BAILII is celebrating 24 years of free online access to the law! Would you consider making a contribution?
No donation is too small. If every visitor before 31 December gives just £1, it will have a significant impact on BAILII's ability to continue providing free access to the law.
Thank you very much for your support!

IN THE SUPREME COURT OF JUDICATURE QBCMF 95/1688/B

COURT OF APPEAL (CIVIL DIVISION)

ON APPEAL FROM THE HIGH COURT OF JUSTICE

(Mr Justice Tucker )

Royal Courts of Justice

The Strand

London

Wednesday 26 March 1997

B e f o r e:

THE PRESIDENT

(Sir Stephen Brown )

LORD JUSTICE HOBHOUSE

and

LORD JUSTICE WARD

B E T W E E N:

PROMET ENGINEERING (SINGAPORE) PTE LTD

(Formerly Self-Elevating Platform

Management Pte Limited ) Plaintiff/Appellant

- v -

NICHOLAS COLWYN STURGE & ors Defendants/Respondents

_______________

(Computer Aided Transcription by

Smith Bernal, 180 Fleet Street, London EC4A 2HD

Telephone 0171 831 3183

Official Shorthand Writers to the Court)

_______________

MR STEPHEN RUTTLE (instructed by Messrs Norton Rose, London EC3)

appeared of behalf of THE APPELLANT

MR D MILDON and MR NIGEL EATON (instructed by Messrs Clyde & Co,

London EC3) appeared on behalf of THE RESPONDENTS

_______________

J U D G M E N T

(As Approved by the Court )

_______________

LORD JUSTICE HOBHOUSE: In September 1983 the platform "Nukila" went into service in the Ardjuna Field in the Java Sea. She had just been built by a yard in Singapore. She was a mobile self-elevating accommodation and work platform measuring about 136 feet by 122 feet supported on three tubular leg columns. Each leg had an overall length of 208 feet with a diameter of 8 feet. The sea bed in the Ardjuna Field is soft. To stop the legs sinking into the silt there was connected to the bottom end of each leg a 28ft square spudcan having a depth of 4ft. Each spudcan was effectively a large steel box strengthened with internal bulkheads and brackets. They were accessible through manhole covers on the upper side. Piping systems in the legs and the spudcans enable them to be ballasted.

The platform was mobile because, when the legs were raised, the platform was effectively a large barge or pontoon with sufficient buoyancy to float. Once it had been towed to and positioned at its working location the legs would be lowered so as to rest upon the sea bed and the body of the platform then jacked up on the legs so as to stand clear of the water. The three legs were arranged in a triangle, one at the front of the platform, one at the port side aft and one at the starboard side aft of the platform. The legs passed through apertures in the platform where there are jacks for raising and lowering then and for raising and lowering the body of the platform on the legs. To move the platform or take it back to port the platform body would be lowered into the water and the legs raised. In operation the platform was used as a working platform but it also provided accommodation for some 100 people.

The legs were steel tubes or cylinders made of steel 1¼" thick. The spudcan were constructed around the outside of the foot of the legs so that the legs were continuous through the spudcans. The circular apertures in the top and bottom plating of each spudcan (also 1¼" thick) was welded circumferentially to the steel of the leg. The internal bulkheads of the spudcan were themselves welded to the top and bottom plating and, if they abutted onto the leg, to the steel of the leg as well leaving scallops (apertures to relieve stress) at the top and the bottom. Inside the tube of the legs corresponding structures were welded to provide internal support and continuity.

The spudcans and the legs were liable to be subjected to quite major stresses. The sea bed would not be uniform. There might be movement. The effects of the wind and the sea and currents on the upper parts of the structure would also create stresses on the spudcans and where they joined the legs. The design of and specification for the legs and spudcans took account of these stresses. But the actual construction was not in all respects carried out in accordance with the specification.

The circumferential welds attaching the top plates of the spudcans to the legs were not properly profiled. The classification society under whose supervision the platform was built was ABS and all the relevant welding was required to be in accordance with their rules. In fact the welds were not properly profiled that is to say, to quote the report of Dr Timothy Baker dated 27 September 1995, which was the evidence accepted by the trial judge Mr Justice Tuckey, "the profile on the weld between the leg column and the underside of the spudcan top-plate was poor in that there was an abrupt change between the surface of the weld and the parent metal". (p.5) The weld profile should have been controlled and the weld-toes should have been ground. The weld was in a location where there was already a high stress concentration and an inadequately profiled weld would increase this concentration excessively and be likely to shorten the fatigue life of the structure and lead to fatigue cracking. ( ib.)

Within the past 40 years metal fatigue has become a well understood process. It arises from repeated cycles of stress exceeding a certain value. It is important to design marine structures so that the stresses in any given location are not unnecessarily raised or (which is the same thing) concentrated. A badly designed or made weld may lead to a concentration of stress which will then over a period of time cause the condition of metal fatigue to arise. Metal fatigue starts with microscopic changes in the structure of the metal which lead to the formation of minute cracks which then grow in size. As they develop over time these fatigue cracks progressively become more detectable, initially by scientific investigation and eventually by the naked eye. Assuming that the cycles of stress continue and that the process of cracking does not itself relieve the stress, the fatigue crack will continue to grow until the metal shears or some other failure of the structure occurs. It will be appreciated that the presence of a fatigue crack will itself concentrate stresses at its tip and thereby lead to an extension of that crack unless and until those stresses are relieved. Similarly the presence of a fatigue crack will weaken the structure and therefore tend to cause other fractures or failures of the structure.

From 1983 to February 1987 the Nukila operated without any untoward incident. But in February 1987, whilst a routine inspection of the legs and spudcans was being carried out by divers, they observed serious cracks in the top-plates of all three of the spudcans. Closer examination revealed that the metal of the legs themselves also contained serious cracks as did some of the internal bulkheads of the spudcans. The condition revealed was dangerous and threatened the whole safety of the Nukila. The platform body was lowered and the legs raised. The state of the legs and the spudcans was fully investigated both at the site and thereafter once she had been towed back to Singapore. The results of the investigation are contained in various reports of the marine technology consultants Veritec, in particular in their report No.87050 dated 7th April 1987.

The findings of that report were not disputed at the trial and were accepted by Tuckey J. The report which includes numerous drawings and photographs details what can only be described as damage to the legs and spudcans. There were circumferential full width cracks having a gap of between 10mm and 16mm in the metal of the legs themselves. In one leg only 30% of the circumference remained intact; in another it was 20%; and in the third only 10% remained. These cracks had themselves led to other large cracks which branched out in different directions through the steel of the legs and the top-plating of the spudcans. There were also cracks, some of them major, in the bulkheads of the spudcans and in the vicinity of some of the piping holes. Some of these fractures were attributed by the report to mechanical stresses created by the weakening of other parts. Some cracks were described as shear fractures, that is to say fractures which resulted from stresses which exceeded the tensile or torsional strength of the metal.

Extensive repairs had to be carried out. It was agreed that they cost S$903,148. At the material time the Nukila was owned by Promet Engineering Singapore PTE Ltd. The owners had insured the Nukila on the London market under time policies for the 12 months from 9th September 1986. The Defendant in the action, Mr Sturge, is the representative underwriter. The time policies were hull and machinery policies of marine insurance; the interest insured is described as "Hull and Materials, Engines and Machinery etc, and everything connected therewith". The Institute Time Clauses Hull (1/10/83) were incorporated with immaterial amendments. These clauses extend the ordinary marine cover so as to include risks which would not otherwise be covered. ( Thames and Mersey Marine Insurance Co v Hamilton Fraser and Co , 1887, 12 App Cas 484.) Thus, in the clause known as the "Inchmaree" clause:

"6.2. This insurance covers damage to the subject matter insured caused by ...

6.2.2. Bursting of boilers breakage of shafts or any latent defect in the machinery or hull

.... "

The policies also incorporated the Institute Additional Perils Clauses - Hulls (1/10/83) which in consideration of an additional premium extended the insurance to cover -

"1.1. The cost of repairing or replacing

1.1.1. Any boiler which bursts or shaft which breaks

1.1.2. Any defective part which has caused loss or damage to the vessel covered by clause 6.2.2 of the Institute Time Clauses Hulls 1/10/83,

....

2. Except as provided in 1.1.1. and 1.1.2 nothing in these additional perils clauses shall allow any claim for the cost of repairing or replacing any part found to be defective as a result of a fault or error in design or construction and which has not caused loss of or damage to the vessel."

The owners claimed under the policy for the cost of repairing the legs and spudcans. Their primary claim was made under the Inchmaree clause. They put it very simply. What was repaired was 'damage to the subject matter insured caused by latent defects in the hull'. The damage was the fractured metal. The evidence, which was accepted, was that the fractures which necessitated the repairs had occurred subsequent to 9th September 1986. They were damage to the subject matter insured, that is to say the Nukila. They were caused by latent defects in the hull of the Nukila. The latent defect in the case of each of the three legs and spudcans was the defective circumferential weld. That defect had initiated very small fatigue cracks by September 1986. But as at that date the exercise of due diligence by the owners would not have discovered the defect in the weld nor the minute cracking that existed at that time. Insofar as the costs of the repairs included the cost of repairing as well any defective part which had caused that damage, such cost was also recoverable under the additional perils clauses.

The underwriters denied liability. Their answer to the primary way in which the owners have put their case is to say that the subject matter insured was not damaged. There was no consequential damage. All that occurred was that the latent defect in each leg manifested itself. The cracks discovered in each leg in February 1987 were no more than the discovery of the latent defect that existed in that leg. Therefore there was no loss which the owner is entitled to recover under the policy.

The action was tried in the Commercial Court before Mr Justice Tuckey. There were issues both of fact and law. He heard expert evidence. The reports of Veritec were accepted as was the evidence that the cracking observed in February 1987 had all occurred during the period of cover. He rejected the Defendant's expert evidence that all that had occurred was solely the result of fair wear and tear. He did not accept that the fatigue life of these structures, if properly constructed, was as short as 4 years. He accepted the evidence of the Plaintiffs' expert Dr Baker that the cause of the cracks was the improperly profiled weld between the lower side of the top-plate of the spudcans and the steel column of the legs.

The Judge preferred the submissions of the Defendant on the primary point. He referred to Oceanic v Faber 11 Com Cas 179, Hutchins v Royal Exchange Ass Co [1911] 2 KB 398 and Scindia SS Co v London Assurance Co [1937] 1 KB 639. The proposition of law which he derived was "if all that has happened is that a latent defect has become patent, there is no cover under the Inchmaree clause". He said:

"To speak of damage to the part as damage to the vessel begs the question and "produces confusion of thought and language" as Mr Justice Branson said in Scindia. The Court is still left with the task of seeing if there is consequential damage: whether there is or not does not necessarily depend upon the degree of damage to the part itself." (p.88)

"Just because something has a name it does not become a separate part for this purpose. .... [Mr Mildon, for the defendant,] suggested that a part for this purpose was one which was physically separable and or performed a separate function from other parts. On this analysis, neither welds, or the spudcans, or the columns could be considered separate parts. The part was the leg, comprising the column and the spudcan and the means by which one was attached to the other.

I accept Mr Mildon's submissions about this. I think it would be quite artificial to describe a weld as part of a vessel. At the foot of a leg, neither the columns nor the spud can have separate functions. They are physically joined together by welds supported by internal diaphragms and bulkheads to form one structure which enables the platform to stand on the sea bed. If the weld is not a part, the assured's case depends upon damage (cracking) spreading from the column to the spudcan. Ex hypothesi there is no cover before this has happened and yet it is cracking in the columns which is the most serious. This is highly artificial. It illustrates perhaps that this approach to the question is too analytical. What happened here is that there were flaws in the weld which developed into cracks which spread into the immediately adjoining structures which the weld was meant to hold together. As a matter of common sense it is impossible to see that at this stage anything consequential has happened which can be characterized as damage to the vessel.

I therefore conclude that the assured has failed to establish that there was damage to the vessel so their claim under the Inchmaree and additional perils clause fails." (p.89)

The critical point therefore is whether the owners were entitled to say that what was observed and reported by Veritec was damage to the Nukila caused by a latent defect or, as held by the Judge, was simply the latent defect itself.

Cover in relation to inherent defects and their consequences raise problems for the law of marine insurance. In principle the policy does not cover such a risk. Section 55(2)(c) of the Marine Insurance Act 1906 states that:

"Unless the policy otherwise provides, the insurer is not liable for ordinary wear and tear .... inherent vice or nature of the subject matter insured ..."

Insurance covers fortuities, not losses which have occurred through the ordinary incidents of the operation of the vessel. Similarly the insurance does not cover the costs of maintaining the vessel or running it. As the Judge held to be the case in the present action, the cracking occurred as a result of the ordinary working of the platform at sea and the presence of the latent defects in the welds. There was no external accident or cause. Correcting latent defects is, as a matter of principle, an expense to be borne by the shipowners and not by underwriters. Similarly, the pre-existing defective condition of the subject matter of the insurance (be it hull or cargo) can be said to have made the loss something which was bound to happen and therefore not fortuitous. These types of question have been addressed in a number of the cases on cargo insurance and, in relation to hull insurance, in The Caribbean Sea [1981] Lloyds 338. But, as s.55 of the Act recognises, if the parties make a specific agreement, a policy can cover such risks. The presence or absence of a latent defect in the hull or machinery of a vessel is, by definition, unknown to the assured and whether or not there is such a defect and whether or not it will during a given period of time or maritime adventure have an impact or cause any damage is fortuitous from the point of view of the assured. As is demonstrated by the Inchmaree clause and other similar clauses which have been introduced into policies following the series of decisions in the House of Lords in 1887, there is both a market need for such cover and a willingness to provide it.

However there are further difficulties. A policy of insurance does not cover matters which already exist at the date when the policy attaches. The assured if he is to recover an indemnity has to show that some loss or damage has occurred during the period covered by the policy. If a latent defect has existed at the commencement of the period and all that has happened is that the assured has discovered the existence of that latent defect then there has been no loss under the policy. The vessel is in the same condition as it was at the commencement of the period. Therefore, in any claim under the Inchmaree clause or any similar clause, the assured has to prove some change in the physical state of the vessel. If he cannot do so, he cannot show any loss under a policy on hull. We are not concerned in this case with any insurance of different interests: the present policy is a hull and machinery policy. If, however, damage has occurred, that does involve a physical change in the condition of the vessel and can be the subject of a claim under the policy. Similar difficulties for the assured arise under s.69 and the measure of indemnity for a partial loss. If the vessel has not been damaged there can be no cost of repairing damage and insofar as the relevant condition of the vessel is the same as that which existed at the commencement of the cover there has been no depreciation in the insurable value of the vessel.

A further factual difficulty arises when the latent defect is some feature of the hull or machinery which creates an excessive stress concentration and therefore will lead to a condition of metal fatigue and the formation of fatigue cracks. No clear dividing lines can be drawn. A crack is itself one of the forms of discontinuity that can concentrate stress. It can be both the consequence and the cause of metal fatigue. Fatigue cracks not detectable by the exercise of due diligence are a typical example of latent defect. But it is equally accurate to say that fatigue cracks are the consequence of metal fatigue. The Judge's finding was that the relevant cracking (whether it amounted to damage or not) was caused by the badly profiled welds which were latent defects in the hull and accordingly an insured peril. (p.89)

Mr Mildon specifically argued before us that no question of degree is involved. The feature is either the latent defect (or its manifestation) or it is damage. He submitted that the test is whether the crack relied upon was in the defective part or some other part of the hull. The Judge accepted his argument that "a part for this purpose was one which was physically separable and performed a separate function from" the other part. It is this reasoning, which Mr Mildon submitted was decided to be correct in the Scindia case and was decisive in the present case. The Judge clearly agreed with him. He held that all the cracks observed by Veritec existed in the same part that is to say in the structure of the relevant leg and spudcan which were not physically separable and performed no separate function. Mr Mildon accepted that on his argument the hull of a ship would for this purpose be a single part and that no system of fatigue cracking however extensive, even if extending from one compartment to another or from one strake to another and whether or not leading to shearing or other tensile failures in the hull could, short of a total loss, amount to damage to the hull of the vessel for the purposes of the Inchmaree clause. This submission which flies in the face of the ordinary use of language and which is said to have been dictated by the decision in the Scindia case is one to which I will have to return.

The reliance in this argument upon the concept of what is a "part" does not derive from the language of the Inchmaree clause itself. That clause does not use the word. It is used in the additional perils clause but only in a context which does not affect the primary, and relevant, question whether or not there has been damage to the subject matter of the insurance. The argument of Mr Mildon recognized that the concept upon which he relies has to be derived from what has been said in the cases about the language of the Inchmaree clause rather than directly from the clause itself. It must also be added in fairness to Mr Mildon and the reasoning of the Judge that the argument of Mr Ruttle for the owners both at the trial and in this Court was prepared to travel, albeit only as an alternative, a fair way down the same path as that argued for by Mr Mildon. As I will explain, I do not consider this approach helpful. Indeed it is open to the reproach quoted by the Judge and much relied upon by Mr Mildon of disclosing "a confusion both of thought and language".

In my judgment the application of the language of the Inchmaree clause to the facts of the present case is straightforward. At the commencement of the period of cover there was a latent defect in the welds joining the underside of the top-plate of each spudcan to the external surface of the leg tube. By that time that latent defect had also given rise to minute fatigue cracks in the surface of the tube in the way of the weld which could also properly be described as latent defects. Those features during the period of cover caused extensive fractures in the full thickness of the tube extending in places both above and below the defective weld, extensive fractures in the metal of the top-plating and bulkheads of the spudcans and other fractures at other locations. This was on any ordinary use of language damage to the subject matter insured, the hull etc of the Nukila. It was, as the Judge found, caused by the condition of the Nukila at the commencement of the period, that is to say by the latent defects I have identified. Therefore, subject to authority, the arguments of the owner should be accepted and the claim should succeed.

Turning to the authorities it must at the outset be recognised that, whether or not they are strictly binding upon us, they must, insofar as they represent the existing authoritative statements of the law only be departed from if they are clearly wrong. This principle has been stated on a number of occasions in the field of commercial law where it is recognized that the parties enter into contracts on the basis of the law as it has been stated in the applicable authorities. For a court, in deciding a dispute under a commercial contract, later to depart from those authorities risks a failure to give effect of a contractual intention of those parties as evidenced by their contract entered into on a certain understanding of the law. As Lord Dunedin said in Atlantic v Dreyfus [1922] 2 AC 250 at 257:

"My lords in these commercial cases it is I think of the highest importance that authorities should not be disturbed and if your lordships find that a certain doctrine has been laid down in former cases and presumably acted upon you will not be disposed to alter that doctrine unless you think it clearly wrong."

(See also per Lord Wilberforce in The Aries [1977] 1 WLR 185 at 190-1.) I examine the authorities relied upon by Mr Mildon and referred to in the judgment with this principle in mind.

The first is the Faber case, (1906) 11 Com Cas 179, Walton J; (1907) 13 Com Cas 28, CA. The policy was a time policy on hull covering (at the relevant time) port risks including an earlier version of the Inchmaree clause. The policy attached in May 1902. On a routine dry docking in October 1902 the tail shaft was drawn and a surface crack was observed. The shaft was condemned and the ship owners had to pay for a new shaft. They sought to recover the cost of doing so from the underwriters. The examination of the shaft disclosed that 11 years earlier in 1891, when a new section had been welded on to the shaft the work had not been done properly and a void had been left in the shaft. This was not thereafter detectable until the void led to a crack becoming visible on the surface of the shaft. There was no evidence or proof that the condition of the shaft, and this crack, had been any different in May 1902 from that which was discovered in October of the same year. Thus the owners were faced with the immediate, and in truth insuperable, difficulty that they could not show that the condition of the vessel had in any way changed during the period of the cover or that anything physical had happened during that period which had depreciated the value of the vessel or had necessitated a repair. These points were forcibly made by counsel for the underwriters Mr Scrutton KC (p.183). The owners, whose argument was presented by JA Hamilton KC (later to become Lord Sumner) was (p.102) -

"A latent defect is a peril insured against. The flaw was a latent defect which naturally extended itself to the surface and the assured sustained damage to his machinery. There was a damage other than the defect itself, namely, that caused by the condemnation of the shaft within the currency of the policy. Through a latent defect the shaft was condemned and that condemnation brought about a damage to the assured."

Rejecting this argument, Walton J said at p.188:

"When the underwriters underwrote the steamer for 12 months from May 1902 to May 1903 with this clause in the policy, I am satisfied that they did not mean to say: 'If during that year you find any latent defect in the machinery or hull, we undertake to make that latent defect good.' I am quite satisfied that the underwriters by this clause did not undertake any such liability. If the latent defect during the year of the policy causes any loss or damage then they do undertake to indemnify the ship owners against that loss or damage. That is enough to dispose of this case, and if I am right in that the Plaintiffs are not entitled to recover.

But supposing I am not right in saying that there was no damage caused by a latent defect and supposing I ought to find that the fracture, the crack, was a consequence of the latent defect, the flaw, that they are not one thing, but that one is the consequence of the other, and that there is a crack caused by the latent defect, and that the crack is damage to the machinery, within the clause, have the plaintiffs made out any case that the underwriter is liable under this policy? He is not liable unless the damage was caused by the latent defect during the currency of this policy. Was it caused between May 18 1902 and October 30 1902 when it was discovered? Did the crack appear between those two dates? Did the flaw cause the crack after May 18 1902? There is no evidence to show that it did. There is no evidence to show that the vessel was in any way in a different condition on October 30 when the crack was discovered from that she was in on May 18 1902 when the policy attached."

The help which Mr Mildon seeks to obtain from this case arises from what was said by Walton J in the earlier parts of his judgment. At p.185 he refers to the terms of the clause and the necessity of showing that there is damage caused by a latent defect:

"A claim does not fall within the clause unless there is some loss of or damage to hull or machinery or some part of the hull or machinery and there is no claim unless the damage has been caused through a latent defect, or through one or other of the causes that are mentioned in the clause - in this particular case by a latent defect. Therefore there must be a latent defect causing loss of or damage to the hull or machinery and causing that loss of or damage to the hull or machinery during the currency of the policy under which the claim is made. .... It seems to me, that the loss or damage here is the fracture, the crack. Was that caused in consequence of a latent defect? There was an imperfect welding causing a flaw in the shaft, the flaw being the imperfection of the joining of 2 pieces which were welded together: that imperfect welding or, in other words, that flaw did not become visible on the surface, until some time between April 1900 and October 1902. That is the inference of fact which I draw from the evidence.

The crack which is the damage, the only damage which is proved, is really nothing but the development of the flaw - that is of the latent defect. In my opinion such development of a latent defect is not 'damage to the machinery through a latent defect'. In such a case I think the damage is not damage caused by latent defect, but is the latent defect itself and nothing more; a latent defect becoming patent is all that has happened, and it seems to me that the latent defect becoming patent is not within the words of this clause 'damage to the machinery through a latent defect'. The argument for the plaintiff is this, that the loss is a pecuniary loss which the owner suffers when he has to replace his defective shaft, and that therefore the underwriters of a policy which is current when the defect is discovered and when the owner therefore has to replace the shaft are the underwriters who have to pay; .... "

He then commented upon the unreasonableness of that argument and how it would impact upon successive underwriters when in truth there was no change in the condition of the vessel from one period to another. Later on p.187 when considering the phrase "damage to the hull or machinery caused by breakage of shafts", he said that that means that-

"If a shaft breaks and in consequence of that any other parts of the machinery or the hull get damaged then the damage caused by the breakage of the shaft is covered by this clause."

Walton J used the word part to help him to distinguish between what was the latent defect and what could be described as damage caused by the latent defect. This is the same as what he was saying in the passage I quoted earlier

"Supposing I ought to find that the fracture, the crack, was a consequence of the latent defect, the flaw, that they are not one thing but that one is the consequence of the other ....."

The ability to distinguish between the cause and the consequence is a necessary ingredient in the conclusion that one thing was caused by another. His view of the case was summarized in the words: "I think the damage is not damage caused by the latent defect but is the latent defect itself and nothing more." This is a statement of fact which involves the court in making a finding of fact.

The ship-owners appealed to the Court of Appeal. The arguments of counsel were the same and the appeal was dismissed. The reasons given by the members of the Court differed and the majority view corresponded to the narrower argument of Mr Scrutton and is probably inconsistent with some of what Walton J had said in the earlier passages of his judgment which I have quoted. For example Buckley LJ said at p.36:

"I do not accept what Walton J said as to the loss being confined to the injury occasioned to some other parts of the ship by the breaking of the shaft."

Similarly Fletcher Moulton LJ said at p.34:

"In my opinion these words support the view that the loss or damage in question means actual loss or actual damage. But I may say here that if the shaft breaks it appears to me that that is actual loss caused by the actual breaking of the shaft, and I see no reason why that should not be covered by the policy, though I do not decide it, as it is not necessary to do so in this case."

Fletcher Moulton LJ went on to develop the lack of fortuity point which I have referred to earlier:

"A defect initially latent but spreading until it becomes a patent defect is an ordinary incident in all machinery. .... That is a case of a latent defect developing into a patent defect. But it is so ordinary an instance that it is one of the commonest forms in which the economic wearing out of a part of machinery occurs. I do not believe for one moment that this clause means that the machinery is insured against the existence of latent defects. ..... There was simply a condemnation of a shaft which had shown that it was no longer fit to be used."

The decision of the Court of Appeal does not provide any support for Mr Mildon's arguments.

The next case is Hutchins of which the facts are even more striking. It concerned a time policy on hull including the Inchmaree clause. The vessel was built in 1906 in England but her stern frame was a casting which had been supplied by a continental foundry. The casting had been improperly made using an inclusion from a separate batch of metal which, when it cooled, caused shrinkage cracks. The foundry was well aware of what it had done and took steps to conceal it by covering over the evidence of the defect. Their deception was successful and the casting was passed by the classification society and incorporated into the hull of the vessel. In March 1909 it happened that she was docked at Barry for painting. The steel structure was scraped and the defect in the casting was uncovered. The frame was condemned by the surveyor and had to be replaced. The owners claimed the cost of the repair from underwriters. The evidence was that the casting was in precisely the same defective condition as it was when the vessel was built. All that had happened was that the defect had been uncovered and discovered. It is hard to see how the claim against underwriters could ever have been expected to succeed. It was truly a case where on any view there was nothing more than the latent defect itself. No damage had ever been caused to the vessel. Nothing had occurred during the currency of the policy other than the discovery of the defect. The trial judge was Scrutton J who adopted that reasoning. He summarized the decision of Walton J in the Faber case as having "held that the latent defect becoming patent was not damage to the machinery through the defect". (p.404) However in a passage at the conclusion of his judgment he used the phrase "damage to other parts of the hull happening during the currency of the policy through a latent defect" as justifying a claim under the Inchmaree clause. This phrase and the use of the word part occurs in a paragraph in the judgment which has been correctly questioned for other reasons but which in its choice of language provides some support for Mr Mildon's argument.

The shipowners appealed. Vaughan Williams LJ and Farwell LJ expressly adopted the judgment of Walton J in the Faber case. In particular Vaughan Williams LJ adopted his statement: "The damage is not damage caused by a latent defect but is the latent defect itself and nothing more; a latent defect becoming patent is not within the words of this clause 'damage to the machinery through a latent defect'." Fletcher Moulton LJ was again a member of the Court. His reasoning was essentially unchanged:

"It is suggested that this was a 'loss of or damage to hull through a latent defect in the hull' within the meaning of the Inchmaree clause. It was in my opinion nothing of the kind. It was not loss or damage caused by a latent defect, but the latent defect itself. To hold that that clause covers it would be to make the underwriters not insurers, but guarantors and to turn the clause into a warranty that the hull and machinery are free from latent defects, and, consequently, to make all such defects repairable at the expense of the underwriters. The fact that it begins with a word "insurance" negatives in my opinion the possibility of its being so interpreted."

The facts of the Hutchins case were clear and straightforward and fell on the wrong side of the line. It was the latent defect itself and nothing more. The adoption by the Court of Appeal of the judgment of Walton J is indicative of their opinion as to the factual situation to which it applied.

The third case is the Scindia case decided by Branson J. It is this case which Mr Mildon submits establishes the proposition for which he contends and necessitates the drawing of a distinction between one part of the hull or machinery and another. Again the policy was a time policy on hull including the Inchmaree clause. The vessel had traded for a number of years without relevant incident and in December 1931 was in dry dock at Bombay for the purpose of renewing the lower half of the wood lining of the stern bush. For this purpose it was necessary to remove the propeller and draw the tail end shaft. While the propeller was being wedged off, the shaft broke at the point where the cone begins owing to a latent defect in the shaft at that point. The propeller and the tapered end of the shaft fell to the bottom of the dock and one blade of the propeller was broken. Underwriters admitted liability to pay for the repair to the propeller but refused to pay for a new shaft. The surveyor had reported that the shaft had "broken off at the end of the liner exposing a deep smooth flaw extending downwards from the top" of the shaft as he was observing it: "approximately one half of the material was sound and had been newly fractured." The contemporaneous records treated it simply as the discovery of an "old flaw". The broken shaft was disposed of as scrap.

The shipowners put their argument in three ways: Where a policy covers breakage of shafts, the fact of the breakage of a shaft is sufficient to give the insured a claim under the policy whatever may have been the cause of the breakage: If the clause does not cover breakage of shafts simply, then, at all events, on its true construction it covers loss or damage to hull or machinery through breakage of shafts, and since in the present case the shaft was part of the machinery, the damage through the breakage of the shaft is covered by the clause: Further, the clause covers loss or damage to hull or machinery through any latent defect in the machinery, and, since the shaft was part of the machinery, and the breakage of the shaft is clearly shown to have been due to a latent defect therein, that damage is covered by the clause.

Branson J rejected the first argument as being directly contrary to the wording of the clause. He also rejected the second argument saying:

"That, it seems to me, is a forced construction of the language and not the ordinary meaning which, reading the clause as a piece of English prose, one would be inclined to put upon it. It follows other clauses in which obviously the loss or damage happens to something different from the thing by which the damage is said to be caused. The first clause is "caused by accidents in loading", and so forth: the next is "caused through the negligence of master, mariners" and so forth. Both those clauses obviously envisage, as it seems to me, a state of affairs in which the main cause produces damage which has an effect on something else: and I see no reason why, when after these two clauses one comes down to the one with which I have particularly to deal, one should read it in any other way. It seems to me, therefore, that the proper reading is that the breakage of the shaft itself is not covered, nor can it properly be said that the breakage of the shaft is a loss of or damage to machinery caused by the breakage of the shaft. The breakage of the shaft is the breakage of the shaft, and if, by reason of the breakage of the shaft, the machine is torn to pieces, then one would get damage caused by the breakage of the shaft. But in this case the only damage beyond the damage to the propeller, which has been paid for, is the actual damage which happened to the shaft itself, to wit, the breakage of the shaft. To speak of that as damage to the machinery which the breakage of the shaft has caused, seems to me to produce a confusion both of thought and language, which I think should not be introduced in the construction of a clause of this kind."

Turning to the third way in which the shipowners had put their argument he rejected it in very much the same terms:

"The plaintiffs say further that was here a latent defect in the shaft and that damage arising from breakage of the shaft is damage which has been caused to the machinery, to wit, the shaft, by reason of a latent defect in the machinery, to wit, the shaft. In this case the latent defect had proceeded so far from the ordinary wear and tear and use to which a propeller shaft is subjected that it actually resulted in a solution of the continuity of the shaft and it is said that by reason of that fact this case can be distinguished from the case of Hutchins v Royal Exchange . The plaintiffs asked me to say that this curious position arises: that because it is now known that this tail shaft was in such a state owing to this latent defect that had the defect been discovered it would immediately have been appreciated that the tail shaft was only worth what it would fetch as scrap metal and because the defect had become so serious that a perfectly ordinary operation had caused the shaft to break in two, the loss is thrown upon the underwriters and they must not only provide a new shaft, but bear the cost of installing it in the ship. ..... It seems to me that, although in this case the latent defect had gone so far as to cause a complete severance of the shaft into two pieces, yet all the considerations which were applied in Hutchins v Royal Exchange apply equally in the present case. What I have already said in regard to the breakage of the shaft and the necessity for their being some damage caused by the breakage to the shaft other than the breakage of the shaft itself, seems to me to apply also to the case of a latent defect. Damage to hull or machinery caused through a latent defect in the machinery is something different from damage involved in a latent defect in the machinery itself. .... All that has happened is that [the old flaw] has gone on developing, and although the shaft was not subjected to anything in the shape of a peril, but to an ordinary operation of ship repairing, that operation caused its breakage."

He described his reasons as being effectively those of Walton J in the Faber case.

In my judgment the decision in the Scindia case and what was said by Branson J does not alter the question which we have to consider nor does it require a different construction to be placed upon the wording of the Inchmaree clause than that to be derived from the ordinary meaning of the words used. The question remains one of fact; was damage to the subject matter insured caused by the latent defect? The facts in the Scindia case assisted Mr Mildon's argument in that they provided an example of a shaft which contained a defect, the "old flaw", which then had lead to the fracturing of the remainder of the shaft. He was able to say that Branson J did not regard this fracture as involving any damage to the shaft distinct from the pre-existing latent defect in the shaft. But the facts of that case were clearly very different from those of the present case. Branson J was clearly justified in reaching the conclusion which he did. No loss by a peril insured against had been proved. The shaft was already in a condition which required it to be condemned and its value was already no more than its scrap value. The breakage of the shaft was simply the breakage of a part that was already without value and required to be replaced. There was no insured loss. This is what is reflected by the language which Branson J used when asking whether there was something different from the damage involved in the latent defect itself. This conclusion is reinforced by his treatment of the case as being on all fours with that in the Faber case. The loss which the shipowner had suffered was a financial loss arising from the discovery of the latent defect not from the latent defect having caused any damage.

In my judgment Mr Mildon's argument cannot be accepted. There is potentially a factual problem involved in distinguishing between an embryonic fatigue crack as a latent defect and a system of cracking of such magnitude and severity that it significantly weakens the integrity or tensile or shear strength of the structure of the vessel. This distinction is not made easier by the knowledge acquired over the last 40 years of the fatigue characteristics of metals and the propagation of fatigue cracks. Contrary to the submission of Mr Mildon it must, at least in part, be a question of degree. The word defect does not as such connote damage. Similarly, a damaged article may properly be described as being defective but that may be because it is defective in function. On the facts of the Faber and Hutchins cases, the shaft and frame would properly have been described as defective but would not be described as damaged. The view of Branson J was that the shaft in the Scindia case was in reality to be described as defective and not as damaged. In the present case, having read the Veritec report, it would be an abuse of language to describe the legs and spudcans as merely defective; on any ordinary use of language they were damaged. They were damaged by being subjected to stresses which they were unable to resist due to the latent defects, that is to say the wrongly profiled welds and the incipient fatigue cracks. In my judgment the Scindia case does not require a different conclusion. Indeed to read the Scindia case as having the effect for which Mr Mildon contends would be to contradict the language of the clause.

The submission based upon the use of the word part is in my judgment open to the same objections. It leads to absurd results. It provides no criterion for distinguishing between what is and what is not damage. The suggested criteria - what can be physically separated, what performs a separate function - are not derived from anything contained in the clause nor from anything said by Walton J and Branson J. The word part is capable of being used in a whole variety of ways depending upon the context. Its use provides no answer to any relevant question. The weld is a part just as much as is a bracket or bulkhead or plate or the totality of the leg structure.

The use of the word part in the additional perils clause is normally simply to avoid the need to exclude from the indemnity to which the assured is entitled if he proves a claim under the Inchmaree clause of deducting the cost of repairing or replacing the originally defective part. For this purpose there is no need to define what is meant by the word part. It creates no problem in the present case. (Steering gear cases may raise difficult questions with which we are not concerned.) It provides no guidance to the construction to be placed upon the Inchmaree clause itself beyond emphasising the need under that clause to prove that damage to the subject matter insured has been caused.

The Judge rightly criticized the artificiality of some of the almost metaphysical arguments which were addressed to him; indeed, each counsel before us criticized the over-elaboration and subtlety of the other's arguments. Both referred during the course of their argument to cases upon the limitation of actions and the accrual of causes of action, for example, Pirelli v Oscar Faber [1983] 2 AC 1, Murphy v Brentwood DC [1991] 1 AC 398 and Invercargill CC v Hamlin [1996] AC 624. I agree that these authorities do not help. They are addressing a different and far more difficult problem. The question in the present case is simply one of applying a clearly expressed clause to the facts of the case.

In my judgment the questions to be asked are:

(1) Was there damage to the subject matter insured?

(2) Did that damage occur during the period covered by the policy?

(3) Was that damage caused by a latent defect in the machinery or hull of the vessel?

In the present case the answer to each of those questions is in the affirmative. It follows that the owners are entitled to recover an indemnity from the underwriters.

In conclusion I should briefly refer to certain other arguments of Mr Ruttle which in my judgment the Judge correctly rejected. Imminence of loss or damage is not the same as damage: damage is physical damage which has occurred. Similarly, there is no significance in the omission of the word 'of' after the word 'loss' in the 1983 print of the Institute Clauses. Finally, the Judge correctly rejected the alternative argument based upon the sue and labour clause. Assuming that the owners' primary argument failed, the Judge was right to consider what would have been necessary for the purposes of avoiding an insured loss and concluding that the expenditure would not have exceeded the deductible under the policy.

But for the reasons I have given I consider that the primary argument of Mr Ruttle succeeds and the appeal should be allowed accordingly.

LORD JUSTICE WARD: I agree that the appeal should be allowed for the reasons given by Hobhouse LJ but as we are differing from Tuckey J, I state shortly the reasons for my conclusion.

The relevant clause is the "Inchmaree" clause 6.2 which provides:-

"6.2. This insurance covers damage to the subject matter insured caused by...

6.2.2 ....any latent defect in the...hull..."

On an ordinary construction of that clause the issue whether or not the particular loss was covered would depend on the answers to the following questions:-

(1) Was there actual damage to the subject matter i.e. the hull?

(2) Was the direct cause of that damage some latent defect in the hull i.e. a defect which could not have been discovered at the time by a reasonably careful examination?

Because this is a time policy, the further question is whether the loss occurred within the period covered by the policy.

So to view the issue seems to me entirely to follow the analysis which Walton J gave the question in Oceanics Steam Ship Co v Faber (1906) 11 Com. Cas. 179, approved as it was by this court in Hutchins Brothers v Royal Exchange Assurance Corporation [1911] 2 KB 398. Walton J said at p.185:-

"...I have to construe the clause. It seems to me quite plain that the effect and sense of this clause is not that the underwriters guarantee that the machinery of the vessel is free from latent defects, or undertake, if such defects are discovered during the currency of a policy, to make such defects good....The underwriters agree to indemnify the owner against any loss of or damage to the hull or machinery through any latent defect, so a claim does not fall within the clause unless there is loss of or damage to hull or machinery or some part of the hull or machinery and there is no claim unless that damage has been caused through a latent defect...Therefore there must be a latent defect causing loss of or damage to the hull or machinery, and causing that loss of or damage to the machinery during the currency of the policy under which the claim is made."

If one is considering whether there is damage to the hull and whether such damage is caused by a latent defect in the hull, it follows that the damage must be something different from, something over and above and incrementally greater than the latent defect itself. Where the line is to be drawn is a matter of fact and degree. It requires a factual assessment of, on the one hand, the nature of the latent defect and all that is inherent in it, and, on the other hand, the nature of the damage to the hull. The latter must be directly caused by the former. A careful consideration of the facts of the Oceanics Steam Ship Company case are important in understanding how that distinction came to be made. Walton J found this at p.184:-

“In October, 1902, during the currency of the policy in question, the vessel was docked at San Francisco for the purpose of being overhauled. The propeller was removed, the tail shaft drawn into the tunnel for examination and then it was discovered that there was a fracture about three feet forward of the after liner in the shaft. It was a fracture, a crack, extending for the length of about, one of the surveyors says eight inches, and another says thirteen inches.”

It is to be noted that the judge is using “fracture” and “crack” synonymously. He went onto to find at p.185:-

"...I think from the evidence that the fracture was caused by an imperfect welding made in 1891. Apparently the flaw arising from the imperfect welding had not made itself visible on the surface in 1900. At some time between the survey in April 1900, and October, 1902, the flaw which was not visible in April 1900, became visible on the surface in the form of a crack such as has been described.”

As I read that finding the flaw which became visible in the form of the crack/fracture was the same fracture/flaw which was caused by the welding. So he found at p.186:-

"The crack which is the damage, the only damage which is proved, is nothing really but the development of the flaw - that is of the latent defect. In my opinion, such development of a latent defect is not “damage to the machinery through a latent defect.” In such a case I think the damage is not damage caused by the latent defect, but it is the latent defect itself and nothing more; a latent defect becoming patent is all that has happened, and it seems to me that the latent defect becoming patent is not within the words of this clause “damage to the machinery through as latent defect.””

As I understand that judgment the “development of the flaw” was not (or, at least, not to any material extent) a progression of or a worsening of the flaw but merely its becoming visible - “a latent defect becoming patent is all that has happened.” The only damage proved was the crack but the crack was not new: it was only the latent defect made manifest.

When the case went on appeal Lord Alverstone CJ and Fletcher Moulton and Buckley LJJ all entertained some doubt about a suggestion that had the shaft broken from the latent defect within it, it would not have been covered. In the judgment of the Court of Appeal, it was unnecessary to pursue that question since the claim failed because there was no evidence that the loss occurred during the time when the policy attached. In the course of his judgment Fletcher Moulton LJ said this at p.34:-

"A defect initially latent, but spreading until it becomes a patent defect, is an ordinary incident in all machinery.”

One must look carefully to see what he meant by “spreading”. He continued, now at p.35:-

"A person may carefully examine a cylinder cover on one day and find no trace of any defect in it. A week later he may find a trace of a crack. It is his duty, of course, then to replace it if he can do so. He may be perfectly certain in his mind that the reason that the economic life of that cylinder cover has come to an end is because there was initially something weak in it, and as is always the case the weak point is the first to give in. That is a case of a latent defect developing into a patent defect....(This clause) only means that, if through their latency those defects have not been guarded against, and actual loss of the hull or machinery, or damage to the hull or machinery, arises from those defects, the insurers will bear the burden of that loss.”

As I understand his judgment, he is distinguishing between what is inherent in the latent defect, and further damage caused by it. Buckley LJ drew the same distinction at p.36:-

“I do not think that the risk which was covered included the risk of discovering, during the currency of the policy, that a latent defect, which had been existing for some time previously, was there.”

The next case is Hutchins Brothers v Royal Exchange Corporation [1911] 2 KB 398. On the facts found by Scrutton J at p.400 it appeared that during the casting of the stern frame, a V-shaped wedge of metal of different composition from the original mould was run in and a cooling crack took place in that wedge which he was disposed to think spread into the original material. That crack showed markedly on the surface and the manufacturers covered it with some steel wash. It was therefore undetectable on ordinary examination. In time the vessel had to force its way through ice in the Black Sea. The judge found at p.402:-

"I am not satisfied that the ice either made the fracture or increased it. In the result I find that there was an obvious defect in the stern frame at (the place of manufacture); that it was covered up by the makers and remained undiscoverable by reasonable inspection until after the commencement of the policy; and that its becoming visible was due to the ordinary wear and tear of the ship’s life.”

He found at p.405:-

"The only damage is, in my view, the latent defect itself, which by wear and tear has become patent...The stern frame has not been lost in fact; it is there as it was before the policy began; the only change is that a previous latent defect has by wear and tear become patent.”

He was upheld by the Court of Appeal. Fletcher Moulton LJ again drew the distinction at p.411:-

"It was not loss or damage caused by a latent defect, but a latent defect itself.”

In the third case, Scindia Steamships (London), Ltd v The London Assurance [1937] 1 KB 639 Branson J described what happened as this (p.648):-

"During the operation of wedging off the propeller the shaft was being subjected to an ordinary operation of repair which any shaft of proper strength and construction would be able to sustain without any difficulty, but, owing to what is described as a “smooth flaw extending downwards from the top as the shaft then lay” deep into the metal, involving about one half of the material, the other half of the shaft remained and was broken.”

It is apparent from further examination of the facts as set out on page 643 that the “deep flaw” was “found to be the old flaw” when the breakage was examined. The surveyor found that there was a pre-existing deep smooth flaw (which seems then to have been in no different condition from what it was in from the start). Branson J made the distinction at p.651:-

"Damage to hull or machinery caused through a latent defect in the machinery is something different from damage involved in a latent defect in the machinery itself.”

On the facts he found:-

"I the present case, the tail shaft broke owing to a latent defect existing in itself, and...existing before the underwriters ever came on risk. There is nothing to show that this latent defect developed or came into being during the currency of the policy. The fact that it is spoken as “an old flaw,” whether it had previously been discovered or not, tends to show that it had existed before this policy was taken out. All that has happened is that it has gone developing, and although the shaft was not subject to anything in the shape of a peril, but to an ordinary operation of ship repairing, that operation caused its breakage.”

My conclusion from a close consideration of those cases is that the crucial test is to draw the distinction between what constitutes the latent defect itself, and whether damage to the hull arising during the operation of the policy was caused by that latent defect.

Tuckey J at p.88 directed himself:-

"...before there is any cover...there must be damage to the vessel in the sense considered in (these) cases...”

Mr Mildon, counsel for the underwriters submitted to him and repeated the submissions to us that all that had happened was that the defect had developed and had become patent and there was no “consequential damage” to the vessel. The judge found at p.89:-

"What happened here is that there were flaws in the weld which developed into cracks which spread to the immediately adjoining structures which the weld was meant to hold together. As a matter of common sense, it is impossible to see that at this stage anything consequential has happened which could be characterised as damage to the vessel.”

That is a conclusion of fact and since the facts were not in dispute we are in as good a position in this case to form a view about those facts as the trial judge was and the usual inhibitions against interfering with findings of fact do not fetter our assessment of the first question: was there damage to the hull?

To answer that question one needs to identify what the latent defect was. The evidence establishes that the circumferential welds on the underside attaching the top plates of the spudcans to the legs of the platform were not properly profiled in that there was an abrupt change between the surface of the weld and the parent metal. This caused minute fatigue cracks which were latent. These minute cracks, and others that had developed by then, remained latent at the commencement of the policy. By the time the damage was discovered the cracks were so large as to be visible even on the underwater inspection by the diver. There were circumferential cracks in the legs as wide as 10 to 16 mm in the metal of the legs extending around 70% of one leg, 80% of another and 90% of the third. There were cracks in the top plating and bulkheads of the spudcan. They can be seen in the photographs. The legs and spudcan are part of the hull. Applying the judge’s common sense test to this question, it seems to me plain that this was not simply a latent defect becoming patent. It was well beyond that. Metal had fractured. That was damage to the hull.

The second question is whether or not that damage was caused by a latent defect. The judge found at p.89:-

"The assured has established that there was something wrong (defective) with the welds. They were badly profiled, so that fatigue cracks developed in them much more rapidly than should have been the case. Contrast the welds on the upper side of the top plate which were properly profiled and did not crack. The defect was latent and within the hull, so it was an insured peril within the inchmaree clause. The cracks were not ordinary wear and tear.”

That finding was also inevitable on the facts of this case. The conclusion is inescapable that there was damage to the hull which was caused by a latent defect in the hull. Accordingly the assured was entitled to recover the cost of repair.

I turn briefly to the submission that the claim fails because the plaintiff had not shown that some other part than the defective part of the hull suffered damage.

Nothing in the unambiguous language of clause 6.2 compels that construction. The use of “any defective part” in clause 1.1.2 of the Institute Additional Perils Clause does not justify the incorporation of that distinction into clause 6.2. The Additional Perils Clause serves another purpose which is to extend the cover. I do not overlook the summary at the conclusion of Scrutton J’s judgment in Hutchins Brothers at p.406 when he says:-

"...What is recoverable under this part of the Inchmaree Clause is:...(3) Damage to other parts of the hull happening during the currency of the policy, through a latent defect...”

Since he goes on immediately to observe “the pre-existing latent defect itself is not damage, indemnity for which is recoverable”, it seems to me that he is in fact doing no more than re-state the essential distinction upon which his judgment depends which is between what is a latent defect in the hull and what is damage to the hull. He had on p.405 posed the question:-

"Has any part of the hull been lost in fact during the currency of the policy?”

I have added the emphasis to make the point that he did not ask whether any other part had been lost. In Oceanics Steam Ship Company Walton J had posed the question in equally broad terms at p. 188:-

"If the latent defect during the year of the policy causes any loss or damage, then they do undertake to indemnify the ship owner against loss or damage.” (my emphasis added)

At p.186 he had said:-

"...A claim does not fall within the clause unless there is loss of or damage to hull or machinery or some part of the hull or machinery, and there is no claim unless that damage has been caused through a latent defect...”

Once again the emphasis is added to make the observation that he did not refer to some other part of the hull. He did use that language when dealing with the separate question of the breakage to the shaft when at p.187 he said:-

"...if a shaft breaks and in consequence of that other parts of the machinery or hull get damaged, then the damage caused by the breakage of the shaft is covered by this clause.”

He was making a very different point.

Consequently I find nothing in the language of the clause itself nor in any of the judgments to which we have been referred which justify the enquiry which the judge was invited to undertake which was to decide whether the weld was a separate part from the legs and the spudcan which were united by that weld or whether the united legs and spudcan together constituted the defective part. If I were driven to it, I would find the weld was a separate part. To say by what criteria one draws the line in defining a “part”, is not easy. By contrast if, as I find, the questions are - to put them in a different order - (1) was there a latent defect in the hull and if so what constituted that defect, (2) did the defect cause damage to the hull, then the answers, being matters of fact, can be given without conceptual difficulty by properly instructed experts applying the established authorities.

I would therefore allow the appeal.

THE PRESIDENT: I have had the advantage of reading in draft the judgments of Lord Justice Hobhouse and Lord Justice Ward. I agree that the appeal should be allowed for the reasons which they give.

ORDER: (1) Appeal allowed;

order of Tuckey J to be set aside;

(2) judgment for the plaintiff against defendants in the sum of S$1,030,419.09 and US $12,000 with interest thereon from 14.4.87 to date at 1.5% over the Singapore Prime Rate on Singapore Dollars; liberty to apply in respect of S$32,273.40;

(3) costs of appeal and at first instance to be paid by the defendants to the plaintiff, such costs to be taxed if not agreed;

(4) costs of the action paid by the plaintiff to the defendants to be repaid by the defendants forthwith;

(5) matter remitted to Commercial Judge for rates of interest to be determined;

(6) leave to appeal refused;

(7) stay pending appeal, subject to payment of money either into joint account or to the plaintiff's solicitors, and defendants' undertaking to petition House of Lords within six weeks.

______________________________