BHP Billiton Petroleum Ltd. & Ors v Dalmine SpA

The appellant, Dalmine SpA, is an Italian steel-maker which manufactured and provided the 12" diameter steel pipes used in the construction of a sub-sea gas reinjection pipeline serving the Lennox and Douglas oil and gas fields in the Liverpool Bay area of the Irish Sea. The pipeline ran between the Douglas and Lennox platforms. The respondents were at the relevant time co-venturers and participants in the Liverpool Bay Development, which embraced those fields and the pipeline. The first respondent, BHP Billiton Petroleum Limited, acted on behalf of the co-venturers as the “Operator” of the Development. We shall refer to the respondents simply as “BHP”. This litigation has arisen out of the failure of the pipeline, which has had to be replaced.

The pipeline was laid between 30 April and 27 June 1994 but did not enter service until April 1996.The discovery of a problem occurred on 7 June 1996, when gas bubbles were noticed on the surface of the sea. Investigations ensued, and by 22 June 1996 leaks had been identified at six failure sites, referred to as Leak Sites A-F, within a portion of the pipeline known as the “Failure Zone”. The distance between sites A-F was 1.15 kms. The Failure Zone comprised 144 pipes, each of 12 metres. The total length of the pipeline was 31.7 kms. It was established not only that the pipeline had failed at each of the leak sites A-F, but also that it had not failed elsewhere: at one time another twelve sites came under suspicion, but in the event were cleared.

The reasons for the failure of the pipeline were investigated. It was established that cracks had developed in the roots of welds which joined the pipes together. At leak sites A-F these cracks had propagated from the weld roots into the parent metal of the adjacent pipe and had developed into through-wall cracks, linking the interior and exterior walls of the pipe. The cracks had initiated because of a combination of excessive hardness of the weld root metal and because the pipeline was subject, as was expected, to sour service conditions, ie the combination of hydrogen sulphide and water. This has the effect of releasing hydrogen atoms which permeate the crystal lattice of the steel and embrittle it. The effect is known as sulphide stress corrosion cracking or SSCC. The propagation of the cracks into the parent pipe metal was due to the fact that the force exerted by the tip of the crack (expressed as “Kapplied” or the applied stress intensity factor) exceeded the resistance of which the parent pipe steel was capable (expressed as “KIscc” or the threshold stress intensity factor). The resistance of the pipe metal in turn depended on the steel’s carbon equivalent value or “CEV”. For that or other reasons the specification under which the steel was manufactured stipulated a maximum CEV of 0.40%. The lower the CEV, the more resistant the steel is to crack propagation. The higher the CEV, the less resistant it is. When the CEV of the pipes adjacent to the leak sites was investigated it emerged that in the case of each of sites A-F where the through-wall cracks had occurred, at least one of the pipes either side of the weld had a CEV greater than 0.40% and was therefore in that respect out of specification (“non-compliant pipe”). The average CEV of the non-compliant pipes at the leak sites ranged from 0.419% to 0.447%.The pipeline did not fail at any welded joints where both of the pipes were compliant.

As a result of the failure of the pipeline BHP commenced this litigation against both British Steel plc and Dalmine. British Steel was the supplier to BHP of the pipes under contract and entered into a sub-contract with Dalmine for their manufacture. There were no contractual relations directly between BHP and Dalmine, but the same data sheet and specification operated under both contract and sub-contract. BHP sued British Steel in contract and Dalmine in negligence. BHP’s case against British Steel in contract was abandoned after this court’s decision adverse to BHP on the effect of British Steel’s limitation and exception clauses: see BHP Petroleum Ltd v. British Steel plc [2000] 2 Lloyd’s Rep 277. It emerged, however, in the course of discovery that Dalmine had fraudulently misrepresented the CEV of certain pipes at the inspection and certification stage of the manufacture and supply process. Pipes are manufactured in batches of fifty at a time, and these batches are known as “heats”. Dalmine was required to test each heat for compliance with the data sheet and specification and to certify compliance. For these purposes two pipes out of each heat had to be tested. Pipes from six heats tested non-compliant. However, a senior member of Dalmine’s quality control department deliberately changed the results and produced false inspection reports and a false certificate of compliance. Dalmine has admitted responsibility for that fraud and that BHP was entitled to and did rely on the false documents.

It is therefore common ground that in reliance on Dalmine’s deceit BHP accepted and utilised the pipes by incorporating them into the pipeline in circumstances where if it had known the truth the pipes would have been rejected.

Nevertheless there remained an issue or issues of causation between the parties in relation to BHP’s case in deceit which was fought out at a trial before Cresswell J in February 2002. The judge expressed the issue(s) in this way (at para 33 of his judgment, [2002] EWHC 970 (Comm), unreported 16 May 2002):

“Did the incorporation of non-compliant pipe cause the pipeline to fail (as the claimants say) or would it have failed anyway (as the defendant says)?”

The issue(s) grew out of an earlier order dated 24 November 2000, which had made directions for a trial covering the questions of reliance and causation, inter alia in the following terms:

“Was the Claimants’ loss caused by their reliance on the Defendant’s fraudulent misrepresentations? Did the pipeline fail for the reasons pleaded in paragraph 25 of the Re-Re-Amended Points of Claim?”

In the run-up to that trial, however, certain agreements and concessions were made. By a memorandum of agreement dated 17 January 2002 the parties agreed that the causation issue should proceed on the basis that, if no false representations had been made, only compliant pipes would have been incorporated into the pipeline. And on 5 February 2002 Dalmine conceded reliance by BHP. That left only the issue of causation.

The parties are agreed (see the memorandum of agreement) that if the remaining issue of causation is determined in BHP’s favour, BHP will be entitled to judgment on liability with damages to be assessed and it will not need to pursue its action in negligence.

The judge resolved the issue of causation in BHP’s favour. As a result his order provides for judgment to be entered against Dalmine on liability in deceit, with damages to be assessed.

Among the questions debated before Cresswell J was an issue as to burden of proof. BHP accepted that it bore the burden of proving that the incorporation of non-compliant pipe caused the pipeline to fail, which was the first part of the issue stated by the judge, but submitted that Dalmine bore the burden of proving that the pipeline would have failed in any event, ie even if it had been made solely of compliant pipe, which was the second part of the issue stated by the judge. In the end Cresswell J did not have to determine the burden of proof issue, for he found that BHP had succeeded on both parts of the causation issue. Thus he concluded (at para 280):

“For the reasons set out above I find that on the balance of probabilities the incorporation of non-compliant pipe caused the pipeline to fail. For the reasons set out above I find on the balance of probabilities that the pipeline would not have failed anyway.”

For the purposes of this appeal both parties made detailed submissions on the burden of proof issue in their skeleton arguments. At the hearing we therefore asked the parties to deal with it first. Following the overnight adjournment at the end of the first day of the appeal hearing, we indicated that we had decided that the burden of proving that a pipeline made only of compliant pipe would have failed in any event rested on Dalmine. We reserved our reasons.

Later that day Mr Christopher Hancock QC, counsel on behalf of Dalmine, informed us that Dalmine recognised that in the circumstances it could not sustain that burden. Mr Hancock explained that he did so because he accepted the judge’s finding at para 213 that it was not possible to say what the CEV of untested pipes on the pipeline was. In fact the judge’s finding was more complex than that, for his agnosticism was qualified (“save to say”) by what he went on to state in sub-paragraphs (1) and (2) of para 213. Be that as it may, the position is that Dalmine has conceded that, if the burden of proving what would have happened to a hypothetical pipeline made up solely of compliant pipe rested on it, it could not sustain that burden. For the purposes of that concession, plainly made in the face of the court, it ultimately does not matter whether the reasons for it lie in some finding of the judge which Dalmine considers that it cannot displace, or in some more general view of what it could achieve. Dalmine came to this appeal asserting that it could, if necessary, show that even a pipeline made up solely of compliant pipe would have failed, but it now accepts that it cannot meet that burden.

The sole issue on which it is therefore necessary for us to give our reasons is the burden of proof issue.

Although Mr Hancock submitted that the pleadings could be in no way conclusive on what was ultimately a question of law, he did not go so far as to say that they were irrelevant. He could hardly do so, since he relied on a passage on the burden of proof taken from Phipson on Evidence, 15^th ed, 2000, at paras 4-02/06, in which the legal burden, or what the learned authors of Phipson call the “persuasive burden”, is discussed (at para 4-03) by reference to the following statement by Lord Maugham in Constantine Line v. Imperial Smelting Corporation [1942] AC 154 at 174:

“The burden of proof in any particular case depends on the circumstances in which the claim arises. In general the rule which applies is “Ei qui affirmat non ei qui negat incumbit probatio” (Footnote: ¹). It is an ancient rule founded on considerations of good sense and it should not be departed from without strong reasons.”

Phipson continues:

“This rule is adopted principally because it is just that he who invokes the aid of the rule of law should be the first to prove his case; and partly because, in the nature of things, a negative is more difficult to establish than an affirmative. The burden of proof is fixed at the beginning of the trial by the state of the pleadings, and it is settled as a question of law, remaining unchanged throughout the trial exactly where the pleadings place it, and never shifting.”

In the event it was Mr Mark Howard QC, counsel for BHP, who took us to the pleadings. In BHP’s particulars of claim the plea of causation was contained in paras 24/25. Although the precise language of the plea altered over a series of amendments and re-amendments, the essence of it from the beginning is fairly reflected in the terms of the final pleading that –

“As a result of further investigation, six leak sites were identified at which the pipeline had failed due to SSCC…As for the cause of the failure of the pipeline at these particular sites:

(1)

SSCC initiation in the weld roots is now known to have been caused by a combination of excessive hardness of parts of the weld produced by the welding procedure applied by McDermott and the use of the Pipeline for Sour Service.

(2)

The Pipeline failed where one or more of the Incorporated Pipes either side of the weld had a CEV in excess of the specified maximum of 0.40% because the relatively low resistance to crack propagation of the material from which those Pipes were made allowed the cracks to extend through the thickness of the Pipes.”

Following the discovery of Dalmine’s deceit, a new para 30A was added to the particulars of claim. Sub-para (s) pleaded that if BHP had known the true test results for the heats in question, it would have rejected all pipes from those heats (that was no longer in dispute) and sub-para (t) continued:

“The incorporation of the non-compliant Pipes caused the Pipeline to fail. The Claimants will rely on the facts that:

(i)

as shown in Schedule 3, at each of the six welds where a leak occurred, at least one of the Pipes either side of the weld came from one of the heats…and as such did not comply with the Specification and/or the Data Sheet in that it had a CEV in excess of 0.40%; and

(ii)

the higher the CEV of any particular Pipe, the less resistant it is to the propagation of cracks initiated by SSCC.”

Dalmine’s points of defence also went through a number of versions. The original response to paras 24/25 of the particulars of claim had been to make no admissions and to put BHP to strict proof that the cause of the cracking was excessive hardness resulting from the CEV in certain pipes being greater than 0.40%. The final response, however, deleted that plea and substituted the following (in para 11):

“11.1

It is admitted that the pipes cracked.

11.2

The cause of these cracks was sulphide stress corrosion cracking.

11.3

The cracks initiated in the weld root, and then propagated through the weld metal, the heat affected zone (“HAZ”) and the parent metal.

11.4

As regards the crackinitiation:-

11.4.1

The welding procedure adopted for the pipeline caused the weld metal at the weld root to develop excessive hardness.

11.4.2

This excessive hardness in the weld root was notcaused by excessive hardness in the parent metal.

11.4.3

Under the extremely sour conditions to which the pipeline was exposed, the excessive hardness in the weld root caused hydrogen cracks to initiate. Cracks initiated both in welds which joined pipes which complied with the CE specification and in welds which joined pipes which did not comply.

11.5

As regards the crackpropagation, the cracks would have propagated irrespective of whether the parent pipes were within the CE specification.”

For present purposes the critical pleading is that contained in para 11.5, for if the cracks would not have propagated in compliant pipe, then the differences between the parties as to crack initiation (on which there was some common ground, such as that it was at least in part due to the welding procedure and to the process of SSCC) would not have mattered.

A similar plea as to crack propagation irrespective of whether pipes were compliant or not was contained in para 13A.8 (in response to BHP’s plea relating to reliance on Dalmine’s deceit) and in para 14 (in response to BHP’s plea of loss and damage). Thus –

“13A.8…Dalmine will say that, if no false representations had been made, the Claimants would still have suffered the loss pleaded, because:

13A.8.1 Compliant pipe would have been required and delivered.

13A.8.2 The same weld procedure would have produced excessive hardness in the weld root metal.

13A.8.3 The same weld procedure would have been followed as was in fact followed.

13A.8.4 Cracks would have initiated in the weld root.

13A.8.5 The stress intensity applied by the cracks so initiated would have been above the resistance to crack propagation of compliant pipes.

13A.8.6 Cracks would have propagated in such compliant pipes, leading to leakage”

and –

14.1

It is admitted that the Claimants have replaced the pipeline and that they were prompted to do so by the cracking in the pipeline. It is denied that the cause of the cracking was as alleged by the Claimants. Dalmine’s case as to causation is set out in paragraph 13A.8 above. In the premises and by reason of the facts and matters set out in paragraph 14.1B below, it is averred that the pipeline would have been required to be replaced irrespective of anything done or not done by Dalmine.” [emphasis added]

The “facts and matters set out in paragraph 14.1B below” were another denial of reliance on Dalmine’s deceit and by reason of Dalmine’s pre-trial abandonment of that denial does not affect the position.

In essence, therefore, Dalmine’s case on causation was that a hypothetical pipeline built entirely of compliant pipes would have failed in any event. That, it may be noted, is “averred”. On that basis, Dalmine would be able to submit that the real cause of the failure of the actual pipeline, properly understood, was not the presence of non-compliant pipes, but the other factors which had contributed to the initiation of the cracking in the weld roots, such as the weld procedure, the excessive hardness of the weld roots, and the sour conditions promoting SSCC.

Pursuant to those pleadings the following issues were agreed to arise to enable the court to deal with Dalmine’s case that even a pipeline made entirely of compliant pipe would have failed in any event. I take them from paras 60/70 of Cresswell J’s judgment:

“I. The resistance of Compliant Pipes within the Failure Zone to Crack Propagation

60.

Issue 1. How resistant to crack propagation would compliant pipes have been within the Failure Zone?

61.

Sub-Issue 1.1: What was the upper limit of the CEV of “compliant pipes” according to the Specification and Data Sheet?

62.

Sub-Issue 1.2: What was the actual CEV of each of the pipes within the Failure Zone?

63.

Sub-Issue 1.3: Which heats fall to be excluded from consideration in the light of the parties’ agreement of 17 January 2002?

64.

Sub-Issue 1.4: What, if any, figure should the Court take as being the CEV of pipes which Dalmine would have supplied (instead of those pipes which it would not have supplied had it not been fraudulent) or which fall to be excluded as a result of the agreement of 17 January 2002?

65.

Sub-Issue 1.5: On the assumption that only compliant pipes had been incorporated into the pipeline, what was the resistance to propagation of compliant pipe within the Failure Zone?

II. Cracks in Welds in the Failure Zone

66.

Issue 2: What range (as regards depth) of cracks was present, or was likely to appear, in welds between compliant pipes within the Failure Zone and what was their likely propensity to propagate?

67.

Sub-Issue 2.1: What size cracks had initiated and/or propagated pre-failure and what was their distribution between the welds?

68.

Sub-Issue 2.2: What scope was there for the further initiation and/or propagation of cracks had the pipeline not been de-pressurised, given further time and/or changes in environmental conditions?

69.

Sub-issue 2.3: What was the stress intensity factor likely to have been at the tip of the crack of the size which Dalmine contends would be sufficiently large to cause failure of compliant pipes?

III. Approach to Probability

70.

Issue 3: Can the Court assess the probability of the combination of factors required to produce the situation in which a crack would develop in a weld between compliant pipes within the Failure Zone, which would be sufficiently large to overcome the resistance offered by (compliant) pipe either side of the weld, to the propagation of that crack through the thickness of the pipe material? If so, how should it go about doing so?”

It follows from this statement of the issues raised that Dalmine’s case on causation could not be answered without conclusions inter alia not only as to the actual CEV of each of the compliant pipes within the Failure Zone (that issue was limited to the Failure Zone because Dalmine did not in the event seek to say that the pipeline would have failed outside the Failure Zone) but also as to the hypothetical CEV of compliant pipes which Dalmine would have supplied in place of non-compliant pipes; not only as to the actual cracks which had developed in the pipeline, but also as to what further cracks would have initiated or propagated if the pipeline had been longer in service (because it had only been in service from April to June 1996, and only at a pressure of 95 bar whereas full operational pressure would have been at 128 bar, and it was then de-pressurised and taken out of service following the discovery of the leaks); and not only as to the future service in operation of the existing compliant pipe, but also as to the condition of hypothetical compliant pipe which would have been supplied in place of non-compliant pipe (because even Dalmine’s case was that it was only compliant pipe at or near the maximum CEV allowed under the specification which could possibly present a risk of failure).

It was not in dispute before us that the burden of proving causation is, as a general rule, on the claimant. Mr Hancock submitted that this was so whether the claim was in tort or in contract, and whether the tort was negligence or deceit.

Of course, causation, for all that it is a matter of common sense (see Allied Maples Group Ltd v. Simmons & Simmons [1995] 1 WLR 1602) can be a difficult concept. The requirements of the duty of mitigation are bound up with causation, but the burden of proving a failure to mitigate rests on the defendant. In the present case the foundation of Mr Hancock’s argument is the proposition that to prove causation a claimant such as BHP must meet the “but for” test, and that that test not only reflects the general rule that the burden of causation is on a claimant, but requires of him the proof of a negative, namely that but for the matter of which complaint is made, the claimant would not have suffered the alleged loss. Thus Mr Hancock relied on a passage in Clerk & Lindsell on Torts, 17^th ed, 2000, at para 2-06, as follows:

“The first step in establishing causation is to eliminate irrelevant causes, and this is the purpose of the “but for” test. The courts are concerned, not to identify all of the possible causes of a particular incident, but with the effective cause of the resulting damage in order to assign responsibility for that damage. The “but for” test asks: would the damage of which the claimant complains have occurred “but for” the negligence (or other wrongdoing). Or to put it more accurately, can the claimant adduce evidence to show that it is more likely than not, more than 50 per cent probable, that “but for” the defendant’s wrongdoing the relevant damage would not have occurred. In other words, if the damage would have occurred in any event the defendant’s conduct is not a “but for” cause.”

Mr Hancock therefore submits that the onus of proving that but for the incorporation of non-compliant pipes the pipeline would not have failed in any event rests and remains on BHP. If this is so, it would seem to conflict with the indication of where the burden of proof lies to be derived from the pleadings: for there it appears that the plea that a pipeline built only of compliant pipe would not have failed in any event is raised by Dalmine, not by BHP. Mr Hancock explains that apparent contradiction as follows. He submits that when a claimant pleads that his defendant caused his loss he implicitly pleads that nothing else did and that his loss would not have occurred in any other way. If nothing further is said by the defendant, then the claimant and the court can ignore the burden of those implicit and silent negatives or rather regard them as fulfilled. If, however, the defendant can properly raise a case for argument that some other cause was operative, or that the claimant’s loss would have been suffered in any event, then he throws back on the claimant the underlying and inherent onus of disproving the negative in order to prove his positive case of causation.

In our judgment, however, this is an unrealistically theoretical approach to the problem. We think, with Lord Maugham, that the pleadings are likely to be a more accurate guide to where the burden lies. It is true that they cannot be definitive, for a party does not, we think, take upon himself a burden which the law does not impose on him merely by volunteering a pleading which he could properly have covered by a mere denial; nor will a claimant free himself from a burden of proof which rests on him merely by pleading his case inadequately. However, in most cases the pleadings are likely to be a good guide, and a fortiori where, as here, no one has been able to complain that the pleadings have fallen below the mark.

In this connection we think that the role of the “but for” test should not be exaggerated. As the learned authors of Clerk & Lindsell stated at the beginning of the passage cited above, the purpose of that test is to eliminate irrelevant causes. The point is emphasised again later on in the same paragraph, where the text continues –

“It is worth bearing in mind that the “but for” test functions as an exclusionary test, i.e. its purpose is to exclude from consideration irrelevant causes.”

If, therefore, the sole dispute between the parties had been whether the welding procedure (which it was alleged had led, by a process of over-rapid cooling, to excessive hardness of the metal in the weld root), or SSCC, had by themselves or in combination with one another been the cause of the pipeline failure, it may well have been that the burden of such a dispute would have remained on BHP. However, that was not what the dispute was about, for the plain facts were that the pipeline had not failed at any point other than where the pipe on one or both sides of the weld had been non-compliant. SSCC, possibly in combination with the welding procedure, may have caused some cracks to initiate, but it had nowhere caused such cracks to propagate sufficiently to cause the pipeline to fail. If the pipeline had failed at some welded joint adjacent to a pair of compliant pipes, then we think that BHP may well have borne the burden of showing that the cause of the pipeline’s failure was non-compliant pipe rather than the welding procedure and/or SSCC, for both of which Dalmine was of course not responsible.

In the present case, however, the issue under discussion is not whether the welding procedure and/or SSCC as distinct from non-compliant pipe caused the loss of the actual pipeline, but whether they would have caused the loss of another pipeline, a hypothetical pipeline, even if that had been constructed solely out of compliant pipes. Dalmine’s plea is that a pipeline built of compliant pipes would have failed in any event. It is not said when a pipeline built only of compliant pipes would have failed, but it seems to us that on the facts it must necessarily be at some time subsequent to the failure of the actual pipeline, and that was highlighted in sub-issue 2.2 above.

If, in such circumstances, one can imagine that Dalmine had succeeded in proving that the pipeline would have failed in any event, even if the deceitfully imposed non-compliant pipe had all been replaced by compliant pipe, then a court would be entitled to infer from that fact once established, if in all the circumstances it seemed right to it so to infer, that BHP had failed to prove that even the loss of the actual pipeline had been caused by the imposition of non-compliant pipes. The court might conclude that the effective cause of the loss of the pipeline was in that case not the non-compliant pipes but the welding procedure and/or SSCC. Or it might nevertheless still conclude that the loss of the pipeline at the time it was lost was still due to the non-compliant pipes, but that a pipeline built only of compliant pipes would have failed one, or five, or how ever many years later, and in the light of that finding adjust the quantum of the claimant’s loss. None of this means, however, that the burden of proving Dalmine’s negative hypothetical case would have rested on BHP.

The distinction which we are seeking to draw above may be illustrated by a decision which is cited in para 2-06 of Clerk & Lindsell, namely that of Barnett v. Chelsea and Kensington Hospital Management Committee [1969] 1 QB 428. The plaintiff there was the widow of a man who died of arsenic poisoning: he had attended the defendant’s hospital, but was negligently sent home without adequate treatment. The question was what would have happened if he had been properly treated. Nield J held that the burden of proving that he could have been saved remained on the plaintiff (but that in any event it had been shown that on the balance of probability he could not have been). That was a case, however, where the cause of death was arsenic poisoning, and it was for the plaintiff to prove that nevertheless the law should regard the negligence as the operative cause. She could only do that by proving that but for the negligence her husband would probably have been saved. In the present case, however, the cause of the loss was pipe failure solely where non-compliant pipe was in place, and it was for Dalmine to show that the law and common sense should nevertheless regard the operative cause to be some other condition of the pipeline by proving that compliant pipes would have failed in any event; or that BHP’s losses otherwise properly recoverable on the rule of damages for deceit should be curtailed because the pipeline would therefore have had to have been replaced before the end of its natural life in any event.

We think that this approach, which we would regard as consistent with principle and common sense, is also supported by the recent decision of this court in Chester v. Afshar [2002] EWCA Civ 724, [2002] 3 WLR 1195. The claimant there sought damages for the consequences of post-operative paralysis from a surgeon who had negligently failed to warn her of the small risk of such paralysis which the operation involved. The claim was based on the failure to warn, not on any inadequacy in the performance of the operation. The judge found that the operation would not have taken place when it did had the claimant been warned of the risk, that there was therefore a sufficient causal connection between negligence and loss, and that that link was not broken by the possibility that the claimant might have consented to the operation in the future with the same result. This court upheld the defendant’s liability.

This court’s judgment contrasted three situations (at paras 38/39). If the evidence had shown that the claimant would have had the operation performed then and there even if she had been warned of the risk, then she would have failed to prove that the defendant’s negligence had caused her loss. If on the other hand she would never have had the operation, had she been warned of the risk, then her case on causation was proved. The third case, with which her appeal was concerned, arose where she could not show that she would never have had the operation with its attendant risk. Did it follow that she could not prove causation? No, for the defendant by his negligence had changed the risk in a material way by causing her to undergo the operation then and there. In such a situation, causation was proved and the rest was a matter of quantum:

“On the assessment it will be open to the defendant to argue (and prove so far as he may be able to do so) that (i) it is more likely than not that the claimant would have undergone an operation with the same or similar risks in the future; and (ii) it is more likely than not that the same risk would have eventuated” (at para 43).

So in this case we think that causation is proved once BHP has shown that the reason why the pipeline failed when it did was because of the failure of non-compliant pipe which but for Dalmine’s deceit would have been rejected. BHP has shown that the pipeline failed only where one or both of the pipes was non-compliant and at no other welded joint. In such circumstances, if Dalmine wishes to show that a hypothetical pipeline made up only of compliant pipe, given more time and the operation of the pipeline at the ultimate working pressure of 128 bar, would have failed in any event, then it bears the burden of proving that on the balance of probabilities. For these purposes, a mere possibility of such failure would not be enough. However, Dalmine concedes that it cannot sustain that burden.

The present appeal was not concerned with the particular difficulties which can arise in cases concerned with industrial diseases and therefore it was entirely understandable that a famous pair of decisions, recently joined by a third, were not cited to the court: we refer to McGhee v. National Coal Board [1973] 1 WLR 1, Wilsher v. Essex Health Authority [1988] AC 1074 and Fairchild v. Glenhaven Funeral Services Ltd [2002] UKHL 22, [2002] 3 WLR 89. Nor was the court concerned with the special considerations which arise out of successive conversions: see Kuwait Airways Corporation v. Iraqi Airways Co (Nos 4 and 5) [2002] UKHL 19, [2002] AC 883. These last two recent decisions of the House of Lords have shown that the “but for” test of causation is not of universal applicability: see the discussion added to paras 2-06 and 2-08 of the main text of Clerk & Lindsell by the Second Supplement to its 18^th ed, 2002 and the general statements of principle set out in the speech of Lord Bingham of Cornhill in Fairchild at para 12. There was no submission, however, that the “but for” test did not apply or did not apply in its normal way in the tort of deceit. Nor was there any submission based on the consideration that Dalmine’s duty not to mislead BHP as to the proper certification of the CEV of the pipes was directly related to the importance of the pipes being sufficiently resistant to cracking in a sour environment. We have nevertheless borne those decisions in mind in considering the normal requirement for a claimant to meet the “but for” test, but regard the analysis which we have carried out and the conclusion which we have derived in this case to be consistent with the principles there set out.

It follows that this appeal must be dismissed.