 393 [sic] and NOT 5191 as per CPC. C of Q checked and passed by local inspection office and C of Q states 393. This has been queried but I have not responded as yet.

AIC called in Saybolt to witness ITS in New York. Galaxy then appointed SGS.

Vessel has discharged premium pcl but Galaxy refuse to accept reg pcl. Vessel remains alongside on demurrage.

Late PM yesterday AIC served written telex notice of claim against Mobil London with cc copy to me at West Thurrock.

To date no admission of liability by ITS has been made.

Have managed to obtain original RVP samples ex Coryton and whilst these have been broached for original tests have had Grabner RVP's conducted at West Thurrock and overall average for four tanks loaded ex shore onto 17 ship tanks find RVP to be 9.33.

Am trying to stall AIC/Mobil but things becoming very heated."

85. Mr Lucas had also been out of his office on 16 April, and, unlike Mr Chalmers, had not been contacted by telephone. He had been in Aberdeen. He returned to West Thurrock by an early flight on the morning of 17 April. He thought he might have reached his office by 1100. He knew nothing of the Kriti Palm.
86. At just before 1100 Mr Rackham received by fax from ITS USA its analysis report on samples taken from the vessel's tanks at New York. RVP (by D5191) was given at 9.11 ranging to 9.56. At almost the same time Mr Rackham received another call from Mr Whitaker. Mr Rackham's note reads: "AIC $1 million – vessel sitting on berth." Mr Whitaker demanded to "speak to somebody NOW".
87. That appears to have been the catalyst for the involvement of Mr Lucas. As Mr Rackham said in his witness statement: "I briefed [Mr Lucas] about the matter and that Mr Whitaker wanted to speak to somebody more senior than me". Mr Chalmers was still at his course.
88. Mr Lucas gave evidence about the circumstances of that briefing. In his witness statements he said that he had not seen Mr Rackham's email to him of 16 April. Mr Rackham came to see him and told him that in 15 minutes he would be getting a call from Mr Whitaker of AIC. He was briefed as quickly as possible. He learned that the cargo had been tested on arrival and been found out of specification for RVP both by ITS and SGS. Mr Rackham had a file of documents with him, from which he showed him some, including the certificate itself, pointing out the error over the test method used. That error was the one concrete fact he had in his mind. He also had in mind that one of the shore tank certificates had found the relevant parcel to be out of specification (the one ex tank 61 x 4, with an RVP result at 8.85/9.01 but also out of specification results for MON and Olefins: see para 54 above), but that AIC had ordered loading nevertheless. He did not know and was not told of the Cooper retest or its results and he had not seen Mr Rackham's early morning e-mail of 17 April. In his cross-examination he said that the telephone conversation had taken place immediately after his briefing, with Mr Rackham still in his office with him.
89. The judge found that "Mr Rackham informed Mr Lucas" of the Cooper retest and its results by the time of his telephone conversation with Mr Whitaker. He said that the "overwhelming probability" was that Mr Rackham had told him of those matters in his briefing prior to the call (at para 274). He recalled Mr Rackham's witness statement that he was "sure that I gave Nigel as full a briefing on events as possible, but I cannot now recall the detail" (at para 270); and also that, when asked if "the pertinent points would obviously include the facts of the re-tests and the re-tests results?", Mr Rackham had answered "I would presume so, yes" (at para 272). He referred to the note in Mr Rackham's log ("JC re Kriti Palm. Harass UK. Need low RVP") as demonstrating the "importance and significance attached by ITS to those results" (at para 273). However, the judge made no findings as to the urgency with which, on the evidence, Mr Lucas had had to be briefed. The judge did not expressly find that, in addition to Mr Rackham informing Mr Lucas of the Cooper retest and its results, Mr Lucas knew, that is took in and assimilated, the information and its contextual significance. On behalf of ITS, Mr Gaisman complains of those omissions.
90. The judge's findings of Mr Lucas's knowledge, however, went well beyond being informed of the Cooper retest and its results. At an earlier and critical stage of his judgment, in the course of answering a technical issue, viz "4. Were the RVP results by test method ASTM D323 stated in the ITS certificates…(i) accurate and/or accurately stated? (ii) wrong?", the judge said this:

"198. The experts agreed that in the light of the ship/shore quantity figures, any contamination would have been very limited and would not have caused a significant increase in the Vapour Pressure.

199. I accept Mr Revell's evidence [he was the expert witness for AIC], in relation to the broached/opened samples re-tested by Mr Cooper, as follows. The Vapour Pressure of the samples when tested would probably have been lower than when they were first tested at Coryton. The results of the Cooper re-tests established on a balance of probabilities that the tests ITS carried out before loading significantly understated the Vapour Pressure of the regular grade gasoline. (In cross-examination Mr Lucas agreed that because of the loss of light ends in the case of a broached sample, one would expect the RVP figure on re-test to be lower).

200. I accept Mr Revell's opinion that the information now available (particularly the Cooper re-test results) indicates that the original ASTM D323 tests carried out by Mr Mailey of ITS were not in accordance with the stated test procedures and that at least some of the ASTM D323 results, as reported by ITS, were probably incorrect and significantly understated the Vapour pressure in at least two of the shore tanks at Mobil Coryton.

201. The ASTM D5191 results reported by both ITS (US) And SGS at disport were broadly comparable. These two sets were also broadly similar but marginally higher than the Cooper re-test results, which were performed on broached samples. I accept Mr Revell's opinion that the ITS (US) and SGS Vapour Pressure results broadly corroborated each other and when viewed together with the Cooper re-test results confirmed that the regular motor gasoline on board the Kriti Palm was on a balance of probabilities off-specification.

202. I find that the Vapour Pressure results by test D323 stated in the ITS certificates of quality for the regular cargo were probably wrong. The results of the Cooper re-tests were a key piece of evidence, highly relevant to this conclusion.

203. I find that Mr Lucas and Mr Chambers knew, understood and appreciated the matters set out in the last four paragraphs."

91. Mr Gaisman criticises this passage, or at any rate the last paragraph of it, on the basis that the judge was here conflating two different things. One was the answer to the question posed in issue 4, which had to be answered at trial, with the help of a full and probing examination of the accumulated evidence and the assistance of expert witnesses. The other was what Mr Lucas or Mr Chalmers knew at the material time, nearly ten years earlier. Mr Gaisman submits that this conflation is unfair to Mr Lucas and Mr Chalmers. Ex hypothesi they could not have known of Mr Revell's opinion. Dr Marshman, who gave expert evidence for ITS, disagreed with Mr Revell in significant respects. Moreover, the judge's conclusion on issue 4 was expressed in somewhat cautious terms ("on a balance of probabilities…probably…on a balance of probabilities"). This, he submitted, is to be contrasted with the cogent evidence which is needed to prove a guilty mind for fraud.
92. At this stage, I would merely observe as follows. First, the judge does not say that Mr Lucas or Mr Chalmers knew, understood and appreciated the matters set out in paragraph 198 of his judgment. This is possibly important because, unless one has been able to eliminate contamination from a comparative exercise, it is impossible to find that a retest of loading samples and comparable disport figures are mutually corroborative. Secondly, the judge at this point does not explain why he states that Mr Lucas and Mr Chalmers knew, understood and appreciated what Mr Revell had opined or what the judge found to have occurred on the balance of probabilities. It remains to be seen whether that is subsequently explained. Thirdly, the judge does not say at what stage Mr Lucas and Mr Chalmers had the requisite knowledge.
93. In this last respect, the judge returned to this subject under (inter alia) issue 28, "Was Mr Lucas dishonest in what he said and did not say during the telcon on 17 April 1996". He there found, at para 283, as follows as to Mr Lucas's knowledge as of 17 April, in terms which are a close parallel to his earlier findings at paras 199/203:

"d) Mr Lucas knew that the results of the Cooper re-tests established on a balance of probabilities that the tests ITS carried out before loading significantly understated the Vapour Pressure of the regular grade gasoline; and

e) Mr Lucas knew that the Cooper re-test results indicated that the original D323 tests carried out by Mr Mailey were not in accordance with the stated test procedures and that at least some of the D323 results, as reported by ITS, were probably incorrect and significantly understated the Vapour Pressure of the gasoline in at least two of the shore tanks at Mobil Coryton; and

f) Mr Lucas knew that the ITS (US) and SGS Vapour Pressure results broadly corroborated each other and when viewed together with the Cooper re-test results confirmed that the regular motor gasoline on board the Kriti Palm was probably off-specification; and

g) Mr Lucas knew that the Vapour Pressure results by test method D323 stated in the ITS certificates of quality for the regular cargo were probably wrong and that the results of the Cooper re-tests were a key piece of evidence, highly relevant to the conclusion".

94. The judge then went on to make two further findings:

"h) Mr Lucas knew from the results of the Cooper re-tests that the cargo would probably have been shown to be off-specification, if DVPE had been tested in accordance with D5191; and

i) Mr Lucas knew that in all the circumstances then known to him it was wrong for ITS to maintain that a certificate which said "Fuel meets Specification" was and remained a good and valid certificate."

95. The issues which the judge was answering at this stage of his judgment were concerned (inter alia) with Mr Lucas's honesty at the time of his telephone conversation, and not directly with Mr Lucas's state of knowledge. No previous issue had asked a question as to Mr Lucas's state of knowledge at that time, except issue 23, which was limited to the Cooper retest and its results: "Did Mr Lucas know of the re-tests and the results thereof (i) by the time of the telcon on 17 April 1996 and (ii) thereafter?" The judge had dealt with that issue briefly, making the findings set out in para 89 above. It appears therefore to have been treated as axiomatic or implicit in those findings, relating to what Mr Rackham told Mr Lucas, that Mr Lucas assimilated the information, contextualised it, and immediately and consciously derived all the knowledge which the judge had considered was to be derived on the balance of probabilities with the help of expert witnesses by the conclusion of the trial.

12. 17 April: the telephone conversation. What Mr Lucas said.

96. The telephone conversation between Mr Lucas and Mr Whitaker was in the presence of Mr Rackham, still in Mr Lucas's office. The conversation, or most of it, was recorded by Mr Lucas. Mr Lucas said that there came a moment in the conversation when he realised that he ought to record it, and he switched on his dictaphone recording device and held it close to the telephone. He presumably did so for his own, or ITS's, protection.
97. Within 24 hours Mr Lucas caused a transcript to be made from the recording. His secretary typed it up. When, as they did almost immediately, the parties resorted to their lawyers, Mr Lucas gave a copy of the transcript to ITS's lawyers. The transcript survives and therefore provides a verbatim account of the conversation, or most of it. A copy of the transcript is annexed to this judgment as Annex A. It can be read for itself.
98. There was no evidence at trial from Mr Whitaker. Apparently, he was no longer employed by AIC and declined to become involved.
99. There is some dispute about who initiated the call, and also how much of the conversation was not recorded.
100. Mr Lucas's recollection was that Mr Rackham had told him that he would be receiving a call within 15 minutes. The transcript does not record the opening of the conversation. However, at a very early stage of the transcript [at 1B] Mr Whitaker says: "why I am calling you on a fairly urgent way is…". That, it seems to me, is a clear indication that it was Mr Whitaker who initiated the call.
101. As for the unrecorded part of the conversation, Mr Lucas's estimate was that his recording began some 5 or 6 minutes into it. That might be right, but there is no real way of telling. All one can say is that the transcript, from where it begins, reads as though it is quite close to the start of the serious part of the conversation, for Mr Whitaker's first comment is to make the point that D323 had been used instead of the CPS requirement for D5191, and his second comment is that already cited of "why I am calling you on a fairly urgent way". Once started, the transcript reads as though it is complete, and it ends with Mr Lucas saying "Bye".
102. In his evidence Mr Lucas may have elaborated somewhat on the transcript, for instance, in answer to Mr Hamblen's question "Q. You say, "We will be standing by that certificate"?", saying "A. That is what I said, several times." (Day 6.156). If he meant that he had said that several times in the conversation, as distinct from agreeing several times in his cross-examination that he had said that, then there is no evidence from the transcript itself that he did so.
103. What did Mr Whitaker and Mr Lucas say to one another? I will leave for a later stage the question of what Mr Lucas may have thought that he was saying, or what Mr Whitaker understood him to be saying. What did they say?
104. The first thing to note is that it was common ground between them that the wrong test method had been used and that in that respect ITS could assist AIC with a written statement. In that connection, Mr Whitaker said this:

"I guess what I am looking from you is a confirmation that, written confirmation that Caleb Brett has indeed used the wrong test method…and I am going to have to be looking for some sort of compensation from yourselves…Are you prepared to give me that information?"

To that Mr Lucas replied:

"I could certainly give you a statement of fact, a written statement of fact what happened on the events and I would get that to you by fax if you wish or line mail."

105. Mr Whitaker then made a further request:

"I would then also need a statement saying that the said cargo on the Kriti Palm does not meet and I don't believe it does meet the Colonial M2 grade."

106. To which Mr Lucas said:

"Ok, Tom, there is two things there. The statement of facts as to what we actually did I can produce that. Whether the material was offspec leaving Coryton or not we cannot really comment on that because our work was on the shore tanks. Ok.

107. Mr Whitaker persisted, saying that he had an off-spec cargo "verified by Caleb Brett and SGS" and that he believed that D5191 ("the other method") would give a higher RVP reading, to which Mr Lucas responded:

"Ok well I don't know whether that is true or not, I don't know whether there is any correlation between the two methods that you are talking about at this time. Our problem now with RVP is that it is impossible to go back into any of the samples because no samples are kept under ice…"

108. It is submitted by Mr Hamblen that that was a lie, coming from someone who knew that the Cooper retest had been performed. If so, it is a lie which the judge did not find, although he did refer to the remark twice (at paras 281 and 283). It will be recalled that Mr Rackham had said in his first email (that of 16 April) that Mr Sampino had reported to him that it was not possible to perform a RVP test "as all samples comped in tin cans". Mr Gaisman submits that the impossibility of which Mr Lucas was speaking was the impossibility of a certifiable or reliable further certificate. Mr Lucas was not denying the existence of available samples, and all parties would have known about the formally retained samples, some of which had already been tested at New York by ITS USA (Caleb Brett), something to which Mr Whitaker had just referred. All this was in the context of a request by Mr Whitaker (his second request) for a statement from ITS that the cargo was out of specification, in effect a request for a new certificate.
109. A little later Mr Whitaker put his point like this:

"So I have to find out some way of proving that it actually is off specification. You are telling me that I cannot do that."

110. There then followed what the judge regarded as the critical exchange:

"NL. Not to the load port but you can at the discharge port.

TW. Well, I have a quality certificate from you that says it is on specification.

NL. We will be standing by that certificate."

111. Mr Whitaker persisted, confirming that what he was concerned with was conflicting statements from independent inspectors, with ITS UK saying one thing and ITS USA saying the other: "I have got two, I have got the same company independent…both saying they are right. Where do I go from here?" The transcript then reads:

"NL. I can't give advice on that Tom, all I know is that the loading sorry the tanks at the loading ports were analysed. The certificates were produced on a tank by tank basis and decisions were made on that information.

TW. Alright I know that that happened but that was inaccurate information.

NL. Well I can't comment on that over the telephone. I really don't know."

112. There was a similar exchange a little later:

"TW. Or was it inaccurate information?

NL. I can't say that. I don't know.

TW. Well if the test 323 was done where test 5191 should have been done, do you not see that as inaccurate?

NL. I can't comment on that. I can't say whether it is inaccurate or not."

113. Mr Whitaker next came very close to taking the Veba Oil point. In effect, speaking as a layman, he did so. He said:

"I have hired you to give me a colonial specification to prove the colonial test, you did not give me that according to your employee Chris Rackham. You gave me tests 323 which is not the colonial test."

114. To which Mr Lucas said "Right". There followed this:

"TW. Which, is therefore, which is therefore inaccurate. Now I have a certificate from you saying that it meets colonial specification.

NL. You have a certificate from us. I can't recall what it says."

115. Mr Hamblen criticised Mr Lucas for that answer, saying it was disingenuous. However, the certificate did not say that it met colonial specification. It said that the RVP met specification on test D323. The certificate did not refer to the CPS. The trouble was, as Mr Whitaker and Mr Lucas had already agreed, that the reference to D323 was not in accordance with colonial specification.
116. The next important exchange was as follows:

"TW. And I have a certificate that says that on that ship the shore tank composite meets colonial M2 and R2 grades and that is something that is issued by Caleb Brett and what I am finding out at the disport by Caleb Brett that I don't have colonial so there is something there. I don't know all I am asking you to do is give me a statement saying that it does not meet. That's what I need and if you are not willing to give me that then I am going to have to get with your legal counsel quick because I have got a ship waiting to complete the discharge.

NL. Well, I guess the answer to that is Tom I can give you statements of fact which is what we have done. It is probably not going to save what you are looking for in this case or I can refer you onto somebody who would provide a service on behalf of legal counsel.

TW. What's that?

NL. Well you need to talk to somebody who can deal with this in a legal way. I think that's what you are saying to me isn't that Tom?

TW. Yea I am going to have to at this point. It is a law case for me right now. I mean I have a ship waiting to discharge."

117. In that exchange, Mr Lucas repeats that there is no problem in giving Mr Whitaker his first request, but avoids responding to Mr Whitaker's second request, for a statement that the cargo failed to meet specification. Mr Lucas also recommended Mr Whitaker to do what Mr Whitaker said he needed to do, which was to obtain legal advice.
118. Finally, Mr Whitaker came straight to the point and said:

"You either can change the certificate and tell me what I bought. If you can't do that then I am going to have to sue you…My legal counsel is a New York firm called Colt, Mallet- Prevost. My lawyer's name is Robert Gruendel."

119. The judge's findings on this conversation were as follows:

"283. The telephone conversation must be considered as a whole. I accept Mr Hamblen's submission on behalf of AIC that the overall impression conveyed by Mr Lucas during the telephone conversation was that the certificate for the regular cargo was and remained a good and reliable certificate…

284. The representation that the certificate (which stated 'Fuel meets Specification') was and remained a good and valid certificate, was a false representation made by Mr Lucas who was reckless as to its truth. I find that Mr Lucas intended that AIC should act in reliance on the representation and I find that AIC did so…

288. For the reasons set out above, in my judgment the representation that the certificate (which stated 'Fuel meets Specification') was and remained a good and valid certificate, was false and misleading.

289. Further, I find that Mr Lucas was reckless as to the truth of this representation made during the telephone conversation…

292. For the reasons set out above I hold that ITS are liable to pay damages to AIC for deceit."

120. Mr Gaisman submits that the judge had failed to consider the telephone conversation as a whole; that the overall impression conveyed by Mr Lucas could not be said to be that the certificate remained a good and reliable, or good and valid certificate; that it was not right to dissipate the requirement of a clear and unequivocal representation into an overall impression; that it was wrong to speak about what was conveyed or what impression was given without considering carefully three separate matters, namely what was said, what was intended and what was understood, and that it was only where all three matched that a finding of fraud could be made; that in reaching his conclusions the judge paid either no or wholly insufficient regard to either Mr Lucas's own evidence, or to the lack of any proper challenge to it on the issue of his honesty, or to contemporaneous evidence of the parties' understanding of the conversation.
121. Mr Hamblen submits that the judge was fully entitled to come to these conclusions, and in circumstances where Mr Lucas knew of the Cooper retest and its results, but said he did not, that the judge should have gone further.
122. I therefore turn to set out the circumstances relevant to these submissions, beginning with the contemporaneous evidence following the telephone conversation.

13. The follow up correspondence

123. Later that afternoon Mr Lucas sent a fax to Mr Whitaker, referring to their conversation. He said:

"You requested that we change the certificate in relation to the above cargo. This we cannot do as this would be a fraudulent act.

You indicated that the alternative position is that you will take legal action against us. Sadly we await the next communication from you on this matter."

124. That is contemporaneous evidence as to how Mr Lucas regarded the telephone conversation. It was in that context that he had said that ITS would be standing by its certificate. It was perhaps unfortunate that Mr Lucas sent so indignant a written response. Yet that was how he committed himself in writing. He clearly felt strongly about the request to change the loading certificate, ie provide a certificate that the cargo was off specification in respect to RVP. There would never have been any difficulty, as the telephone conversation demonstrates, about a written statement as to the use of the wrong test.
125. On 18 April Mr Whitaker replied to Mr Lucas's fax in a telex which included the following:

"1. We were informed yesterday am by Chris Rackham and confirmed by you that Caleb Brett used the incorrect test method on RVP at the loading of the Kriti Palm. Our question yesterday was very simple – was the original binding irrevocable quality certificate issued by Caleb Brett still valid or does it need to be reissued with corrected results due to Caleb Brett's error? We under no circumstance would make a request for a fraudulent document. You are as likely to incur legal action against yourself by suggesting the same as you are by providing inaccurate test information…

3. We understand that you "stand by your previous results" even though you cannot now be certain that it does or does not meet colonial specifications per Lucas/Whitaker phone conversation…

6. We hold you responsible for all costs and consequences in this matter and reserve all our legal rights and remedies in this matter.

7. Our legal counsel in London and New York is Robert Gruendel of Colt, Mallet-Prevost…"

126. This is a significant document, for it is Mr Whitaker's immediate statement of what he took away with him from the conversation. In the light of the absence of Mr Whitaker's evidence from the trial, it is also Mr Whitaker's only statement. Like Mr Lucas's fax to which it is a reply, it is not at any rate in substance mentioned in the judgment below. It is quite likely to have been drafted with legal assistance. It will be noted that the issue raised in the first paragraph is an issue of validity of the certificate – the Veba Oil point. The paragraph seeks to downplay Mr Whitaker's telephone demand for a new certificate stating the cargo to be off specification to a mere enquiry: if the old certificate is not valid does it need to be reissued with corrected results for the right test? As for the critical third paragraph, this properly links Mr Lucas's "stand by" comment with his repeated comments that he cannot be certain about the quality of the goods tested by the right test. It is somewhat inaccurate, however, in glossing what Mr Lucas said about standing by the certificate with standing by its results.
127. On 19 April Mr Lucas faxed his response to Mr Whitaker's telex:

"We, as an inspection company, can only issue certificates of quality stating what we found using the test methods we used.

In the present case, we issued certificates of quality only in relation to the shore tanks…

In all cases, it was plain on the face of the certificates that the test method was ASTM D323.

The results reported were what we found. We are unable to produce a reporting stating a result which we did not obtain by a test method which we did not use."

128. Mr Lucas also supplied details of his solicitors, Hill Taylor Dickinson (Mr Pople).
129. The judge mentioned these three communications, but only in passing, quoting very briefly from Mr Lucas's faxes and not at all from Mr Whitaker's telex.

14. The premium parcel RVP

130. A small point on Mr Lucas's fax of 17 April emphasises something about the telephone conversation to which so far I have not drawn attention. Mr Lucas's fax is headed "Cargo: Reg. Unleaded/Prem. Unleaded", ie referring to both parcels of gasoline. This was the "above cargo" of which he spoke in the fax. This was in one sense an error, because the RVP problem had only arisen about the regular gasoline parcel. That had been clear to Mr Rackham, see, for instance, his resumé email of early on 17 April. Nevertheless, in the telephone conversation Mr Whitaker had more than once referred to a problem with both parcels. Thus, Mr Whitaker had spoken of a certificate for both M2 (regular) and R2 (premium) grades as having erroneously stated the relevant parcels to be on specification when the wrong test had been used: and it was in that context that he had said that "I don't have colonial" and asked for a new certificate. In an important sense he was right, for of course Galaxy was also making a fuss about another property of the premium parcel, although not about its RVP.
131. When Mr Rackham asked Mr Cooper to retest the residues by D5191, the retest had been performed on the premium parcel shore tank sample residues as well as on the regular parcel shore tank sample residues. Mr Cooper had found the premium parcel RVP result by D5191 to be 8.12/8.17. Mr Rackham had recalculated it (erroneously) at 8.34/8.54. The original premium parcel certificate of quality had given the D323 result at 8.12/8.26. At New York, the ship's composite blend sample for the premium parcel had been tested by the D5191 method at 8.11 and the on discharge test had resulted in readings of 8.37/8.43: in terms of RVP there was no specification problem, for the limit for premium R2 grade was again 9.

15. Other witnesses of the telephone conversation

132. Since the transcript of the telephone conversation survives, other witnesses of it (apart from Mr Lucas and Mr Whitaker), such as Mr Rackham, are not needed to help to identify what was said. But any such witnesses may be relevant to an understanding of how the conversation was understood; and in general to the question of deceit.
133. Mr Rackham was a witness to the conversation. He said that he could hear both sides of it, because Mr Lucas was using the speaker facility on his telephone. Up to that time he had the conduct of AIC's complaints and of Mr Whitaker's urgency. He above all undoubtedly knew of the Cooper retest and its results.
134. By the time of the trial he had left ITS (in October 2003) and was international co-ordination manager at another inspection house. He gave three statements. The first, made on 27 September 2004, was provided to ITS. It gives the impression of being in large part a reconstruction from the documents. As for the Cooper retest and results, he could not remember who had asked for them, he was inclined to think it was Mr Chalmers rather than himself; they were done for internal purposes; he would not regard them as valid for testing as they had already been broached. He said that he did not think that he told Mr Lucas about them, and he repeated that evidence at trial. He also said in his first statement that he was sure he gave Mr Lucas as full a briefing on events as possible but could not recall the detail. The judge preferred Mr Rackham's general remark as supporting a finding that he did tell Mr Lucas about the Cooper retest and results. On that basis, Mr Rackham above all, although the more junior manager, could have been expected to be fully complicit with any dishonesty involved in the conversation. He said of the conversation, after being shown the transcript and confirming that it accorded with his recollection:

"Nigel was very calm and very careful during the conversation. He obviously did not wish to give the client incorrect information but also did not want to implicate the company in respect of any liability…I am also informed that AIC say in their Points of Claim that ITS knew that the two tests gave different results but did not reveal this to the Claimant. Neither Nigel Lucas nor I knew whether there was any correlation between the two methods. We did not conceal anything from AIC."

135. Mr Rackham's second and third statements, made on 7 and 16 October 2004, were given to AIC. In the second statement, he said nothing about the Cooper retest or results, presumably because he was not asked about them. He said that he knew about the transcript of the conversation, but had not read it: that was incorrect, he had read it and commented about it in his statement to ITS two weeks earlier. In his third statement, he said that he had been shown some further material, and commented on the Cooper retest and results. His response was to distance himself from them, eg

"My role in relation to the retests on the cargo was simply to pass on the results to Mr Chalmers and Mr Loughead. Mr Chalmers, as Head of Compliance, would have taken over responsibility for handling the matter after AIC began complaining about problems with the cargo. It would have been for Mr Chalmers, along with Mr Lucas, to decide whether to inform AIC about the retests and the results of them…Mr Chalmers and Mr Lucas would certainly have been anxious to obtain the results of the retests because by 16/17 April 1996 ITS were coming under pressure from AIC and Mobil."

He appears to have forgotten that both Mr Chalmers and Mr Lucas were out of the office for much of this time. There was further oral evidence at trial of this nature: thus he thought that Mr Lucas's main briefing would have come from Mr Chalmers rather than himself, and he could not really recall his own briefing of him. He thought that his "Harass UK" log entry was Mr Chalmers chasing for a low RVP result.

136. Mr Rackham was cross-examined on behalf of both parties at the trial. He was cross-examined by Mr Hamblen about the telephone conversation. The relevant passage is as follows (Day 5.92/93):

"Q. So [Mr Lucas] is saying in essence, "You cannot now check the RVP of the samples". Now bearing in mind that ITS had done exactly that -

A. Yes

Q. – were you not a bit uncomfortable about Mr Lucas telling Mr Whitaker, "There is nothing you can do about it" even though ITS themselves had just carried out a check?

A. No, not at all. Any retests that ITS did on those samples was purely for their in-house satisfaction. The results that they gave would not be – could not be considered binding by anybody. They are RVP samples. They had previously been broached.

Q. As we discussed this morning, they were a test which you regarded as being important tests.

A. We wanted to see what we were getting, but there was no way that we could actually use those results in the public domain. They were an in-house operation.

Q. But you told us this morning, Mr Rackham, that if the retests had shown that or had supported the figures in the load port certificates –

A. Yes.

Q. – then that information would have been communicated to AIC probably through SGS, through ITS (USA).

A. We would have informed the New York office more than likely, yes.

Q. Or AIC would have been told direct?

A. Yes, but we could not have put any confidence in those results.

Q. Given that it is likely that AIC would have been told either directly or through ITS (USA) if the retests had shown that or indicated that the RVP was less than 9 and supported load certificate of quality, did you not consider they should also be told if the results showed something else?

A. That was not a decision that I was empowered to take.

Q. I understand that. It was not your call.

A. No.

Q. But did you not feel that really in fairness they ought to be informed?

A. No, I would have waited for an instruction from senior management."

137. That was the limit of his cross-examination on the telephone conversation. It was not explored with him that Mr Lucas had made a false representation during that conversation; nor put to him that Mr Lucas had, to his, Mr Rackham's, own knowledge, been dishonest in that conversation. In essence, he was asked about what Mr Lucas had not said, not about what he had said. He was asked, according to his own standards, to criticise Mr Lucas, but declined to do so.
138. In dealing with Mr Rackham's evidence (at paras 129/139), the judge said nothing about the telephone conversation itself. He regarded Mr Rackham's critical evidence as consisting in that part of it from which it could be inferred that he had briefed Mr Lucas about the Cooper retest and its results (although Mr Rackham thought that he did not) and that part of it which suggested that Mr Chalmers and he had thought that the Cooper retest was potentially important ("Harass UK"). In that latter connection the judge said this:

"132. Mr Rackham said that he regarded the Cooper re-tests as being an important matter and that he hoped and anticipated that the results would lead to the situation being resolved. He added that had the re-tests come out in the way that he had hoped, he would have informed AIC of the results.

133. At a later stage of his evidence Mr Rackham added that "we could not have put any confidence in those results" because the samples had been broached. This answer was out of line with some of his earlier evidence."

139. The evidence referred to at para 132 of the judgment was in the context of Mr Rackham's hope, at any earlier stage, that good RVP results on the D5191 Cooper retest would have resolved matters. So it may have done. The evidence referred to at para 133 was taken from the passage at the later stage (which I have cited above), when Mr Whitaker was requiring a new certificate and Mr Rackham said in that context that the figures could not have been relied upon. The questions arise: Was that inconsistent? Was Mr Rackham's view (or Mr Chalmers') at the earlier stage to be ascribed to Mr Lucas? Mr Gaisman submits not.
140. Mr Robert Hatcher had been managing director of AIC's associated service company from January 1996 to July 2004. He was also a senior trader. He sat at adjacent desks with Mr Whitaker and others in Monaco. In his witness statement for AIC (admitted under a hearsay notice and otherwise not commented on by the judge), he said this about the telephone conversation, which he had obviously witnessed:

"I can recall a heated conversation between [Mr Whitaker] and Nigel Lucas of ITS. At one point it appeared that ITS were recording the conversation, and I was concerned that Mr Lucas was trying to set up AIC in some way. I have seen an incomplete transcript of that conversation prepared by ITS. I cannot remember the precise words used, but the transcript broadly reflects my own recollection – ITS were determined to maintain their position and somehow put AIC in the wrong. My fears were confirmed when, the following day Mr Lucas sent a message accusing AIC of fraud. In fact Mr Whitaker had simply been asking Mr Lucas either to confirm that the certificates issued by ITS saying "fuel meets specification" were reliable or if they were wrong to reissue them after the correct test had been done."

141. This of course is not a contemporaneous gloss of the telephone conversation, but one produced in the context of this litigation. There is no suggestion here that AIC got a positive answer to the former alternative posed by Mr Hatcher at the end of that citation: the latter alternative, Mr Hatcher infers, resulted in Mr Lucas's accusation of fraud (in fact with the assertion that it would be fraud on the part of ITS to produce a new certificate). The differing strands of reliability, mentioned here by Mr Hatcher, and validity, with which Mr Whitaker was concerned (see at paras 125/126 above), may be noted.
142. Mr Hatcher (together with Mr Whitaker) had also participated in a telephone conversation with Mobil on 18 April 1996. The transcript of that conversation, recorded by Mobil, has also survived. Mobil informed Mr Hatcher that they would be retesting their own sealed loading samples. Mr Hatcher in that context referred to "the retained samples at the Caleb Brett lab" and said –

"the sealed sample[s] at Caleb Brett lab belong 50% to AIC and they will have to be retained until other parties will arrive and the seals will be broken."

143. For some unexplained reason, however, AIC and Mobil did not arrange for the testing of the retained loading samples.
144. That conversation with Mobil was a free-flowing discussion about the tactics of dealing with Galaxy. During the course of it Mr Hatcher referred to the fact that ITS had "come to us and said that…they did err in the test report for RVP".
145. I will deal with Mr Lucas's and Mr Chalmers' evidence about the telephone conversation below. Although Mr Chalmers was not present at the telephone conversation, being still away at his course, he saw a transcript of it at an early stage after his return. AIC says, by its respondent's notice, that he (as well as Mr Lucas) was guilty of deceit by failing later to correct what Mr Lucas had said in it.

16. After the telephone conversation

146. Later on 17 April Mr Sampino telephoned to request that all loading samples be sealed in the presence of SGS. Mr Rackham was asked to arrange this, which he did. On 18 April the retained samples and broached testing samples were all sealed and seal numbers assigned both by ITS and by SGS.
147. On 18 April AIC paid Mobil.
148. On 20 April the vessel, which had been moved to a different berth to continue her discharge for the account of AIC, completed discharging.
149. On 22 April AIC sought payment from Galaxy under its letter of credit, but Galaxy obtained a temporary order from a Swiss court restraining payment.
150. On 23 April Galaxy sent to AIC a hard-hitting telex in which it laid stress on the out of specification findings which had been obtained at New York by both ITS USA and SGS, and in particular relying on the results for the sealed shore tanks and ship's tanks samples carried on the vessel (9.04 and 9.3). It commented that these results "would be low due to the fact that they were improperly stored on board the vessel" (ie the containers were metal and not chilled). Galaxy also complained that AIC had failed to supply the certificate of quality for shore tank 61 x 4 (the certificate which said that parcel had failed to meet specification). On 26 April, AIC refused to supply that certificate, relying on the composite blend certificate.
151. In the meantime cargo underwriters had commissioned a survey from Minton Treharne to check whether there had been contamination on board the vessel. Their conclusion, in their report dated 24 April, was that there had been no in-transit contamination.
152. On 29 April, after making enquiries of Mr Mailey, Mr Rackham produced a "Non-Conformance Report" for ITS. The "Description of Non-Conformance" box was filled in: "Export gasoline testing at Coryton. Vapour pressure test method by incorrect interpretation of specification." "Proposal for Improvement" read: "Full copy of test requirements to be placed with lab tech Coryton, inspection coordinator Grays, & lab manager." "Final action" read: "ITS placed on notice of claim. Lab tech interviewed and requirements redefined. Directive re copy of information flow issued." On the same day Mr Rackham issued that "Directive", which referred to "a weak link whereby the actual method requested is not always cross referenced". There was no suggestion in these papers that there had been any error in performing the D323 test in itself.
153. On 3 July 1996 the Swiss court set aside the temporary restraining order obtained by Galaxy, on the ground that it was inappropriate to interfere with a documentary credit save where there was manifest fraud. Thereafter AIC was paid by Galaxy, and Galaxy took delivery of the regular cargo, as well as that part of the premium cargo which it had not so far discharged.

17. AIC's Swiss and Galaxy's English litigation

154. Later in 1996 both AIC and Galaxy issued legal proceedings against one another. AIC's action was in Switzerland, for damages arising out of Galaxy's failed restraining order. Galaxy counterclaimed for breach of the contract specification. In England, Galaxy sued both AIC and ITS (1996 Folio No 3182): AIC, for breach of the sale contract, and ITS for negligence in failing to ensure that its certificate was fair and accurate. AIC did not third party ITS.
155. In April 1997 Mr John Pople, ITS's solicitor, was considering discovery in Galaxy's proceedings. The documents he was looking at included those relating to the Cooper retest, such as Mr Rackham's e-mails of 16 and 17 April 1996 and Mr Cooper's manuscript note of his results. On 17 April 1997 Mr Pople wrote to Mr Chalmers for his assistance in clarifying whether such documents were privileged or not as having been brought into existence in anticipation of litigation. He asked Mr Chalmers to check with Mr Rackham as to "what was in his mind when he created these documents". He wrote –

"These results are "invalid" because the samples had been broached for testing at an earlier stage, but still somewhat alarming because they showed a number of results over 9 (when one would have expected the pressure to have reduced consequent upon the samples being broached)…I don't consider that any of these documents are very damaging. They clearly acknowledge our error in using the wrong test method, but this is apparent and undeniable. The ASTM D5191 test would have produced some results over 9 if we had used it, so the tests run at Chris's instigation on the broached retained samples don't seem to me to add anything."

156. On the next day, 18 April 1997, Mr Pople confirmed his advice in a telephone conversation with Mr Chalmers, of which a note by Mr Pople survives. On the basis of information from Mr Rackham that the relevant documents had been brought into existence in anticipation of litigation, Mr Pople advised that they were privileged. In the current litigation, it was accepted, in ignorance of this 1997 advice, that the documents were not privileged, and disclosure was made of them. That was on the advice of counsel. It is accepted that Mr Pople's advice in 1997 was in error. It is not suggested by AIC that the 1997 failure to disclose these documents amounted to deliberate concealment.
157. In March 1998 ITS issued a third party notice against AIC. In September 1998 the English proceedings were all stayed, pursuant to article 22 of the Lugano Convention, to await the outcome of the Swiss proceedings.
158. In September 2000 the Swiss court of first instance found in favour of AIC and against Galaxy, but Galaxy appealed. On 19 April 2002 the Geneva court of appeal reversed the lower court and upheld Galaxy's counterclaim, in the sum of $1,165,037.62 plus interest of $51,036.72. The reasoning of the Swiss court of appeal as found by Cresswell J (at para 65) was to this effect. There was no contractual provision that the shore composite blend would be determinative. The cargo supplied was therefore uncontractual, as shown by the discharge port tests. AIC therefore had to demonstrate an absence of fault, but it could not do so since (i) AIC knowingly loaded out of specification cargo, since the 61 x 4 shore tank parcel was out of specification; (ii) AIC had refused to produce the certificate in respect of that parcel, a certificate which Galaxy only obtained through ITS's discovery in the English proceedings; (iii) Mr Whitaker knew that the composite blend certificate had involved the wrong test; (iv) despite the results obtained at New York by ITS USA, AIC refused any further testing or discussions, discharged the cargo, and enforced its contract with Galaxy; (v) AIC acted contrary to the most elementary rules of good faith.
159. Much of that reasoning would not be familiar to English law, although I am confident that the result would be the same in England. On any view the certificate, whatever its status under the AIC/Galaxy contract, was invalid because it had involved a test outside the governing specification. Therefore the question was simply whether the goods delivered were in accordance with specification. Even if the cargo quality, in the absence of any valid loading certificate, was to be judged at loading, the only evidence of any tests by the contractual test method showed the cargo to be out of specification. Therefore Galaxy must succeed.
160. The judge added, in an aside at the end of his para 65:

"[It should be remembered that the Geneva Court of Appeal (and AIC and Galaxy) did not know of the Cooper re-tests and the results thereof]."

161. That is true; but I am not sure what the judge thought followed from it. What the Swiss judgment indicates is that, even without knowledge of the Cooper retest and results, the evidence of contractual non-conformity was all one way, and in a case where the load port certificate was not determinative.
162. AIC appealed to the Swiss Supreme Court, but in December 2003 lost its appeal.
163. In the meantime AIC had commenced litigation first against Mobil (on 26 April 2002) and then against ITS herein (on 17 May 2002). Its claim against Mobil was compromised on 5 May 2005, on the payment to AIC by Mobil of $200,000 and no order as to costs.

18. The expert evidence

164. I have already set out the judge's findings on the expert evidence at para 90 above. He had earlier in his judgment set out the experts' evidence itself at paras 140/147 and 198/202 of his judgment.
165. Mr Revell, a consultant chemist employed by Minton Treharne & Davies, gave evidence for AIC. His experience included investigation of contamination claims and working for an inspection house, and in that capacity acting as a consultant for Mobil's Coryton refinery. Dr Marshman, who gave evidence for ITS, was described as a highly experienced petroleum chemist with extensive experience of quality disputes. But she had never worked on the inspection side of things. The judge therefore considered that Mr Revell was better placed to assist the court. He described Mr Revell as an impressive witness.
166. The experts made two joint reports, which the judge cited. They agreed: that if the D323 and D5191 tests had been performed correctly the results should be "comparable"; but that D323 would be prone to greater variations as indicated by its higher reproducibility; any contamination would have been very limited, but there was in any event no evidence of any contamination from the shoreline system or during the voyage; in general results on broached samples will be lower (however, Dr Marshman said that in some cases results have been found to be higher); and the original certified Mailey test results and the Cooper retest results "were so far apart that both should be considered suspect".
167. However, the experts did not agree on the following. (1) Mr Revell was of the opinion that the cargo was off-specification at loading because it was off specification at discharge and there was no evidence of contamination. Dr Marshman, however, could not confirm whether or not the cargo was off-specification at loading. (2) Mr Revell considered that the Cooper retest results gained credibility by comparison with the results found at disport. Dr Marshman was of the view that the Cooper retest results were invalid. (3) Mr Revell considered that the Mailey tests were not performed in accordance with D323 procedures with respect to calibration and in some cases sample handling. Dr Marshman thought that these criticisms were wrong.
168. It is not clear from the judge's judgment, either in his treatment of the experts' evidence, or from his treatment of Mr Lucas's knowledge, why he, Mr Lucas, was dishonest, when Dr Marshman, who was not criticised as an expert, thought that it was not possible to confirm that the cargo was off-specification at loading and that the Cooper retest results were invalid.

19. International and ITS internal guidelines

169. Both the judge in his judgment and Mr Hamblen in his submissions on this appeal have placed considerable stress on passages from international and ITS internal guidelines. Mr Gaisman, however, submits that the judge adopted inapplicable guidelines.
170. The question of guidelines arose late in the trial, and was expressly raised by the judge. The guidelines were assembled and the experts considered them and wrote an agreed report, but by then the witnesses of fact had already given their evidence, and there was no application to recall any of them.
171. The judge dealt with the guidelines in two passages of his judgment. The first, headed "International standards" at paras 150/163, records the agreement of the experts. He there finds, according to that agreement, that NAMAS M10 applied to ITS West Thurrock, ISO 9002 (1994) applied to the ITS Inspection Division, and ISO 9001 (1994) applied to the Coryton laboratory at Mobil's facility: that is to say that each of those units had accreditation with respect to the relevant standards. Accreditation is not mandatory from a regulatory point of view. However, lack of accreditation or withdrawal of accreditation may result in a company experiencing loss of credibility, subsequent loss of business and being prevented from applying for certain business. The judge went on to explain that NAMAS (now ISO 17025) dealt specifically with laboratory facilities; and that the ISO 9000 series stated the requirements for a quality management system applicable to organisations concerned with production, testing, research and inspection activities.
172. In the second passage (at paras 186/191) the judge set out what he considered were relevant passages from the applicable international standards, reflected in ITS's own internal Quality Control Manual. He repeated (at para 187) the experts' agreement that ITS West Thurrock was accredited to NAMAS M10 and ITS Inspection was accredited to ISO 9002. However, he then proceeded to ignore ISO 9002 and to refer only to NAMAS. The only explanation given by the judge for doing so is at para 191 where he says that "The certificates were issued by the Gray's office which is based at West Thurrock. The relevant provisions of NAMAS Accreditation Standard M10 that applied to ITS West Thurrock represented internationally accepted practice at the material time…"
173. On this appeal Mr Gaisman submits that the judge has erred. The original testing by Mr Mailey was done at the Coryton laboratory, but the inspection certificate was issued by Mr Twyford as part of the ITS Inspection Division. This division was described as "Grays" (so-called after its previous location and referred to by the judge at his para 191 just cited), but was then based at West Thurrock where ITS had a building known as Caleb Brett House. There was also a separate laboratory at West Thurrock, which was where Mr Cooper worked. The critical standards were those which applied to the Inspection Division, which had issued the certificates. Thus Mr Rackham's 16 April 1996 email was headed "Possible claim for Grays Inspection" and referred to "ITS Grays loadport C of Q". However, the judge, without explanation, ignored the separate standards which applied to the Gray's Inspection Division and instead applied standards which related specifically to laboratories. The laboratory at West Thurrock had neither issued the certificates nor performed the Mailey vapour tests. The possibility of confusion between the two separate elements both in West Thurrock, viz the Inspection Division ("Gray's") and the West Thurrock laboratory should be noted.
174. Mr Hamblen, however, submits that the judge did not err but proceeded on the basis of Mr Mildon's acceptance at trial that the West Thurrock laboratory accreditation (NAMAS) was the relevant one. Thus during final oral submissions (at Day 11.120/122) there is a discussion between the judge and Mr Mildon about the applicable standards. It is initiated by the judge saying –

"If I am to set out in a judgment the duties that an inspection agency owe when something goes wrong, should I follow the professional standards as identified in NAMAS…or what duties do you contend for? This is a matter of wide importance and plainly I need as much help as I can get."

175. It will there be observed that the judge (Mr Gaisman would say, correctly) was concerned with the duties of an "inspection agency". To this inquiry, which the judge flagged up as being important to him, Mr Mildon responded by referring to the two laboratories involved, Coryton and ISO 9001 (Mr Mailey) and West Thurrock and NAMAS (Mr Cooper), but not, as Mr Gaisman submits that he should have done, to the ITS Inspection Division and ISO 9002. In the end, after expressing some preference for the Coryton laboratory as being the place where the original tests referred to in the certificates themselves had been performed, Mr Mildon was content to rest on NAMAS:

"In terms of the general guidelines, I would accept that the general guidelines are for West Thurrock and NAMAS in the form in which your Lordship has now got it. Does that help?"

176. The judge can therefore hardly be criticised for adopting NAMAS, which is what he did.
177. Mr Gaisman nevertheless perseveres in submitting that Mr Mildon's error has led the judge astray. It is unfortunate that this should still be a matter of controversy at this stage, especially as, as will now appear, NAMAS and ISO 9002 differ in a critical aspect.
178. Thus NAMAS contained the following passages (for a lengthier citation, see the judgment below at para 189):

"12 CALIBRATION CERTIFICATES, TEST REPORTS AND TEST CERTIFICATES

12.2 The certificate or report shall be factually correct and shall be checked before issue.

12.12 Material amendments to a calibration certificate, test report, or test certificate after issue shall be made only in the form of another document, or data transfer including the statement, "Supplement to Calibration Certificate, Test Report or Test Certificate, serial number --- (or as otherwise identified)", or equivalent form of wording. Such amendment shall meet all the requirements of 12 of this Standard.

12.13 The Laboratory shall notify clients promptly, in writing, of any event such as the identification of defective measuring or test equipment that casts doubt on the validity of results given in any calibration certificate, test report or test certificate or amendment to a report or certificate.

13 HANDLING OF COMPLAINTS AND ANOMALIES

13.2 Where a complaint, or any other circumstance, raises doubt concerning the Laboratory's policies or procedures, or with the requirements of this Standard, or otherwise concerning the quality of the Laboratory's calibrations or tests, the Laboratory shall ensure that those areas of activity and responsibility involved are promptly audited in accordance with 4 of this Standard.

13.3 Where the audit findings cast doubt on the correctness or validity of the Laboratory's calibration or test results, the laboratory shall immediately notify, in writing, any client whose work may have been affected."

179. It will be observed that NAMAS is written in terms of a "Laboratory" as the subject of its procedures.
180. ISO 9002, on the other hand, contains none of these passages. It is structured, rather, to require the accredited company to develop its own management procedures. For instance –

"4.14 Corrective and preventive action

4.14.1 General

The supplier shall establish and maintain documented procedures for implementing corrective and preventive action…

4.14.2 Corrective action

The procedures for corrective action shall include:

(a) the effective handling of customer complaints and reports of product nonconformities;

(b) investigation of the cause of nonconformities relating to product, process and quality system, and recording the results of the investigation…

(c) determination of the corrective action needed to eliminate the cause of nonconformities;

(d) application of controls to ensure that corrective action is taken and that it is effective."

181. ISO 9002 therefore refers its reader in effect to internal standards produced in order to comply with the requirements of the international standards. ISO 9001 follows a similar format, and does not contain the NAMAS provisions cited by the judge.
182. ITS had internal manuals, but different ones for its "Inspection Division" and for "West Thurrock" respectively. The judge quoted only from the (January 2003) West Thurrock manual, for instance (at para 190):

"11.1 Policy

The results of each test or series of tests shall be reported accurately, clearly, unambiguously and objectively to the Client in the form of a test report or test certificate…

11.4 Supplementary Certificates/Reports

Where amendments to test reports or test certificates are required after issue these shall be in the form of a separate document…and should be clearly identified. Supplementary reports and amendments shall meet all other requirements…

11.5 Validity of Certificates/Reports

The Laboratory shall notify the clients in writing of any circumstance, which casts doubt on the validity of results given in any test report or certificate, and, where possible, following corrective action, repeat the tests affected and re-issue an amended report or certificate.

HANDLING OF COMPLAINTS AND ANOMALIES

12.2 Procedure

…

c) Subsequent investigation of the complaint shall take place as soon as possible. All findings and resulting actions shall be recorded and the client notified appropriately, preferably in writing. If necessary, corrected versions of reports shall be submitted…"

183. It is clear from this manual as a whole, starting with its introductory "Quality Policy Statement" that it is directed to laboratories. It was issued under the authority of a laboratory QA manager.
184. However, the ITS manual relating to the Inspection Division (dated April 2003) contained none of the above passages. It begins –

"1.0 Scope

The procedures in this manual apply to field locations of the Company's Inspection Division. They encompass the organisation of operations at locations with and without laboratories, but matters concerning laboratory testing will be found almost exclusively under separate control in files on the appropriate company internet site for the particular facility."

185. The index to this manual illustrates its contents, eg sampling, client reporting, sample handling, calibration, non-conformance action etc. The non-conformance action section (dated April 2003) demonstrates the basis on which Mr Rackham completed his non-conformance report dated 29 April 1996 (see para 152 above). Thus –

1.0 PURPOSE

1.1 To define the responsibilities and actions required in the event of a reported incident construed as a non-conformance with Company policy and procedure.

2.0 SCOPE

2.1 The procedure is designed to record, investigate and action complaints and deficiencies which may develop during the operation of the Company quality system.

5.0 PROCEDURE

5.1 Non-conformances shall be categorised as follows:

5.1.1 Category A

Failure to carry out specific requirements of a nomination…

5.3 Any event that can be classified as a possible claim against ITS…must in the first instance, be reported to the UK Regional Inspection Manager or his nominated deputy. In this case, a fully documented report shall be formulated and sent to the UK Regional Inspection Manager or his nominated deputy, as soon as possible. No other parties shall be advised until this procedure has been completed and there is agreement by the UK Regional Inspection Manager or his nominated deputy.

5.4 The distribution of completed Non-Comformance Reports (all boxes, 1 to 6), for Categories A and B only, will be…

5.5 Where corrective action is required and can be initiated locally, it shall be described in Box 6…"

186. The judge clearly built his implied obligations (set out at para 192 of his judgment and referred to at para 51 above) on the model of NAMAS standards. He cites no other basis for obligations which he finds to have existed. Para 192 of his judgment comes immediately after his citation from those documents. He said (at para 186) that

"The best guide to what is comprised in the duty owed by an inspection agency (whether in contract or tort) to take reasonable care to ensure that any certificate issued is accurate as to the matters on which it is instructed to report, is found in the International Standards at the material time".

He then proceeded to quote extensively from NAMAS M10 (and from ITS's own laboratory manual which reflected it). At para 191 he said that NAMAS applied to Gray's office at West Thurrock and represented internationally accepted practice. He immediately went on to conclude (at para 192) that ITS's implied obligations to both Mobil and AIC included all those obligations set out at para 192(i) to (viii).

20. ITS's practice and duties with regard to sample retention

187. The judge relied on ITS's internal manuals in one other important respect, and that was with regard to the retention of samples. He considered this under issues: 15. What was ITS' standard practice as regards sample retention? What material was subject to ITS' sample retention? 18. Was the material re-tested by Mr Cooper being held by ITS on behalf of AIC and Mobil? 19. Was the material re-tested by Mr Cooper properly to be regarded as the joint property of AIC and Mobil? 20. Should ITS have sought AIC's and Mobil's permission before conducting such re-tests and was ITS in breach of duty in failing to do so? The judge considered these issues at paras 242/244 and 246/259.
188. In sum, the judge found it was ITS's practice to retain all samples taken, including broached testing samples, for 90 days. It appears, but it is not entirely clear, that he considered that the RVP testing samples (used by Mr Cooper) were "retained samples", a submission made to him by Mr Hamblen. Thus he referred to them as "retained and broached" samples (at para 252) and regarded it as significant that in the draft report dated 3 April 1996 (see at para 57 above) these samples had been listed as being retained for 90 days (at para 253 of his judgment). Regardless of the precise legal status of those samples prior to AIC's complaint, they ought to have been recognised as "important evidence which should in the circumstances have been held to the order of Mobil and AIC" once a complaint or doubt had been raised about the original tests (para 255). He cited Mr Lucas's evidence that these samples were potentially important and that it was not appropriate to take them and use or retest them without informing the clients (para 256: see, however, herein below under paras 214/215). He therefore found that ITS was "in breach of a duty to take reasonable care" in failing to seek permission from AIC and Mobil before conducting the retest. He did not, however, expressly find whether or not the samples were the joint property of AIC and Mobil. He concluded by stating:

"258. In any event it was quite unacceptable that ITS should carry out the re-tests when

(i) if the results supported the contention that the cargo was on-specification, the results would be disclosed to Mobil/AIC, but

(ii) if the results did not support the contention that the cargo was on-specification, the results would not be disclosed to Mobil/AIC."

189. In adopting these conclusions, the judge had regard, it would seem, to the ITS manual (dating from September 1998) from which he cited at para 242. This was the manual which related to ITS's Inspection Division (which the judge ignored for the purposes of finding its implied obligations: see above under "International and ITS internal guidelines").
190. The section of this manual from which the judge quoted is headed "Sample Handling". It reads inter alia as follows:

"1.0 PURPOSE

1.1 To describe the actions and responsibilities necessary to ensure the controlled handling of samples from the time of their being drawn to the disposal stage.

2.0 SCOPE

2.1 This procedure covers the transportation, labelling, registration and delivery of samples either for testing or for storage.

4.0 RESPONSIBILITY

4.1 It is the responsibility of the Chemist/Inspector to transport samples safely to the area office reception point. It is also his responsibility to carry out registration, labelling and analysis requirements documentation. If samples are not required for testing, it is the responsibility of the Inspector to place them in the locations's sample store. It is the laboratory supervisor/chemist's responsibility to receive samples for testing and comply with laboratory booking arrangements. Subsequent to testing, it is the laboratory staff's responsibility to place unused surplus sample material in the locations's sample storage.

5.0 PROCEDURE

…

5.6 Samples shall be retained in storage for a minimum period of 90 days, unless the client's agreement to change this period can be obtained…

5.8 Where a client requests an extended period of retention beyond the 90 days normally granted and where this is agreed, such samples shall be segregated and clearly marked for extended retention.

5.9 Agreement to extend the normal retention time beyond the ninety day period normally granted shall be in writing and the disposal date now accepted shall be stated in the agreement."

191. Mr Hamblen submits that these provisions, covering retained and testing samples, make no distinction between them so far as a basic duty of 90 day storage is concerned. He refers to what he describes as Mr Chalmers' acceptance, on the basis of these provisions, put to him in cross-examination, that that was so. He submits that the judge's conclusions were justified. He goes further and submits that the broached Cooper samples always were and remained the property of AIC (and Mobil).
192. Mr Gaisman submits that the manual is not a contractual document, and that these provisions beg the question of what samples are required to be kept for the client, for an initial period of 90 days, subject to further agreement. He submits that "retained" samples in a contractual setting are samples which have been agreed to be retained (for 90 days) and that in that context the phrase is a term of art (as Mr Chalmers said). The "Responsibility" clause merely assigns internal responsibility for the handling of samples. In any event, the broached Cooper samples were not "unused" surplus sample material. The judge's conclusions could not be justified on the basis of this material.

21. Mr Lucas's evidence

193. Mr Lucas's evidence has to be seen against the background of the above material and of the judge's findings on it. It is necessary to encapsulate the evidence given by Mr Lucas, and the judge's findings on that evidence. That will leave Mr Chalmers' evidence to consider. However I take Mr Lucas first, for two reasons. Mr Lucas held his telephone conversation with Mr Whitaker before Mr Chalmers returned to the office from his course: and there is no evidence that there was any contact between the two on the matter of the Kriti Palm prior to the telephone conversation. Secondly, the judge made no finding of deceit against Mr Chalmers. Ultimately, however, it will be necessary to consider Mr Lucas's evidence in the light of other evidence including that of Mr Chalmers, and vice versa, especially for the purposes of AIC's respondent's notice concerning an on-going duty on the part of both Mr Lucas and Mr Chalmers to correct what Mr Lucas had said to Mr Whitaker on the telephone and in that context concerning the further allegation of deceit levelled at both of them.
194. The essence of Mr Lucas's evidence was this. On the morning of 17 April 1996 he came entirely new to the problem of the Kriti Palm. He had only 15 minutes to be briefed on the problem by Mr Rackham. Mr Rackham did not tell him about the Cooper retest. He had faith in Mr Mailey as a chemist. Although it was clear that Mr Mailey had used the wrong test, the certificates issued by ITS remained the proper historical record of what Mr Mailey had found (using the wrong test). He, Mr Lucas, did not know whether there was a correlation between D323 and D5191. Mr Whitaker told him that D5191 would give a higher RVP reading than D323. What the loading RVP of the cargo was he did not know. He was happy to state that the wrong test had been used, but he was not prepared to issue a new certificate or equivalent statement to say that the cargo was off-specification when it had been loaded. He did not know whether or not it was, and there were no samples available which could properly permit him to certify that the cargo had been off-specification. This was despite the disport results, of which he knew. He was wary and upset that Mr Whitaker was asking him to do something that he thought that ITS could not properly do. It was in that sense, and because he regarded the existing certificate as the historical record of what ITS had done, that he said that ITS would be standing by it. It was a case of "publish and be damned". He was also wary because he was facing a claim situation where legal action had already been threatened. In those circumstances he was not prepared to admit that there had been any fault on the part of ITS – other than that the wrong test had been used. His wariness had led him to record the conversation at the time.
195. After the conversation, he was very upset. That had led him to write in the terms of his fax later that day. He also instructed Mr Pople that day, the only time he had instructed lawyers. As for the Cooper retest, he did not recall when he found out about it, but he assumed he would have learned of it from Mr Chalmers at some later time. He was uncertain why Mr Rackham had ordered the retest. Opened samples were unreliable. They could not be made the basis of any new certificate. He still did not know the correlation of D323 and D5191, but he accepted it would be logical to assume that they would produce comparable results. If he had known of the Cooper results he would ("of course") have had to tell Mr Whitaker of them, but with a warning that he should not have been relying on them. So, the fact that he had told Mr Whitaker that it was impossible to go back into the samples seemed to him now to be clear evidence that he did not know about the Cooper retest at the time, for such a statement "would have been … misleading". And it would also have been misleading simply to say he did not know whether the certificated result was inaccurate, if he had known of the Cooper retest result. On the other hand, there were other things he did not then know which he could have discussed, such as the correlation between the original and Cooper figures for the premium cargo. In retrospect that was significant.
196. As for the certificates, they had been performed by the wrong test, but he felt that in other respects they were a reliable account of what had been found, by D323: the shore tank analysis showed that the cargo was on specification and he was convinced that the certificate was reliable; in saying that ITS would stand by the certificate he was telling Mr Whitaker that as far as he was concerned it remained a good certificate.
197. He was asked why he did not tell Mr Whitaker about the Cooper retest results when he found out about them, and said: "Once you are under notice of claim, that is the end of communications, you just stop. That is how we do things in Caleb Brett…We call it the 'stone wall'". It was put to him that at some time he and Mr Chalmers had taken a conscious decision that AIC should not be informed of the Cooper results, and he said he had no recollection of such a decision.
198. He was asked about retained samples. He said that RVP samples would normally be taken for testing and not for retention. He was astonished at the suggestion to him that the Cooper samples had been formally retained samples. However, on the assumption that they were, he agreed that clients should have been informed that they were being tested. He agreed with the view expressed by Mr Loughead in a witness statement that these were "potentially important samples", on the basis that "Any sample concerned with this ship is important, yes". Nevertheless, he thought that Mr Rackham had been misguided to commission the Cooper retest. If he had known about it in advance, he would have stopped him: "It is ridiculous, do not do it". The tested samples were two weeks old, they were unstable, and he would have regarded them as totally unreliable.
199. The end of his cross-examination was reached without Mr Hamblen having put to him that he had been deceitful in anything he had said or done. He had not even been directly challenged in his evidence that he had not known about the Cooper retest at the time of his conversation with Mr Whitaker. (He had been questioned, at the outset of his cross-examination, as to whether he had read Mr Rackham's email to him of 16 April (which did not mention the Cooper retest) or Mr Rackham's second email sent early on 17 April (which did mention the retest and its results), a copy of which had Mr Lucas's initials (VNL) on it in manuscript. He was questioned on the basis that the latter email had been handed to him by Mr Rackham as part of his briefing, but he said he had not seen it until later.) He was not cross-examined on the basis of Mr Rackham's evidence as to the nature of his briefing (see above at para 89). In one sense that had been equivocal, but it provided no direct evidence that Mr Rackham had told Mr Lucas about the retest: on the contrary, Mr Rackham had said that he did not think that he had told Mr Lucas. So that may explain the line that cross-examination had taken. When Mr Hamblen came to question Mr Lucas about his telephone conversation with Mr Whitaker, he did it expressly on the assumption that he then knew about the Cooper retest and results.
200. So, at the end of the cross-examination, Mr Mildon rose to "put down a marker" that no case in deceit had been put by reference to the telephone conversation of 17 April. The judge said that he thought it very important that "the case is fairly and fully put to every witness". Mr Hamblen therefore resumed his cross-examination, and put a case of fraud in the most perfunctory of ways, and not by any specific reference to the Cooper retest results, nor by reference to the implied representation subsequently found by the judge, nor even by reference to any express words contained in the transcript of the telephone conversation. Thus,

"Q. What I put to you earlier this morning, Mr Lucas, was that at the time of the conversation with Mr Whitaker, you appreciated that there was clear evidence that the cargo was off-specification and that had it been tested with the correct method, it would have been shown to be off specification and that, in the light of that understanding, you knew that ITS was in no position to state that it was standing by its results. I would suggest that you deliberately made that statement intending it to be relied upon but knowing it was untrue.

A. I disagree with that proposition, and I would actually like to answer each of those questions one by one, if I may…As your client has brought an action of fraud against me I feel entitled.

Q. Of course, I am not stopping you in any way. The question was this: "What I put to you earlier this morning, Mr Lucas, was that at the time of the conversation with Mr Whitaker, you appreciated that there was clear evidence that the cargo was off specification and that had it been tested with the correct method, it would have been shown to be off specification…"

A. Can I stop you there? I did not have that information and I did not have that view…

Q. I put to you that in the light of that understanding, you were in no position to state that ITS were standing by the results.

A. We would always have to stand by our results if that is the information we had at the time.

Q. And that you deliberately made that statement intending it to be relied upon but knowing it was untrue.

A. I did not know it was untrue and I did make that statement intending it to be relied on."

201. There then followed a lengthy series of questions for Mr Lucas from the judge. First, the judge was interested to know about "written professional standards" and in particular what ITS should do "if it has applied the wrong test to its knowledge". The judge returned to that question again and again. The guidelines referred to above had not as yet been disinterred or analysed by the experts. Mr Lucas's answers were therefore rather in the abstract. Some answers are:

"The onus I think is on communicating what you have done wrong, which is what I did to Mr Whitaker…I do not think the documents go into that, that level of detail. I think they assume a level of perfect performance…I was aware I was not talking to Mobil at the same time, so I tried to maintain a very clear independent position all the way through…The key thing here is the certificate, what you have actually signed off…"

202. The judge next turned to the certificate and its statement "Fuel meets specification" and asked "Was that right or wrong?" Mr Lucas said it was a matter of checking the analysed numbers against the specification. Thus 8.22 for RVP was within specification. The judge then put (what I think of as the Veba Oil point): "But the specification required a particular test…so the fuel on that view did not meet the specification?" To which Mr Lucas replied that he did not know: "I could not look at this and say that the fuel did not meet specification because D323 was used…The question is: was it 8.22 and would it have been 8.22 by ASTM D5191? My feeling is, looking at this, I could not say it would not be…" (Mr Lucas thus reversed the burden of the legal effect of the error, but that is a matter of law.) The judge then obtained Mr Lucas's agreement that an inspection house was like an "independent arbitrator", and asked: "This may not be fair and I am simply trying to understand what happened, but the position that you took in this telephone conversation…was that you maintained that the certificate…was a valid certificate and that the fuel met specification", to which Mr Lucas replied: "That was my view at the time and is still my view today."
203. The judge asked some questions on the hypothesis that he had known of the Cooper retest results. Mr Lucas said:

"I would still look at these certificates as being valid but I would have discussed the issue of Cooper with him, what it really meant. The problem of going into a discussion about the Cooper results is that AIC would have required a certificate, which of course we could not produce."

204. Finally, so far as this summary is concerned, the judge asked about situations where a certificate is withdrawn. Mr Lucas said:

"That is unusual. It is unusual to withdraw certificate unless you really have evidence that the wrong sample has been taken. For example, the wrong tank has been analysed. That bizarrely can happen…you would have to say the certificate is invalid…"

205. (In my judgment, that is correct. If the wrong sample is taken, or the wrong goods are analysed, the inspecting house has not carried out its instructions and the certificate is invalid. If that error is known to the inspecting house, it should withdraw its certificate. If it is unknown, then of course the certificate will not be withdrawn, but it will be invalid nevertheless. "Invalid" in this sense also refers to the position between the traders involved, who so commonly contract, as all inspecting houses know, on the basis that the certificate is binding. The same was true of the certificate in this case, where the wrong test was performed, even if Mr Lucas did not recognise that.)

22. The judge's findings about Mr Lucas

206. The judge referred to Mr Lucas's evidence in essentially two parts of his judgment: at paras 101/106, where, in a section devoted to witnesses called by ITS, he sets out what he clearly considered to be relevant and important parts of his testimony; and in paras 280/292 where he found Mr Lucas to have been deceitful in his telephone conversation with Mr Whitaker.
207. (a) Paras 101/106: I will set out the judge's description of Mr Lucas's evidence in this passage, in full:

"101. Mr Nigel Lucas

102. Mr Nigel Lucas has a degree in chemistry. In 1996 he was General Manager of ITS and was based at West Thurrock. In that role he had management responsibility for every Caleb Brett site in the United Kingdom (plus Norway and Nigeria). He held that position from 1990 to 1998. Mr Lucas reported to Mr Loughead, the Regional Director for Europe, Africa and the Middle East.

103. Mr Lucas said that at the time of his conversation with Mr Whitaker on 17 April he believed that the certificates that ITS produced were reliable certificates. When giving evidence Mr Lucas said that he understood that Mr Mailey was working 100% for Mobil in the blending programme and then worked 50/50 for Mobil and AIC in the independent inspection and testing. Mr Lucas explained "if one was looking at RVP being measured and the blender making decisions about what extra components to put in as the hours go by, you would have a breadcrumb trail of RVP testing which would be a nice smooth curve up to the target point he was looking for. That would have told me a volume of information about how reliable that sample would have been." Mr Lucas said that on 17 April 1996 he thought that Mr Mailey was involved in the blending exercise. If in fact Mr Lucas was mistaken as to Mr Mailey's role in April 1996 he cannot have made proper enquiries."

208. It emerged at trial that Mr Lucas was mistaken about Mr Mailey being involved in blending the gasoline. Mr Gaisman submits that this is an irrelevant and unfair criticism. At the time of his conversation, Mr Lucas had only a very short time to be briefed to talk to Mr Whitaker. He said in his evidence that, short as that time was, he was interested to learn from Mr Rackham as much as he could tell him about Mr Mailey's role as the tester.
209. I continue:

"Mr Lucas said if he had known (a) that Mr Mailey was not involved in the blending exercise and (b) of the Cooper re-tests and the results thereof, "he would have stood by the certificate, but we would have got into a much bigger argument then about what the Cooper results actually meant. We would have had to get into a three-cornered discussion with Mobil, AIC and ourselves to try and see what this actually would mean by way of a decision."

104. In evidence for the Swiss proceedings before District Judge Silverwood-Cope on 11 November 1999, Mr Lucas was asked the following question – "…please describe why it was (or was not) possible to check the actual RVP…?" Mr Lucas answered as follows "…it makes complete sense. I did not think you could analyse RVP 15 days later because the samples are unstable. The logic being they would not be representative of the same period after 15 days." The Swiss courts did not know that the Cooper re-tests had been carried out by ITS. It was suggested to Mr Lucas that his answer for the purposes of the Swiss proceedings was misleading because no mention was made by him of the Cooper re-tests."

105. It is appropriate to make (and I do) due allowance for the difficulties Mr Lucas faced when giving evidence because of the passage of time. Nonetheless I have marked reservations about important parts of his evidence.

106. I will return to consider Mr Lucas's evidence in some detail later in this judgment."

210. The judge did not state in that passage what the marked reservations he had about Mr Lucas's evidence were. Nor did he return to that subject when he reverted to Mr Lucas and proceeded to his finding that he had been deceitful in his telephone conversation. The judge clearly thought that Mr Lucas had been briefed on the Cooper retest by Mr Rackham, and on that basis that Mr Lucas had immediately understood its significance. However, the judge did not say that Mr Lucas had lied in giving evidence that at the time of the conversation he did not know of the Cooper retest.
211. (b) Paras 280/292: I have already set out above the judge's findings as to what Mr Lucas said and knew in his telephone conversation with Mr Whitaker. As for what he said, the judge appears to have been primarily influenced by Mr Lucas's single answer to his own question cited at para 202 above: for at paras 282/3 the judge said:

"282. When giving evidence Mr Lucas confirmed that on 17 April in the course of the telephone conversation he was telling Mr Whitaker that the certificate was and remained a good and valid certificate and that the fuel met specification. He added "That was my view at the time and is still my view today."

283. The telephone conversation must be considered as a whole. I accept Mr Hamblen's submission on behalf of AIC that the overall impression conveyed by Mr Lucas during the telephone conversation on 17 April was that the certificate for the regular cargo was and remained a good and reliable certificate."

212. As for what Mr Lucas knew, the judge found that he was told by Mr Rackham during his briefing of Mr Lucas about the Cooper retest and its results; and otherwise primarily or wholly based himself on what his own conclusions on the balance of probabilities were as to the facts relating to the cargo, its samples, and the two test methods, having heard expert evidence at trial. He then ascribed to Mr Lucas the same knowledge as of the time of his telephone conversation with Mr Whitaker. He did so, in para 283 of his judgment, before any further consideration of Mr Lucas's evidence. He immediately proceeded to a finding of deceit against Mr Lucas in para 284 ("reckless as to its truth"). This is set out in detail at paras 89/95 above.
213. It was only in his next para 285 that the judge said that "Some of the answers which Mr Lucas gave in cross-examination serve to confirm these conclusions". He then cited some portions of the cross-examination which figures among the evidence I have referred to above. Mr Gaisman submits that there is nothing in the cross-examination there cited by the judge which in itself throws light, without further explanation, on Mr Lucas's knowledge or reckless unconcern with the truth.
214. I should also mention in passing, the judge's reliance on Mr Lucas's evidence, in the following passage of his judgment

"239. Mr Lucas (in my view rightly) accepted in cross-examination that the residue of the RVP samples tested by Mr Mailey were potentially important samples, and that ITS should not have done anything to them (including the Cooper re-tests) without informing their clients, Mobil and AIC."

215. That reference was clearly not intended as a criticism of Mr Lucas's evidence. Mr Gaisman submits that this part of Mr Lucas's evidence was on the hypothesis that the residue samples were formally retained samples, something which he said would have astonished him.

23. A third element

216. There is a third element to his findings about Mr Lucas which is very much connected with Mr Chalmers and that is his findings related to Mr Lucas's failure to mention in the telephone conversation, and their deliberate decision thereafter, not to disclose the Cooper retest and results to AIC and Mobil. I will deal with that separately below, after considering Mr Chalmers' evidence. The judge's views in this respect are based on his findings that Mr Lucas was told of the Cooper retest results by Mr Rackham prior to his telephone conversation with Mr Whitaker, and that, even if he did not then know of them, he and Mr Chalmers were aware of them very shortly thereafter (para 269). His views proceed, however, also on the basis that Mr Lucas and Mr Chalmers had a duty to disclose these retests to AIC (and Mobil), and knew that they did so (at paras 311/312). This is particularly relevant to the second main issue in the appeal, that of section 32 and deliberate concealment (the context of paras 311/312). Thus his findings in this respect cannot be separated from his views as to ITS's legal obligations.
217. Nevertheless, there remains for consideration, both on ITS's appeal and on AIC's respondent's notice, whether failure to disclose, also put as a failure to correct what Mr Lucas had said in his telephone conversation, was in the judge's judgment, or should have been, an element in a finding of deceit.
218. The judge made no express finding that a failure to disclose, or a failure to correct, or deliberate concealment, was in itself deceitful or entered as an element into his finding of deceit against Mr Lucas. As expressed in his judgment, Mr Lucas's deceit consisted in what he had said, not in what he had failed to say. On the other hand, submissions on this appeal have indicated that there is an element of ambiguity or opacity in this connection. It is connected with the judge's express finding that Mr Lucas had been deceitful because he had been "reckless as to the truth of this representation" (not that he had been knowingly deceitful). The judge does not explain this finding further, and Mr Gaisman submits that the judge, in that silence, had erred in adopting an essentially negligence test rather than a test of dishonesty. Thus the judge expressed his critical findings about deceit as follows:

"288. For the reasons set out above, in my judgment the representation that the certificate (which stated that 'Fuel meets Specification') was and remained a good and valid certificate was false and misleading.

289. Further, I find that Mr Lucas was reckless as to the truth of this representation made during the telephone conversation…

291. If, contrary to my finding set out above, Mr Rackham did not inform Mr Cooper of the Cooper re-tests and the results thereof in the course of briefing Mr Lucas prior to the telephone conversation on 17 April, he and/or Mr Chalmers should have informed AIC and Mobil of the same as soon as they became aware of the results. Mr Chalmers was provided with a transcript of the telephone conversation when it had been transcribed.

292. For the reasons set out above I hold that ITS are liable to pay damages to AIC for deceit."

24. Mr Chalmers' evidence

219. Mr Chalmers' evidence is thus relevant to deliberate concealment and to a penumbra of interface between that topic and deceit. It is also relevant to the nature of an inspection house's duties.
220. The essence of Mr Chalmers' evidence was as follows. In 1996 he was ITS's claims manager. He was not a chemist. He was away at a course on 16 and 17 April 1996, and believed (on the basis of a reconstruction of events) that he remained away for the rest of the week, only returning on Monday 22 April. He stamped the copy of Mr Rackham's 17 April email which has "VNL" written on it in manuscript with his personal stamp on Sunday 21 April: showing that he was then working at home catching up with correspondence. That was the email which referred to the Cooper retest and the figure of 9.33. He accepted that the "VNL" looked like his handwriting. He was at no time in contact with AIC or Mobil.
221. It is known from the documents that on 16 April Mr Rackham sent Mr Chalmers an email (which he may have read that day) and that he spoke to him at 1333 ("Harass UK"). He and Mr Chalmers were friends. By 1333 Mr Rackham had had an early and at that time promising report from Mr Cooper and thought that matters would be resolved. In his witness statement, Mr Chalmers said in general that he did not recall events clearly, and he did not remember the 1333 conversation. He did not think that "harass" was a word that he would use. He thought that "UK" was a reference to West Thurrock. He repudiated any suggestion that he would put pressure on a chemist for any particular result. He suggested the possible interpretation of Mr Rackham's note that it referred to pressure on ITS from AIC, who did need to be able to justify a low RVP, but that could only be speculation.
222. Mr Chalmers referred to the Cooper retest and results in his first statement. He did not identify when he first heard of them, but in general the inference of what he there said would be that he did not know of them for a little while (possibly until he came to read Mr Rackham's email of 17 April). Thus, "I would have given no credence to the Mike Cooper tests…" and

"I would have placed no reliance on the Cooper results…I do not know who asked for these tests to be run. I did not and certainly would not have done so. The tests appear from the email to have been done at Chris Rackham's initiative and out of curiosity. The tests were unofficial and in-house…Had I seen the results at the time and had I been in the position of deciding whether or not to publish them to Mobil and AIC, I would not have published them for these reasons. The tests were not valid by reason of the history of the samples and could not properly be given ASTM D5191 certification."

In a subsequent statement, he said:

"It may well have been my call, ultimately and at some stage, not to release the Cooper test results and most likely was. However, the better way of putting the issue is that no decision not to release the result would have fallen to be made. It just would not have occurred to me to release them. Had it occurred to me that a decision needed to be made, I certainly would not have released them for the reasons given in my first statement."

223. In connection with a late amendment of AIC's pleadings to allege that ITS owed duties not to touch the samples which Mr Cooper had tested and in any event to report any findings, Mr Chalmers rejected the allegation. He pointed out that the Cooper samples were not "retained samples", which was a term of art, but the residues or "slops" of samples taken originally for the purpose not of retention but of testing. After their original testing, they were kept, but ultimately only for disposal. He had never come across another case in which residues had been retested for vapour pressure, and such testing was invalid under all tests for vapour pressure. He said that the normal thing to do when there is a dispute is to draw fresh samples or, where appropriate, use the retained samples, and have them tested by an independent laboratory in the presence of representatives of both or all parties to the dispute, something called "referee testing".
224. In his statements, Mr Chalmers also gave evidence about the frequency with which his office had to deal with claims, from all over the world. Since about 1991 there had been some 850 matters recorded. The Kriti Palm affair had been routine. Even in terms of AIC's large claim figure presented after ten years of litigation, the Kriti Palm claim did not stand out as exceptional. In his experience RVP claims did not result in large figures, because the problem can be resolved by reblending. It was unrealistic to allege, as AIC had done, that this matter would have caused major consternation to his department.
225. In his cross-examination, Mr Chalmers said that "if it is obvious that we have done something wrong, we would tell" the client, and if there were joint clients, both of them.
226. As for the 1333 telephone conversation with Mr Rackham, he did not remember it, but thought that he "would blow up" on hearing of the retesting of residues. He put it in this way:

"My Lord, retesting in the way that we now understand that it was done and the way that I would have understood as soon as I got back to the office and spoke to everyone is simply not on. It is against the method, and we are either a testing house or we are not a testing house. If we are a testing house then we test things according to the method and we do not test spent samples and produce a test that we cannot stand by or give a test number to and therefore cannot use or report. It is simply a botched operation."

227. He accepted, however, that Mr Rackham was fairly experienced and would know whether it was appropriate to retest.
228. He accepted that it seemed logical that testing on a broached sample for RVP would give a lower result because of the loss of light ends, but he did not accept that such a test could still be a reliable test (as giving at least a minimum). He pointed out that the test methods all said that broached samples may not be used: it was not for him to go behind that. He would not even regard the Cooper retest as being valid evidence. On the other hand, the tests conducted at the discharge port did show that something was wrong with the original result, because that was the one result that was out of line. He was pressed to say that, even though the Cooper results might not be publishable, they were still highly relevant information, but he would not agree: he said –

"I do not see myself going to AIC and telling them, "We took an inappropriate test, got inappropriate results, we cannot stand by them, but here they are". That would be highly irresponsible. It would be mischief in my eyes…

229. Mr Chalmers was asked about the transcript of Mr Lucas's telephone conversation and the concept of standing by the certificate's results. He pointed out (correctly) that Mr Lucas had referred to standing by the certificate. He put the matter in this way:

"Yes. The whole thing is of concern. I was under the impression, speaking to Mr Lucas, that he was being asked to change the certificate. That was my impression talking to Mr Lucas. Well he cannot do that, so he is in a sense stuck with the certificate that he has unless by some way it can be replaced. The certificate, as it stands, is disputed. The certificate is itself the source of the dispute, if you like. To alter that certificate would be wrong. It is valuable evidence of what was said at the time and what was tested at the time and what were the results at the time, and he may not alter that."

230. The judge then intervened with what I think of as the Veba Oil point. He asked:

"One factor, of course, is that the wrong test has been used so it cannot possibly meet specifications, can it?"

231. To which Mr Chalmers replied:

"Well the certificate is in a sense contradictory on its face anyway because of the method number. I am not sure you can actually say that the fuel meets specification on the basis of the certificate as it stands, no. I would have expected that to be the grounds of the dispute."

232. He went on a bit later to say this:

"My strong impression at that time and today is that what was being asked was that a published certificate was to be changed and somehow made better. We may not do that. At this time we are already in a dispute, we have already appointed lawyers, the opposition already have in their hands a certificate with obvious problems on it and they have all the other information to enable them to handle their dispute."

233. Mr Hamblen then asked:

"whether it is acceptable for ITS to say that it is standing by the certificates in circumstances where ITS knows that there is something wrong with one of the published results. Is that acceptable or not?"

and Mr Chalmers replied:

"Phrased like that, no."

234. Mr Hamblen strongly relies on that answer, taken by itself. Mr Gaisman submits that it should be understood in context: Mr Chalmers accepted that, because the wrong test had been used and stated on the certificate, it had "obvious problems on it"; what Mr Chalmers did not accept was that the Cooper retest added anything to what the parties knew. A few lines later Mr Chalmers was saying this:

"We are already in contention. Everybody knows the whole surrounding situation. We have appointed lawyers and we were handling matters from there. We have a certificate which does not look correct because it is out of line with the other figures, we have no way of rechecking that is valid in my eyes and we do not know whether the test is the problem or not, but we cannot find anything wrong with the test. There is at this point a great deal more work to be done, in other words."

235. A bit later Mr Hamblen put to Mr Chalmers the one question which he relies on as amounting to challenging Mr Chalmers as to his honesty, thus –

"Q. I would also suggest to you, Mr Chalmers, that you deliberately did not disclose them in the knowledge and the intent that AIC would be misled thereby."

A. Again, I think you are wrong. No, I do not agree with you."

236. There were some questions about samples. Mr Chalmers distinguished between formally retained samples and testing samples ("two different things here"). If there was a dispute, in general reliance was placed on "retention samples", taken in duplicate or triplicate, with some sent with the vessel. He was neither a sampler nor an analyst, but his understanding was they were relied on because they were in the original condition with no danger of contamination. As for testing samples, they might very often be kept, but not for RVP because that test could never be repeated on the same sample. As for an RVP testing sample, that was spent, a slop, exhausted for the purpose it was drawn for, it could not be re-used, it was rubbish, a waste sample. He did not agree with Mr Loughead, although he was the "head man", in his view that the RVP testing samples had been "potentially important".
237. (I interpolate here that Mr Loughead had made a witness statement in January 2005, in Houston, where he was chief executive officer of Caleb Brett globally. He gave a statement because AIC had pleaded that he had been guilty of fraud because he knew by 17 April 1996 that the certificate was false (in certifying that the cargo met specification) or was reckless as to whether it was true or false. It will be recalled that Mr Rackham's email of 17 April, with the Cooper retest result of 9.33, had been copied to him. He said he remembered nothing about this, but added:

"Hearing about the matter now, I would say that:

a) I would have regarded the samples as potentially important and would not have regarded it appropriate to take them and use them.

b) It is not correct, technically, to retest a previously opened sample for RVP. Samples intended for RVP testing must be taken specifically for that purpose and used only once.

c) It would not be surprising that different results would be obtained if previously broached samples were used for testing."

It is plain from the whole of that passage that "potentially important" cannot be extracted alone. It appears to mean that the testing samples could possibly be important (in the event of a dispute?), but that that possibility must be qualified by the facts that the samples cannot properly be revisited and could throw up misleading discordant results. It will be recalled that Mr Lucas, unlike Mr Chalmers, had agreed with Mr Loughead's comment: see para 198 above.)

238. Mr Chalmers was shown the ITS manual for sample handling, and agreed that it provides for testing samples to be retained for 90 days as well. (I have referred to this manual above: see paras 189/192. Mr Chalmers was the only witness to be questioned by direct reference to any of the manuals or guidelines.)
239. At the end of Mr Hamblen's cross-examination, the judge had some 20 minutes of questions for Mr Chalmers. The judge again began by asking about relevant professional standards and whether they touched on the problem of using the wrong test method. Mr Chalmers said he would get hold of them, but that he did not think there was anything in them. The judge observed that in an international dispute any decision may have marked implications and the court would expect to be shown the market material.
240. The judge asked about the certificate subscript "Fuel meets specification", and Mr Chalmers confirmed that it was most unusual to find it on a certificate, he could not recollect seeing it on one before. The judge asked if, when it had been realised that the wrong test method had been used, it was "commercially appropriate" to stand by a certificate which said that fuel meets specification. Mr Chalmers said:

"I have thought about this a whole lot. Most of my thinking was based on what Mr Lucas understood about what was being asked of him. In the sense that I do not think they could change the certificate to make it better, which is what they were apparently asking, or what Mr Lucas believed they were asking, I do not believe they could have changed it. By the 17^th, we were in a full-blown dispute…Sadly there is no way of retesting…There is nothing that we can point to that Mr Mailey obviously did wrong in the test, which is why in my statement I said that the figure is right…It is a very difficult position to answer a question straightly on…I think we would have to set it aside as being a certificate in dispute. I think that would be my approach…What I mean is we would have to acknowledge, "Look, there is a dispute. The certificate you have is the subject of a dispute". I am not sure that we can simply scrap it, as it were, because Mr Lucas – in a sense I agree with him, this refers to shore tanks. There is a tremendous amount happens between the shore tanks and the ship's tanks…"

241. The judge returned to the same question a few pages further on:

"Q. Now, leaving aside the position that Mr Lucas was in, what do you consider should have happened in relation to this certificate? In appropriate commercial terms, should ITS have stood by it or should they have written a letter withdrawing it or what should they have done?

A. I think we should have been much clearer in what we were saying. We did tell them the wrong tests had been used. Sitting here today, I think I agree with you that we could have written them a letter spelling out exactly that, that the wrong tests had been used and that there was therefore, in view of the other tests in America and the result we have here, an area of doubt.

Q. And "the result we have here" means what?

A. The RVP result, my Lord.

Q. In this certificate?

A. Yes.

Q. So you are recognising – all I am asking for is your help – but you say in your witness statement quite rightly that you have enormous experience of the professional standards that apply in this important area.

A. Yes my Lord.

Q. Are you saying that as of today you recognise that you should have written a letter saying, "We can no longer stand by this certificate"?

A. Sitting here today, I think I would have to recognise that."

242. Sitting in court that day, and ignoring the position Mr Lucas had been in, Mr Chalmers recognised that the certificate should have been withdrawn since it had employed the wrong test and there were discordant results at New York. It seems to me that that is, in essence at any rate, the Veba Oil point. If the certificate should have been withdrawn, then that is because the wrong test had been used. Of course, if there had not been discordant results at New York, then in practice there would have been unlikely to have been any problem.
243. After Mr Chalmers had explained again how disputes of this kind were normally worked out, with referee testing of the discharge quality, or of the ship's retained samples, the agreement of a definitive quality, the striking of a deal and a financial claim negotiated and settled with ITS, the judge continued to worry at the problem:

"Q. Maybe this is unfair, and I certainly do not intend to be unfair, but it might be said that one problem was that ITS was standing by a certificate which said "Fuel meets specification"?

A. I would have to accept that. I would have to accept that, sir.

Q. So you would recognise that was unfortunate?

A. It is very unfortunate. The whole Whitaker conversation set the tone of things and created the understanding that what they were wanting us to do was something that we could not."

244. That answer parallels the answer he had given to Mr Hamblen cited at para 229 above. The contrast Mr Chalmers, as it seems to me, was making was between withdrawing the certificate (because it was based on the wrong test in the face of discordant results), something he would now accept as having been an appropriate way forward, and making a new certificate, ex post facto, which was something that he continued to think ITS could not do.
245. It will be seen below that some of the answers given by Mr Chalmers to the judge were regarded as significant.

25. The judge's findings as to Mr Chalmers

246. The judge set out the evidence of Mr Chalmers at paras 107/114 of his judgment. He made a number of findings about Mr Chalmers elsewhere in his judgment, but he never returned in substance to Mr Chalmers' evidence.
247. (a) Paras 107/114. I will in his case too set out the description of Mr Chalmers' evidence, as found in the judgment, in full:

"107. Mr John Chalmers

108. Mr Chalmers has a marine background. In March 1996 he was employed by ITS as Claims Manager. In addition he had responsibilities for claims for the Group at large. He also served as UK Quality Assurance Manager and Safety Adviser and had a wide role in Europe in terms of Quality Assurance. His title was Technical Manager Eastern Hemisphere/Safety Manager UK.

109. It is appropriate to make (and I do make) due allowance for the difficulties Mr Chalmers faced when giving evidence because of the passage of time. Nonetheless I have marked reservations about certain parts of his evidence.

110. The entry in Mr Rackham's logbook at 13.33 hours on 16 April ("JC re Kriti Palm. Harass UK need low RVP") I find probably refers to a telephone conversation between Mr Rackham and Mr Chalmers. In cross-examination Mr Rackham accepted that the note suggested that Mr Chalmers was anxious to get the results. The contemporary note in Mr Rackham's logbook is to be contrasted with Mr Chalmers' evidence that the Cooper re-tests were a "botched up job on a spent sample" and that he would probably have "blown up all over the telephone" when he became aware of the re-tests.

111. Mr Chalmers agreed that Mr Rackham was experienced and would know whether it was appropriate to re-test or not.

112. Mr Chalmers accepted that he was of the view that something was wrong with the result at Coryton because that result was out of line with other results.

113. When asked whether it was acceptable for ITS to say that it was standing by its certificate in circumstances where ITS knew there was something wrong with one of the published results, Mr Chalmers said "phrased like that, no". At a later stage his evidence was as follows:

Q. …should ITS have stood by [the certificate] or should they have written a letter withdrawing it or what should they have done?

A. I think we should have been much clearer in what we were saying. We did tell them the wrong tests had been used. Sitting here today, I think I agree…that we could have written them a letter and that was therefore, in view of the other tests in America and the result we have here, an area of doubt.

Q. And "the result we have here" means what?

A. The RVP result…

…

Q. Are you saying that as of today you recognise that you should have written a letter saying, "We can no longer stand by this certificate"?

A. Sitting here today, I think I would recognise that."

114. I will return to consider Mr Chalmers' evidence in some detail later in this judgment."

248. I observe: (1) The judge did not say what marked reservations he had about what parts of Mr Chalmers' evidence. It is possible to infer that the judge had reservations, in the light of his findings concerning the "Harass UK" telephone call, about Mr Chalmers' evidence concerning that and the value of the Cooper retest. The judge expressly contrasts the log entry with that evidence. It is not easy, however, to see what reservations the judge could have had about the other extracts from Mr Chalmers' evidence highlighted in the above citations. (2) Those citations could leave a reader of that part of the judgment with the impression that Mr Chalmers was accepting that it was the Cooper retest RVP result which caused Mr Chalmers to be recognising that ITS should have written a letter saying it could no longer stand by the certificate. However, a fuller reading of the relevant parts of Mr Chalmers' evidence (set out in the preceding section), given in answer to the judge's own questions, shows that Mr Chalmers is considering, at trial, the consequences of a certificate which had been based on the wrong test method and of the discordance of disport results with "the result we have here"… "The RVP result…" … "In this certificate" (see para 241 above and the critically partial citation by the judge reproduced at para 247 above).
249. (b) Other parts of the judgment. The judge said he would return to consider Mr Chalmers' evidence in some detail later in his judgment. That is an over-statement. (i) He did not do so when dealing with the significance of international and internal guidelines for the implied obligations he found applicable to ITS's contract (at paras 178/192). (ii) In finding, on the basis of Mr Revell's expert evidence, that the cargo would on the balance of probabilities have been off-specification if tested at loading by D5191 and that the certificate's D323 result was probably wrong, the judge also concluded, without further elaboration, that Mr Chalmers, as well as Mr Lucas, "knew, understood and appreciated" these matters (at paras 198/203). (iii) In repeating that finding (at paras 204/205), the judge added a reference to Mr Chalmers' acceptance that it was very unfortunate that Mr Lucas had said that ITS "will be standing by that certificate", without commenting further on the context of that acceptance. (iv) At para 275, in answering the issue "24. What was Mr Chalmers' knowledge of the re-tests and the results thereof by 17 April 1996 and thereafter?" the judge merely found that if he did not learn the results prior to his return to the office, he would certainly have known them upon his return. (v) I have already cited the judge's finding (at para 312) that Mr Chalmers made a deliberate decision not to disclose the Cooper retest and results knowing that he was under a duty to do so.

26. The issues considered by the judge

250. A list of 46 main issues, agreed by the parties prior to trial, was set out by the judge at para 164 of his judgment. From para 175 onwards, the judge considered these issues seriatim. On this appeal, however, the parties have concentrated mainly on the issues of deceit and deliberate concealment, and in that connection the question of what duties were owed by ITS.

27. Deceit: the law

251. The elements of the tort of deceit are well known. In essence they require (1) a representation, which is (2) false, (3) dishonestly made, and (4) intended to be relied on and in fact relied on. Each of those elements may of course require further elaboration. We are not on this appeal really concerned with the element of reliance, but we are concerned with the other three and in particular with the identification of the relevant representation and the question of dishonesty. These are the principal matters put in issue (see paras 21 and 24 above for the rival cases made on appeal).
252. At the basis of any claim in deceit is the representation in question. Its falsity, and the honesty of the representor, cannot begin to be considered until the representation in question has been identified. In the case of a written document, the representation can usually be pinpointed (unless questions of implication arise), but of course context remains everything. In the case of an oral representation, the identification may be a more difficult process, involving disputed testimony, but again context remains everything. In the present case, the oral conversation in question was recorded, so that evidential disputes as to what was said have not in practice arisen, despite the great length of time between the April 1996 telephone conversation and trial. However, two factors remain important. The first arises because the representation found by the judge, although based on express words ("We will be standing by that certificate"), was derived by implication as a matter of "overall impression" (judgment, para 283). The second is a general point relating to human conversation: although it is of course possible to be more or less deliberate about one's speech, nevertheless the natural ebb and flow of conversation as part of an essentially interactive process means that it differs significantly from a written document. It does not necessarily have a single writer's logic, it is not composed, and it cannot be read as a whole before its communication. For both these reasons, evidence of contemporary views of what the parties to the relevant conversation understood themselves to be saying or hearing may be of special importance.
253. It is sometimes said that the necessary representation must be unequivocal. That is too broad a statement to be accurate. Because dishonesty is the essence of deceit it is possible to be fraudulent even by means of an ambiguous statement, but in such a case it is essential that the representor should have intended the statement to be understood in the sense in which it is understood by the claimant (and of course a sense in which it is untrue) or should have deliberately used the ambiguity for the purpose of deceiving him and succeeded in doing so: see Clerk & Lindsell on Torts, 19^th ed, 2006, at paras 18-23 and 18-33; Akerhielm v. De Mare [1959] AC 789 (PC). As Cotton LJ said in Arkwright v. Newbold (1881) 17 Ch D 301 at 324:

"In my opinion it would not be right in an action of deceit to give a plaintiff relief on the ground that a particular statement, according to the construction put on it by the Court, is false, when the plaintiff does not venture to swear that he understood the statement in the sense which the Court puts on it."

254. It remains true, however, that in any case of fraud the dishonest representation must be clearly identified.
255. It is also standard law that to found an action in deceit the representation relied on must be one of fact. A statement of opinion will not suffice unless the deceit is in the fact that the opinion was not, or not honestly, held or in some further implicit dishonest misrepresentation of fact to be derived from the statement of opinion; and neither will a misstatement of law suffice save on the same ground that it involves implicitly a misstatement of fact, viz that the representor did not in fact entertain the opinion of law which he expressed. In this context Clerk & Lindsell state (at para 18-12):

"the test as to whether a statement of opinion involves such a further implied representation [of fact (e.g. that the person stating the opinion has reasonable grounds for his belief)] will involve a consideration of the meaning which is reasonably conveyed to the representee; and the material facts of the transaction, the knowledge of the parties respectively, their relative positions, the words of the representation and the actual condition of the subject-matter are all relevant to this issue"

citing Smith v. Chadwick (1884) 9 App Cas 187, Bisset v. Wilkinson [1927] AC 183, and Brown v. Raphael [1958] Ch 636 (CA).

256. As for the element of dishonesty, the leading cases are replete with statements of its vital importance and of warnings against watering down this ingredient into something akin to negligence, however gross. The standard direction is still that of Lord Herschell in Derry v. Peek (1889) 14 App Cas 337 at 374:

"First, in order to sustain an action in deceit, there must be proof of fraud and nothing short of that will suffice. Secondly, fraud is proved when it is shown that a false representation has been made (1) knowingly, (2) without belief in its truth, or (3) recklessly, careless whether it be true or false."

257. In effect, recklessness is a species of dishonest knowledge, for in both cases there is an absence of belief in truth. It is for that reason that there is "proof of fraud" in the cases of both knowledge and recklessness. This was stressed by Bowen LJ in Angus v. Clifford [1891] 2 Ch 449 where he said (at 471):

"Not caring, in that context, did not mean not taking care, it meant indifference to the truth, the moral obliquity of which consists in a wilful disregard of the importance of truth, and unless you keep it clear that that is the true meaning of the term, you are constantly in danger of confusing the evidence from which the inference of dishonesty in the mind is to be drawn – evidence which consists in a great many cases of gross want of caution – with the inference of fraud, or of dishonesty itself, which has to be drawn after you have weighed all the evidence."

258. And in Armstong v. Strain [1951] 1 TLR 856 at 871 Devlin J, after a full citation of passages in earlier authorities which stress the need for dishonesty (also called actual fraud, mens rea, or moral delinquency), said this about the necessary knowledge:

"A man may be said to know a fact when once he has been told it and pigeon-holed it somewhere in his brain where it is more or less accessible in case of need. In another sense of the word a man knows a fact only when he is fully conscious of it. For an action of deceit there must be knowledge in the narrower sense; and conscious knowledge of falsity must always amount to wickedness and dishonesty. When Judges say, therefore, that wickedness and dishonesty must be present, they are not requiring a new ingredient for the tort of deceit so much as describing the sort of knowledge which is necessary."

259. Moreover, whether it is in the matter of identifying the relevant misstatement or in the finding of a dishonest mind, it is necessary to bear in mind the heightened burden of proof which bears on the claimant, as discussed in cases from Hornal v. Neuberger Products Ltd [1957] 1 QB 247 to In re H (Minors) [1996] AC 563. In the latter case Lord Nicholls of Birkenhead said this (at 586):

"Built into the preponderance of probability standard is a generous degree of flexibility in respect of the seriousness of the allegation. Although the result is much the same, this does not mean that where a serious allegation is in issue the standard of proof required is higher. It means only that the inherent probability or improbability of an event is itself a matter to be taken into account when weighing the probabilities and deciding whether, on balance, the event occurred. The more improbable the event, the stronger must be the evidence that it did occur before, on the balance of probability, its occurrence will be established. Ungoed-Thomas J. expressed this neatly in In re Dellow's Trusts [1964] 1 W.L.R. 451, 455: "The more serious the allegation the more cogent is the evidence required to overcome the unlikelihood of what is alleged and thus to prove it."

260. None of these matters of law are in the least abstruse, and prima facie it can be assumed that a judge of the great experience of Cresswell J would have them all in mind. It is nevertheless to be observed that his reference in his judgment to what a finding of deceit requires is contained in a single short paragraph (para 279), with citation to an essentially inconsequential passage in Standard Chartered Bank v. Pakistan National Shipping Corporation (No 2) [1998] 1 Lloyd's Rep 684 at 704. Mr Gaisman makes complaint of this as something undermining the judge's findings and conclusions and rendering them flawed. It will be necessary therefore to consider carefully whether the terms in which the judge came to his conclusions on the case in deceit withstands or supports Mr Gaisman's complaint.

28. Deceit: the representation

261. I have set out in detail above the evidence and the judge's conclusions. I will now seek to summarise this material.
262. First, what was the representation relied on? The pleaded representations (para 23A of the re-re-amended particulars of claim) are complex and various. I will attempt to simplify, for the purposes of this appeal. AIC pleaded in the first instance that the certificate (not the telephone conversation) contained a false representation in saying that "Fuel Meets Specification". Next, as to the 17 April conversation, AIC pleaded that Mr Lucas expressly or impliedly repeated that representation "by refusing to accept that the Certificates were inaccurate and by stating that they would be standing by the Certificates". I would understand that pleading as essentially putting forward the certificate's "Fuel Meets Specification" as the critical representation, implicitly repeated in the telephone conversation by means of the matters just quoted. There was no pleading of a representation, express or implied, made in the telephone conversation, that the certificate was good, valid or reliable.
263. Secondly, how was AIC's case on representation put at trial, as reflected in the judgment? The judge set out AIC's case at paras 73/86 of his judgment. A case on negligent misrepresentation is referred to (at para 78) by reference to 17 April, that "ITS continued to represent that the fuel met specification and that Mr Mailey's results were reliable". That is, I think, the first reference to a representation as to the reliability of the certificated results. (Previously, AIC had merely pleaded, in support of its case that the critical representation was that made through the certificate's statement that "Fuel meets Specification", that in the telephone conversation Mr Lucas had "refus[ed] to accept that the Certificates were inaccurate".) Thus the judge summarised AIC's case on deceit at para 81 as follows: "Mr Lucas was at the very least reckless in maintaining that the fuel met specification and that the RVP result of 8.22 psi was reliable". That seems to parallel, as one would expect, the case on negligent misrepresentation.
264. Thirdly, how did the judge record his findings on representation? When he turned to set out his conclusions on deceit (at paras 276ff), the judge said (at para 283) that he accepted Mr Hamblen's submission that "the overall impression conveyed by Mr Lucas…was that the certificate for the regular cargo was and remained a good and reliable certificate". That reflects the way in which Mr Hamblen seems to have put his case at trial. However, the judge went on (at para 284) to speak of a representation in different terms, saying that "The representation that the certificate (which stated 'Fuel meets Specification') was and remained a good and valid certificate" was false. And (at para 288) he referred again to "the representation that the certificate (which stated 'Fuel meets Specification') was and remained a good and valid certificate".
265. Fourthly, did the judge err in this conclusion? Was he entitled to come to this conclusion?
266. In my judgment, the judge was not entitled to come to this conclusion. It was an impossible conclusion. If it was possible, the judge nevertheless plainly erred. My reasons are as follows.
267. (1) The representation found was imprecise and, in an action in deceit, unfairly so. As ultimately stated, the representation found was that the certificate "was and remained a good and valid certificate". But along the way the submission that was said to be accepted was that the representation was that "the certificate was and remained a good and reliable certificate". In either event, the representation was not, and could not be said to have been, expressly made, nor was it even said to have been implied, but it was derived as a matter of "overall impression". For that reason it may therefore not be altogether surprising that the representation vacillated between the "good and reliable" and the "good and valid" forms. It has to be observed, however, that in this context "reliable" and "valid" are not prima facie co-terminous. (The polymorphous word "good" adds nothing and has to take its shade of meaning from either "reliable" or "valid".) The certificate was undoubtedly invalid, not valid, because the wrong test had been used (Veba Oil), but that of course was not the complaint made on AIC's case in deceit. AIC knew that the wrong test had been used. It was not deceived in any way as to this aspect of the facts. If that had been the representation alleged (it was nowhere pleaded), the case in deceit would have been bound to fail, for, although a statement that the certificate was valid would have been incorrect (if one likes, false, but it seems better to say incorrect), that was a statement of law, not fact, and in any event, rightly or wrongly, Mr Lucas seems to have thought (but it is not entirely clear) that the use of the wrong test did not in itself invalidate the certificate. At any rate it was never suggested that Mr Lucas was dishonest in thinking, if he did, that the certificate survived invalidity despite the use of an admittedly wrong test. Moreover, it was AIC's case, accepted by the judge in the face of Veba Oil, that the invalidity of the certificate for present purposes was irrelevant (Mr Hamblen's reply submissions at Day 11.134).
268. As for "reliable", however, what does it mean to speak of a reliable certificate? The judge has not explained this, nor has AIC on this appeal. Were it to mean a valid certificate, one that is reliable to work its effect in the contractual context, then its meaning parallels that of "valid". The fact that the judge appears to have used the words "reliable" and "valid" as synonyms suggests that that was how he regarded the words, as being without distinction. Alternatively, and perhaps more literally, "reliable" could mean a certificate that was safe to rely on, in the sense that it was a certificate which could be used in the trade for the purpose for which it was intended without any reasonable doubt as to its effectiveness. The whole purpose of such certificates is to enable international traders to buy and sell certificated goods as being the goods so certificated. If a question mark opens up about a certificate, then those that deal in it may find themselves involved in a dispute, which is exactly what the certificate is designed to avoid. If that is the word's meaning, however, then the situation remains the same as above. The certificate was not reliable. The wrong test had been performed. The certificate spoke only to that test and no other. No one knew, by reference to the certificate, what result would be produced by a different test. And in any event the required specification test had not been performed. But, of course, the fact that such a question-mark had opened up about the certificate was in the open. That was not AIC's complaint in deceit.
269. A further alternative, and what I think Mr Hamblen perhaps had in mind in shifting AIC's ground from that pleaded to a representation about reliability, whatever the judge may have thought, is that "reliable" was being used in the sense of "accurate". Thus the positive representation, that the certificate was reliable, was substituted at trial for the pleaded representation, that fuel meets specification, to be derived from Mr Lucas's refusal to accept that the certificate was inaccurate. The complaint was that Mr Lucas, by saying that ITS would be standing by a certificate which itself recorded that "Fuel meets Specification" was implicitly misstating the facts, since the fuel did not meet specification and the certified RVP result of 8.22 was materially wrong, in that it should have been over 9. That would at least be consistent with the case made by AIC on dishonesty, accepted by the judge, that Mr Lucas was fraudulent by reason of his knowledge that, if retested by D5191, a result of over 9 would probably have been obtained and that therefore the certified result of 8.22 by D323 was "probably wrong" (since Mr Lucas also knew that the tests were comparable): see at para 283 of the judgment).
270. Among the difficulties with that further alternative version of the representation, however, are: (i) the judge's representation is stated in different terms ("good and valid"); (ii) the certificate in speaking of an RVP result and in saying that the fuel met specification was speaking to a particular test (D323) at a particular time, and was saying nothing about any result by reference to a different specification test, D5191, at a different time; (iii) Mr Lucas, by saying that AIC would be standing by its certificate, was saying nothing expressly, or, it seems to me implicitly, about what a different certificate, recording the result of a different test, might have said; (iv) it would be necessary to argue that Mr Lucas had, at least impliedly, represented that the certificated test result was a correct result, but such a representation, even if made, could only in context be a representation of opinion, for Mr Lucas, as Mr Whitaker knew, was not speaking to his own findings; (v) the question would then be whether or not Mr Lucas was dishonestly representing his opinion, a matter which was never pleaded nor investigated; and (vi) the matters discussed below.
271. (2) When raising the question of what could be involved or implied in the remark that "we will be standing by that certificate", or how the "overall impression" given by the telephone conversation might be encapsulated, consideration needs to be given, as the judge himself said, to the conversation as a whole. But a review of the conversation as a whole demonstrates a number of significant matters. First, it was common ground that the wrong test had been performed. Secondly, it was known and discussed that tests at New York had shown the fuel to be out of specification. Thirdly, there were demands by Mr Whitaker for a further certificate or statement that the fuel was out of specification; and in any event for compensation. Fourthly, the consequences of those factors were recognised as something on which AIC might well wish to be advised by lawyers, and Mr Lucas himself suggested as much to Mr Whitaker and offered to assist Mr Whitaker in obtaining such advice (see para 116 above). Fifthly, Mr Lucas said that he did not know whether there was any correlation between the two tests (in response to Mr Whitaker's statement that D5191 would produce a higher reading than D323), and (although the judge was to find that Mr Lucas did know that the tests were comparable) there was no pleading of misrepresentation in this regard. Sixthly, and of particular importance, in answer to Mr Whitaker's direct probing as to the inaccuracy of the D323 test, Mr Lucas repeatedly said he did not know and could not comment (see paras 111/112 above: "Well I can't comment on that over the telephone. I really don't know…I can't say that. I don't know…I can't comment on that. I can't say whether it is inaccurate or not"). None of those statements were the subject matter of complaint as being a false representation of Mr Lucas's thinking. Seventhly, and also of importance, given the stress placed by AIC on the circumstance that the certificate stated "Fuel meets Specification", is the fact that when Mr Whitaker put to Mr Lucas that he had a certificate saying "that it meets colonial specification" (the nearest that the conversation comes to dealing expressly with the certificate's wording), Mr Lucas replied "You have a certificate from us. I can't recall what it says" (in fact the certificate did not say that the fuel met colonial specification).
272. Pausing there, I would comment as follows. If one is seeking the "overall impression" of this conversation, in my judgment one would be bound to conclude that, whatever "We will be standing by that certificate" may mean or have been intended to mean, Mr Lucas was making it plain that he did not know whether or not the certificate was accurate or not (that was on the D323 test), and that he did not know how the two tests correlated. In those circumstances, he was expressly saying that he did not know whether or not, whatever the certificate said, the fuel had been within specification or would have been shown to have been within specification if tested by the right test. If that is what Mr Lucas was saying expressly, then it seems to me to be impossible to derive, from any combination of what Mr Lucas said about standing by the certificate and what the certificate said about fuel meeting specification and any overall impression of the conversation, an implied or derived representation that the certificate was good and reliable or good and valid in any meaning of those terms relevant to AIC's complaint in deceit. This factor is highlighted by AIC's pleading that the critical misrepresentation (that fuel meets specification) was to be derived from Mr Lucas's refusal to accept that the certificates were inaccurate. He did not refuse to accept that the certificate was inaccurate so much as state his personal agnosticism.
273. Of course, given the judge's ultimate findings about Mr Lucas's knowledge, AIC might possibly have made a case in deceit to complain about Mr Lucas's representations that he did not know whether or not the certificate was inaccurate and that he did not know how a D5191 test would correlate with a D323 test. Or AIC might have complained that the crucial representation to be derived from the conversation was that Mr Lucas knew nothing to suggest that a D5191 test would have produced a result out of specification. But that was not the case that was made. If it had been, it might have led to a more detailed examination of Mr Lucas's knowledge (as to which see below) than the basic trial inquiry as to whether he knew about the Cooper retest results. It might also have led to a critical examination (which never appears to have taken place) as to why the Cooper retest on broached and unsealed samples was of greater or particular importance given what was already known by all parties about the formally retained and tested samples which travelled with the ship. Or an entirely different case in deceit might have been developed (somewhat analogous to the subsidiary case on continuing duties which AIC made on the hypothesis that Mr Lucas, and Mr Chalmers, had only learned the Cooper retest results after the telephone conversation, as to which see below), to the effect that there was a positive duty on ITS to disclose to AIC the Cooper retest results: so that Mr Lucas's deliberate failure to disclose the results in some way amounted to a dishonest misrepresentation or other fraudulent conduct in itself.
274. (3) Given the difficulty of identifying with the necessary precision the representation upon which Mr Lucas has been found guilty of deceit and the possible ambiguities of the situation, I judge it to be particularly important to consider the contemporaneous communications between the parties immediately following the conversation in question, to see whether they throw any light on the above matters. What did Mr Whitaker take away with him as his understanding or overall impression of the conversation? What did Mr Lucas think that he had been saying to Mr Whitaker? I refer to paras 123/129 above. It is clear from those communications that the critical question for both parties was Mr Whitaker's request for a further or amended certificate (or other statement) to the effect that the fuel was out of specification. Mr Whitaker had regarded that as his right (for otherwise litigation was threatened) because he knew that the wrong test had been performed by ITS, considered that that threw a question mark over the validity of the existing certificate, and also considered that the New York tests had shown the cargo to be non-contractual. Mr Lucas was put out, even outraged, by the request for an amended certificate. But for present purposes perhaps the most significant passages in the exchange are these. Mr Whitaker said:

"We understand that you "stand by your results" even though you cannot now be certain that it does or does not meet colonial specification per Lucas/Whitaker phone conversation…"

275. Although that is in several respects a mis-gloss of the conversation, we know, by reference to the transcript, that it is essentially correct – and in any event it is what Mr Whitaker took away with him – in linking the remark "We will be standing by that certificate" with Mr Lucas's repeated statements of agnosticism as to any particular results on either of the tests. Mr Lucas replied:

"We, as an inspection company, can only issue certificates of quality stating what we found by the test methods we used…In all cases, it was plain on the face of the certificates that the test method was ASTM D323. The results reported were what we found."

276. That is saying: "You know, from the face of our certificate, that we have used the wrong test. The results are what we found. But we are unable to certify retrospectively."
277. No complaint of deceit is made about what Mr Lucas said in these written communications. Given the imprecision of a lengthy telephone conversation, and Mr Whitaker's own précis of it in his telex, I would think that if a case of deceit was to be made it would have to be made on those written documents, which are intended to sum up the parties' respective positions. AIC's and the judge's virtual disregard of these communications in my judgment shows the difficulty of the judge's finding of the representation upon which he has founded liability in deceit. No one has suggested that, whatever errors may have led to ITS's certifying an RVP of 8.22, that result was not honestly made. Ultimately, all that Mr Lucas was saying, as a matter of overall impression, was: these are our findings, made by the wrong test, we cannot rewrite them; we do not know whether the fuel meets specification by the right test; you say you will sue us, but we do not admit liability.
278. In these circumstances, I think that the absence of Mr Whitaker's evidence is in itself critical, although I am in any event sceptical as to how his evidence might have assisted. It is a case where the dictum of Cotton LJ which I have cited (at para 255) above from Arkwright v. Newbold is wholly apposite. In my judgment, the judge's finding of a misrepresentation on which to found a case in deceit is an impossible one; alternatively, it is a case in which he has plainly erred. In any event, on this aspect of the matter, nothing in particular turns on oral evidence.
279. (4) In the whole of this connection, I have not overlooked evidence given by Mr Lucas to the judge such as that described at para 202 above: where Mr Lucas said it was his view at the time and still remained his view that the certificate was valid and that the fuel met specification (an answer relied on by the judge at para 282). That evidence, however, cannot affect what it was that Mr Lucas said to Mr Whitaker, for which we have the contemporary transcript; and it is certainly not evidence on which any case of dishonesty can be built. It would be unfair to regard that answer as being something that Mr Lucas was maintaining at the time, especially given his overall evidence and the terms of the transcript. Moreover, it is clear from other answers there discussed that Mr Lucas's views in this context are involved with questions of law and opinion.
280. The primary case in deceit must therefore fail. Mr Lucas did not make the representation which the judge has found that he made; and in any event, the judge has not found with sufficient clarity a representation of fact which was false in a sense which AIC sought to make the basis of its claim in deceit. That in my judgment is an end to the case in fraud. But I would go on briefly to express my grave concerns about the judge's finding of dishonesty. I do so, on the hypothesis of the representation that he has found, and on the further hypothesis that "good and valid" involves a representation of fact of a kind which in some way was undermined by the Cooper retest results. I confess that there is artificiality in those assumptions.

29. Deceit: dishonesty

281. The first question which arises under this heading is whether Mr Lucas was told of the Cooper retest and its results. Despite the evidence of both Mr Rackham and Mr Lucas to the contrary, I would not disagree with the judge's conclusion, if indeed it was open to me to do so, that the former did brief the latter, prior to the telephone conversation, about the Cooper retest and the resultant figure of 9.33. That is essentially because of the probabilities of the thing. It was Mr Rackham himself who had instituted the Cooper retest. And it was Mr Rackham himself who had calculated the final figure of 9.33 (albeit inaccurately). I would also agree with the judge that the better construction of the entry in Mr Rackham's log at 1333 on 16 April (the "Harass U.K." entry) was that it was a note of Mr Chalmers' reaction on hearing of the Kriti Palm problem and the Cooper retest (at a time, I would however emphasise, when it was expected that the retest would confirm an RVP figure under 9). In those circumstances, it would be probable that Mr Rackham would tell Mr Lucas about the Cooper retest. Although Mr Rackham said in evidence that he did not tell Mr Lucas, he accepts that he sought to give him as good a briefing as possible. Nearly ten years after the event, it would be very difficult for any witness to be clear in his own mind on such a matter: and Mr Rackham's memory was obviously not of the best, because he had even forgotten that it was he who had set up the Cooper retest himself. Mr Rackham was not criticised for these difficulties about his evidence: he was by the time of trial independent of ITS and he had given a witness statement to both sides.
282. Mr Lucas, of course, also said that he had not been told by Mr Rackham about the Cooper retest. He also said that if he had been told, he would have told Mr Whitaker, and he used that as support for his evidence on this topic. It was perhaps a natural thing to say, despite other, inconsistent, evidence he gave about the "stone wall" technique in responding to notice of a claim. It is a natural thing to say, because it is a kind of bright line test of one's recollection, when being pressed on a matter as old as this. However, it has the flavour of reconstruction, indeed it is reconstruction. By the time of trial, all the possible implications of these events had no doubt been endlessly replayed in his mind.
283. Assuming, however, that Mr Rackham told Mr Lucas of the Cooper retest, the critical, second question, however, is whether the full implications of that information were conscious to his mind (in the important sense discussed by Devlin J in Armstrong v. Strain) so as to render him potentially dishonest in anything he said. For these purposes, and especially where a representation derived by implication or impression is concerned, a judge must make a careful assessment of all aspects of the situation. If Mr Lucas was told, did he take it in? Did he understand the full implications of what he was told? He would have been given a considerable amount of information in a hurried, urgent, briefing. There were two parcels involved, even if, for the purposes of this litigation, all attention is now concentrated on the regular parcel alone. That was not the position at the time, with complaints being made about both parcels (see paras 130/131 above). There had been a number of D5191 tests on the fuel. Of particular importance, one might have thought, was the test of the retained composite blend sample, ex shore tanks, which had travelled with the ship and had been tested in New York. That was far more a suitable candidate for consideration than anything done internally by ITS alone on a sample which had been used, was broached, had been left unsealed, was open to contamination and could not be properly used for re-certification (as everyone agrees). As for ex ship samples, there was always the uncertainty of contamination, at any rate until such stage (not yet reached) as that could be confidently eliminated. If the uncontractual D323 tests mattered at all, what was the relationship between that and any D5191 test? What in any event was the position where ITS had carried out an uncontractual test? That test may have found what it found, but what was the value or significance of that? Whatever its significance or validity, could it be subsequently falsified by any subsequent test by a different method? And even if it could, could any inference of negligence or responsibility be attached to the original tester?
284. What is remarkable is that even after all the time, effort, expense, collection and recollection of evidence, legal expertise and analysis, the trial displays, to my mind, a still inchoate understanding of the problems set up by an uncontractual certificate which speaks to a test which does not conform to the test required by the ruling specification. Given these difficulties in the leisure of litigation, what might occur in the immediacy of an urgent briefing of a person entirely new to the problem, followed by a telephone call from a demanding client?
285. What would have been Mr Lucas's motive for being dishonest? What had he personally, or ITS, to gain by any dishonesty? He had nothing whatsoever to gain, he was being involved in the Kriti Palm for the first time. As for ITS, this was one among countless disputes. What had he, or ITS, to lose by dishonesty? Everything. Motive, of course, is unnecessary for dishonesty; but motiveless dishonesty by a senior and experienced manager is difficult to comprehend.
286. What was the context of this dispute? Whatever be the disputed status of the broached testing samples retested by Mr Cooper, the as it were official retained, listed, sealed samples sent with the ship, or kept in duplicate by ITS, were the samples which the parties must have intended to be consulted in the event of a problem such as that which occurred in the case of the Kriti Palm. Such consultation of the samples (it had already begun in New York) was or would be performed by an independent tester, in the presence or with the knowledge of all parties who might wish to be involved ("referee testing"). It would also involve Mobil as well as AIC. ITS could not deal unilaterally with either client. Mr Lucas (and Mr Chalmers) would know this and expect it (as was their evidence). In such circumstances, if there was any responsibility on the part of ITS, its managers could not hope to evade it. But equally, ITS could not deal with AIC unilaterally, without the full involvement of Mobil.
287. In these circumstances, what was the importance of the Cooper retest? It was initiated by Mr Rackham, and, on the basis of "Harass U.K." (but that alone), Mr Chalmers showed interest in it. Mr Lucas, however, had not been involved in it at all. Moreover, it is one thing to show interest in a different D5191 test which would confirm a low reading (under 9) by a D323 test, at a time when that was the expected result (as Mr Chalmers appears to have done), and another thing to have a test result consistent with the high reading of the D5191 tests conducted in New York. ITS might feel comforted by the former, and disappointed by the latter. But in either event, there were discordances which could not immediately be explained. And if a new D5191 test was relevant, then for the reasons explained above there were other samples of more significance than those tested by Mr Cooper. None of this seems to me to say anything about what ITS's obligations, if any, in respect of the Cooper retest were (but see below). The judge described the Cooper retest results as a "key piece of evidence, highly relevant to the conclusion" and regarded as such by Mr Lucas. It is not apparent why the Cooper retest was "key" in a sense beyond the more formal tests conducted at New York, or which could have been carried out in the UK. It is not apparent either why the judge found that Mr Lucas then and there concluded, as the judge did after trial, that the Cooper retest results were "key". It is entirely unclear upon what basis the judge so found. Even upon the reconstructed hypothesis that he had known (ie in the important sense of the word) about the Cooper retest, Mr Lucas in his evidence had repeatedly stated the very limited matters which could be derived from it and its unreliability (in its proper sense and in the sense of not being certifiable); and Mr Chalmers gave similar evidence. On the same basis of his hypothetical knowledge, Mr Lucas was prepared to accept that his conversation would be misleading without raising the matter of the Cooper retest, but that tells one nothing about his true knowledge and understanding at the time.
288. Thirdly, what were the judge's findings in respect of all these matters? In my judgment, the judge does not engage with these questions, or does so barely. I have pointed out above that the judge's findings both as to Mr Lucas's knowledge and as to his dishonesty are dealt with in a single passage at paras 276ff and essentially at paras 283/289. His findings as to knowledge are, to my mind, seriously undermined by the fact that they essentially reproduce his conclusions on the balance of probabilities as to the facts relating to the cargo, its samples, and the two test methods, having heard expert evidence at trial: see his judgment at paras 198/203 and my judgment at paras 90/95 and 212 above, and also at paras 164/168. At paras 198/203 he proceeds directly from his conclusions on the expert evidence to his brief finding (at para 203) that Mr Lucas and Mr Chalmers "knew, understood and appreciated the matters set out in the last four paragraphs", but does so without any explanation or reasons at all. In effect he does the same at para 283 (d) to (g), where he repeats his findings at paras 198/202, again ascribing this knowledge to Mr Lucas and Mr Chalmers as of 17 April, but again without any reasons; and on this basis extending his findings to the matters in para 283 (h) and (i) as well. In effect, this is mere recapitulation. I repeat what I have just said about the judge's finding that Mr Lucas regarded the Cooper retest results as "key". His findings as to dishonesty, moreover, are essentially non-existent: for he proceeds directly from his recapitulation of Mr Lucas's knowledge to his conclusion that Mr Lucas was reckless as to the truth, ie was dishonest (at para 284). In doing so, he discusses none of the issues raised above. In effect, he derives both knowledge and dishonesty of the protagonist of the telephone conversation of 17 April 1996 directly from his own preference for one expert over another at trial in 2005, essentially without further reasoning. Moreover, he does not explain why his finding of dishonesty is expressed in terms of recklessness rather than knowledge, which raises a concern that negligence rather than dishonesty might have intruded into his thinking.
289. At paras 285/7 of his judgment, having already come to his conclusion as to dishonesty, the judge cited some extracts from Mr Lucas's cross-examination as serving to confirm that conclusion (see at para 213 above). The essence of those passages, however, is that Mr Lucas was there accepting, on the hypothesis of his knowledge of the Cooper retest results, that his conversation would have been misleading without some, albeit highly qualified, reference to them. However, I am unable to understand how the mere citation of such passages explains Mr Lucas's knowledge or dishonesty. In effect, neither his knowledge nor his honesty were being challenged in these passages: as the judge himself accepted towards the end of the cross-examination when Mr Mildon laid down his marker about the absence of such challenge. The judge had previously said that he had "marked reservations about important parts of his evidence" (at para 105), but he never identified what those parts were. It might be inferred that it was those parts of his evidence where Mr Lucas did not accept that he had been told of the Cooper retest, for the judge found that he had been told and had understood its full ramifications, as the judge had found them to be at trial. However, the judge never states that he considered Mr Lucas had lied to him, and the language of unidentified "marked reservations" falls a long way short of a finding that he had lied on such a central point.
290. I repeat what I have said above at para 168: it is not clear from the judge's judgment why Mr Lucas was dishonest, when Dr Marshman, who was not criticised as an expert, thought that it was not possible to confirm that the cargo was off-specification at loading, and that the Cooper retest results were invalid.
291. As for Mr Chalmers: although he was not a party to the conversation of 17 April, he was within at any rate some five or six days (see para 220 above) back at his office and catching up with the Kriti Palm affair. (It is inappropriate, if it matters, to regard him as being back in the office almost immediately after the telephone conversation.) He was also accused by AIC of deceit, on the basis of a continuing duty to correct the false representation made by Mr Lucas on the telephone. Although the judge does not appear to have gone so far as accepting that case against Mr Chalmers, nevertheless the case in deceit against Mr Lucas cannot be dealt with isolated from the evidence of Mr Chalmers. As it is, the judge made the same findings of knowledge against Mr Chalmers, equally unreasoned, for the findings were made in the same passages highlighted above, and also spoke of marked reservations about his evidence as well, again without identifying the passages in question.
292. I would refer to paras 219/249 hereof above, where I deal with Mr Chalmers' evidence and the judge's findings in reaction to it. In my judgment, I can find very little, if anything, in that evidence to begin to support a view of Mr Lucas as having been dishonest in his telephone conversation, or of what amounts to the alleged complicit climate of dishonesty existing between those two executives. Time may have led to a process of reconstruction in places in Mr Chalmers' thoughts about the subject, but on the whole I find it difficult to read his evidence without being struck by Mr Chalmers' moderate (and to me persuasive) objectivity about the problem. He said that, in the context of the issue with Mr Whitaker and his demand or request for a new certificate, the Cooper retest was inappropriate and its results were of no value. He accepted the hypothesis, in the context of the analysis exposed by the litigation, that if ITS knew that a certified result was wrong (my emphasis: Mr Chalmers accepted neither hypothesis), then it would not be acceptable for ITS to say that it was standing by its certificate (see para 233 above). Given what I have said above about the issue of knowledge, however, that hypothesis is hardly a basis for any findings of dishonesty (compare Mr Chalmers' answer to the sole question in which his honesty was challenged, at para 235 above): and particularly because in context the whole problem was overlaid with Mr Chalmers' attempts to see his way towards an analysis of the situation in terms of the problem that was obvious, that of the wrong test having been performed: see paras 234, 240/244. In my judgment, it is of no assistance in the prosecution of a case in deceit, to talk merely in terms of what might be "commercially appropriate" (see para 240 above) or of what "should" (with hindsight) have been done (see para 241 above): both in principle, because what one needs is a contemporaneously guilty mind, and above all because, in this context, the underlying problem which colours everything is the performance of the uncontractual test. However that problem was analysed at the time or in retrospect, it was always there as uncontractual conduct which went to the heart of the certificate and draws the disquiet of disease over the situation. Of course it would be inappropriate or wrong in some sense of those words to insist on a certificate which, because it was based on an uncontractual test, was in truth invalid and therefore, however viewed, raised an issue. However, as stated above, the disease was patent, as were the discordant results in New York, and that is not what the case in deceit was concerned with. There are remedies enough for such an uncontractual certificate.
293. In all these circumstances, if there had been clarity as to the fraudulent representation alleged, I would have had such serious concerns as to the finding of dishonesty, ie dishonest knowledge or recklessness, in the absence of adequate reasoning, as to be deeply troubled as to whether the finding of dishonesty could stand on appeal: despite the authorities which remind the court of appeal of its limited role in reviewing such findings of a trial judge, of which I have at all times been conscious. As it is, given the view I have taken of the inadequacy of any proper identification of the underlying representation, the finding of deceit cannot stand in any event. It therefore seems to me to be all the more difficult to state a definitive position in connection with a finding of dishonesty in respect of a hypothetical misrepresentation which has not been identified or proved. In effect, without a relevant misrepresentation, dishonesty cannot be asserted let alone proved. Ultimately, in the absence of a properly reasoned finding of deceit, it seems to me that the judge's conclusion cannot stand. My own judgment would have been that there was no deceit: more muddle and fallibility than fraud. In these circumstances Mr Lucas and Mr Chalmers are in any event fully entitled to consider themselves acquitted of any shadow of a finding of any dishonesty on their parts.

30. A third element; and the respondent's notice

294. I refer to paras 216/218 above, and for the present address the question raised on AIC's respondent's notice as to whether Mr Lucas and/or Mr Chalmers were guilty of deceit in failing to correct the terms of Mr Lucas's pleaded misrepresentation (during his telephone conversation) at some shortly later date when, if not known before, each of them learned of the Cooper retest and its results.
295. Mr Hamblen accepts (para 220 of the respondent's skeleton argument) that the judge "did not make an express finding as to whether Mr Lucas and Mr Chalmers were dishonest in failing to correct, qualify or withdraw the certificates". As Mr Gaisman fairly observed, that is not a good start for AIC to have to make. In such circumstances it ought to be extremely difficult to go beyond the findings of the judge, irrespective of but also a fortiori because of the deep concerns I have expressed above.
296. As for Mr Chalmers, the submission is in any event an impossible one. He made no representation of any kind to Mr Whitaker. He never had a conversation with AIC. He was in no way responsible for Mr Lucas's conversation, even if he received a transcript of it. In the absence of any representation by him, he could not be guilty of deceit. It is true that a false representation, honestly made at the time of its making, but subsequently discovered to be false, may impose the obligation on its maker to correct the falsehood, if it is still timely to do so: see Chitty on Contracts, 29^th ed, 2004, at para 6-017, Clerk & Lindsell at para 18-21. But that only applies to the representor. That obligation will need to be considered in the case of Mr Lucas. Mr Hamblen may also possibly rely in this context on his submission, accepted by the judge, of a duty on the part of ITS to inform AIC of the Cooper retest. However, even if such a duty was owed, so that there was a duty to speak, that could not in my judgment impose on Mr Chalmers, who had said nothing and had no contact with AIC, a duty to correct Mr Lucas's statements.
297. As for Mr Lucas, the position might arguably have been different if he had made the misrepresentation which the judge found him to have made. But that necessary substratum has fallen away. In any event, it would be necessary to show that, even when Mr Lucas had learned of the Cooper retest and its results, he had done so in circumstances which answered all the deep concerns which I have expressed above about the proof of any dishonesty. It would be necessary for him to have been conscious of the representation in question, and conscious of or at any rate reckless about its falsity. All this in circumstances where his conversation had been overtaken by the exchange of written communications between him and Mr Whitaker. In my judgment, that dishonesty was never proved, if indeed it was ever made the subject of proper challenge.
298. As for the duty to inform AIC: if the duty existed, a question which I will discuss below in the context of deliberate concealment, then I would be prepared to assume, for the sake of argument, that a case in deceit might be made against ITS where there had been a dishonest failure to fulfil that duty: something which would I think require a conscious evasion of that duty, dishonest concealment, or what Lord Bingham of Cornhill described as "deliberate and dishonest or reckless non-disclosure": see HIH Casualty and General Insurance Ltd v. Chase Manhattan Bank [2003] UKHL 6, [2003] 2 Lloyd's Rep 61 at para 22. But, for the reasons already advanced, I do not consider that such an additional conclusion is justified on the limited, and in my judgment flawed, findings and reasoning of the judge.
299. It follows that the respondent's notice in this respect fails.
300. Further aspects of the respondent's notice, in particular the submission as to continuing duties (outside deceit) designed to carry AIC beyond the date of 17 May 1996 which is the prima facie cut off date for limitation purposes, will be considered below, following discussion of the second main issue, that of deliberate concealment and section 32 of the Limitation Act 1980.

31. Deliberate concealment

301. It has to be acknowledged that the issue of deliberate concealment arises in a surprising context. It was known almost immediately upon arrival of the Kriti Palm at her destination that the certificate had been based upon the wrong test, an uncontractual test, and that the specified test which was vital to the contractual specification of the fuel had not been used. Whether or not the legal consequences of that fact were also immediately apparent to the parties, they were certainly capable of being so, and at that time. In truth, the certificate was invalid, the fuel was uncertificated, whether or not it met its specification had yet to be established, and ITS was therefore in breach of contract and had conducted itself negligently. AIC knew everything it needed to bring its claim. AIC had six years to bring that claim, to which there was, as it seems to me, no adequate defence.
302. AIC also knew upon the vessel's arrival in New York that the fuel was testing out of specification on the correct D5191 test. That was the result, whether the test was made on the fuel ex the vessel's tanks at disport, or, more significantly, on the fuel ex ship at loadport, or, most importantly, on the ex shore tank samples which had been sealed at the loading port and sent with the vessel as official samples of the contractual quality. No question of contamination could arise on the sealed samples (and the question of contamination on vessel or voyage was quickly dispelled: see para 151 above).
303. In these circumstances, it is surprising that AIC did not bring a claim against ITS within the limitation period of six years. While the Cooper retest provided further D5191 results which were also above 9, AIC's ignorance of those results until June 2004, when they emerged in disclosure in these proceedings, neither prevented AIC from bringing those proceedings (which were already on foot), nor explains AIC's failure to bring those proceedings until after the prima facie limitation period had expired.
304. Of course, those proceedings could not have referred to the Cooper retest results until they were known about following disclosure. But that is the common occurrence of material being disclosed in the course of proceedings which assists the claim that has been made.
305. It might have been thought that these were unpromising circumstances for a case being made to extend the limitation period by reference to section 32(1)(b) of the 1980 Act and the concept of deliberate concealment. In order to try to manoeuvre around these difficulties, AIC has gone to intricate, almost exquisite, lengths to attempt to plead new causes of action which seek to depend not on the real gravamen of AIC's complaint, viz that ITS had provided an invalid certificate and thus had failed to certify the fuel which it had been required to test, or had negligently performed even the uncontractual D323 test which it had employed, but on the non-disclosure of the Cooper retest and its results.
306. It also has to be acknowledged that the issue of deliberate concealment arises on facts and pleadings of considerable complexity and uncertainty. The judge appears to have treated the issue as obiter (see para 296 of his judgment: "In view of the finding of deceit it is not strictly necessary to consider s 32(1)(b) and s 32(2). Nevertheless I add the following for completeness."); and perhaps for that very reason he did not consider it upon any alternative hypothetical factual basis, eg that he had been wrong to make a finding of deceit. It follows that he was approaching the issue of deliberate concealment on the basis that Mr Lucas had been dishonest. If that finding cannot stand, as in my judgment it cannot, that makes for grave uncertainty and difficulty about the legitimacy of his conclusions under this heading.
307. A further difficulty is that in his conclusion on deliberate concealment the judge nowhere identifies in respect of which pleaded causes of action he finds that deliberate concealment prevents the running of the limitation period. On the contrary, he appears to have regarded the Cooper retest and results as critical to all AIC's various pleaded causes of action (see generally at paras 316/331 of his judgment). This is important because the purpose of section 32(1)(b) appears to be designed to cater for the case where, because of deliberate concealment, the claimant lacks sufficient information to plead a complete cause of action (the so-called "statement of claim" test). It is therefore important to consider the facts relating to an allegation of deliberate concealment vis a vis a claimant's pleaded case. However, the judge appears to have regarded the deliberate concealment of the Cooper retest and results as relevant to any claim at all. Thus:

(1) at para 317: he said that ITS deliberately concealed "a fact relevant to AIC's right of action against ITS". The point is made quite generally.

(2) at para 324: he said that AIC's "allegations of breach of contract/negligence…are now to a critical extent premised on the Cooper re-tests and the results thereof". Again the point is made quite generally.

(3) also at para 324: he said that AIC "were not able to plead in 1996" that the certificated results were probably wrong. I am unable to understand why the judge makes that assertion, given the disport results obtained by test D5191. It appears that this assertion may be involved in the judge's mind with the idea that in 1996 AIC believed that the D5191 test might have given a higher reading than the D323 test and thus that D5191 results did not by themselves invalidate the certificated D323 results as such. If so, this is irrelevant, since the Cooper retest, like the disport retests, were by D5191 and thus were neutral on that question. In any event, AIC was able to plead, and did plead prior to learning of the Cooper retest, that the certificated results were probably wrong, eg (at para 22 of the amended particulars of claim dated 1 August 2003) that "In breach of duty the Defendant failed to exercise reasonable care in carrying out the analysis and in issuing the Certificates in respect of the cargo…Further, had the RVP been tested in accordance with ASTM D5191 it would have shown a result in excess of the contractual maximum of 9.0 psi". I will revert below to the puzzle created by the judge's assertion. For the present, it is again to be observed that the judge's point is wholly general.

(4) at para 325: the judge said that the Cooper retest results "established on a balance of probabilities" that the certificated tests significantly understated the RVP. But nothing was "established" until trial. Establishment of facts on the balance of probabilities is the burden of trial. The question would prima facie appear rather to be what could have been pleaded: again, I will revert to this below. Again, the point made is entirely general.

(5) at paras 326/329: the judge made further points as to the effect of available evidence on the establishment of facts on the balance of probabilities. In particular, at para 328 he repeated his finding, made earlier, that the Cooper retest results were "a key piece of evidence, highly relevant" to the trial conclusions, and that "viewed together" with the disport results they confirmed that the certificated D323 results were probably wrong (para 327). Again, the points made are entirely general.

308. It would appear therefore that the judge was applying a test relating to the usefulness and importance of the Cooper retest results as evidence for the proof of AIC's case at trial. In my judgment, as will appear, that is a wrong test.
309. A further difficulty is that the judge never really identifies the factual basis of his finding of deliberate concealment. Of course, I quite see that, on the finding of dishonest knowledge against Mr Lucas for the purpose of his reckless fraud, and a finding of complicit knowledge against Mr Chalmers, it is an understandable and perhaps easy progression, even if the detail of it was never spelled out, to infer that, if the Cooper retest was not subsequently disclosed, then that was due to a conscious and deliberate decision not to do so. If, however, as I have sought to explain above, the findings of conscious knowledge and dishonesty can not be supported, the basis of that inference goes. Even so, the inference would struggle against the circumstances relating to the decision taken in April 1997 with Mr Pople's assistance and on his advice that the matters relating to the Cooper retest were privileged from disclosure (see paras 155/156 above).
310. As it is, how did the judge put his finding of deliberate concealment? Earlier in his judgment the judge had held that ITS owed a (legal) duty to AIC to inform it about the fact of the Cooper retest and its results: at paras 260/269, in answer to trial issues 21 and 22 (Should ITS have informed AIC and Mobil about the fact of the re-tests and the results thereof and was ITS in breach of duty in failing to do so? Should ITS have informed AIC about the fact of the re-tests and the results thereof in the telcon on 17 April 1996, and/or subsequent to the telcon, and was ITS in breach of duty in failing to do so?), he said that such a duty existed and had been breached. He repeated (at para 263) that the results of the Cooper retest were "a key piece of evidence", highly relevant to his trial conclusions about the RVP quality of the fuel, and that "Mr Lucas and Mr Chalmers knew, understood and appreciated" such matters. His finding of the duty and its breach appears to have been premised on such knowledge, for he only finds that Mr Lucas and Mr Chalmers "should have informed" AIC and Mobil, and therefore breached that duty, as from the time of that knowledge, whether that was during the telephone conversation of 17 April or "as soon as they became aware of the results" (at para 269). He also found that both Mr Lucas and Mr Chalmers were aware of the retest and its results at latest "very shortly" after the telephone conversation (ibid).
311. Later in his judgment, when considering deliberate concealment, the judge said (at para 312):

"I find that Mr Lucas and Mr Chalmers made a deliberate decision not to disclose the fact of the Cooper re-tests and the results thereof to AIC and Mobil in circumstances in which the retests and the results should have been disclosed to AIC and Mobil, and Mr Lucas and Mr Chalmers knew that the re-tests and the results thereof should have been disclosed to AIC and Mobil. I further find that Mr Lucas and Mr Chalmers were aware that ITS was under such a duty and that Mr Lucas and Mr Chalmers made a deliberate decision not to disclose the re-tests and the results thereof."

312. Thus the judge found knowledge of the Cooper retest and results and a full understanding of their significance as key evidence, a duty to inform born of that knowledge, knowledge of that duty, and a deliberate decision not to disclose what was known ought to be disclosed. The suggestion of these findings appears to be that, irrespective of any decision taken by Mr Lucas personally during his telephone conversation with Mr Whitaker, thereafter the decision not to inform AIC, or rather the decision to conceal the relevant information from AIC, was the joint complicit decision of both Mr Lucas and Mr Chalmers. Other than the fact of the absence of disclosure to AIC in the light of the knowledge found, I am left wholly uncertain of the basis of the judge's finding of such complicit deliberate concealment on the part of Mr Lucas and Mr Chalmers.
313. I have already said that in my judgment the judge's findings of dishonest knowledge as of the time of his 17 April telephone conversation cannot stand against Mr Lucas; and that there is no reasoned explanation of a finding of total knowledge on the part of Mr Lucas or Mr Chalmers of the judge's post-trial conclusions as to the quality of the cargo and Mr Mailey's D323 certificated tests. A fortiori, there is no explanation of a finding of such total knowledge recognising the Cooper retest as a critical piece in the evidential jigsaw.
314. In these circumstances, I think that the judge's conclusions as to deliberate concealment cannot stand in any event. They are too seriously undermined by his errors in relation to the case in deceit and his findings of dishonest or guilty knowledge. However, a more detailed consideration of the elements of the statutory test of deliberate concealment demonstrates, in my judgment, that the errors go further and in any event do not permit his conclusion on this issue.

32. Section 32(1)(b)

315. It is convenient to state the critical language of section 32(1)(b):

"(1)…where in the case of any action for which a period of limitation is prescribed in this Act, …

(b) any fact relevant to the plaintiff's right of action has been deliberately concealed from him by the defendant,…

the period of limitation shall not begin to run until the plaintiff has discovered the…concealment…or could with reasonable diligence have discovered it."

316. In Cave v. Robinson Jarvis & Rolf [2002] UKHL 18, [2003] 1 AC 384, Lord Millett (with whom Lord Mackay of Clashfern and Lord Hobhouse agreed) said (at para 25):

"In my opinion, section 32 deprives a defendant of a limitation defence…where he takes active steps to conceal his own breach of duty after he has become aware of it…But it does not deprive a defendant of a limitation defence where he is charged with negligence if, being unaware of his error or that he has failed to take proper care, there has been nothing for him to disclose."

317. Lord Scott (with whose speech Lords Slynn, Mackay and Hobhouse concurred) said (at para 60):

"…deliberate concealment for section 32(1)(b) purposes may be brought about by an act or an omission and that, in either case, the result of the act or omission, i.e. the concealment, must be an intended result…A claimant who proposes to invoke section 32(1)(b) in order to defeat a Limitation Act defence must prove the facts necessary to bring the case within the paragraph. He can do so if he can show that some fact relevant to his right of action has been concealed from him either by a positive act of concealment or by a withholding of relevant information, but, in either case, with the intention of concealing the fact or facts in question. In many cases the requisite proof of intention might be quite difficult to provide. The standard of proof would be the usual balance of probabilities standard and inferences could of course be drawn from suitable primary facts but, nonetheless, proof of intention, particularly where an omission rather than a positive act is relied on, is often difficult. "

318. In Williams v. Fanshaw Porter & Hazelhurst [2004] EWCA] Civ 157, [2004] 1 WLR 3185, Park J said (at para 14):

"(iv) The requirement is that the fact must be "deliberately concealed". It is, I think, plain that, for concealment to be deliberate, the defendant must have considered whether to inform the claimant of the fact and decided not to. I would go further and accept that the fact which he decides not to disclose either must be one which is his duty to disclose, or must at least be one which he would ordinarily have disclosed in the normal course of his relationship with the claimant, but in the case of which he consciously decided to depart from what he would normally have done and to keep quiet about it."

319. In the same case, Mance LJ discussed the possibility that whereas Lord Millett in Cave v. Robinson had regarded active concealment (in relation to a recognised breach of duty) as indispensable to the operation of section 32(1)(b) (a "narrower" formulation), Lord Scott had also regarded a mere omission to be within the subsection (a "wider" formulation). Mance LJ said it was unnecessary to determine which formulation prevailed, since in that case, involving a solicitor's deliberate concealment of what he knew himself to be under an ongoing duty to disclose, the deliberate concealment fell within either formulation (at para 36). But he also said (at para 29):

"But in many cases there may be no running relationship, and, even where there is, it may not involve any general duty to inform the other party of relevant facts. On the face of it, "concealment" in such a context might seem to require active conduct, rather than a mere decision to remain silent – even in circumstances where it would be normal or moral to speak."

320. Brooke LJ agreed on the basis that the defendant solicitor had "deliberately concealed its existence from her although he was under a duty to tell her about it" (at para 51).
321. It appears therefore that there must be either active and intentional concealment of a fact relevant to a cause of action, or at least the intentional concealment by omission to speak of a fact relevant to a cause of action which the defendant knew himself to be under a duty to disclose. There is no decision that anything less than a duty to disclose will suffice in the absence of active concealment.
322. The subsection speaks of "any fact relevant to a plaintiff's right of action". What is meant by that? The judge's treatment of this issue (see above) speaks of the Cooper retest results as "evidence": also as "a key piece of evidence", and as confirming when viewed together with the disport tests that the fuel was off-specification, and as also highly relevant to the conclusion that the D323 tests were probably "wrong" or "incorrect" or understated the RVP. All of this suggests that the judge was reading the statutory test as being a very wide one simply referring to any fact evidentially probative (or at any rate significantly so) to AIC's right of action. In only one sentence (at para 324) did the judge speak of AIC's pleadings, to say that in 1996 AIC was "unable to plead" that the D323 test results were probably wrong ("and other matters referred to below"). In context, however, even that reference to AIC's pleadings was simply to one particular of inaccuracy among others and indicated no particular legal approach to the significance of any pleadable factual assertion. Of course, the importance of the certificate and its results was not that the D323 test gave rise to inaccurate results, but that the D323 test was the wrong test and that the fuel was in truth out of specification (on the right test).
323. In this connection it is clear from authority that the statutory words "any fact relevant to a plaintiff's right of action" are to be given a narrow rather than a wide interpretation. Thus in Johnson v. Chief Constable of Surrey (CA, unreported, 19 October 1992), where the claim was in false imprisonment and the police had deliberately concealed facts relevant to the absence of reasonable cause, this court accepted the defendant's submission that "the relevant fact must be a fact without which the cause of action is incomplete", contrasting a fact relevant to an action and to a right of action (5A, 6C). Thus Rose LJ said "Facts which improve prospects of success are not, it seems to me, facts relevant to his right of action" (at 6E). He accepted that the interpretation was a narrow one (at 6G). Russell LJ agreed, saying (at 7E): "Accordingly, whilst I acknowledge that the new facts might make the plaintiff's case stronger or his right to damages more readily capable of proof they do not in my view bite upon the "right of action" itself. And Neill LJ emphasised that although absence of reasonable cause was an element in the tort of false imprisonment, the "gist of the action" is in the imprisonment itself, which establishes a prima facie case and puts the burden of proving justification on the defendant. Therefore the statutory words "must mean any fact which the plaintiff has to prove to establish a prima facie case" (at 8E/H).
324. Moreover, in C v. Mirror Group Newspapers [1997] 1 WLR 131 (CA), where the same words fell to be applied, this time as found in section 32A of the 1980 Act, this court again applied the narrow test determined in Johnson. Neill LJ, with whom Morritt and Pill LJJ agreed, said "The relevant facts are those which the plaintiff has to prove to establish a prima facie case" (at 138H). He again contrasted such facts with evidence which relates "to the proving of the case rather than the existence of the right of action", citing as further authority (at 138D) a dictum of Sir John Donaldson MR in Frisby v. Theodore Goddard & Co (CA, unreported, 7 March 1984).
325. The judge, who cited both Johnson and C v. Mirror Group Newspapers, nevertheless plainly did not apply a "statement of claim" test but an evidential one. He therefore erred. The question is whether, on this appeal, Mr Hamblen can make good that error.

33. The four tests

326. In his skeleton argument, Mr Gaisman submitted that for the requirements of the deliberate concealment principle to be met, four points had to be answered, in each case against ITS, namely:

(1) Did ITS owe a duty to disclose the relevant information?

(2) Were Mr Lucas and Mr Chalmers aware of that duty?

(3) Did Mr Lucas and Mr Chalmers take a deliberate decision, in spite of their knowledge of ITS's duty, not to disclose the relevant information?

(4) Was the information relevant to AIC's right of action?

327. In his judgment below, Lord Justice Buxton adopts three tests which essentially cover the same ground (see at para 427 below). I have been considering the last of them. I revert to the first, the existence of a duty of disclosure.

34. (1) Did ITS owe a duty to disclose the relevant information?

328. Mr Hamblen submitted that there was no need for this ingredient to be met. He contended that it would be sufficient if the relevant facts would have been disclosed in the ordinary course of the relationship between the parties, citing the dictum of Park J in Williams v. Fanshaw Porter at para 14 (see at para 321 above), where Park J suggested that as an alternative. However, the other members of the court in that case did not adopt the same broader language, and on the facts a continuing duty to disclose was found on the part of the solicitor defendant. As I understand the analysis of Brooke and Mance LJJ in Williams, what is needed is either a duty to disclose and a conscious omission to disclose, or at least active concealment. AIC has not alleged active concealment.
329. Indeed, AIC's pleading of deliberate concealment, which is brief (see para 23 of the amended reply) appears to be essentially premised on pleadings in the statement of claim which assert duties of disclosure "as a matter of law" or contract (see now paras 23A1/23B3) and, in a response to ITS's request for further information, to stress "in particular" the deliberate concealment of "the fact that had the cargo been tested in accordance with ASTM D5191 it would have been, or may well have been, shown to be off-specification as regards RVP" (at para 5 thereof); something which, by virtue of the disport tests, AIC always knew or were in a position to know.
330. As for a duty to disclose, the judge found that such a duty existed, founding himself in the main on the NAMAS guidelines and ITS's internal guidelines, but also referring to passages in Mr Lucas's and Mr Chalmers' evidence where they accepted (on certain factual hypotheses and/or subject to hindsight and/or qualifications) that they should have informed AIC and Mobil of the test and its results: see paras 195, 225 and 241 above. (I would also refer, however, to paras 248, 282 and 292 above). In my judgment these are in themselves an unsatisfactory basis upon which to find the existence of legal duties of disclosure.
331. I have already set out the circumstances concerning the judge's consideration of international and internal guidelines (see at paras 169/186 above); and also matters relating to ITS's practice and duties with regard to sample retention (at paras 187/192 above). In the light of the controversies concerning these aspects of the case, I would prefer to consider the question of an inspection house's duties of disclosure as a matter of principle.
332. It will be recalled that the purpose of such certificates in international trade is to provide an independent tester, certainty and finality. Such certificates are very commonly provided on the basis, as here in the case of AIC's purchase, of a final and binding determination in the absence of fraud or manifest error. Formally retained samples are sealed and sent with the vessel and/or kept by the tester: in the event of any controversy they are available to the parties and may be made subject to "referee testing". The process is governed by contract. Duties of care are implied in that contract, but, as a matter of classical legal analysis, will go no wider than is required by law or business necessity. Any duty of disclosure will similarly need to be explicable as an implied term required by business necessity or law.
333. In such a context, what duties of disclosure are required? We have not been referred to any authority on the subject, and, given the importance of this area, I wish to be cautious. Intuitively and commercially, I am sceptical of a duty expressed in the judge's wide and unqualified terms, viz that "where a complaint or any other circumstance raised doubt concerning the quality of the tests, to ensure that the relevant work/tests were promptly audited/reviewed. Where the audit/review cast doubt on the correctness or validity of the test results such as to necessitate a Supplement to the certificate, to write to Mobil and AIC immediately enclosing the Supplement…" (para 192(viii)). As I have said above, that sounds to me like legislation rather than a process of implication. If a complaint or any other circumstance raising doubt could require a fresh review and a supplement to the certificate, then a certificate intended, as it so commonly is, to be final and binding save for fraud or manifest error, would be in practical terms worthless. There would be endless disputes.
334. Subject to the width of the "manifest error" exception, and certainly in the absence of that exception, a certificate remains binding between parties to a sale contract even if it is not disputed to be inaccurate and even if the seller acknowledges that the goods are of inferior to contractual quality, and even if the certifier acknowledges a mistake: see Toepfer v. Continental Grain Co [1974] 1 Lloyd's Rep 11 (CA). As Cairns LJ there said (at 15):

"For the sake of achieving certainty in the great majority of cases, it is worth while to take the risk that occasionally a wrong decision will be given and that there will be no means of reconsidering it."

335. In a typical case, a cargo might be tested at destination and found to have a different quality from that certified, or even to be out of specification. The contract contemplates that, if a query is raised, then further testing will be done by mutual agreement upon the formally sealed and retained samples, and probably by an independent referee. It may be concluded that the certificate is arguably or probably inaccurate, and possibly even negligently so. Nevertheless the certificate will bind, where it is expressed to be final and binding, in the absence of fraud or manifest error. If the certificate is not expressed to be final and binding, then the matter will probably be resolved on a testing of the formally retained samples. What is not contemplated as a matter of contractual implication (internal self-governance is of course another matter) is an unsupervised internal review by the testing house alone, without the mutual agreement of the client or clients, on broached samples which cannot be the basis of a further certificate.
336. In the present case, of course, the wrong test had been performed, as was demonstrated by the certificate itself. The certificate was therefore invalid (Veba Oil). That is the law, binding in this court. For present purposes, where we are concerned with implied duties, it is plain that we have to be guided by the law and not by the parties' possibly mistaken views as to the law. The fact that the wrong test had been performed was patent on the face of the certificate. It was known about and acknowledged by at latest 17 April 1996. If there was a duty to disclose that the wrong test had been performed (if that had not been patent and known), as there very well might be, since it fatally destroyed the validity of the certificate and also in my judgment amounted to manifest error, then there was no breach of that duty. That disclosure, that knowledge, was or ought to have been sufficient to obviate any reliance on the certificate.
337. If a loading certificate is known to be invalid after the vessel has loaded and left the loading port, what can be done? In my view, nothing – except own up. It is too late to provide a loading certificate. (By agreement, of course, anything can be done and other devices deployed, but that is another matter.) Whether withdrawn or not, the certificate is invalid. The certifier is in breach, irremediable breach, actual repudiatory breach. There may be remedies, but the time for performance is past. The sale contract which had been intended to be governed by a binding certificate of quality obtained at loading (in the present case at a pre-loading stage, for the certificate was to be on the fuel ex shore tanks), becomes uncertificated. The buyer is entitled to reject the documents, for he has bought inter alia a certificate. The goods may or may not be within specification, but the buyer has not only agreed to buy goods of a certain specification, but goods certified to be of a certain specification. Therefore, even if the buyer could be satisfied that the goods were in truth within specification at the critical moment under the sale contract (eg on loading or, as here, ex shore tanks), he would be entitled to reject the documents and/or the goods.
338. If the goods are accepted, there may in such circumstances still be a dispute about their quality (and consequential compensation). That dispute may in practice be resolved by the agreed testing of loading samples; or simple negotiation; or the dispute may have to go to law. In such a dispute, the certifier may be allowed, or requested, to assist, but he does so outside his original contract, or at any rate in mitigation of his breach. Toepfer is again authority for the proposition that in such circumstances, once a certificate has been given, the inspecting house, like a judge, is functus. He is not like a solicitor, with that continuing obligation to advise his client and a continuing obligation of disclosure acknowledged in Williams.
339. Where an invalid certificate is given, either party to the sale may have consequential losses which it will seek to recover from the certifier, in contract in the case of the client. Even a third party buyer in chain may have remedies against a negligent certifier, in tort rather than in contract. What the damages for such parties may be, whether clients of the certifier or not, will depend on the circumstances. Since, in the absence of fraud or manifest error, a valid certificate remains effective between the parties to a sale even if it is in fact inaccurate and possibly negligently so, the possibility of a third party suing the certifier for negligent misstatement or breach of a duty of care, opened up by Niru Battery, confirms the possibility of a new area of liability (cf Montrod Ltd v. Grundkötter Fleischvertriebs GmbH [2001] EWCA Civ 1954, [2002] 1 WLR 1975, which was distinguished in Niru Battery). However, I do not see that as adding to the inspection contract a new implied term of an ongoing duty of disclosure. That was in no way the subject matter of Niru Battery, where the inspection house had negligently facilitated a fraud by certifying that they had supervised the loading of the goods (relying on a false bill of lading) when they had not.
340. My intuitive understanding of the situation therefore does not encompass an implied term on an inspection house to disclose any private test it makes on residues on the goods in question or the results of such a test, a fortiori not in circumstances where no valid certificate has been issued and it is too late to certify anew.
341. What then could be the basis for an implied term such as that found by the judge? In such circumstances what arguments have been or could be deployed to suggest otherwise?
342. One possibility might be that the contract for certification of the goods is still open for performance and thus in everything that the certifier does he acts or is still obliged to act for his client(s) in execution of his contract. This has not been advanced by AIC as the explanation, and yet sub silentio it appears to lie at the root, or at any rate to be a necessary foundation, for its, and the judge's ultimate position. It is some such idea that was behind the thought (eg of Mr Whitaker, but also perhaps of the judge's broad implied term) that there was still time to issue a new certificate, or an amended or supplementary certificate. (I accept, of course, that anything can be done by fresh agreement of the parties; but that is a different matter.) Since the argument has not been made expressly, it is difficult to know what would be said in support of it. Mr Hamblen on the contrary accepted that the Cooper retest could not be made the basis of a fresh certificate, and that a test on such broached samples was not a legitimate, publicly reliable, basis for certifiable results. In my judgment, however, in any event the contract is not a continuing one to provide advice or ongoing evidence of the quality of the goods, but to certify it on or before loading. A valid certificate, even if erroneous, cannot be withdrawn (save by consent of all the parties) in the absence of fraud or manifest error; and an invalid certificate, whether withdrawn or not, is ineffective. It is too late to adjust the position after time for the certification has passed.
343. It is true that the consequences of a breach (the provision of an invalid or otherwise negligently obtained certificate) may be ongoing, even after the certifier's contract has been performed or misperformed, in ignorance of some underlying defect, even if it is patent: for instance, if, rightly or wrongly, the certificate continues to be used among traders as an item in commerce. That, however, is a common consequence of a breach of contract: it may affect third parties. Where at any rate the error is manifest, as here, the danger of such consequences ought to be slight.
344. A second possibility, therefore, is that utility and morality might combine to suggest that, when once any issue about a certificate has been identified, the certifier of a possibly inaccurate certificate owes a duty to investigate and report: not because the original certificate can be changed or substituted, but simply to throw further light on the question-mark that has been raised. Even though the original certificate, if valid, cannot be withdrawn, amended or substituted, nevertheless the danger of inaccuracy or negligence requires the certifier at least to issue a sort of health warning about it, if he cannot confirm its accuracy in full measure. Since the investigation is required for the purpose of a further report, it follows that there must be disclosure of any investigation; indeed, there must be further investigation. That is in one respect what the judge has found by his implied term, although to my mind he has done so in the belief that the purpose of the further investigation and report is the fulfilment of the contract, about which I have already stated my scepticism above.
345. This is an important argument. It is floating, silently, in the background, but has not been clearly articulated: for the term which the judge has derived has been obtained from the international and internal guidelines, and/or from evidence from the witnesses as to what "should" be done, without any particular analysis. Nevertheless, Lord Justice Buxton in a powerful passage in his judgment (centred at para 439 below) has, I think, adopted it, stating that the converse challenges reality. It is important because, if it is correct, it is in theory applicable to every conceivable kind of contract. The issue would be: if the provider of goods or services under contract suspects or has reason to think that the service or goods which he has provided has some defect or flaw in it, does he have an ongoing obligation, and if so, of what kind, to disclose his suspicion or concern, or to investigate it, and to report? There are of course standard remedies for standard causes of action: for breach of contract, for fraud, for negligence including negligent misstatement. If there is a danger of future harm, investigation and/or disclosure may avert it, and thus contingent liabilities may be entirely averted. If not averted, the remedies on underlying causes of action remain. But is there a separate continuing duty to investigate and/or disclose? The argument has not been articulated. No authority has been cited for it. As far as I know, authority is against it: see Bell v. Lever Bros Ltd [1932] AC 161. Thus there is no general principle of disclosure. A party must not deceive; but in the absence of a rare duty of disclosure (as exists in the case of contracts uberrimae fidei, or within litigation), may remain silent. Of course, if he speaks, he may become liable in misrepresentation or Hedley Byrne negligence: but that is another question. There is no general principle that the wrongdoer comes under an additional obligation to disclose his wrongdoing. We are not here concerned with dangers to life or health. I do not think it would therefore be right to find that an unmade argument supports the implied term found in this case.
346. In any event, an implied term should go no further than is necessary and reasonable. No such traditional analysis was performed by the judge. In the present context, it seems to me that, for reasons already discussed above, such an implied term would be destructive of the utility of such certificates in international trade. Therefore it is both unreasonable and contrary to business efficacy. Nor is such a term necessary to meet this case, for in truth the certificate was invalid, because of use of the wrong test, and ought to have been recognised as such. Therefore, ITS could not escape such liability as might be brought home to it for its uncontractual and possibly negligent issue of an invalid certificate. That would concentrate the mind on matters which were directly relevant to all aspects of the breach. Since the error was manifest, this approach would also concentrate attention on aspects of causation and remoteness. And since the certificate was invalid and manifestly so, it matters not a jot whether the D323 test was performed incorrectly, or negligently, or not. That is a will of the wisp. The contract specified D5191 because only that test could satisfy the requirements of the Colonial Pipeline Specification, since in the USA all gasoline has, according to its grade, to be fungible for the purpose of transport through pipelines.
347. In exceptional situations, of which this may well be a unique or very rare example, where an invalid certificate is not dealt with as such even though the cause of its invalidity is known, there may be a need for investigation: but that was to be achieved through the contemplated means of the consensually retained samples. The provision and existence of those samples to my mind excludes other surviving material such as the broached testing samples as being subject to the same contractual significance. Contractually contemplated tests, carried out mutually and supervised independently had been performed at the discharge port, and similar tests were contemplated (and discussed between AIC and Mobil, see paras 142/143 above) on samples retained in the UK: but unaccountably those tests were never carried out. If the certifier in addition conducts private tests, I see no need to imply a parasitic duty to reduplicate the underlying duty to take care to provide a valid and accurate certificate. In any event, the certifier should be entitled to investigate its potential liability for itself, a fortiori in circumstances where its client has made a claim against it. The posited duty is neither necessary nor reasonable.
348. It is true that to some extent the witnesses accepted, on certain premises which are themselves in general unclear, a form of obligation of disclosure. However, the premise appears to have been either that the samples were contractually retained samples (a hypothesis which the witnesses doubted or disputed and which I reject, see below) or that the certificate was valid but challengeable for error (a hypothesis which seems to me to be wrong in law), or simply on the basis of hindsight. Such evidence in any event has to be seen in the context of the evidence as a whole. In general, the witnesses did not accept that they had acted in breach of any duty, let alone consciously so. There was, nevertheless, as it seems to me, a natural sensitivity, which could be played on in cross-examination, as to the consequences of the provision of a certificate which, whatever the analysis, had been produced on the basis of the wrong test. A good example is Mr Chalmers' evidence set out at paras 240/241 above.
349. A third possibility, which the judge found and Mr Hamblen defended on this appeal, is to imply the duty of disclosure from the guidelines relied on by the judge. However, in my judgment, for the reasons stated above where I have dealt with those guidelines, as well as the further reasons discussed under this heading, the judge erred, or was misled, in finding assistance in them. They are in truth designed to assist laboratories, in general, with the self-regulation and management of their business: they cannot control the implication of terms in specific contracts. In any event, the judge was taken, in error and contrary to the evidence before him, to the wrong guidelines.
350. A fourth argument was that the Cooper retest was performed on AIC's property, or on fuel held to its (and Mobil's) order in some way, and that for that, technical, reason, the fact of it and its results ought to have been disclosed. The judge accepted the alternative formulation of this argument (see his para 252, and my para 188 above), saying nothing about the property aspect. In my judgment, however, the submission does not support the implied term in either formulation. The testing samples were provided to ITS by Mobil for ITS's purposes. They were never placed on board the vessel, so that title never passed to AIC. I accept Mr Gaisman's submission that the testing samples were not contractually retained samples. In my judgment they had become ITS's property, for the purpose of their tests. They could have been thrown away. Since they survived, they might have been kept and were at most "potentially important" as reference evidence, although they could not have been made the subject matter of certifiable tests or even of so-called "referee testing", and there were more important contractually retained samples: see inter alia paras 214/215 and 237 above. The judge found that ITS was "in breach of a duty to take reasonable care" in failing to seek permission to perform the Cooper retest. In my judgment no permission was needed, and the status of the Cooper retest is that of an internal test on ITS's own property. Although the judge was not attracted by Mr Rackham's suggestion that the Cooper retest results might have been divulged if they had come in under 9, it must often be the case that parties drawn into dispute may make their own tests or seek their own evidence, and that disclosure, where such material is not privileged, comes into effect only in the context of litigation.
351. In sum, I do not think that there was a contractual duty of disclosure. In effect, AIC has pleaded such parasitic implied duties in a number of different ways, in essence because it was conscious that breach of the underlying causes of action were or were in danger of being time barred in the absence of deceit or fraudulent suppression. It therefore sought to present its claim not only or so much on the straightforward bases of an invalid or inaccurate or negligently produced or misstating certificate, but rather or also on the basis of subsequent and parasitic breaches of a duty of disclosure, variously described (see below). But, in truth, there was no shortage of remedies available to AIC under the contract. Primarily, there was the provision of an invalid certificate (Veba Oil). But even if the answer lay in proving negligent testing or inaccurate certifying, the remedies lay at hand. However, AIC allowed six years to go by. The complications of the case are entirely created by that simple fact. That is the "necessity" which has bedevilled the case and caused fraud and deliberate concealment to be pleaded. That leads directly to the fourth of the four issues itemised above.

35. (4) Was the information relevant to AIC's right of action?

352. I have stated above that the judge erred in applying a wide evidential test of relevancy for the purposes of the statutory phrase in section 32(1)(b), rather than the narrow statement of claim test required by two decisions of this court in Johnson and Mirror Group Newspapers. Can Mr Hamblen make good the position on this appeal?
353. AIC submits, as part of its respondent's notice, that the Cooper retest is relevant to a number of causes of action identified in paras 3/8 of Annex 1 to AIC's respondent's notice. These are (a) torts of deceit or negligence in failing to correct, qualify or withdraw the certificate and/or statements of the April 17 telephone conversation; and (b) negligence or breach of contract or duty in failing to disclose the Cooper retest and its results, or in failing to obtain permission for the Cooper retest. In his skeleton argument, Mr Hamblen submitted that the judge had applied the right test (viz as to material facts which have to be pleaded, as distinct from evidence) and also submitted that the dividing line was illusory and that the judge's view should therefore be accepted. He submitted that the post-disclosure amended pleadings were "premised squarely on the Cooper retests".
354. In my judgment, these submissions fail. The case in deceit has already failed. The certificate was invalid. If anything turns on what it said, viz that RVP was 8.22 on the D323 test or that "Fuel meets specification", then relevant causes of action could be and were pleaded before disclosure of the Cooper retest. The certificate had indeed been negligently issued since it used a D323 test. The Cooper retest for its part was a D5191 test, not a D323 test. At best, the Cooper retest was simply a further test, more evidence, that something had gone wrong in the issue of the certificate. If anything is alleged to turn on what was said in the April 17 telephone conversation, then it is necessary to repeat the analysis above which concludes that there was no relevant representation which refreshed the certificate's representations. And even if there had been, the position remained what it was: that the certificate had been negligently issued.
355. I therefore turn to the pleaded causes of action which were particularly relied on, namely those which depend on alleged duties of disclosure about the Cooper retest (including the failure to obtain permission to make the test). I have said above that those duties did not arise. But even if they had, it seems to me that Mr Gaisman is correct to submit that section 32(1)(b) is not fulfilled by the pleading of causes of action which are artificially designed to attempt to turn a parasitic question of knowledge of further evidence into an essential basis of the claimant's right of action. The Cooper retest was simply further evidence that the goods were out of specification and that something had gone wrong in the original test and resultant certificate.
356. That is well and sufficiently indicated inter alia by AIC's particulars (served June 2005, ie at a late stage after disclosure of the Cooper retest) of ITS's "failure properly and/or accurately to test the regular cargo". AIC said, twice, that it was "not in a position to say precisely what went wrong", but that errors were to be inferred from a comparison between the original test results and later retests, citing "the Cooper retests and as further borne out by (1) the various pressure results obtained by analysis of samples at the discharge port and (2) the lack of any evidence of contamination on the vessel". In other words the Cooper retest results were simply further evidence.
357. I think these general points cover the various formulations of the post-disclosure amended pleadings, but I give some examples below, with apologies for my inadequate paraphrasing.
358. Thus at para 23A1 of the particulars of claim, it is said that in the light of the Cooper retest and results the cargo was off-specification at loading and that that would have been shown if properly tested by the D5191 test: that ITS was bound to disclose "those truths", and to communicate the fact of the Cooper test and its results, a fortiori in the light of the April 17 telephone conversation. But the essence of the complaint remains that the cargo was off-specification and that could and was pleaded before the Cooper retest disclosure.
359. At para 23B1, it was pleaded that, because it was to be implied that all samples taken at loading would be retained for 90 days, therefore permission had to be obtained for using or disposing of the samples utilised in the Cooper retest and AIC/Mobil had to be informed of what had happened. In my judgment, there was no such implied term – on the contrary the contract expressly contemplated officially retained samples – but in any event the complaint is merely an indirect route back to the position obtaining in the previous paragraph. A similar plea was made in para 23B2 by reference to the Cooper retest samples being the property of AIC/Mobil.
360. Although I would accept that loss and damage is probably not the focus of the statement of claim test, I think that it is relevant to observe that the primary loss and damage claimed by AIC in these proceedings is that of its liability to Galaxy in the Swiss proceedings, viz a liability for supplying an uncontractual cargo to Galaxy, a liability established (despite ignorance of the Cooper retest) on the basis of the disport tests, AIC's knowledge that the certificate had involved the wrong test, and AIC's refusal of further testing or discussions (see para 158 above).
361. What, at the end of the day, did the Cooper retest and its results add to the pleadable case? In my judgment, nothing save a modicum of further evidence: see in this connection Mr Revell's opinion summarised at para 167(1) above. AIC already knew that the disport results showed the fuel to have an RVP in excess of 9 psi by the right test, D5191. It knew that there was no contamination during the voyage. In particular, it knew that the official, sealed, composite sample of the fuel loaded ex shore tanks and carried with the vessel, tested by D5191, was over 9. That test, being carried out on a loading sample, likewise excluded the possibility of contamination. The Cooper retest results, 9.1 as properly calculated, but not on a contractually retained sample but on a broached and used testing sample, and not independently reviewed, was at best simply in line with the disport results. It is and always was simply a matter of inference that the D323 tests were inaccurately and/or improperly performed (or some of them were: the premium cargo D323 tests appear to have been spot on). If AIC at one time believed that the D5191 test would give a higher reading than the D323 test, rather than an essentially identical (or "comparable") reading, that belief was not induced by anything said on behalf of ITS. At all times AIC had access to legal or expert advice. The Cooper retest said nothing, and is not alleged to have said anything, about the comparability of D323 and D5191. The D323 tests are ultimately uncontractual and irrelevant. AIC knew that that was so or had the means of knowing it, or ought to have known it. It knew that D323 had been the test used, and that it was an uncontractual test. The Cooper retest results produced no falling of scales from AIC's eyes. The incompetent use of a comparable test (D323) could always have been pleaded, but in truth was irrelevant. The Cooper retest was just another D5191 test, and on a sub-standard, broached, unsealed, uncontractual sample, which the relevant tests stated could not be used for testing.
362. In my judgment, neither the judge's findings nor AIC's further submissions pursuant to its respondent's notice establish that any deliberate concealment of the Cooper retest and its results involved "any fact relevant to the plaintiff's right of action". If I am wrong about that and the issue of deliberate concealment generally, it will be necessary ultimately to identify the relevant cause or causes of action in respect of which section 32(1)(b) operates to extend the limitation period.

36. (2) and (3): consciousness of duty and deliberate decision to conceal

363. If there was no duty to disclose, and if the Cooper retest and its results in any event added at most further evidence in support of causes of action which AIC was always in a position to plead, then these further ingredients do not matter. I revert to them for the sake of completeness.
364. In my judgment, for the reasons discussed under the heading of deceit and again referred to at the beginning of my discussion under the issue of deliberate concealment, the judge's assertions of a consciousness of a duty to disclose and of a deliberate decision to conceal are not adequate to sustain his findings. They are based on his view of Mr Lucas as deceitful and dishonest, and of Mr Chalmers sharing a form of unarticulated complicity with him, even if not himself responsible for the telephone conversation to which he was not of course a party. The assertions are based on findings of knowledge on their part which, as reasoned, are a repetition of findings which the judge derived after trial with the assistance of experts and on the balance of probability (without reference to the especial care needed in the context of allegedly unconscionable conduct). And if there was no deceit in the context of the 17 April conversation, there is no explanation of how or when Mr Lucas and Mr Chalmers took their apparently joint and complicit decision to conceal the Cooper retest results in consciousness of ITS's duty to disclose them: and that remains the case even if one considers that each of them knew of the Cooper retest results on 17 April or shortly thereafter. The transcript of the conversation was immediately provided to ITS's lawyer, Mr Pople. This is because the parties were then and there in an adversarial legal situation, with lawyers appointed on both sides. When in 1997 Galaxy commenced its English proceedings against AIC and ITS, Mr Pople considered the Cooper retest papers for the purposes of disclosure in that litigation. In his view they were privileged; and even though it is now accepted by ITS that that was an erroneous decision, there has been no suggestion by AIC that that amounted to deliberate concealment on anyone's part. It is not easy to see how Mr Lucas and Mr Chalmers took a deliberate decision to conceal the Cooper retest from AIC in consciousness of their duty to reveal it, when the whole matter was put before Mr Pople for his decision in 1997. Both Mr Lucas and Mr Chalmers denied any decision deliberately to conceal the information. If such a decision had been taken, it would be highly likely that some positive trace of it would survive in documentary disclosure. At any rate one might have expected such matters to be discussed by the judge.
365. I do not say that unconscionability is a formal requirement of section 32(1)(b) deliberate concealment. That was made clear by Lord Scott in Cave v. Robinson (at para 65), despite what was said by Lord Browne-Wilkinson in Sheldon v. R H M Outhwaite (Underwriting Agencies) Ltd [1996] AC 102 at 145H to the effect of "unconscionable behaviour by deliberately concealing the facts relevant to the plaintiffs' cause of action" being the "underlying rationale" of section 32. Nevertheless, Lord Scott also said:

"I respectfully agree that it is difficult to think of a case of deliberate concealment for section 32(1)(b) purposes that would not involve unconscionable behaviour and that most cases of deliberate commission of breach of duty for section 32(2) purposes would be in the same state."

366. In sum, in the absence of fraud, I do not think it is possible to sustain, in the light of the judge's reasons or lack of them, his conclusion that Mr Lucas or Mr Chalmers can properly be found responsible for deliberate concealment.

37. Section 32(2): a fallback case.

367. Another aspect of AIC's respondent's notice is a fallback reliance on section 32(2). The point was relied on at trial, but the judge made no finding on it.
368. Section 32(2) provides as follows:

"For the purposes of subsection (1) above, deliberate commission of a breach of duty in circumstances in which it is unlikely to be discovered for some time amounts to deliberate concealment of the facts involved in that breach of duty."

369. There is a discussion of the relationship between section 32(1)(b) and section 32(2) in Cave v. Robinson. In any event, it seems to me that, whatever might be said in favour of section 32(2) providing an alternative avenue for avoiding limitation, had the judge's reasoning survived, the analysis which I have preferred above simply makes such reliance impossible.

38. Continuing duties

370. Finally, AIC relies on the concept of continuing duties as a means of side-stepping limitation. Mr Hamblen submitted that ITS's duties to correct or qualify false representations contained in the certificate and/or to issue a supplement to the certificate and to disclose the Cooper retest and its results were continuing duties which subsisted and were breached on and after 17 May 1996.
371. The equivalent submission was made at trial (judgment, at para 75). The judge set out some law relating to continuing breaches, in effect a citation from Chitty at para 28-035 (at para 294 of his judgment), but that was in the context of issues (nos 31/34) relating to deliberate concealment. The judge otherwise has no discussion of continuing duties. However, at para 332 of his judgment, in answer to issue 35 (Are AIC's claims time-barred…?), he said this:

"332. I answer this issue in the negative. AIC's claim in deceit succeeds. In addition to AIC's points about continuing duties, facts relevant to the claimant's right of action were deliberately concealed by ITS in the circumstances and for the reasons set out above (issues 31-34)."

372. However, nowhere in his judgment does he discuss or accept or explain why he accepts, if he was accepting, "AIC's points about continuing duties". This may be because the point was raised only in AIC's closing written submissions at trial. So much so, that Mr Gaisman complained that there was a serious procedural irregularity in allowing the point to go ahead. It seems fair to say that there is no agreed issue which plainly raises the point.
373. Mr Hamblen relied, but not strongly, on para 332 as a finding in his favour on the submission of continuing duties. He submitted in any event that ITS's duties continued because the relationship between AIC and ITS was a continuing one, because of the 90 day sample retention period, because of the continuing discussions relating to the certificate, and the continuing possibility that third parties would rely on it.
374. In my judgment, however, the relationship was not a continuing one, nor was there any breach of a continuing duty. I have pointed out above that the inspection and certifying contract was not a continuing relationship, such as that (subject to the extent of a retainer) of a solicitor and client. ITS had no responsibility as AIC's adviser. It was not like a tenant, with its continuing obligation to keep in repair (an example given by Chitty). Contractual samples were to be retained for 90 days, but the Cooper retest was not performed on those samples; the samples retained on the vessel were tested early on at the discharge port, and there was never any request for referee testing of the UK retained samples. AIC paid for the fuel on 18 April 1996. Any cause of action was complete and final on the issue of the certificate, or at latest on or immediately after 17 April. The certificate spoke only to a particular test at a particular time. In any event, it seems to me that the relationship had ended at latest when, immediately on or after 17 April, the parties told each other that they were instructing their respective lawyers. Moreover, even in the case of a relationship such as that of solicitor and client, a failure to do something is not a continuing breach which constantly refreshes the limitation period: see Bell v. Peter Browne [1990] 2 QB 495.

39. Causation, remoteness and loss

375. It was common ground that, in the absence of a liability in deceit, questions of causation, remoteness and loss, if they remained relevant for other reasons, such as deliberate concealment, would have to be remitted back to the commercial court.

40. Conclusion

376. In sum I would allow this appeal on both main issues. There is no liability on the part of ITS in deceit, and otherwise AIC's potentially meritorious claims are time barred. As to the issue of deliberate concealment I have read in draft and gratefully taken into consideration the judgments of Lord Justice Buxton and Sir Martin Nourse, but would respectfully dissent.
377. I would respectfully endorse what Lord Justice Buxton has said at the start of his judgment about the procedural difficulties of this appeal.

Sir Martin Nourse:

378. I have had the advantage of reading in draft the judgments of Lords Justices Buxton and Rix.

Deceit

379. I agree with them that ITS did not commit the tort of deceit, either through Mr Lucas in the telephone conversation of 17 April 1996 or subsequently through Mr Lucas and Mr Chalmers. While I recognise that our decision on this question involves the reversal of findings of primary fact made by the judge, I am satisfied that the evidence before him was insufficient to establish either the necessary representation or the necessary state of mind. In regard to this question there is nothing I can usefully add.

Deliberate concealment

380. The question whether it is open to this court to reverse the following findings made by the judge in paragraph 312 of his judgment is less straightforward:

"I find that Mr Lucas and Mr Chalmers made a deliberate decision not to disclose the fact of the Cooper re-tests and the results thereof to AIC and Mobil in circumstances in which the re-tests and the results thereof should have been disclosed to AIC and Mobil, and Mr Lucas and Mr Chalmers knew that the re-tests and the results thereof should have been disclosed to AIC and Mobil. I further find that Mr Lucas and Mr Chalmers were aware that ITS was under such a duty and that Mr Lucas and Mr Chalmers made a deliberate decision not to disclose the re-tests and the results thereof."

381. That paragraph contains a holding that Mr Lucas and Mr Chalmers were under a duty to disclose the re-tests and the results thereof to AIC. It also contains clear and specific findings of the primary facts which are necessary to establish a case within section 32(1)(b) of the Limitation Act 1980. In my judgment it is not possible to say either that the findings were not open to the judge on the evidence before him or that they are undermined by his errors in relation to the case in deceit and his findings of dishonest or guilty knowledge. With regard to the latter point, the concealment, in order to be "deliberate", did not have to be dishonest, and the facts supporting the judge's findings were substantially different from those on which he relied in holding that the case in deceit was established. Accordingly, in agreement with what is said by Lord Justice Buxton in paragraph 446 of his judgment, I respectfully disagree with the views expressed by Lord Justice Rix in paragraphs 314 and 366 of his judgment.
382. The two remaining questions under section 32(1)(b) are, first, whether ITS was under a duty to disclose the Cooper re-tests and the results thereof to AIC; second, whether the re-tests were relevant to any of AIC's rights of action that are otherwise statute-barred.
383. As to the duty of disclosure, I am in general agreement with the reasoning of Lord Justice Buxton expressed in paragraphs 428 to 443 of his judgment. For myself, I would be content to base the existence of the duty on the simple commonsense of the notion that a certifier who has acquired knowledge of a material inaccuracy in his certificate is obliged to disclose it; see paragraphs 439 and 440.
384. With regard to the question whether the Cooper re-tests were a "fact relevant to the plaintiff's right of action", the decision of this court in Johnson v Chief Constable of Surrey binds us to interpret those words as denoting a fact without which the cause of action is incomplete. That makes it necessary to identify the cause of action.
385. Again, I am in general agreement with the reasoning of Lord Justice Buxton as expressed in paragraphs 451 to 466 of his judgment. In particular, I agree that it is incorrect to look simply at the umbrella complaint of breach of contract and negligence. What must be looked at are the individual complaints, each of which raises a separate cause of action. The complaint that ITS deliberately concealed the Cooper re-tests and their results from AIC raised a new and separate cause of action.

Conclusion

386. For these reasons, in company with Lord Justice Buxton, I would allow the appeal in part by reversing the judge's decision on the claim in deceit. I would affirm his decision on deliberate concealment.

Lord Justice Buxton:

The course of the appeal

387. The preparation of this appeal took a most unsatisfactory course.
388. The judge delivered judgment, and gave permission to appeal, on 7 October 2005. Grounds of Appeal, supported by a 42 page skeleton argument drawn by leading and junior counsel who had acted at the trial, were served on 11 November 2005, and met by a respondents' skeleton argument dated 16 December 2005. It was agreed between the parties that three days would be required for the hearing of the appeal. The Lord Justice supervising commercial appeals, appreciating from the documents and skeletons that the appeal was likely to be burdensome, ordered that in addition to those three days there should be allocated to the constitution hearing it two days of prior reading time, and three days at the close of the appeal for consideration of the judgment. The appeal was duly listed under those arrangements for hearing on 16-18 May 2006.
389. At some time in, it would appear, December 2005 those advising ITS decided to withdraw instructions in the appeal from leading counsel who had appeared at the trial, and instructed in his place Mr Gaisman QC. The court (and, we understand, those advising AIC) first heard of this when Mr Gaisman wrote to the Civil Appeals Office on 6 April 2006 announcing that he needed to express the points arising on the appeal in his own way, and that therefore he would be serving a new skeleton argument in substitution for that previously served. There would be no need to amend the notice of appeal.
390. That comparatively bland statement did not cause alarm bells to ring. They however rang out loud and clear on 26 April 2006 (by then, fifteen working days before the date for the hearing of the appeal) when a further letter arrived from Mr Gaisman stating that 5-6 days would now be required for the hearing of the appeal, and enclosing a replacement skeleton argument that extended to no less than 168 pages, or some three times longer than its predecessor.
391. Had it not been for the reservation of days for pre-reading and for writing of the judgment, it would have been impossible to accommodate the appeal in its revised form, and the matter would have had to be adjourned, very likely for several months or more, with very serious waste of the court's resources. In order to avoid that outcome the court was able, despite other commitments, to find five days within the total period allocated to the appeal, although that meant the surrender of all of the days originally allocated to judgment-writing.
392. Mr Hamblen QC nonetheless, and understandably, applied for the appeal to be adjourned in any event. We held a detailed hearing (again, at some considerable inconvenience to the members of the court) on 5 May 2006, at which we dismissed that application, but made such arrangements as we were able to ensure that Mr Hamblen was not handicapped in the presentation of his clients' case. In the event, we are satisfied that both parties have had a fair hearing, however much the achievement of that has placed unnecessary burdens on Mr Hamblen and his team.
393. I am obliged to make four comments. First, the rules do not provide for the late submission of a completely revised case, and Mr Gaisman did not seriously suggest that they do. Second, CPR 52PD 15.11A, which provides for the service of a supplementary skeleton argument by the appellant 14 days before the hearing, does not permit, or envisage, anything remotely like what happened in this case. "Supplementary" means what it says. The rule is designed to provide for lately decided or, within limits, discovered authority; changes of underlying factual circumstances; or the brief answering of points in the other side's argument that genuinely do not arise out of the original grounds of appeal. It does not permit the wholesale revision of the original case, particularly at three times the length of that original. Third, we were exercised as to whether we should admit the new skeleton at all, as opposed to making the appellants stand on their original document. It was principally the fact that Mr Gaisman's clients stood convicted of fraud that persuaded us to extend the latitude that we did.
394. Fourth, should a party in the future find itself in this position, wishing to make a very much expanded case and to extend the time for the hearing, it is essential, as a matter of practical administration and not just of courtesy, that the court is approached, with full details, at the earliest date at which that becomes apparent. Any such approach should make entirely clear what is wanted and why it is wanted. That will enable the matter to be placed promptly before a Deputy Master or, in appropriate cases, Lord Justice, and avoid the unseemly scramble that was the precursor to this appeal.
395. I turn to the substance of the appeal. It would be a work of supererogation to emphasise the debt that is owed by the court, and by the parties, to Rix LJ for the comprehensive account of the facts and issues in the case that is to be found in §§ 1-250 of his judgment. In what follows I draw very heavily and gratefully on that account.

AIC's claims

396. AIC asserted two main claims before the judge. The first was that in the conversation of 17 April 1996 ITS, through Mr Lucas, committed the tort of deceit. By a respondent's notice in this court that claim was added to by a separate assertion that, even if Mr Lucas was not deceitful in that conversation, shortly thereafter both he and Mr Chalmers acquired sufficient knowledge to make ITS guilty of deceit when they failed to correct the statements that had been made in that conversation. The second was that the limitation difficulties attaching to its claims in negligence and breach of contract, described in §15 above, could be cured by the fact that both Mr Lucas and Mr Chalmers had deliberately concealed facts relevant to those causes of action, so as to attract section 32(1)(b) of the Limitation Act.
397. I deal with the two main issues, as they were before the judge, the tort of deceit and the claim under the Limitation Act. I do not address the allegation under the respondent's notice of deceit through failure to correct a previous statement, because I respectfully agree with the observations on that issue that are to be found in §§ 294ff of the judgment of Rix LJ.

The claim in deceit

The elements of the tort

398. Clerk & Lindsell (19^th edition) §18-01 says of the tort of deceit:

The tort involves a perfectly general principle. Where a defendant makes a false representation, knowing it to be untrue, or being reckless as to whether it is true, and intends that the claimant should act in reliance on it, then in so far as the latter does so and suffers loss the defendant is liable for that loss.

And in more detail as to the mental element of the tort, no-one has improved, or sought to improve, on the analysis of Lord Herschell in Derry v Peek (1889) 14 App Cas 337 at p376:

First, in order to sustain an action of deceit, there must be proof of fraud and nothing short of that will suffice. Secondly, fraud is proved when it is shown that a false representation has been made (i) knowingly, (ii) without belief in its truth, or (iii) recklessly, careless whether it be true or false. Although I have treated the second and third as distinct cases, I think the third is but an instance of the second, for one who makes a statement under such circumstances can have no real belief in the truth of what he states. To prevent a false statement from being fraudulent, there must, I think, always be an honest belief in its truth.

399. Additionally, for a case of fraud certain other requirements must be fulfilled. Those are:

i) The false statement must be clearly identified.

ii) The primary facts relied on as establishing the terms of the statement, and the falsity of it, must be proved by cogent evidence.

iii) Where knowledge of a fact held by the speaker is relied on to make his statement deceitful, he must be "fully conscious" of that fact and have conscious knowledge of the falsity of his statement: Armstrong v Strain [1951] TLR 856 at p 871, per Devlin J.

The judge's findings

400. The judge concluded that Mr Lucas committed deceit in his conversation with Mr Whitaker on 17 April 1996. The elements of that finding were:

i) "the overall impression conveyed by Mr Lucas during the telephone conversation on 17 April was that the certificate for the regular cargo was and remained a good and reliable certificate": Judgment, §283.

ii) "the overwhelming probability" was that Mr Rackham informed Mr Lucas of the results of the Cooper re-tests in the course of briefing Mr Lucas before the telephone conversation: Judgment, §274.

iii) Mr Lucas therefore knew from the Cooper re-tests that the RVP results stated in the ITS certificates were probably wrong; and that the Cooper re-tests were a key piece of evidence, highly relevant to that conclusion: Judgment §§ 202-203 and 283 (d)-(h), set out in §§ 90 and 93-94 above.

iv) Mr Lucas therefore also knew that in all the circumstances known to him it was wrong for ITS to maintain that a certificate which said "Fuel meets Specification" was and remained a good and valid certificate: Judgment, § 283(i).

v) The representation that the certificate was and remained a good and valid certificate was a false representation made by Mr Lucas who was reckless as to its truth: Judgment, §284.

vi) Mr Lucas intended that AIC should act on that representation and AIC did so: Judgment, §284.

The approach of this court

401. This court is subject to well-recognised inhibitions when considering findings of primary fact by the trial court. It is only necessary to set out two statements of high authority:

The principle is well settled that where there has been no misdirection on an issue of fact by the trial judge the presumption is that his conclusion on issues of fact is correct. The Court of Appeal will only reverse the trial judge on an issue of fact when it is convinced that his view is wrong. In such a case, if the Court of Appeal is left in doubt as to the correctness of the conclusion, it will not disturb it: per Lord Steyn in Smith New Court v Scrimgeour Vickers [1997] AC 254 at pp 274H-275A

The need for appellate caution in reversing the Judge's evaluation of the facts is based upon much more solid grounds than professional courtesy. It is because specific findings of fact, even by the most meticulous Judge, are inherently an incomplete statement of the impression which was made upon him by the primary evidence. His expressed findings are always surrounded by a penumbra of imprecision as to emphasis, relative weight, minor qualification and nuance…..of which time and language do not permit exact expression, but which may play an important part in the Judge's overall evaluation: per Lord Hoffmann in Biogen v Medeva [1997] RPC 1 at p 45.

402. That diffidence must be contrasted with the court's role assessing the conclusions drawn by the judge from the facts that he does find, or from sources, usually documents, that are equally accessible to this court as they are to the trial judge. Those principles can be illustrated from the issue, crucial in the present case, of whether Mr Lucas, on the assumption that he then knew about the Cooper re-tests, consciously made a false representation in the conversation of 17 April 1996. Where, as in the present case, the representation alleged is not in the form of a verbatim statement, but of an "overall impression" (see § 400(i) above), then that conclusion has to be drawn from the primary facts, and the judge's view on it, though of course very valuable, cannot be dispositive. And further, as to the speaker's state of mind, and motives, although the judge will have had a great advantage in seeing and hearing the witness in assessing the validity of his denial of dishonesty, in many cases, of which this is one, that will not be the only or even the primary source on which the judge bases his conclusion. Rather, the judge will base himself on the logical consistency and plausibility of the witnesses' claims in the context of the primary facts. This court, though again finding the judge's conclusion of great value, will not be inhibited from subjecting that process of logic and deduction to logical scrutiny. That is particularly so in that serious category of cases where the judge has made findings of fraud. Where a charge of fraud against a witness whom the judge has seen and heard has been rejected it is unlikely that this court would feel able to reinstate the charge; but the converse does not follow.
403. With those principles in mind I review the elements in the judge's findings.

The representation

404. The importance of clearly identifying the terms of the representation that is alleged is that it is that representation, and no other, that must be shown to have been made with knowledge of its falsity. Accordingly, not only must the representor have been aware that he was saying what the court finds him to have said, but also he must have been aware that in saying that he was not telling the truth. The principal complaints made by ITS of the judge's approach were: (i) it was, if not plainly wrong, then at least hazardous to identify a representation that is subject to the foregoing rules on the basis of "general impression"; (ii) that approach made it the more difficult to conclude, with the required degree of certainty, that Mr Lucas knew that he was making that particular representation; (iii) the facts strongly militated against Mr Lucas having created the impression found by the judge.
405. Before examining these complaints, it will be convenient to consider how the matter was understood by the recipient of Mr Lucas's remarks; and in that connexion to say something of the impact of Veba Oil.
406. Although Mr Whitaker did not give evidence, the judge in his §221 accepted evidence from another AIC executive:

Mr Sampino of AIC said in his witness statement "… we found out … that the method used to test the cargo's RVP at the load port was not the method stated in the CPS. However, I also remember Mobil insisting that the cargo was on-spec and ITS saying they would stand by the certificates of quality. ITS even accused Tom Whitaker of asking them to commit fraud, just because he asked them to correct the certificates of quality if they were not accurate … in the circumstances, despite Galaxy's complaints, we had no choice but to pay Mobil. Mr Hatcher and Mr Whitaker took that decision, but I would not have acted differently. A company like AIC does not refuse to pay an oil major without a very good reason, and certainly not when it is being told by the oil major that there is nothing wrong with a cargo and cargo inspectors are standing by certificates of quality that are to be final and binding.

Further, there was unchallenged evidence from a Mr Payne, a person intimately involved in dealing with the Kriti Palm, that having been told that ITS was standing by its results, and that it was impossible to retest the samples, AIC felt obliged to pay Mobil, and thereafter became drawn into litigation with Galaxy on the basis of the certificate: see §§ 154ff above.

407. All this, however, gives rise to a considerable difficulty in the light of what we now know. On the facts well-known to AIC, the wrong test had been used, and therefore (leaving aside entirely the Cooper re-tests or even any question as to what the proper test would actually show as to RVP) the certificate was invalid: see the exposition of the Veba Oil issue in §§ 29-44 above. So if Mr Lucas said that the certificate was good and valid, that was untrue; and if, as they did, AIC relied on the certificate in their dealings with Mobil and Galaxy, they did so under a mistake of law of their own making. However, not only AIC itself, but also its lawyers at trial had failed to grasp the implications of the errors in the certificate: see §39 above. And the parties in this appeal were so far from thinking that Veba Oil had any relevance to the issues before us that they did not include the case in their joint bundle of authorities: a compilation that, whatever else might be said about it, revealed no diffidence about putting before the court anything that might even remotely bear on the issues in the appeal. Although the position may not be entirely satisfactory, the only possible course for this court is to proceed on the basis adopted by the parties in their actual dealings, that as between them the use of the wrong test did not in itself destroy the certificate.
408. I consider, therefore, that a representation that the certificate "was and remained good and reliable" would have made sense between the parties in their state of understanding of the case. I am unable to agree with what I understand to be Rix LJ's view, set out in §§ 267-268 above, that such a representation would have been incoherent in any event. But, nonetheless, was that a representation that Mr Lucas in fact made?
409. The course of the conversation between Mr Lucas and Mr Whitaker is described in detail in §§ 104-118 above. What the judge regarded as the critical statement is recorded in §110: "we will be standing by that certificate". If that statement alone stood, I would regard it as validating the judge's construction of the conversation. The whole context of the conversation was that Mr Whitaker wanted to know whether he could rely on the certificate, and rely on ITS as certificator. However little Mr Lucas knew about the history of this particular dispute, as a senior executive of a testing house he must have known of the role that certificates would play in his client's affairs. Arguments such as those pursued before us that all that this statement meant was that the readings on the certificate would not be altered pay insufficient respect to the context in which the statement was made. Taken on its own, this statement did give the comfort that AIC drew from it and to their reliance on it: see §406 above.
410. But the statement did not stand on its own. When Mr Whitaker pressed Mr Lucas as to whether or not the actual values in the certificate were correct, Mr Lucas was notably reluctant to commit himself: see §§ 111-112 above. That attitude may not have been very attractive, and might cause ITS difficulty in a contractual or negligence claim, but it plainly was not a positive assertion of the kind, or in the terms, that the judge construed out of the conversation. And it was the values that mattered, not only to Mr Whitaker, but also in the context of the present claim; because the foundation of the case against Mr Lucas is that when he spoke he knew of the Cooper re-tests, whose importance is as to the light they cast on the accuracy of the values: see § 400(iii) above.
411. The position that Mr Lucas took in the conversation as to the values in the certificate in my view makes it impossible to read the conversation taken as a whole as containing the assertion found by the judge. And I agree that that is borne out by Mr Whitaker's understanding expressed on the following day: see §125 above. Others in AIC may have taken comfort from ITS's position, including its decision not to withdraw the certificate, and might well be able to complain of the damage that they suffered in trying to escape from the dilemma into which ITS's attitude placed them. But that does not lead to a conclusion that in the conversation itself Mr Lucas made the assertion that the judge found.
412. If the statement on which the finding of deceit was based by the judge is not sufficiently shown to have been made, then that finding of deceit plainly cannot stand, however much it may be claimed that the speaker misled his audience in other ways. That is also the conclusion reached by Rix LJ in §280 above. However, like my Lord, I will go on and review the other elements in the judge's finding of deceit, not least because some parts of that enquiry are also relevant to the second part of the case relating to deliberate concealment.

Mr Lucas's knowledge

413. The judge made findings as to the knowledge of Mr Lucas (and of Mr Chalmers) at two places in his judgment. The first was in his §§ 198-203, the passage set out in §90 above. As indicated in §91 above, that passage is open to criticism, because it indicates or might indicate that the judge was attributing to Messrs Lucas and Chalmers in 1996 knowledge and understanding that had only been exposed before him after elaborate expert evidence at a trial in 2005. I think that there is some force in that complaint. It should perhaps be remembered that in that passage the judge was answering the question posed for him by the parties as to whether the RVP results in the certificate were accurate. His observation in his §203, after he had answered that question, that Messrs Lucas and Chalmers knew the matters set out in his answer, could perhaps at that stage of the judgment be characterised as gratuitous.
414. However, the judge returned to that issue in his §283, when answering the specific question: Was Mr Lucas dishonest in what he said and did not say during the telecon on 17 April 1996? The answer that he gave to that question is set out in §93 above. In that part of the judgment there is no direct reference, and almost no inferential reference, to the view of the experts. Almost the whole of the judge's emphasis is on Mr Lucas's knowledge of the Cooper re-tests and of their importance. To those subjects I now turn.

Were the Cooper retests regarded as important?

415. I take this issue first, because it is of crucial importance in assessing the actions and evidence of Messrs Lucas and Chalmers.
416. At the trial ITS gave evidence, particularly through Mr Chalmers, that the Cooper retests had been an inappropriate step, involving the plainly unreliable testing of residues or "slops"; they were a private initiative of Mr Rackham; and Mr Chalmers if he had known about them would have rebuked Mr Rackham in forceful terms for engaging in that operation. That evidence is summarised in §§ 222-226 above. If reliable, it lends credibility to the claim that, contrary to the judge's findings, ITS through Mr Lucas did not appreciate that the Cooper retests cast serious doubt on the values stated in the certificate. The judge did not accept Mr Chalmers' evidence on this issue. I understand from §§ 291-293 above that Rix LJ considers that the judge was wrong in so holding. I cannot, with deference, accept that that position is open to us in the light of the principles set out at the start of this judgment. However, if that seems too simplistic a way of addressing this point I would mention a number of matters that support the judge's conclusion.
417. First, Mr Rackham regarded the step that he took as appropriate and important. In his early morning e-mail of 17 April he reported the findings of the Cooper retests, which were detrimental to ITS, not only to Mr Chalmers but also to Mr Loughead, who was the senior manager of all the people involved in this matter: see §§ 83-84 above. Second, on the previous day Mr Rackham had had his 1333 conversation with Mr Chalmers, noted by the enigmatic "JC re Kriti Palm. Harass UK. Need low RVP". The parties' cases as to this note, and the judge's conclusions not only as to the facts but also as to their implications, are set out in §§ 74-76 above. I am afraid that I found ITS's construction, which does not appear to have been advanced at the trial, completely unconvincing. The note is simply not what Mr Rackham would have written if reporting a complaint by him to Mr Chalmers of AIC's aggressiveness, nor did Mr Rackham suggest that it was. And if that had been Mr Rackham's objective it is inconceivable that he would not at the same time at least have mentioned to Mr Chalmers how he had reacted, by commissioning the Cooper retests: at which stage Mr Chalmers, according to his evidence at the trial about his view of the Cooper retests, would have blown up all over the telephone.
418. When reviewing Mr Chalmers's evidence as a whole the judge said that he had had "marked reservations" about parts of that evidence: Judgment, §109. The judge is criticised by ITS for not expanding on those reservations, and I agree that it would have been better, in a case such as the present, if he had said more on the subject. I however remind myself of what Lord Hoffmann said in Biogen, §401 above. And the difficulty already explained of reconciling Mr Chalmers's claims in evidence with the contemporary documents, such as they were, only serves to reinforce the judge's unease.
419. The judge therefore was justified in finding that the Cooper retests were regarded as important by ITS, and were not merely a private exercise initiated by Mr Rackham. He also found, a matter to which we shall have to return, that Mr Chalmers knew about the Cooper retests and their outcome on 17 April.

Was Mr Lucas told about the Cooper retests?

420. The judge held, in his §274, that in his judgement

the overwhelming probability is that Mr Rackham informed Mr Lucas of the Cooper retests and the results thereof in the course of briefing Mr Lucas prior to the telephone conversation on 17 April.

Before us, this finding was strongly attacked. It was said to be inconsistent with the attitude taken by Mr Lucas both in the telephone conversation and in his "follow up" fax (§§ 123-124 above), to the extent that if he then knew of the Cooper retests both the conversation and the fax were cynical exercises in deceit; and inconsistent with, or at least unlikely in the context of, the hurried encounter with Mr Rackham, in a matter about which Mr Lucas previously knew nothing, and under the exigencies of the customer's demands.

421. I would respectfully agree with Rix LJ, §§ 281-282 above, in rejecting those arguments. Just as with the judge's findings in respect of the knowledge of Mr Chalmers, so in the case of Mr Lucas it is very difficult for this court to intervene; and, as in that earlier case, there are a number of commonsense considerations that support the judge's conclusion in any event. Central to investigation of what Mr Rackham told Mr Lucas must be Mr Rackham's reasons for commissioning the Cooper retests, and the weight that he placed on them. We have already explored that issue. It is really very implausible that when he briefed his superior about how he had been holding the fort in his absence he would have told Mr Lucas what the problem was, but not what he had done to try to meet that problem; and almost equally implausible that if he had omitted that last stage, Mr Lucas would not have pressed him on it.

Does that knowledge entail dishonesty on Mr Lucas's part?

422. I would therefore uphold the judge's finding set out in §420 above. That however is very far from the end of the story. We have already noted the nature of the knowledge that is required to establish a case of deceit: see the observation of Devlin J in Armstrong v Strain cited in §399(iii) above. Even if (which for the reasons set out above I do not accept) Mr Lucas can be construed as having in the conversation made the representation found by the judge, it does not follow from his simple knowledge of the Cooper retests that he must have realised, with the degree of immediacy and certainty necessary for a finding of deceit, that what he was saying about the certificate was falsified by the Cooper retests. The points as to the circumstances of the conversation, Mr Lucas's lack of preparation for it, and his reaction after it, listed in §420 above, and the further considerations set out by Rix LJ in §285 above, are directly relevant to that issue, even though they are not dispositive on the prior issue of whether Mr Lucas was told about the Cooper retests.
423. The judge did not address that issue, but moved straight from his finding that Mr Lucas had been told about the Cooper retests to a finding of dishonesty. That process omitted an essential step in the consideration of Mr Lucas's state of mind. Even if there were not the difficulty about the characterisation of the alleged representation, that omission by the judge would in any event prevent me from upholding his finding of deceit.

The limitation issues

The law, and the questions that it poses in this case

424. The way in which these issues arise has been set out in §15 above. The governing law is to be found in section 32 of the Limitation Act 1980:

(1)….where in the case of any action for which a period of limitation is prescribed in this Act, either-

a) the action is based upon the fraud of the defendant; or

b) any fact relevant to the plaintiff's right of action has been deliberately concealed from him by the defendant,….

the period of limitation shall not begin to run until the plaintiff has discovered the fraud [or] concealment….(as the case may be) or could with reasonable diligence have discovered it.

(2) For the purposes of subsection (1) above, deliberate commission of a breach of duty in circumstances in which it is unlikely to be discovered for some time amounts to deliberate concealment of the facts involved in that breach of duty.

425. The judge held that the requirements of the statute were fulfilled in the present case. At his §312 he said

I find that Mr Lucas and Mr Chalmers made a deliberate decision not to disclose the fact of the Cooper re-tests and the results thereof to AIC and Mobil in circumstances in which the re-tests and the results thereof should have been disclosed to AIC and Mobil, and Mr Lucas and Mr Chalmers knew that the re-tests and the results thereof should have been disclosed to AIC and Mobil. I further find that Mr Lucas and Mr Chalmers were aware that ITS was under such a duty and that Mr Lucas and Mr Chalmers made a deliberate decision not to disclose the re-tests and the results thereof.

And he further held, at his §§ 317 and following, that the fact of the Cooper retests was a fact relevant to AIC's right of action against ITS.

426. In reaching those conclusions the judge saw himself as applying two recent authorities, both of them binding upon us, Cave v Robinson Jarvis & Rolf [2003] 1 AC 384, and Williams v Fanshaw Porter & Hazelhurst [2004] EWCA Civ 157. In Cave he relied on what was said by Lord Millett at §25, and by Lord Scott at §60. Lord Millett said:

In my opinion, section 32 deprives a defendant of a limitation defence in two situations: (i) where he takes active steps to conceal his own breach of duty after he has become aware of it; and (ii) where he is guilty of deliberate wrongdoing and conceals or fails to disclose it in circumstances where it is unlikely to be discovered for some time. But it does not deprive a defendant of a limitation defence where he is charged with negligence if, being unaware of his error or that he has failed to take proper care, there has been nothing for him to disclose.

Lord Scott said:

Subsection (2), however, provides an alternative route. The claimant need not concentrate on the allegedly concealed facts but can instead concentrate on the commission of the breach of duty. If the claimant can show that the defendant knew he was committing a breach of duty, or intended to commit the breach of duty - I can discern no difference between the two formulations; each would constitute, in my opinion, a deliberate commission of the breach - then, if the circumstances are such that the claimant is unlikely to discover for some time that the breach of duty has been committed, the facts involved in the breach are taken to have been deliberately concealed for subsection (1)(b) purposes.

In Williams Park J said, at §14(iv):

The requirement is that the fact must be 'deliberately concealed'. It is, I think, plain that, for concealment to be deliberate, the defendant must have considered whether to inform the claimant of the fact and decided not to. I would go further and accept that the fact which he decides not to disclose either must be one which it was his duty to disclose, or must at least be one which he would ordinarily have disclosed in the normal course of his relationship with the claimant, but in the case of which he consciously decided to depart from what he would normally have done and to keep quiet about it.

The last part of that statement, with its reference to disclosure in the ordinary course of duty, is controversial, but the narrower test of breach of duty was endorsed in the same case by Brooke LJ, who said at his §51:

The claimant did not know a fact relevant to her cause of action until a date less than six years before this action was brought, and the reason why she did not know it was that Mr Brown intentionally concealed it from her when he was under a duty to tell her about it. These facts appear to me to fall within the compass of Lord Scott's exposition of the effect of section 32(1)(b) of the 1980 Act in paragraph 60 of his speech in Cave.

427. That exposition, and the judge's application of it, raises the following issues for this court:

i) Was ITS under a duty to AIC to reveal the existence and content of the Cooper retests?

ii) Did ITS, knowing of that duty, decide not to reveal the existence and content of the Cooper retests?

iii) Were the Cooper retests relevant to any and if so which of the rights of action asserted by AIC that are otherwise statute-barred?

ITS's duty to reveal the Cooper retests

428. Before the judge, and to an even larger extent before us, this issue became bogged down in a detailed examination of the duties imposed on ITS by various international standards and practices; the relevance or otherwise of any and if so which ITS internal guidelines; and whether the materials on which the Cooper retests were performed were indeed "retained samples" within the terms of some of the requirements of those guidelines. An account of these exchanges is given in §§ 169-192 above. That account, with respect, says everything that is necessary to understand this issue, although it by no means reproduces every point that was raised before us.
429. The point is fairly made that the judge himself attached importance to the analysis of the international guidelines and to his findings as to the duties of a testing house, to the extent that it was because the judgment contained guidance on those matters, and for no other reason, that he granted permission to appeal. It is, however, relevant to observe that the judge was led into this enquiry by its being included in the question paper set for him by the parties: which perhaps here, as elsewhere, he over-indulgently allowed to dictate the shape of his judgment. The question that the judge was asked is issue 2:

"Was ITS retained by Mobil and AIC in any kind of advisory capacity? What was the proper scope of the contractual duties owed by ITS to AIC?"

430. That question was not framed in terms that are helpful in the enquiry under the Limitation Act. The issue there, and in this case, is not whether ITS acted in an "advisory capacity" to Mobil and AIC, because plainly, apart from its obligations as certifier, it did not do so. Nor is the question of whether in the terms of the Act ITS had a duty to reveal the Cooper re-tests necessarily determined by reference to ITS's contractual duties. However, that question having been posed, ITS answered it by saying that once the inspector had issued his certificate he was functus officio in the same way as an arbitrator after the publication of an award. If the inspector subsequently discovered that he had not performed or had wrongly expressed his mandate (for instance, if he had used the wrong samples; or the certificate misstated the test that had been used), then he should issue a corrective letter to both parties. But if information comes to hand after the giving of the certificate that simply might cause the certifier to alter an opinion as to the values set out in the certificate (a category into which the Cooper retests would appear to fall) there was no obligation to correct the certificate, provided that the opinion expressed in it, in this case by Mr Mailey, had originally been honestly held.
431. As Rix LJ points out in his §§ 334 and 338, Toepfer v Continental Grain Co [1974] Lloyd's Rep 11 is authority for saying that in the absence of manifest error the certificate remains binding between buyer and seller, and the certifier is, unlike, say, a solicitor who, in the terms of the question put to the judge, does have some continuing duty to look after his client's interests. The judge was however taken aback by ITS's extension of that principle to a claim that even in circumstances such as those of the Cooper re-tests there was no question of ITS having to do anything. It was apparently he who required to see ITS's internal documents, having heard expert evidence about international standards that was difficult to reconcile with ITS's approach: see the Judgment at §152, and also §180 above. It is therefore hardly surprising that the judge took his guidance as to the process required from an inspection agency from what he understood was this agency's own internal procedure.
432. It is now complained that the judge was wrong because he referred to the wrong ITS standards, and relied on those that applied to laboratories, rather than to the standards that applied to the function that was performed by Mr Cooper: see §173 above. But I accept Mr Hamblen's submission that the relevance of all of this material is not so much to show failure to comply with a particular and detailed requirement, but rather to demonstrate the general obligations of a testing house when faced with as singular a development as the Cooper retests. And that is how its duties were seen by ITS itself.
433. First, ITS was so far from running its affairs by strict reference to the relevant internal procedure that it was ITS itself that referred the judge at trial to what is now said to be the wrong internal guidance: see §§ 175-176 above. And that the judge had used the wrong internal guidance was so far from the mind of the executives actually running ITS that the point did not feature either in the Grounds of Appeal or even in the 168 page skeleton on the basis of which the appeal was opened before us, but only emerged on something like the second day of the hearing. AIC were justified in complaining that by then the point could not in any event be open to ITS, and I regret that I did not take that complaint more seriously. However, the practical effect of that history is to cast the greatest doubt on whether the relevant ITS executives thought the detail of the various manuals to be relevant to their duty to their client. As AIC pointed out, Mr Chalmers was in court throughout the trial and was aware of the judge's wish to identify all relevant standards and procedures. If the inspection manual now asserted to be the sole source of ITS's duty had been seen by ITS as significant as a matter of commercial reality it would certainly have been relied on before the judge; and even if overlooked then would have emerged much earlier in what has clearly been the most detailed and anxious preparation of the appeal.
434. Second, there is no doubt that it was ITS's evidence that, if the Cooper retests had been known to the relevant executives, then a different attitude should and would have been taken towards AIC (and, by the same token, towards Mobil). That evidence, not only from Mr Lucas and (to a lesser extent) from Mr Chalmers, but also from their superior Mr Loughead, was given in the context of claims that neither Mr Lucas nor Mr Chalmers knew of the Cooper retests at any relevant time, and that in any event the Cooper retests if known to them would have been discarded as useless. As we have seen, the judge rejected all of those claims, in terms that are not open to challenge in this court; but the evidence with which we are concerned was not contingent on those assumptions, and the judge's findings on them cannot affect, and was not suggested to affect, the validity (or the integrity) of the answers that were given.
435. Mr Loughead did not give oral evidence, ITS standing on his witness statement. In that statement he said in relation to the Cooper retests:

Thinking about the matter now, I would say that I regarded the samples as potentially important and would not have regarded it as appropriate to take them and use them.

Mr Lucas agreed, Day 6 page 18 line 22, that as potentially important samples ITS should not have been doing anything to them without informing its client. And he agreed that that information to the client would have included telling AIC what the Cooper retests had shown. That was confirmed in an exchange in cross-examination in the context of the conversation with Mr Whitaker, Day 5 pages 146-147:

Q. Obviously if you had been aware of evidence that the cargo was indeed off-spec in the shore tanks, then you would have been in a position to comment, would you not?

A. If the Mailey results were wrong, then yes, we would have been having a different conversation, Tom Whitaker and I.

Q. If you had had knowledge of the Cooper retests, you would indeed have had evidence, would you not, that the cargo was indeed off-spec in the shore tanks?

A. The point that I would have had to make to Tom Whitaker was that we have got this unusual set of results and I am struggling to rely upon it and I certainly could not produce a certificate from it. That is what I would have had to say to Tom.

Q. But you would have to tell him what you had found?

A. Of course.

Q. With those qualifications?

A. Yes.

436. Mr Lucas's acceptance of a duty to reveal the Cooper retests, if they were known about, is further demonstrated by his use of that duty syllogistically in an argumentative reply when under cross-examination as to his knowledge of those tests. Mr Lucas said, Day 5, page 148, line 6, that the fact that he had a duty to reveal the Cooper retests and had not done so demonstrated that he did not know about the tests: because if he had known of them he would have performed his duty to reveal them.
437. The evidence of Mr Chalmers relevant to this issue is set out in detail in §§ 222-236 above. He dealt with the matter by saying that even if he had known about the Cooper retests, which he did not, he would not have contemplated telling AIC about them. That was not because of any lack of duty to AIC to disclose relevant information, a position that he could not have maintained in the light of his answers recorded at §§ 225 and 233 above; but because if told about the Cooper retests he would have thought them to be worthless. In that he differed from the view of Mr Loughead, §435 above, a position in which Mr Chalmers said he often found himself. And, more pressingly, this part of Mr Chalmers's evidence was rejected by the Judge, in terms with which this court cannot interfere: see §§ 417-419 above. The upshot is that Mr Chalmers gave no evidence that ITS can put against Mr Lucas's acceptance that if the Cooper retests were known about they should have been disclosed to AIC, albeit with whatever "health warning" ITS thought it appropriate to attach to the work of Mr Rackham and of Mr Cooper.
438. It follows from what I have already said that I cannot, with deference, accept Rix LJ's dismissal of these admissions by ITS as having been made on certain factual hypotheses and/or subject to hindsight and/or qualifications: see §330 above. The hypotheses were in fact states of affairs that the judge found to have been proved; and the statements were not accidental or thoughtless, but considered assertions by senior and experienced men in the industry.
439. Not only was a duty to disclose the Cooper retests acknowledged by ITS, but also the existence of such a duty is a matter of commonsense. We have seen the unchallenged evidence of AIC as to the importance to them of the accuracy of the statements of quality on the certificate: see §406 above. The judge also stressed, and no-one has suggested that he was wrong, the importance of the accuracy of a certificate to buyers and sub-buyers: Judgment, §183. Not only as a matter of law, but also commercially, it really challenges reality to think that a certifier, armed with tests that suggested that the tests used to complete the certificate had or might have produced incorrect results, could nonetheless simply do nothing about it; and in particular could properly say nothing about those tests to those who had employed him to certify.
440. That obligation, of not sitting on material of one's own creation that is known to be inconsistent with the certificate, does not lead to the endless uncertainty that Rix LJ fears in his §345. It is for the holder of the certificate to decide what he does with the information once he receives it. What I cannot accept is that considerations of certainty empower the certifier to take that decision for the holder by withholding relevant information from him. Nor does such a duty create a continuing duty of review and disclosure under every conceivable kind of contract: Cf §351 above. This is the specific case of a certificate, where the certifier was in possession of material of his own creation that cast doubt on the certificate that he had given. To hold, as I would, that he was under a duty in Limitation Act terms to reveal that material does not open any floodgate in any other sort of relationship.
441. One further point should be mentioned. When pressed in cross-examination about when he first knew of the Cooper retests, and why on the assumption that he knew of them during his exchanges with Mr Whitaker he did not mention them, Mr Lucas said that in any event it was ITS's policy to adopt a "stone wall" once a claim was notified: see the evidence summarised at §197 above. It was only faintly suggested to us that the existence of a claim would justify non-disclosure, in the sense of taking outside the ambit of section 32 what otherwise would be a relevant breach of duty; but I address the point nonetheless.
442. First, I cannot accept the argument as a matter of law. If ITS did have a duty to tell AIC about the Cooper retests, it cannot exempt ITS from discharging that duty that AIC are threatening to sue ITS on a basis that, by definition, does not embrace the Cooper retests. How ITS protects itself in litigation is its own business, but it cannot use that protection to place its own interests in front of its duty to its client. Second, Mr Lucas could not on his own evidence assert that this was the reason for his decision, because on that evidence he knew nothing of the Cooper retests, and therefore the issue could not have arisen. Third, although in the conversation Mr Whitaker certainly suggested that he might be forced to bring proceedings, it was not the case that during or immediately after that conversation things had reached the stage of a formal "claim", so as to excite the ITS policy referred to by Mr Lucas; nor did Mr Lucas ever break off relations with Mr Whitaker in the way that would apparently have been required by ITS's stone-wall policy had he perceived himself to be "under notice of claim".
443. I am therefore satisfied that in the sense relevant to the 1980 Act interpreted in the authorities set out in §426 above ITS had a duty to reveal to AIC the existence and content of the Cooper retests.

The decision not to perform that duty

444. It will be convenient first to deal with ITS's argument, that if ITS were to be acquitted of deceit, it was most unlikely that the company could not have engaged in deliberate concealment. That contention is wrong, both as a matter of general principle and on the facts of this case.
445. As to general principle, in Kitchen v RAF Association [1958] 2 All ER 241 at p 249C Lord Evershed MR said:

It is now clear…that the word "fraud" in s. 26(b) of the Limitation Act 1939 is by no means limited to common law fraud or deceit. Equally, it is clear, having regard to the decision in Beaman v ARTS Ltd [1949] 1 All ER 465, that no degree of moral turpitude is necessary to establish fraud within the section.

It is accepted that that jurisprudence was successfully captured by the terms of the 1980 Act, referring as it does to deliberate concealment rather than to fraud.

446. Second, as to the facts of this case, the allegation of deceit was that Mr Lucas, in his conversation with Mr Whitaker, made a positive representation about the certificate, which representation was false because of Mr Lucas's knowledge of the Cooper retests. That would have been the case irrespective of whether or not Mr Lucas told AIC about the Cooper retests. That is quite different from a complaint that is limited to saying that Mr Lucas should have told AIC about the Cooper retests so that AIC could make up its own mind as to the course that it should take. I appreciate that Rix LJ does not take that view, and concludes in his §§ 314 and 366 that the rejection of the judge's conclusion in relation to deceit leads to the rejection of his conclusion as to deliberate concealment. For the reasons set out earlier in this paragraph I am with deference unable to agree.
447. Turning therefore to the issue of whether ITS knowingly breached its duty, the judge's conclusion is set out in §425 above. That conclusion was in the face of an express denial by Mr Chalmers (see §235 above); and a slightly more nuanced denial by Mr Lucas, Day 6 pages 20-21:

Q. I suggest to you, Mr Lucas, that you and Mr Chalmers did consider at the time whether AIC should be informed about these retest results and you decided that they should not be.

A. I can tell you that I have no recollection of ever making that decision.

Q. And that you decided that they should not be even though you appreciated, in the light of what you had said to Mr Whitaker and your certificate, that it is something they ought to have been told about?

A. That is not true.

448. The difficulty for ITS is that these denials assumed various facts that the judge found not to have been the case. Given that on the judge's findings both Mr Lucas and Mr Chalmers knew about the Cooper retests; and that, contrary to their evidence at the trial, they had thought the Cooper retests to be significant and relevant to the problem that they were facing; and given that it was accepted by them that on those assumptions there was a duty to communicate with AIC; it is difficult to see how the judge's conclusion as to the decision taken by Mr Lucas and Mr Chalmers does not follow almost as a matter of inevitability. Certainly, it is not open to this court to reverse the judge on this point.
449. It should perhaps be added that there might have been a variety of reasons why Mr Lucas and Mr Chalmers acted as the judge found. I have dealt in §§ 441-442 above with the one matter that has been expressly raised; but it is not possible to explore the general issue further in view of the blanket claims of ignorance of the existence of the Cooper retests that formed ITS's defence on this point at the trial.
450. I therefore conclude that the Judge was correct to hold that for the purposes of section 32 ITS deliberately concealed the Cooper retests and the results of those tests. The next question is therefore whether those facts were relevant to AIC's rights of action.

The relevance of the Cooper retests to AIC's rights of action

451. Rix LJ has, with respect, effectively criticised the judge's handling of this issue. What has to be said to have been the somewhat perfunctory treatment may be attributable to the issue not having been, on the judge's findings as to deceit, of prime importance before him. But, as my Lord says, the issue for us is whether the argument can be made good in this court.
452. Two preliminary points should be made. First, it has not been suggested that "right of action" is different from cause of action. That seems to have been assumed by this court in Johnson v Chief Constable of Surrey (unreported, 19 October 1992), and was confirmed in terms by Neill LJ, one of the judges who sat in Johnson, in C v Mirror Group Newspapers [1997] 1 WLR 131 at p 136H.
453. Second, as Rix LJ emphasises, Johnson stands as authority for the proposition that what must be concealed is something essential to complete the cause of action. It is not enough that evidence that might enhance the claim is concealed, provided that the claim can be properly pleaded without it. The court therefore has to look for the gist of the cause of action that is asserted, to see if that was available to the claimant without knowledge of the concealed material. In Johnson the plaintiff sought to bring a claim of false imprisonment against the police, and discovered more than six years after his arrest that the police had, allegedly, concealed from him material that demonstrated that they had had no genuine belief in his guilt. Neill LJ said at p 8E of the transcript:

the gist of the action of false imprisonment is the mere imprisonment. The plaintiff need not prove that the imprisonment was unlawful or malicious; he establishes a prima facie case if he proves that he was imprisoned by the defendant. The onus is then shifted to the defendant to prove some justification for it.

The same analysis was adopted by Russell LJ at p 7D, holding that the "right of action" was complete at the moment of arrest.

454. It is accordingly necessary to look at the causes of action now asserted by AIC on the basis of their knowledge of the Cooper retests. AIC's pleading, in §§ 21-23B3 of its heavily re-amended Particulars of Claim, is a complex document, difficult to summarise. No criticism is however to be made of the pleader, because he had to work in the context of the serial disclosures as to the relevant events that emerged from ITS in the course of the proceedings. However, in the hope of being able legitimately to shorten consideration of this point I append those paragraphs to this judgment. It should be noted that the pleading includes, without any very clear demarcation, the allegations of deceit that I have rejected earlier in this judgment.
455. ITS attacked the claim that section 32 was satisfied essentially on the basis that throughout the pleader had had available material that enabled him to plead his complaints of breach of contract and of negligence, and that the Cooper retests merely added evidence that strengthened those cases. Before turning to that general submission, it is necessary to refer to one particular complaint.
456. Mr Gaisman accepted, as I understood it, that the plea in §§ 23B1-2 was dependent on knowledge of the Cooper retests. That he agreed followed by definition: one cannot know that x has not been revealed to you if you do not know that x exists. He of course contended that there was no duty to reveal the Cooper retests, a contention that in company with the judge I have rejected. But Mr Gaisman in effect said that it was an abuse of section 32(1)(b) to apply it to such a case. The claim was parasitic on the other claims; added nothing to them by way of loss or damage suffered by AIC; and was simply a device to get some sort of complaint about the concealment of the Cooper retests on its feet. I see the force of those contentions, but no authority was cited for the suggestion that they suffice to exclude the application of section 32 in limine, and such a rule would seem inappropriate if applied to causes of action that are not suggested to be abusive per se. Whether these particular claims add anything to whatever else AIC may recover is of course a different matter.
457. Turning to the wider case pleaded by AIC, at trial ITS appear to have taken an extreme stand, contending that since AIC had been able to plead some claim in contract and negligence in respect of the certification process, and since the Cooper retests had emerged in the course of disclosure during those proceedings, it followed that the latter material was not necessary to complete AIC's cause of action: see the Judgment at §315. I did not understand a view as fundamental as that to be pursued before us. Rather, it was contended that, looking in more detail at the claims now sought to be made, AIC had been able, before it had access to the Cooper retests, to plead, first that the wrong test, D323 rather than D5191, had been used; and second, on the basis of the disport tests (see §61 above) would if so minded have been able to plead that, contrary to the certificate, the cargo was or might be off-specification. The Cooper retests were irrelevant to the first of those claims, and in respect of the second of them only added further evidence.
458. I consider that these arguments do not sufficiently respect the nature of a cause of action in negligence, or breach of contract, when applied to the facts of this case. A party may fail to perform his duty, whether in contract or in negligence, in a variety of different ways. In the present instance, the certifier may breach his duty by negligently reaching a wrong result; or by misinforming his client about some material fact; or by failing to reveal some matter that is relevant to the client's reliance on the certificate. Although each of those complaints relates to a failure of the certifier to perform his duties, the breaches relate to different aspects or heads of those duties, and generate different causes of action: even though all of them are causes of action in negligence, and all of them complain of the certifier's performance of his duties.
459. I am fortified in the view that that is the correct approach by the observation of Hoffmann LJ in Broadley v Guy Clapham [1994] 4 All ER 439 at p 448h, cited with approval by Lord Nicholls in Haward v Fawcetts [2006] 3 All ER 497 at p 501h, that

One should look at the way the plaintiff puts his case, distil what he is complaining about and ask whether he had, in broad terms, knowledge of the facts on which the complaint is based.

It is thus for the claimant to formulate the particular acts of negligence or breach of duty of which he complains. Unless those are incoherent in law, or abusive in that they completely overlap with an already pleaded complaint, so that they would in any event be struck out (on which see §456 above), the claimant is entitled to proceed with them, and to seek the protection of section 32 in respect of those claims if they were made unavailable to him by a breach of duty on the part of the defendant.

460. Looking in that light at the various claims made in this case, the position is as follows.
461. Paragraph 23A1 of the Points of Claim pleads a duty to correct the certificate or to acknowledge the errors in the certificate. That, with deference to what Rix LJ says in his §361, is a different complaint, alleging a different duty on the part of ITS, from the simple complaint that the cargo was in fact off-specification. Before the Cooper retests were available, that claim could be pleaded only in terms of failure to communicate what ITS would or should have known would be the case if the cargo had been tested accurately and by D5191. A breach of duty in terms of failure to communicate what the certifier ought to have known is a different complaint, and asserts a different cause of action, from a breach of duty in terms of failure to communicate what the certifier did actually know, the result of the Cooper retests, which latter breach is pleaded by amendment in § 23A2.
462. Paragraph 23A of the Points of Claim pleads negligent misrepresentation arising out of ITS's dealings with AIC, in particular in the conversation of 17 April 1996. Before the Cooper retests had been revealed, that claim of negligence had to be limited to failure by ITS to take into account what it should have known in general terms. Once the Cooper retests were available to AIC, it was able to plead for the first time that ITS was negligent by failing to give proper weight to the results of the Cooper retests. A cause of action in negligent representation is completed by asserting first the representation; second that it was false; and third the negligent fault on the part of the representor that caused the falsity. Where there are different respects in which the representor was negligent, those different respects generate different causes of action, even though the first and second items listed above are common to all of them. This is such a case. The cause of action that complains that the representation was false because of failure on ITS's part to act on the Cooper retests is different from the cause of action in negligent representation that was pleaded before the Cooper retests were known about.
463. Paragraph 22 of the Points of Claim originally said that the certificate was false because use of the proper test, D5191, would have shown an RVP in excess of 9.0 psi. The complaint then of the use of D323 was that ITS knew that the two tests would give different results, with D323 showing a lower reading that incorrectly made the cargo appear to be within specification, but did not reveal that to AIC. Mr Gaisman argued in his closing submissions that on the assumption that AIC were wrong, and the two tests would in fact give comparable results, then AIC could have used the disport tests, conducted according to D5191, to plead that Mr Mailey's test, using D323, must have been conducted incompetently.
464. I would be very reluctant to adopt that argument. The Judge indeed held, on the basis of the expert evidence at the trial, that D323 would produce results comparable to those produced by D5191. ITS however complained strongly that it had been unfair of the Judge to rely on that finding in the face of very strong evidence that not only AIC but also ITS itself thought otherwise at the time of the events complained of. It is not necessary to set out ITS's evidence to that effect, which is summarised at §§ 109-110 of ITS's opening skeleton in this court. But if that was the general view at the time, AIC can hardly be criticised for not realising, and certainly cannot be criticised for not pleading, that the results of the disport tests in themselves demonstrated that Mr Mailey's D323 tests had very likely been conducted incorrectly. Scales fell from eyes only with the revelation of the Cooper retests, which enabled AIC to plead for the first time not the competent use of a non-comparable test, but the incompetent use of a comparable test.
465. I accordingly consider that all of the parts of AIC's pleading that rely on the Cooper retests assert new causes of action, and not merely better evidence in support of existing causes of action. The Judge was correct to permit AIC to pursue those complaints.
466. All that said, I with respect see the force of Rix LJ's observation in his §363 that the primary loss and damage claimed by AIC in these proceedings is that of its liability to Galaxy in the Swiss proceedings; and that the new causes of action based on the Cooper retests might not add much in terms of actual recovery to the causes of action that could have been pleaded before the Cooper retests were known about. That may be an issue for the future, but it is not the issue before us. We are concerned only with whether the new causes of action are different in law from the old, a question that I answer as set out above.

Disposal

467. I would allow the appeal to the extent only of reversing the Judge's findings as to deceit. Since all of my Lords agree that those findings cannot stand, but the majority of the court would otherwise dismiss the appeal, I would propose (subject to any further submissions by counsel as to the exact form of the order) that the court should dismiss the claim to the extent that it sounds in deceit; and that the remainder of the action should be remitted to the trial judge for him to reconsider its disposal in the light of the failure of the claim in deceit.