[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales High Court (Commercial Court) Decisions
You are here: BAILII >> Databases >> England and Wales High Court (Commercial Court) Decisions >> Versloot Dredging BV & Anor v HDI Gerling Industrie Versicherung AG & Ors (Rev 1) [2013] EWHC 1666 (Comm) (14 June 2013) URL: http://www.bailii.org/ew/cases/EWHC/Comm/2013/1666.html Cite as: [2013] 2 All ER (Comm) 465, [2013] Lloyd's Rep IR 582, [2013] EWHC 1666 (Comm), [2013] 2 Lloyd's Rep 131

[New search] [Printable RTF version] [Help]

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
COMMERCIAL COURT

		Royal Courts of Justice Rolls Building, Fetter lane EC4A 1NL
		14/06/2013

The Hon. Mr Justice Popplewell :

Introduction

1. This is a marine insurance claim for a partial loss. The Claimants ("the Owners") are the owners of a gearless general cargo ship named "DC MERWESTONE" ("the Vessel"). The Defendants ("the Underwriters") are the hull and machinery underwriters of the Vessel under a time policy for twelve months at 1 April 2009. On 28 January 2010 the Vessel was off the coast of Poland, shortly after commencing a laden voyage from Klaipeda, Lithuania, to Bilbao, Spain, when she suffered an ingress of water which flooded the engine room, and incapacitated the Vessel. The Vessel's main engine was damaged beyond repair. The claim by the Owners under the policy is for the resultant loss in the sum of €3,241,310.60.
2. The Underwriters advance three alternative defences to the claim. They deny that the loss was caused by an insured peril. They contend that the loss was caused by the unseaworthiness of the Vessel to which the assured was privy, with the result that no liability attaches by virtue of s. 39(5) Marine Insurance Act 1906. They contend that the claim is forfeit because the presentation of the claim was supported by fraudulent statements. In addition, they challenge one element of the quantum of the claim.

The Vessel

3. The Vessel is a Dutch registered gearless cargo vessel of 2,973 grt, 1,852 nrt and 5,010 dwt. She was built in 1974. In 2001 major renovation work was carried out in Romania and then in the Netherlands (under the supervision of Lloyds Register and the Dutch Shipping Inspectorate). A new bow and midship section were attached to the old aft part of the Vessel; the only original section which remained was aft of the engine room bulkhead at frame 24. She is 90.74m long, 15m in beam and has one hold divided into 3 by 2 non-watertight (moveable) bulkheads. Outboard of the cargo hold are five fuel and ballast wing tanks. There are five port and starboard double-bottom ballast tanks, which are divided longitudinally by a duct keel tunnel.
4. The duct keel, which played a critical part in the casualty, is 62.48m long, 1.8m wide and 1.2m high. It runs most of the length of the Vessel, from the bowthruster room in the foreship to the engine room at the aft of the Vessel. It has what ought to be watertight bulkhead separation from the bowthruster room at frame 112 forward, and from the engine room at frame 24 aft. It runs along the centre line beneath the hold, separating the five pairs of port and starboard double-bottom ballast tanks. Within the duct keel are remotely operated ballast valves, ballast pipes and electrical cables. The tunnel can be entered at three points. At the bowthruster room, which is situated beneath the chain locker at keel level in the bow, there is access through a vertical manhole located beneath the floor plates. The manhole is sealed by a cover which has screw bolts to maintain the watertight integrity of the bulkhead. A similar manhole is located at the aft end of the tunnel beneath the engine room floor plates. This engine room manhole was rarely used because of its inaccessibility. There is a further manhole access in the hold.
5. Once entered, the tunnel can be accessed along its full length, although it is an uncomfortably small space: those of a larger build would find it difficult to enter the tunnel and manoeuvre within it, and those of a smaller build who could enter it would have to travel along it by lying on their back and shifting in that position.

The casualty

6. On 21 January 2010 the Vessel arrived at Klaipeda, Lithuania, to discharge a cargo of soya meal and thereafter to load a cargo of scrap steel. She had her usual complement of six crew, comprising the Master (Captain Loosman), a Chief Officer, a Second Officer, an Engineer, and two ABs. Captain Loosen was the regular master of the Vessel, who had rejoined her at Rotterdam shortly before the commencement of the loaded voyage. The Chief Officer was 21 and newly promoted. The Engineer, Mr Catolin, held a fourth engineer's certificate. He was, however, very familiar with the Vessel, having served on her continuously from 2006 until the time of the casualty as the sole engineer, save for periods on leave.
7. The weather was exceptionally cold and the hatch covers and gangways were covered in ice. The outside temperature varied between minus 10°C and minus 35°C. The crew had to chip ice off the hatch covers before opening them, and used the Vessel's emergency fire pump and lines to blast the chipped ice away before opening the hatches. During this operation Captain Loosman slipped over and broke a number of ribs, as a result of which he was replaced for the subsequent voyage.
8. The emergency fire pump is housed in the bowthruster room. When the crew had finished using the emergency fire pump, they drained the deck lines, but did not drain the seawater from the emergency fire pump or close the sea inlet valve to the pump located in the bowthruster room, as it is common ground they ought to have done.
9. On 24 January 2010 the Vessel commenced loading her cargo of scrap steel and deballasting at the same time, having had to take on ballast for stability reasons when shifting berth prior to loading. Although the Vessel was able to pump out No.3 wing tanks, she was unable (pneumatically) to open the valves to Nos. 1 & 2 wing tanks and No.4 DB tank. All the valves were located in the duct keel. The Engineer entered the duct keel tunnel via the forward manhole in the bowthruster compartment to attempt to release frozen ballast valves. However the valves to No. 1 wing tanks were frozen and he was unable to open them. He assumed that the same would be true of the other valves. As a result a quantity of ballast water remained on board. Upon the Engineer leaving the tunnel, he and/or the Chief Officer replaced the manhole cover. There is an issue as to whether they did so properly and effectively.
10. The Vessel completed loading her cargo of a little over 4,000 mt of scrap steel on 27 January 2010. At about 0230 on 28 January 2010 Captain Lilipaly boarded the Vessel and took over as Master from Captain Loosman, who required medical assistance and returned home. Captain Lilipaly was familiar with the Vessel having served on her as Master on previous voyages.
11. The Vessel left port for Bilbao at about 0930 the same morning with the assistance of a local tug which acted as ice-breaker. Captain Lilipaly noticed that the Vessel was trimmed slightly by the head, which the crew told him was caused by the ballast.
12. At about 2058 that evening, 28 January 2010, the Engineer noticed water rising beneath the floor plates in the engine room. There is a dispute as to whether this was as a result of the engine room bilge alarm sounding. When the Engineer observed water in the engine room he went to the bridge where the Chief Officer was handing over the watch to the Master. The Master and the Engineer went down to the engine room. They saw water below the level of the engine room floor plates. It appeared to be pushing up. At about 2108 the Master ordered the Engineer to start pumping out the engine room bilges. The Master then returned to the bridge and sent out a distress alert at around 2138. Meanwhile, the rest of the crew were mobilised.
13. Despite attempts to empty the engine room bilges using the Vessel's two general service pumps, the water level continued to rise. The crew could not locate the source of ingress, which they believed to be in the engine room. An attempt was also made to use the Vessel's ballast pump to empty the engine room bilges by cutting a hole in the ballast suction line. This improvisation would have been capable of pumping the floodwater overboard but for a mop head which was drawn into the suction line so as to block it.
14. Several vessels responded to the distress call with rescue operations co-ordinated by the Maritime Rescue Co-Ordination Center at Gdynia. Various vessels offered and provided assistance, including the supply of portable pumps, but without success. The Master altered course towards the coast of Poland in case it became necessary to strand the vessel to prevent her from sinking, and continued to run the main engine for this purpose until about 0246 on 29 January 2010 when the main engine became fully submerged and stopped working. The engine room was sealed off and the crew retreated to the bridge, where they spent a cold and fearful night.
15. The tug "AGIS" came alongside at about 1000 on 29 January 2010 and the Vessel was towed to Gdynia. She was pumped out, and in due course it was determined that despite the initial common assumption, there was no leak in way of the engine room. On 22 February 2010 the Vessel was dry docked at Nauta Shipyards S.A. in Gydnia. Two days later the Vessel was refloated, and on 26 February 2010 she was towed to the Bredo Shipyard in Bremerhaven for permanent repairs to be carried out. The Vessel's main engine was damaged beyond repair and was replaced by a new engine and gearbox.

The causes of the casualty

16. The cause and mechanism of ingress of water into the engine room during the casualty is not substantially in dispute. When the crew finished using the emergency fire hose prior to cargo operations at Klaipeda, seawater remained in the emergency fire pump and in the filter located at the inlet side of the pump. The water froze and expanded. The expansion caused a crack in the casing of the emergency fire pump. It also distorted the bar restraining the lid on the filter, so that the lid no longer formed a seal with the filter itself. After the vessel sailed from Klaipeda on 28 January 2010, the ice in the pump and filter began to melt. Once the ice melted, seawater leaked and entered the bowthruster space from the open sea inlet valve through the crack in the fire pump casing and the displaced filter cap.
17. The water then entered the duct keel tunnel through an aperture where the cables ran through the bulkhead from the bowthruster room into the tunnel. The aperture, which was referred to in the evidence as the cable gland, was not packed or sealed and the cables passed loosely through it. The aperture was at a level below the floor plates in the bowthruster room. The water was therefore free to fill the duct keel tunnel once it had entered the bowthruster room.
18. The Owners allege that the manhole cover had not been properly secured by the crew after the engineer had entered the tunnel at Klaipeda, and that this was an additional means of ingress from the bowthruster room to the duct keel. I find that there was no lack of sealing of the manhole cover which played any significant causative part in the casualty. Although the Dutch Shipping Inspectorate ("DSI") report into the incident said that "the front manhole cover was not properly sealed but replaced with a few nuts", and that it "had not been effectively sealed", there was no evidence led by the Owners to this effect by any of the crew; on the contrary, the Chief Officer appears to have qualified his drafted witness statement in manuscript so as to reject the suggestion. If there was any deficiency in the sealing of the manhole cover, its size would not have been significant in allowing transfer of water from the bowthruster room to the duct keel tunnel when compared with the size of the cable gland.
19. At the engine room bulkhead end of the duct keel tunnel at frame 24, the cables emerged from the tunnel horizontally through an aperture into a pipe. This aperture was not sealed at the bulkhead. The cables emerged into the engine room not horizontally but vertically as a result of the pipe being L shaped. The aperture where the cables emerged vertically into the engine room was sealed. However the "pipe" between the aperture in the bilge keel tunnel and the aperture in the engine room was not in fact a continuous watertight pipe but had gaps which did not create a watertight seal. Accordingly water from the duct keel tunnel was free to flow into the engine room through these spaces.
20. In this way entry of water through the emergency fire pump and filter lid in the bowthruster room led to the flooding of the engine room. The Vessel's engine room pumps should have been able to cope with the rate of ingress which led to the loss and damage in the engine room if operating at their rated capacity, but they failed to do so for reasons explored more fully below.
21. The chain of causation was therefore as follows:

(1) Crew negligence. At the end of the operation of using the emergency fire pump system to de-ice the hatches and deck, the crew did not close the sea suction and drain the pump, which is what allowed the water to freeze in the system and the system to remain open to the sea. This is what they should have done and they were negligent in failing to do so.

(2) Unseaworthiness. As a result the Vessel lost her watertight integrity in the bowthruster room because the freezing of the water in the emergency fire pump system caused the pump to crack and the filter lid to be displaced; and because water could enter through both those physical defects as a result of the open suction valve. This rendered the Vessel unseaworthy. The bowthruster room was open to the sea, initially subject to a barrier of ice which would inevitably melt in the course of the contemplated voyage. In simple terms there were two holes in a part of the Vessel's hull and machinery which separated the bowthruster room from the sea.

(3) Ingress of seawater into the bowthruster room, which occurred once the ice started to melt.

(4) Lack of watertight integrity of the bulkheads at frames 24 and 112. The unpacked cable glands at each end of the duct keel tunnel, and configuration of the "pipe" at the aft end, led to the ingress of water into the engine room through the duct keel.

(5) The defective state of the vessel's engine room pumping system. Irrespective of the issue whether or not the engine room bilge alarm went off when it ought to have done, or at all, the crew became aware of the ingress in the engine room at a sufficiently early stage for the pumping system to be deployed. The pumping system should have been able to cope with the rate of ingress which led to the loss and damage in the engine room but it failed to do so.

The Issues

22. The Policy terms included the Institute Time Clauses – Hulls 1/10/83 ("ITC") and the Institute Additional Perils Clauses ("IAPC"). Clause 6 of the ITC provides as follows:

"6 PERILS

6.1 This insurance covers loss of or damage to the subject-matter insured caused by:

6.1.1 perils of the seas…

6.2 This insurance covers loss of or damage to the subject-matter insured caused by: …

6.2.3 negligence of Masters Officers Crew or Pilots…

provided such loss or damage has not resulted from want of due diligence by the Assured, Owners or Managers."

23. The IAPC provide:

"1. In consideration of an additional premium this insurance is extended to cover…

1.2 loss or damage to the Vessel caused by any accident or by negligence, incompetence or error of judgment of any person whatsoever.

2 The cover provided in Clause 1 is… subject to the proviso that the loss or damage has not resulted from want of due diligence by the Assured, Owners or Managers…"

24. The Owners' case is that a proximate cause of the loss was the fortuitous ingress of seawater into the bowthruster room, which is a peril of the seas covered under ITC Clause 6.1.1. In the alternative, the Owners contend that there is cover under the Inchmaree clauses (ITC Clause 6.2.3 and/or IAPC Clause 1.2) because (1) a proximate cause of the loss was crew negligence in relation to the emergency fire pump system, which did not result from want of due diligence by the owners or managers; and/or (2) a proximate cause of the loss (other than that in the bowthruster room itself) was the negligence of contractors in failing to seal the cable duct at each end of the duct keel tunnel, which did not result from want of due diligence by the owners or managers.
25. The Underwriters' case is that the loss was not proximately caused by a peril of the sea. The loss was all proximately caused by the crew negligence in relation to the emergency fire pump. This does not fall within the Inchmaree cover because the loss and damage resulted from want of due diligence by the Owners or managers in:

(1) failing to promulgate appropriate cold weather procedures; and/or

(2) failing to have a proper and effective system for the testing and maintenance of the bilge alarms; and/or

(3) failing to inspect and maintain the forward and aft bulkheads in the duct keel at frames 112 and 24 respectively; and/or

(4) failing to have a proper and effective system for the maintenance of the bilge and ballast pumping system.

26. Arguments (2), (3), and (4) raise issues as to the true construction of the Inchmaree clauses, because they seek to apply the want of due diligence proviso (a) to causes of loss and damage which may not be proximate causes and/or (b) to causes of loss and damage other than the named peril which constitutes the insured peril upon which the assured relies as affording cover under the clause.
27. Alternatively, the Underwriters contend that if the loss was caused by an insured peril, they have a defence under s. 39(5) Marine Insurance Act 1906 because the loss was caused by the unseaworthiness of the Vessel to which the Owners were privy. The relevant unseaworthiness to which this allegation applies is the deficient condition of the engine room pumping system.
28. In the further alternative, the Underwriters contend that if the claim were otherwise valid, the claim is forfeit because the Owners employed fraudulent devices in support of the claim when presenting it to Underwriters in 2010 and 2011.
29. I find it convenient to address the issues in the following order:

(1) Was there a loss from perils of the seas (ITC Clause 6.1.1)?

(2) Did the crew negligence in failing to drain down the emergency fire pump and close its suction valve result from the want of due diligence by the assured in failing to promulgate appropriate cold weather procedures (ITC Clause 6.2.3 and IAPC Clauses 1.2 and 2)?

(3) (a) Was the loss proximately caused by the negligence of persons other than the Owners or managers in failing to seal the cable glands at each end of the duct keel tunnel; and (b) if so, did the deficient state of the bulkheads result from the want of due diligence by the assured (IAPC Clauses 1.2 and 2)?

(4) How are the Inchmaree Clauses to be construed? In particular does the want of due diligence proviso apply (a) to causes of loss and damage which are not proximate causes; and/or (b) to causes of loss and damage other than the named peril which constitutes the insured peril upon which the assured relies as affording cover?

(5) Did the loss result from want of due diligence by the Owners or managers in relation to the bilge alarms?

(6) Was the loss caused by the unseaworthiness of the Vessel to which the Owners were privy (MIA s39(5))?

(7) Did the loss result from want of due diligence by the Owners or managers in relation to the engine room pumping system?

(8) What is the correct quantum of the claim?

(9) Have the Owners forfeited the claim by reason of fraudulent devices employed in supporting it?

Issue 1: Was there a loss by perils of the seas?

30. Subject to particular defences, a marine insurer is liable for any loss proximately caused by an insured peril, even though the loss would not have happened but for the misconduct or negligence of the master or crew (Marine Insurance Act 1906 ss. 55(1), 55(2)(a)). In general terms, therefore, whether or not a loss is covered by a marine policy depends on ascertaining its proximate cause (Global Process Systems Inc. and Another v Syarikat Takaful Malaysia Berhad ("The Cendor Mopu") [2011] 1 Lloyd's Rep 560 at [17] to [18]). In the Victorian era, the proximate cause in marine insurance was readily associated with the last cause in point of time; but the modern focus is on the "real efficient cause". The proximate cause is that which is proximate in efficiency, a question which must be answered applying the common sense of a business or seafaring man (TM Noten BV v Harding [1990] 2 Lloyd's Rep 283, 286-7; The Cendor Mopu (sup.) at [19], [29], [49], [79]). A loss may have more than one proximate cause. If an insured peril is one of the proximate causes, the insurer is liable under a marine policy, notwithstanding that there is another proximate cause which does not constitute an insured peril, unless that other proximate cause is specifically excepted (Wayne Tank and Pump Co Ltd v Employers Liability Assurance Corporation Limited [1974] QB 57; J J Lloyd Instruments Limited v Northern Star Insurance Co Ltd (The Miss Jay Jay) [1987] 1 Lloyd's Rep 32; The Cendor Mopu (sup.) at [22], [77], [88], [96]).
31. There are two elements to what constitutes a peril of the seas. First it must be a "peril", that is to say it must be a fortuity, not something which is bound to happen: "The purpose of the policy is to secure an indemnity against accidents which may happen, not against events which must happen" per Lord Herschell in The "Xantho" (1887) 12 HL 503 at 509. This is reflected in rule 7 of the Rules for Construction Schedule to the Marine Insurance Act 1906 which provides: "The term "perils of the seas" refers only to fortuitous accidents or casualties of the seas. It does not include the ordinary actions of the winds and waves".
32. A fortuity is perhaps most easily defined by its antithesis. As it was put in the judgment of the Supreme Court of Canada in C.C.R Fishing Ltd v Tomenson Inc. (The "La Pointe") [1991] 1 Lloyd's Rep 89, 91: "In general the word "fortuitous", as interpreted by the cases, carries the connotation that the cause of the loss should not have been intentional or inevitable." It excludes intentional ingress, as where the ship is scuttled: P Samuel & Company Ltd v Dumas [1924] AC 431; and it excludes cases where the casualty is inevitable, as where there is a rotten hulk or "debilitated" vessel: E.D. Sassoon & Co v Western Assurance Company [1912] AC 561.
33. I shall have to return to the concepts of fortuity and debilitated vessels in due course. It is important to note for present purposes, however, that the definition in the 1906 Act of perils of the seas as excluding "the ordinary action of wind and waves" is one that draws attention to the question whether the wind and waves have some extraordinary effect, rather than whether they were extraordinary in themselves (The Cendor Mopu at [39], [80], The Miss Jay Jay per Mustill J [1985] 1 Lloyd's Rep 264 at 271). Accordingly the passage of water through a hole in the vessel will be a peril of the seas if the occasion for the water to enter the vessel is a fortuitous external accident, notwithstanding that water would enter the vessel through the hole in any state of wind, sea or weather: Seashore Marine S.A. v Phoenix Assurance Plc (The "Vergina") (No 2) [2001] Lloyd's Rep 698 at [95], [96], [100]. The fortuity may lie in what causes the hole, or what causes the seawater to reach or enter the hole, or a combination of both. If there is such a fortuity, the entry of the seawater is not the ordinary action of the wind and waves because the sea has had an extraordinary effect on the vessel.
34. Secondly the peril which gives rise to the marine casualty must be "of the seas". It is not sufficient for this purpose merely that it happens on board a vessel (Stott (Baltic) Steamers Ltd v Martin (The "Ussa") [1916] AC 304, 311). If it might as well have happened on the land as on sea it would not qualify as a peril of the seas: see Thames & Mersey Marine Insurance Company Limited v Hamilton Fraser & Co (The Inchmaree) (1887) 12 App Cas 484, from which it is clear that it is not sufficient to make the peril "of the seas" merely that it happens whilst preparing for a voyage, or whilst the vessel is at sea, or even that it involves seawater. In that case, Lord Bramwell suggested that perils of the seas encompassed "all perils, losses and misfortunes of a marine character, or of a character incident to a ship as such" (492). Lord Herschell identified the concept as perils to which a marine adventure is exclusively subject, or which are special or peculiar to the marine nature of the adventure (498). Lord McNaghten referred to the concept as confined to "loss ex marinae tempestatis discrimine" (501) and spoke also of "sea damage" (502). In The Ussa Lord Atkinson used the expression "damage of a character to which a marine adventure is subject".
35. The mere fact of ingress of seawater into a vessel is not itself a peril of the seas: Samuel v Dumas [1924] AC 431, 446, 455; Wadsworth Lighterage & Coaling Company Ltd v Sea Insurance Ltd (1929) 34 Lloyd's Rep 285, 287. Where, therefore, a loss arises by reason of ingress of seawater through the Vessel's hull, the search for the proximate cause involves an inquiry into that which occasioned the entry of the seawater.
36. Provided the immediate cause of the ingress is fortuitous, then prima facie the loss will be by perils of the sea. This is none the less so if the cause of the ingress is crew negligence: Davidson v Burnand (1868) LR 4 CP 117; Hamilton Fraser & Co v Pandorf & Co (1887) 12 App Cas 518, 522; The Stranna [1938] P 69, 82; Canada Rice Mills Limited v Union Marine and General Insurance Co Ltd. [1941] AC 55. In Cohen and Sons & Co v National Benefit Assurance Limited and Others [1924] Lloyd's Rep 199, Bailhache J was concerned with a policy covering a U-boat during the period when it was being broken up. The vessel sank as a result of the negligence of those carrying out repairs whilst dismantling pipes, by which they caused a breach of the skin of the vessel. In holding that the loss was by perils of the seas, Bailhache J said:

"In my view, the unintentional admission of seawater into a ship, whereby the ship sinks, is a peril of the sea. There is no warranty in this policy against negligence; there is no exception of negligence; and the fact that the unintentional admission of water into the ship is due to negligence is, in my opinion, totally and absolutely immaterial. There is a peril of the sea whenever a ship is afloat in the sea and water from the sea is unintentionally admitted into her which causes a loss, either to the cargo or to the ship."

The principle may be too widely stated, in that it excludes only intentional admission of seawater, and fails to take account of cases of inevitable admission of seawater due to debility of the vessel. But the decision and the statement of principle accurately reflect the law that an ingress of seawater caused by crew negligence is a fortuitous accident which normally constitutes a peril of the seas.

37. It is equally so if the prior cause of ingress of seawater is the unseaworthiness of the Vessel, unless the unseaworthiness is a debility of a kind which prevents the ingress being fortuitous because it is inevitable in any sea conditions: The Miss Jay Jay per Mustill J at 271-2, and in the Court of Appeal at 37, 39. As Mustill J put it at page 271 col 2, in a passage endorsed by Lord Mance in The Cendor Mopu at [75]:

"Nevertheless it is clearly established that a chain of causation running – (i) initial unseaworthiness; (ii) adverse weather; (iii) loss of watertight integrity of the vessel; (iv) damage to the subject-matter insured – is treated as a loss by perils of the seas, not by unseaworthiness: see, for example, Dudgeon v Pembroke (1874) 9 Q.B. 581, per Mr Justice Blackburn at p.595, and (1877) 2 App. Cas. 284, per Lord Penzance at p.296, and Frangos v Sun Insurance Office (1934) 49 Ll.L.Rep 354, at p.359."

38. It is therefore no bar to the fortuitous ingress constituting a peril of the seas that the crew negligence which has given rise to it renders the vessel unseaworthy. Such was the case in Davidson v Burnand (sup) in which the crew at the loading port were negligent in leaving a cock open in the main discharge pipe in the engine room prior to loading, which allowed ingress of water once the loading of the vessel brought the pipe below the water level. Indeed many cases in which a vessel suffers a loss from ingress on the voyage will involve some defect in the plating or machinery which arose by negligence of someone but has remained undetected. "The incursion of seawater through an undetected defect in the ship's basic plating is a classic case of damage by perils of the sea" (The "Hellenic Dolphin" [1978] 2 Lloyd's Rep. 336, 339 col. 1 (per Lloyd J)).
39. So Lord Wright, in delivering the advice of the Privy Council in Canada Rice Mills Limited v Union Marine and General Insurance Co Ltd. [1941] AC 55, summarised the position in the following terms at pages 67-68:

"Where there is an accidental incursion of seawater into a Vessel at a part of the Vessel and in a manner where seawater is not expected to enter in the ordinary course of things and there is consequent damage to the thing insured there is prima facie a peril of the seas. The accident may consist of some negligent act, such as improper opening of a valve, or a hole made in a pipe by mischance, or it may be that seawater is admitted by stress of weather or some like cause bringing the sea over openings ordinarily not exposed to the sea, or even without stress of weather, by the Vessel heeling over owing to some accident or the breaking of hatches or other coverings. These are merely a few among many possible instances in which there may be a fortuitous incursion of seawater. It is the fortuitous entry of the seawater which is the peril of the sea in such cases……….There are many deck openings in a Vessel through which seawater is not expected or intended to enter and, if it enters, only enters by accident or casualty. The cowl ventilators are such openings. If they were not closed at the proper time to prevent seawater coming into the hold, and seawater does accidentally come in and do damage, that is just as much an accident of navigation (even though due to negligence, which is immaterial in a contract of insurance) as the improper opening of a valve or other sea connection."

40. Applying these principles to the current case, the casualty was a loss proximately caused by a peril of the seas, namely the fortuitous entry of sea water during the voyage. A proximate cause of the loss and damage was the ingress of seawater, and the ingress was fortuitous, having been caused by crew negligence at the loading port. The case falls within the examples of perils of the seas given by Lord Wright in Canada Rice Mills: "The accident may consist of some negligent act, such as improper opening of a valve, or a hole made in a pipe by mischance". There was in this case an improper failure to close a valve, which is no different in kind from an improper opening of a valve. There were two holes made by mischance in the pump casing and filter of a pumping system, which is analogous to a hole made in a pipe. The cause of the accident is no different in kind from one in which damage occurs by water coming in through an opening which has negligently been left open such as a cowl ventilator (Canada Rice Mills) or porthole (The Stranna [1938] P 69, 75; Hamilton Fraser & Co v Pandorf & Co (1887) 12 App Cas 518, 525), or a failure properly to make the hatch covers watertight on completion of loading.
41. Mr Jacobs QC's argument on behalf of the Underwriters that the loss was not proximately caused by perils of the sea has four strands. First it is said that the proximate cause of this loss was not the ingress of seawater but the antecedent negligence of the crew, which rendered the vessel unseaworthy.
42. I can not accept this argument. Where there is a fortuitous ingress of seawater, it is often the ingress itself, rather than that which renders it fortuitous, which is to be regarded as the proximate cause of the loss, or at least a proximate cause of the loss. As Lord Wright said in Canada Rice Mills in the passage quoted above: "It is the fortuitous entry of the seawater which is the peril of the sea in such cases……" This is well illustrated by the decision of the House of Lords in Hamilton Fraser & Co v Pandorf & Co (sup.). In that case the casualty was caused by ingress of seawater due to rats on board a vessel gnawing through a lead pipe. The issue was whether the loss came within a charterparty exception which excluded liability for loss caused by "dangers and accidents of the seas." The House of Lords treated the question as the same as that which would arise under a time policy insuring losses by perils of the seas. It held that the gnawing of the pipe by vermin was not itself an accident "of the seas" (see for example Lord Bramwell at 527, and Lord Herschell at 529). Nevertheless the loss was caused by accidents of the seas: the proximate cause of the loss was the fortuitous entry of seawater. The remoter cause of that fortuitous entry, namely the gnawing of the pipe, did not impact on whether the loss was by was an accident of the seas. Lord Watson said at page 525:

"If the respondents were preferring a claim under a contract of marine insurance, expressed in ordinary terms, I should be clearly of opinion that they were entitled to recover, on the ground that their loss was occasioned by peril of the sea within the meaning of the contract. When a cargo of rice is directly injured by ……… crew of the vessel, the sea has no share in producing the damage, which in that case, is wholly due to a risk not peculiar to the sea, but incidental to the keeping of that class of goods, whether on shore or onboard of a voyaging ship. But in the case where …….one of the crew leaves a port-hole open, through which the sea enters and injures the cargo, the sea is the immediate cause of the mischief, and it would afford no answer to the claim of the insured to say that, ……… had careful hands been employed, the sea would not have been admitted and there would have been no consequent damage."

43. This is the effect of the decisions in Davidson v Burnand, and in Cohen and Sons & Co v National Benefit Assurance Limited and Others, in which the ingress was the result of crew negligence which was not itself an insured peril. So too in Seashore Marine S.A. v Phoenix Assurance Plc (The "Vergina") (No 2) [2001] Lloyd's Rep 698, the sequence of causation was (i) negligence of the crew in leaving open a scupper valve (ii) listing of the vessel as a result of crew negligence in ballasting (iii) entry of seawater through the open scupper valve. Aikens J held that the loss would have been proximately caused by the ingress of seawater as a peril of the seas (see [100] and [107(3)]).
44. This conclusion follows not from some immutable principle of law, but simply by application of the legal test of proximate cause to the facts of each case. In such cases a business or seafaring man would normally regard the fortuitous entry of seawater as being a real efficient cause of the loss. It is the action of the sea on the vessel's hull and machinery which is an effective cause of the loss, irrespective of that which rendered such ingress fortuitous. That is so in the present case.
45. The second strand to the Underwriters' argument is that the crew negligence which caused the ingress was not a peril "of the seas", being one which could as easily have happened on land. This is to introduce an irrelevant consideration. It is not necessary to establish that the fortuity which gives rise to the ingress of seawater is itself "of the seas", or indeed an insured peril in itself. The peril is not the fortuity which gives rise to the ingress, it is the fortuitous ingress itself. Accordingly the fortuity which gives rise to the ingress need not independently be "of the seas". The causative fortuity and the ingress of seawater are both part of the accident or casualty and must be looked at together. It is the accident or casualty, embracing both these elements, which is required to be "of the seas". A fortuitous cause of ingress, which is not "of the seas" in itself, which might as well have happened on land as at sea, but which causes a breach of the watertight integrity of the hull, allowing ingress of seawater, can nevertheless give rise to a loss caused by perils of the seas.
46. Again this is well illustrated by Hamilton Fraser & Co v Pandorf & Co, in which the ingress of seawater was the result of the gnawing of the pipe by vermin, which was not itself a peril of the seas (or other excepted peril under the charterparty). The ingress of seawater was nevertheless held to be "dangers and accidents of the seas", which was equated with perils of the seas.
47. This principle was articulated by McLachlin J in giving the judgment of the Supreme Court of Canada in The La Pointe at page 91 when considering an argument that the corrosion of cap screws, which together with a failure to close a valve had allowed ingress of water into the engine room of a vessel, was not a peril "of the sea". She said:

"In the case at the bar the loss resulted from the sinking of the ship due the ingress of sea-water. This loss would not have occurred on land. The requirement that the accident be "of the sea" is therefore met. The respondents argue that the cause of the loss was the corrosion and that this could have occurred anywhere, including on land. But the test is not whether the defect which started the causal chain which led to the loss is one that could occur exclusively at sea, but rather whether the accident itself – in this case the sinking of the ship – is one which could only occur at sea. Many sinkings result from causes which could occur on land……"

48. Mr Jacobs QC relied on a passage in Colinvaux's Law of Insurance 9^th Ed 1^st Supp. (2012) at paragraph 24-132F, which on one reading might be taken as suggesting that the view that the ingress of water need only be caused by a fortuitous peril, rather than a fortuitous sea peril, has been "robustly rejected in England". The cases cited in support generally go no further than confirming that a mere ingress of seawater is not of itself a peril of the sea and that it must be fortuitous. The only English case cited in the text upon which Mr Jacobs QC relied was the decision of Aikens J in Brownsville Holdings Ltd v Adamjee Insurance Co Ltd (The "Milasan") [2000] 2 Lloyd's Rep 458. In that case the claim failed because the ingress of seawater into the yacht in calm water and during good weather was held to be the result of the deliberate scuttling of the vessel. The assured claimed that the loss was proximately caused by the accidental incursion of seawater into the engine room, and thence into the aft accommodation. By the end of the hearing, the assured advanced no positive case as to why water entered the engine room other than a general allegation that it was due to a "fortuity" (see paragraph 5) and focussed its case on the passage of water from there into the aft accommodation as the proximate cause of the loss. At paragraph 15 Aikens J rejected the assured's contention that the policy was akin to an all risks policy which relieved the assured of the need to identify which particular peril operated to cause the loss. At paragraph 16 he said:

"This conclusion creates an immediate difficulty for the claimants, because they have not pleaded any specific facts that they say were both the reason for the initial incursion of water into the engineroom and which would also amount to one of the insured perils, as they acknowledged in par. 1 of their closing submissions. They are therefore forced to resort to two arguments. The first is that the vessel was seaworthy at the time of her loss and the incursion of seawater into the engine room is unexplained, so therefore there is a rebuttable presumption that the cause of the incursion of seawater is "perils of the sea". This argument necessarily presumes that the initial incursion was the proximate cause of the loss. The second argument (and I think their principal one) takes the opposite line. It is that the initial incursion of seawater into the engine room was not the proximate cause of the loss of the vessel; the proximate cause was the incursion of the water into the aft accommodation and the cause of that is both pleaded and, they say, proved. "

49. Aikens J was there saying that if the proximate cause of the loss were the incursion into the engine room, the owners could not establish that that was a peril of the seas merely by asserting that it was fortuitous without pleading and proving specific facts which would make such an incursion fortuitous, and therefore an insured peril, namely perils of the seas. He was not saying that if the owners had established particular facts which rendered the ingress a fortuity, they would have had to have shown the fortuity itself to have been an insured peril. That was not a live issue before him. He was merely explaining how it had come about that if the ingress into the engine room were the sole proximate cause of the loss, since the owners did not attempt to explain the cause of the ingress into the engine room by any particular theory, they could only establish their claim of loss proximately caused by an insured peril by reliance on the principle that there is a rebuttable presumption of a loss by perils of the sea where a seaworthy ship is lost in unexplained circumstances (Anderson v Morice (1875) LR10 CP 58, Lamb Head Shipping Co Ltd v Jennings (The "Marel") [1994] 1 Lloyd's Rep. 626 at 629, col 2).
50. In the casualty with which I am concerned, the fortuitous ingress of seawater was a relevant proximate cause of the loss and there can be little doubt that that was an accident which was peculiar to the maritime nature of the adventure. Whether the crew negligence which started the chain of causation was "of the seas" is not a relevant inquiry.
51. If I were wrong in this conclusion, and the crew negligence would need to qualify as "of the seas", I would hold that it has that character. The crew were carrying out a peculiarly maritime operation in using the emergency fire hose system to de-ice the hatches and deck in preparation for cargo operations. The procedure which ought to have been followed, namely draining the emergency fire pump system, and closing the sea suction valve, was part and parcel of the wider operation of using the fire hose system for cargo operations. The crew's negligence in these respects was a peculiarly maritime failing which could not have occurred on land.
52. The Underwriters contend that a failure to drain a pump and piping system to avoid freezing damage might as easily happen on land as at sea, for example in the water pipes of a domestic dwelling. But the equipment in question, connected to the sea by a sea suction valve, is a peculiarly maritime piece of equipment which has no relevant comparator on land. Moreover, although the causative negligence included a failure to drain a system of water in circumstances where it might freeze, resulting in fracture or bursting, the negligence was not confined to that failure. A critical element was the negligence in leaving open the sea valve. Had it been closed, the only damage would have been to the fire pump casing and the strainer lid. The quantity of water ingress would have been confined to that in the system, which would have been minimal. The negligence in leaving open the sea valve was a failing or accident which is properly described as exclusively or peculiarly maritime in character. It could not have happened on land.
53. This case differs materially from The Inchmaree which involved a donkey engine of a kind which might as easily be used on land as in a vessel; and which was being used in the same way as would have occurred on land. In The Inchmaree, all activity and damage was internal to the vessel itself, so that the theatre within which the accident was played out could realistically be compared with a theatre on land. No such comparison can be made in this case; here there was a breach in the watertight integrity of the hull as a result of the open sea valve and damage to the pump and filter. The negligence was in operating a system which is of a peculiarly maritime nature.
54. The third strand of Mr Jacobs QC's argument was that it is always open to an underwriter under a time policy to show that the loss and damage was caused by an uninsured peril, in this case unseaworthiness at the commencement of the voyage. In my view this argument falls into two errors. It treats unseaworthiness as the proximate cause of the loss in all cases in which it leads to ingress; yet in such cases it is often the fortuitous ingress which is the relevant peril, as I find it to be in this case. As Mustill J said in the Miss Jay Jay at page 271 col 2, in the passage quoted above, it is clearly established that a chain of causation running – (i) initial unseaworthiness; (ii) adverse weather; (iii) loss of watertight integrity of the vessel; (iv) damage to the subject-matter insured – is treated as a loss by perils of the seas, not by unseaworthiness. The Court of Appeal in that case ([1987] 1 Lloyds Rep 32, 37,41) upheld the finding that the ingress was a proximate cause, whilst recognising that in that case unseaworthiness was also a proximate cause and admitting of the possibility that in some cases unseaworthiness might be the sole proximate cause. It has long been established that a vessel's unfitness to encounter foreseeable weather conditions does not prevent the loss being regarded as one by perils of the seas and will not normally do so: Dudgeon v Pembroke (1877) 2 App Cas 284, 295-6; Whittle v Mountain [1921] AC 615; and The Cender Mopu at [67], [68], [72], [77].
55. The second error is to treat ingress caused by unseaworthiness as uninsured under a time policy. This is to ignore the fundamental distinction between a voyage and a time policy so far as concerns the fitness of the vessel (The Cender Mopu at [40]). As Mustill J put it in The Miss Jay Jay at p270 col 2,

"Under a voyage policy, the assured warrants that the vessel will be seaworthy at the commencement of the voyage. If the warranty is broken, any claim in respect of a casualty occurring during the voyage will inevitably fail, without the need for any complex analysis of the nature of a peril of the seas, or of the doctrine of causation. Under a time policy, by contrast, there is no such warranty, either at the inception of the risk or on sailing. It has been established for more than 100 years that unseaworthiness defeats a claim under a time policy only if the assured knew of it at the time when he sent the vessel to sea: Fawcus v Sarsfield (1856) 6 E.& B. 192; Thompson v Hopper (1858) E.B. and E. 1038; Section 39 (5) Marine Insurance Act 1906. "

56. Some, but not all, kinds of unseaworthiness are capable of defeating a time policy. These are where there is a debility or inherent weakness in the vessel which has come about without any external fortuity. In those cases, of which the rotting wooden hulk in Sassoon v Western Assurance [1912] AC 561 is the classic example, there is no peril of the seas because the loss is dissociated from the action of wind and water; the casualty is simply the result of wear and tear. See also Grant Smith & Co v Seattle Construction and Dry Dock Co [1920] AC 162 per Lord Buckmaster at pages 171-172.
57. I detect two themes in the authorities which distinguish "debility", which defeats a claim under a time policy, from other unseaworthiness, which does not. The first is that debility is to be associated with normal wear and tear, whether caused by the sea or otherwise. The second is that debility is to be associated with cases where there has been no more than the ordinary action of wind and waves, as contrasted with some external fortuitous event which has allowed entry of seawater. Thus debility is sometimes paraphrased as inherent weakness, the distinction being between a deficiency which is inherent and that which arises from some external event: see Merchants' Trading Co. v. Universal Marine Insurance Co (The "Golden Fleece"), (1870) 2 Asp. M.L.C. 431, 432; The Cender Mopu at [39], [80], [81]. The case law is summarised and explained by Mustill J in The Miss Jay Jay at 272 col 1:

"Third, as to "debility". Where a ship sinks through its own inherent weakness, there is no loss recoverable under the ordinary form of policy. It is not enough for this purpose that the vessel is unseaworthy. The loss must be disassociated from any peril of wind or water, even if these form the immediate context of the loss, and constitute the immediate agency (for example, the percolation of water through an existing flaw in the hull) by which the loss takes place. As Lord Buckmaster said in Grant, Smith v. Seattle Construction, sup., the policy is not a guarantee that a ship will float. See also Fawcus v. Sarsfield, (1856) 6 E. & B. 192, in relation to the first loss; Merchants' Trading Co. v. Universal Marine Insurance Co., (1870) 2 Asp. M.L.C. 431, the direction of Mr. Justice Lush approved by the Court of Common Pleas; Ballantyne v. MacKinnon, [1896] 2 Q.B., 455; Sassoon v. Western Assurance Co., [1912] AC 561.

Finally, as to the requirement that a loss by perils of the seas shall be "fortuitous". There may be philosophical problems here, possibly compounded by the placing of more weight than it was intended to bear on the apophthegm of Lord Herschell in Wilson, Sons & Co. v. Owners of Cargo per the " Xantho", ((1887) 12 App. Cas. 503 at p. 509) that –

. . . the purpose of the policy is to secure an indemnity against accidents which may happen, not against events which must happen.

There can be few losses of which it can be said that they must happen, in the sense that this accident is bound to happen in this way at this time. Indeed, in some of the leading cases it could hardly have been predicted that the loss was bound to happen at all, whilst the policy was on risk. It is, however, unnecessary to enter into this problem. When the vessel succumbs to debility, the claim fails, not because the loss is quite unattended by fortuity, but because it cannot be ascribed to the fortuitous action of the wind and waves. A decrepit ship might sink in perfect weather tomorrow, or it might not sink for six months. To this extent a loss tomorrow is not inevitable. But if the ship does sink, there is no external fortuitous event which brings it about. In respect of such losses, the ordinary marine policy does not provide a remedy."

58. Mr Jacobs QC relied on a passage in the speech of Viscount Finlay in Samuel v Dumas at page 455:

"The view that the proximate cause of the loss when the Vessel has been scuttled is the inrush of the sea water, and that this is a peril of the sea, is inconsistent with the well-established rule that it is always open to the underwriter on a time policy to show that the loss arose not from perils of the seas but from the unseaworthy condition in which the Vessel sailed: see Arnould on Marine Insurance, s. 799. When the Vessel is unseaworthy and the water consequently gets into the Vessel and sinks her, it would never be said that the loss was due to the perils of the sea. It is true that the Vessel sank in consequence of the inrush of water, but this inrush was due simply to the unseaworthiness. The unseaworthiness was the proximate cause of the loss. Exactly the same reasoning applies to the case of scuttling; the hole is there made in order to let in the water. The water comes in and the Vessel sinks. The proximate cause of the loss is the scuttling, as in the other case the unseaworthiness. The entrance of the water cannot be divorced from the act which occasioned it."

59. I read the reference to "the well-established rule that it is always open to the underwriter on a time policy to show that the loss arose not from perils of the seas but from the unseaworthy condition in which the Vessel sailed" as referring only to unseaworthiness in the sense of a debility in the vessel. The passage to that effect in the then current edition of Arnould (10^th Ed) supports the proposition by a footnote reference to the debility cases identified in the passage from Mustill J's judgment in The Miss Jay Jay which I have quoted above. Samuel v Dumas was a case of intentional ingress of seawater, and Viscount Finlay was drawing an analogy with an inevitable ingress of seawater as a result of debilitative unseaworthiness. In each case there is no entry of seawater attended by some external fortuitous cause, so that the ingress cannot be a peril of the seas as the proximate cause of the loss. Lord Sumner was using the expression unseaworthiness in the same sense when he said at page 468:

" …. So it is in cases on time policies, where the loss is directly caused by unseaworthiness, for then it is plain that the loss was a certainty, whatever the state of the weather or the sea, and, as has been often said, perils of the sea refer to things that may happen, not to things which must happen in the ordinary course of navigation ….."

60. The fourth strand to Mr Jacobs QC's argument is that the ingress was not a fortuity because the damage to the fire pump housing and filter lid had already occurred before the Vessel left Klaipeda; the ingress was therefore bound to occur once the ice melted, which was inevitable, and the ingress was bound to occur during the voyage even in the most benign conditions. At the time of commencement of the voyage the ingress of water was an inevitable certainty. He argues that the so called "debility" cases are not confined to vessels which have been rendered generally disabled but apply in principle to inevitable loss and damage caused by particular pre-sailing unseaworthiness.
61. Attractively though this point was put, I cannot accept it. I have already explained that the debility cases are cases where the loss arises from the inherent weakness of the vessel due to wear and tear. They are cases where the defect in the vessel's hull does not arise from some external fortuitous event. By contrast, the ingress to the Vessel's bowthruster room did not arise from wear and tear, or any weakness which was inherent, but from an external fortuitous accident, namely the crew's negligence in relation to the emergency fire pumping system. It is a false analogy, therefore, to treat the lack of watertightness of the Vessel at Klaipeda after the crew negligence as a particular instance of the kind of debilitative unseaworthiness which renders a casualty other than fortuitous.
62. It is true that there came a point of time at which the future ingress became a certainty, namely once water had frozen and damaged the fire pump housing and filter cap. But that does not prevent the casualty having been proximately caused by a fortuity. It is clear from the authorities that there may be a loss by perils of the seas under a time policy notwithstanding that the crew negligence which has given rise to the entry of seawater has rendered the vessel unseaworthy at some moment prior to the ingress, and that the entry of seawater thereafter is inevitable. That was the case at the loading port in Davidson v Burnand. That would be the case in the examples given in the Stranno and Canada Rice Mills of an open cowl or port-hole. As a matter of principle I see no reason why it should matter that the relevant moment occurred before the vessel left Klaipeda. This is a time policy not a voyage policy, and the point of time at which the ingress of water became a certainty was after the commencement of the period of cover. In general, the point of time by reference to which any peril under a policy of insurance must be uncertain is the commencement of the policy: an insurer insures risks which may happen during the policy period, not risks which must happen during that period.
63. There is, in my judgment, no logical reason for focussing on the commencement of the voyage in this context. Indeed there may be difficulties in identifying what is meant by the commencement of the voyage. Mr Jacobs QC asserted that what it meant in this case was on sailing from Klaipeda. But the policy responds to losses by perils of the sea in port: if the weather had warmed up in Klaipeda and the ingress of water occurred there, that would surely have been an ingress constituting a peril of the seas, as much as the loss in Davidson v Burnand. A time policy provides continuous cover on and between voyages, and indeed when the vessel might not be said to be on a voyage at all, such as when undergoing repairs or maintenance. If, contrary to my view, it were necessary to seek to identify the moment of commencement of a voyage, which would have to be a concept which included periods in port, why should not the "voyage" in this case be regarded as commencing at least as early as when preparations for cargo operations were being made? One could multiply examples which would lead to surprising distinctions if there were a critical moment of commencement of the voyage in time policies.
64. In this context, Mr Jacobs QC placed particular reliance on Ballantyne v Mackinnon [1896] 2 QB 455. In that case the vessel was insured under a time policy against perils of the seas. The crew failed to load sufficient coal for a voyage from Hamburg to Sunderland. As a result, the vessel ran out of coal before it reached Sunderland. She was proceeding under reduced steam and sail, and was not in any danger, but she sought a tow into Sunderland and claimed the cost as salvage under the hull insurance. The action was tried by Lord Russell of Killowen C.J., who found for the underwriters. The claim failed before him, and in the Court of the Appeal, on the simple ground that the ship had not faced any peril of the seas. She suffered no casualty or danger and could have continued her voyage into Sunderland under sail without any danger. The passage relied on by Mr Jacobs QC is in the judgment of AL Smith LJ giving the judgment of the Court of Appeal at page 461:

"… the loss sustained was not occasioned by a peril of the sea, for in our judgment the loss complained of arose solely by reason of the inherent vice of the subject-matter insured : we mean the insufficiency of coal with which the ship started upon her voyage, the consequence of which was that what in fact did happen must have happened, namely that the ship ran short of coal, no sea peril bringing this about in any shape or way, or placing the ship in a position of danger thereby."

AL Smith LJ was not, however, referring to a loss or casualty which was bound to happen as a result of the condition of the vessel at the commencement of the voyage. He was merely referring to the fact that running out of coal was a certainty and was not itself a peril of the seas, nor did it occasion a peril of the seas. The decision of the Lord Chief Justice, upheld on appeal, was simply that there had been no sea peril which had afflicted the vessel; there was no weather or sea which endangered her, there was no accident or casualty, and no incursion of water into the ship; she could have completed the voyage under sail with no reasonable apprehension of danger (see 460).

65. In support of this argument Mr Jacobs QC referred to passages in Arnould - The Law of Marine Insurance (18^th Ed.) at paras 22-31 and 23-16, in Marine Insurance: Law and Practice at para.14.24 and in The Vergina at [100]. But none of these was addressing the point currently under consideration. He relied also on the passage in the speech of Viscount Finlay in Samuel v Dumas (supra) at 455 "it is always open to the underwriter on a time policy to show that the loss arose not from perils of the seas but from the unseaworthy condition in which the vessel sailed" (Mr Jacobs' emphasis). But that was a reference to cases of debility, properly so called. In those cases, the time when the vessel sets out from port on the voyage during which the casualty occurs is an obvious and appropriate point of time at which to consider the condition of the vessel for the purposes of determining whether the casualty is due to her inherent weakness or an external fortuitous event. In Samuel v Dumas no issue arose as to whether the moment of the sailing of the vessel was the relevant point of time as to when fortuity arose in cases where there was an external fortuitous cause, nor whether it would have been sufficient if the occurrence was a fortuity by reference to an earlier point of time.
66. Mr Jacobs QC also relied on the rebuttable presumption that a vessel is lost by perils of the seas if the owners can show that she commenced the voyage in a seaworthy condition and the loss is unexplained. This principle, going back to Anderson v Morice (1875) LR10 CP 58 was recently expressed in these terms by Dillon LJ in Lamb Head Shipping Co Ltd v Jennings (The "Marel") [1994] 1 Lloyd's Rep. 626 at 629, col 2:

"If it is known that a ship was seaworthy when she set out, and she has never been seen since and nothing has been heard of her crew, then on the balance of probabilities she must have sunk and, on the balance of probabilities, the sinking must have been due to "perils of the sea" because she was seaworthy when she set out But if it was not shown that the ship was seaworthy when she left on her last voyage, the presumption does not apply since it cannot be held on the balance of probabilities that her presumed sinking was due to perils of the sea rather than to her unseaworthy condition."

67. This principle merely takes the time of sailing as the presumed last moment at which the condition of the vessel is ascertainable as a matter of evidence. But if information were available as to her condition at some subsequent intermediate point on the voyage which bore on whether she was seaworthy at the time of her unexplained disappearance later in the course of the same voyage, that intermediate time would be the relevant moment by reference to which to address the issue of seaworthiness for the purposes of the rebuttable presumption. The reference to the time of sailing is no more than an evidential reference, not one addressed to the moment at which fortuity falls to be judged.
68. I am concerned with a casualty in which the chain of causation was (i) an external fortuitous event (crew negligence) (ii) unseaworthiness (but not debility) (iii) the effect of the sea on the vessel (which was not the ordinary action of wind and waves: the wind and waves were not extraordinary but their action on the vessel was) (iv) ingress of seawater (v) damage to the subject matter insured. The casualty was a fortuity because the crew negligence was fortuitous. The fortuitous ingress of seawater was a peril of the seas, which was a proximate cause of the loss and damage.
69. For these reasons I conclude that the loss was caused by perils of the seas. This conclusion renders Issues 2 to 6 immaterial; but I will address them in turn in case I be wrong in my conclusions on Issue 1.

Issue 2: Did the crew negligence in failing to drain down the emergency fire pump and close its suction valve result from the want of due diligence by the assured in failing to promulgate appropriate cold weather procedures?

70. In the proviso to the Inchmaree clauses, a want of due diligence is to be equated with a lack of reasonable care. The test is whether the assured was negligent in the relevant respect. The Underwriters bear the burden of proving the two requisite elements, namely that the assured was negligent and that such negligence was causative of the loss: see The Toisa Pisces [2012] 1 Lloyd's Rep. 252.

Management

71. The Owners bought the Vessel in late 2006, since when she has been managed by Rederij Chr. Kornet & Zonen B.V. ("K&Z"), a Dutch family business which has a 25% beneficial ownership in the Vessel, with offices in Werkendam, Holland. At the time of the casualty the Vessel was one of eight vessels managed by K&Z, operating in the northern European coastal trade. Four of them were newbuildings delivered in 2008 and 2009. The vessels were typically manned by 5 or 6 crew members. They were in and out of port frequently and the workload for the crew was high. Although some maintenance was carried out by the crew, a good deal of the work of inspection and maintenance was carried out by subcontractors. This was particularly true on the engineering side because there was typically only one engineer on board. Most maintenance on the main engines, and all electrical work, was undertaken by outside contractors.
72. The four directors of K&Z at the time were three brothers, Chris Kornet, Gertjan Kornet and Dolf Kornet, together with their brother in law, Kees Parel.
73. Chris Kornet, now aged 48, was the general manager of the company from 2005 or 2006 and the director responsible for administration and finance. He had worked for the company for over 25 years, for 14 of which he had sailed as a master, before coming ashore in 2001. When he first came ashore he acted as a superintendent, visiting the ships about once every 3 months, until he became general manager.
74. Gertjan Kornet was the director responsible for the technical department. He was aged 29 at the date of the casualty. He holds a master's certificate and sailed as captain from July 2003 until January 2007, when he came ashore to act as a superintendent. He was one of two superintendents, each of whom was responsible for half of the fleet vessels. His responsibilities included reviewing the monthly reports and paperwork from the vessels, and visiting the ships when convenient, depending on their trading pattern. His preference was to visit the vessels whenever they came to a German or Dutch port, which typically occurred once every month or 6 weeks. He had last inspected the Vessel on 12 January 2010, about a fortnight before the casualty.
75. Dolf Kornet was the director responsible for crewing.
76. Kees Parel was the director responsible for dealing with safety management issues. He was 41 at the time of the casualty and had served as a master for a total of 11 years, all on K&Z vessels. He came ashore in 2006 to take over responsibility for safety management. He was responsible for writing the safety management system and procedures in force at the time of the casualty. He visited the fleet about 3 or 4 times a year to look at their safety management systems and conducted an annual audit resulting in a written report. The DC Merwestone was one of the vessels on which he had served, being its first captain when bought by the new owners in late 2006.
77. It was common ground in this case that the relevant persons whose conduct and knowledge was to be treated as that of the assured, Owners or managers included each of Chris Kornet, Gertjan Kornet and Kees Parel.

Cold weather procedures and checklists

78. There is no applicable Classification Society or statutory provision which contains a requirement for an owner to have cold weather procedures. Draining the system and closing the sea valve in freezing conditions is a matter of basic good seamanship which the engineer and senior deck officers would be expected to know and understand. The Engineer's evidence to the Dutch Shipping Inspectorate was that he normally closed the valves but that at Klaipeda he forgot to do so. By contrast, in his interview with the Underwriters' solicitor he said that the emergency fire pump suction was always left open. This too was the evidence of Captain Lilipaly at the trial.
79. At the time of the incident, the 2002 edition of the International Management Code for the Safe Operation of Ships and for Pollution Prevention ("the ISM Code") applied to the Vessel. The ISM Code required the managers to develop, implement and maintain a Safety Management System ("SMS"). There is no fixed format for an SMS and it is normal for systems created and implemented by owners or managers to be noticeably different from each other. K&Z had a formally documented SMS which was applicable to their fleet and to the Vessel. Its content and implementation were audited by Bureau Veritas. Bureau Veritas had reviewed and approved the SMS and had carried out a detailed ISM audit on 27 February 2009. Bureau Veritas had issued the Vessel with a Safety Management Certificate valid until 2 April 2012 and the managers with an ISM Document of Compliance valid until 4 December 2010.
80. Although the SMS itself contained no cold weather procedure document, on 19 December 2009, a little over a month before the casualty, Kees Parel sent an email to the Vessel and all other vessels in the fleet with an instruction in the following terms:

"Please make sure that all waterlines are drained and do not contain any water. Also make sure that temperature in the bow thruster room is kept above zero degrees Celsius. This in order to prevent frozen lines."

81. Underwriters contend that this was insufficient and that it was negligent for the SMS not to contain both (1) instructions concerning cold weather procedures and (2) a checklist to be completed whenever cold weather was encountered, so as to encourage compliance with such procedures.
82. The experts had different experiences and different views on this subject. The Owners' expert, Mr Duncan, testified that draining pipes and pumps was a matter of common sense and good seamanship, and that his experience was that some safety management systems have cold weather procedures but others do not. The Underwriters' expert, Mr McFarlane, testified that to the best of his recollection the safety management systems which he had seen in recent years for ships trading in Northern European areas had all had cold weather procedures, and that an SMS should include such a procedure. He expressed surprise at the absence of any requirement by Bureau Veritas that there be cold weather procedures in the SMS for these Owners. When confronted with the fact that Bureau Veritas had approved the SMS without any cold weather procedures despite knowing where the Vessel traded, Mr McFarlane accepted that "two reasonable minds can differ about this".
83. The DSI report into the incident found at paragraph 6.1.1 that " …. Usually when sailing in freezing conditions procedures or checklists are used to prevent damage by freezing" and included within its recommendations to the managers of the Vessel that they include the draining of pumps and pipes in freezing conditions in their working procedures. But the Inspectorate did not recommend that any requirement of this kind be imposed by the Dutch Flag state, and none has been. Its observation that such procedures were "usual" was not expressed to be a criticism of the Owners.
84. Following the publication of the DSI report, the managers implemented the DSI's recommendations by issuing a circular containing a new cold weather procedure and attaching two new checklists. The Underwriters contend that this was an acknowledgment by the Owners that such procedures and checklists should have been in their SMS prior to the casualty. It was not; it was simply a response to the DSI recommendation. I would associate myself with the sentiments of Lord Bramwell in Hart v Lancashire and Yorkshire Railway Company (1869) 21 L.T. 261, 263 that "…....people do not furnish evidence against themselves simply by adopting a new plan in order to prevent the recurrence of an accident. I think that a proposition to the contrary would be barbarous. It would be…to hold that, because the world gets wiser as it gets older, therefore it was foolish before."
85. In those circumstances I am unable to conclude that the managers were negligent in failing to have any formal cold weather procedures in their SMS. Whilst best practice might have involved some instruction or reminder to the crew to drain lines to avoid freezing in cold weather, the Underwriters have not established that such instructions were the universal practice amongst competent owners and managers. I attach particular significance to the fact that they were not required by Bureau Veritas, who were in effect vetting the sufficiency of the SMS of these managers. Kees Parel could reasonably have expected Bureau Veritas to tell him if BV considered that the SMS was deficient in not containing a cold weather procedure. Moreover the managers could reasonably have assumed that their crew knew about this basic aspect of good seamanship: their vessels had been trading to Northern European ports in winter for years and must have encountered freezing conditions not infrequently; but there had not apparently been any previous incident of a similar kind (I reject the Underwriters' suggestion, denied by Owners, that it can be inferred from a single photograph of another fire pump that a similar incident had occurred before).
86. In any event, I would hold that the sending of the 19 December email was sufficient to bring the matter to the attention of the crew on the Vessel. The content of the cold weather instructions which the Underwriters allege should have been included in the SMS was not substantially different (so far as relevant to this incident) from the wording of the 19 December email (although it was common ground that on this Vessel, heating the bowthruster room to prevent freezing was not commercially practicable so that such reference was inapplicable to this particular Vessel). Mr McFarlane accepted that if there were a suitable instruction in place, it would not be necessary to have a checklist in place as well. The question therefore becomes simply whether it was negligent to give the instruction by email rather than by way of a formal procedure within the SMS. Email was a common form of communication with vessels in this fleet in relation to operational matters, and there is no reason to think that the crew were less likely to take account of a short and specific email than of a lengthier protocol which was one of a large number of procedures in the formal SMS documentation. Indeed the reverse may well be true.
87. For these reasons I find that there was no want of due diligence on the part of the Owners or managers in relation to cold weather procedures or checklists. Accordingly, subject to Issue 4, the Owners have established that the loss was proximately caused by crew negligence which was an insured peril under ITC Clause 6.2.3 and IAPC Clause 1.2.

Issue 3 (a): Was the loss proximately caused by the negligence of persons other than the Owners or managers in failing to seal the cable duct at each end of the duct keel tunnel?

88. In my judgment, the answer to this question is yes, in relation to the damage other than that in the bowthruster room itself. The starting point is that the deficient state of the bulkheads at frames 24 and 112 was a proximate cause of the catastrophic damage in the engine room and the debilitating casualty. A business or seafaring man would say that the ingress into bowthruster room, and the negligence which caused it, was not the only efficient cause of the casualty. A relatively slow leak into what should be a watertight compartment in the bow would not be expected to cause catastrophic engine room damage. The lack of watertight integrity between the bowthruster space and the engine room was a real and efficient cause of the casualty, of equal efficiency with the fortuitous ingress into the bowthruster room. Without it, there would have been no damage to the main engine and no debilitating casualty.
89. It follows that if the deficient state of the bulkheads came about as a result of someone's negligence, that negligence is itself a proximate cause of the casualty. If one asks how it came about that the cables were not surrounded by sealant packing in the glands, as it was common ground they should have been, the answer must be either that those responsible for the modification in 2001 failed to install such packing, or that it was subsequently removed. The Underwriters contend that the Owners cannot prove that there was negligence of any person other than the Owners or managers because they cannot identify when and by whom the cable glands were not sealed. But the Underwriters advanced no coherent case that sealing had been removed at any stage by the current Owners or managers. There was no suggestion advanced as to when or why they might have done so. The Underwriters disavowed any suggestion that it might have washed away in the casualty or been displaced by some other accidental cause. The experts agreed that there was no evidence in the photographs taken after the casualty that any packing had been removed. I find that the most likely explanation is that it was never installed at the time of the modifications. It is true that the watertight integrity of the bulkheads should have been checked by Class at survey at the conclusion of the works in 2001 and at Special Survey in 2006; and that according to the Special Survey Report dated 22 April 2006, the duct keel had been examined to the satisfaction of the attending Lloyd's Register surveyor. I would readily accept, however, that such surveys failed to detect the lack of packing in the cable glands at each end. The experts agreed that it was not possible to see that the cable duct at either end was unsealed unless that was being specifically looked for.
90. This was a deficiency in the structure of the vessel as a result of the errors of those who carried out the modifications in 2001, which can be characterised as contractors' negligence. It therefore falls within the scope of IAPC Clause 1.2 unless it can be said that the exercise of due diligence by the Owners or managers thereafter ought to have led to its discovery prior to the casualty.

Issue 3(b): Did the deficient state of the bulkheads at the time of the casualty result from want of due diligence by the assured?

91. The cable duct between the bowthruster compartment and the duct keel tunnel at frame 112 was hidden beneath the floor plates, and was in a location that could only be seen if the floor plates were removed, and the head of the surveyor was brought down level with the penetration. The surveyor would then have had to direct a torch through the opening to see if the opening was sealed further back inside the duct. The lack of packing in the forward bulkhead at frame 112 would not have been obvious looking from the bowthruster room itself or from within the tunnel.
92. At the aft end, the cable gland was sealed where it emerged into the engine room and it would have been a reasonable assumption of any surveyor that the L pipe was continuous, such that there would have been watertight integrity, whether or not sealed also at frame 24 itself. The gaps in the L pipe in the unsealed cable duct between the tunnel and the engine room were in a location under the floor plating of the engine room which was very difficult to access and would not have been seen during normal operations or surveys. The gaps were only discovered after the incident by cutting away the floor plates. The experts agreed that it was not possible to see whether the cable glands were unsealed at either end unless that was being specifically looked for; and that during a typical superintendent inspection one would not expect a close and thorough examination of the water tight bulkheads and their penetrations: in passing through the space, a superintendent would merely assess, so far as practically possible, the overall condition of all readily accessible bulkheads from above the floor plates; he would not be expected to request tank entry or to enter the duct keel to inspect bulkheads from the other side.
93. Accordingly, absent some reason to make a specific inspection of the cable glands, the Owners could not reasonably have been expected to discover the deficiencies prior to the casualty.
94. The Underwriters contend that there were two previous incidents which cast doubt on the integrity of the bulkheads and which should have led to the Owners inspecting them with sufficient intensity to discover the deficiencies. In July 2007, 45m³ of oily water was found in the duct keel and engine room bilges whilst the vessel was in dry dock at the Bolnes Shipyard. This was recorded in Gertjan Kornet's contemporaneous report in the following terms (in translation); "Removed about 45 cbm contaminated oil/water from the shaft alley (last of Koopman's tricks) and the bilge (BV item). Shaft alley does not have a leak from fore and aft bilge". It is not clear how much of this 45 m³ was in the duct keel and how much in the engine room bilges. Captain Lilipaly's evidence was that he had discovered the oily water in the duct keel by entering through the manhole in the hold. The reference to "Koopman's tricks" in Gertjan Kornet's report was explained by him as follows. Koopman was the previous owner of the Vessel and there was a dispute on the Friday before handover about oily water in the engine room bilges which the buyers wanted removed. At handover on the Monday, it was no longer in the engine room bilges, and it was assumed that Koopman had removed it from the Vessel. When oily water was found in the duct keel some 8 months later at dry dock, Gertjan Kornet's conclusion was that Koopman had not removed it prior to handover of the vessel but simply pumped it into the duct keel.
95. The Underwriters suggested that it would have been negligent to treat this as a credible explanation because the crew would have sounded the duct keel in the meantime on at least a few occasions if not daily or weekly, and the presence of such oily water would not have remained undetected in the 8 months or so since the purchase. But as Gertjan Kornet explained, he did not know whether or how often the void spaces had been sounded, and there does not appear to have been a regular practice of sounding the duct keel tunnel. In any event, without knowing the quantity of such oily water found in July 2007 in the duct keel itself, it is impossible to tell whether the sounding would have given a reading which caused any concern, given that the quantity would have been distributed over the length of the tunnel subject to the Vessel's trim.
96. I am not persuaded that he was incorrect in the conclusion which he reached as to the source of the water in the duct keel, still less that he was negligent in reaching that conclusion. There is no reason to think that the oily water got in from the bowthruster room, there being no evidence suggesting sufficient water there to have that consequence; and the obvious source for such oily water was the engine room bilges, rather than the bowthruster room, because it would be unlikely (although not impossible) for the bowthruster room to contain water which was mixed with oil. The water cannot realistically have got in from the engine room through the deficiency in the L pipe because that would have required water in the engine room bilges to a very high level, just under the floor plates, which would not have gone unnoticed and which would have triggered the bilge alarm. If there were some other breach in the watertight integrity of the bulkhead at frame 24, such as in the manhole cover, the problem would have recurred, whereas there is no suggestion that it did (prior to 2009 when a leaking manhole cover was discovered). On the evidence before me Gertjan Kornet's conclusion that it was "Koopman's tricks" seems the most likely explanation.
97. The Underwriters' expert Mr Tanner accepted that if Gertjan Kornet were convinced that the water had been put in there on purpose by Koopman, he would not have been acting imprudently in failing to carry out any further investigation. I find that there was no negligence in failing to carry out the sort of comprehensive inspection of the bulkheads which might have revealed the cable gland deficiencies. Moreover had Gertjan Kornet carried out any further inspection, he could not reasonably have been expected to discover the cable gland defects. It would have been reasonable to confine any such inspection to the engine room end of the duct keel, and he would have been reasonably entitled to assume that the cable duct was sealed, given (a) the presence of the obvious sealing where the cables emerged from the L pipe in the engine room, (b) the reasonable assumption that the L pipe was continuous and (c) the extreme improbability of the cable duct being the source of ingress from the engine room because of its height above the shell plating. The deficiencies in the L pipe were only discovered after the casualty by cutting away floor plates in the engine room, which it would not have been reasonable to expect as part of an investigation into the presence of some oily water at Bolnes.
98. More oily water was discovered in the duct keel in October 2009. The oil record book records 200kg of slops removed from the engine room bilges and duct keel and a further 10.93m³ of liquid removed from the duct keel whilst the vessel was at the Hardinxveld shipyard. This was "washwater" as a result of cleaning, including steam cleaning in the duct keel, but it suggested that some water had got from the engine room into the duct keel through the bulkhead. Gertjan Kornet's evidence, which I accept, was that the aft manhole was hose tested, found to be leaking, and resealed by replacement of the gasket. He did not check for any other source of leakage. The Underwriters suggested that he was negligent in failing to check for any other source of the leak, but this suggestion was convincingly rejected by the Owners' expert Mr Barker, and was not really supported by their own expert, Mr Tanner. It was reasonable to treat the defective manhole cover gasket as the obvious source of the leak and to look no further. There was no negligence in failing to carry out the sort of comprehensive inspection of the bulkheads which might have revealed the cable gland deficiencies.
99. For these reasons I find that there was no want of due diligence on the part of the Owners or managers in relation to the defective cable duct glands. Accordingly, subject to Issue 4, the Owners have established that the loss (other than that in the bowthruster room) was proximately caused by contractors' negligence, which was an insured peril under IAPC Clause 1.2.

Issue 4: How are the Inchmaree Clauses to be construed? In particular does the want of due diligence proviso apply (a) to causes of loss and damage which are not proximate causes and/or (b) to causes of loss and damage other than the named peril which constitutes the insured peril upon which the assured relies?

100. ITC Clause 6.2 is significantly different from IAPC Clauses 1.2 and 3. ITC Clause 6.2 is concerned with a number of specifically identified perils: crew negligence, barratry, bursting of boilers, accidents in cargo or bunkering operations etc; whereas IAPC Clause 1.2 is wider, including cover simply for "any accident", as well as for the negligence, incompetence or error of judgment of any person whatsoever. However the parties did not address the construction question on the footing that there was a material difference between the clauses, and I will assume that the proviso falls to be construed in the same way in both. I was not referred to any authority which might assist on the question of construction.
101. I have little hesitation in concluding that the proviso is confined to causes of loss and damage which are proximate causes. The words "resulted from" are synonymous with "been caused by", and the latter expression was used interchangeably with the former in the Underwriters' submissions. Proximate cause is the theory of causation applied to insured perils in marine insurance; it would be surprising if standard clauses dealing with insured perils were intended to introduce some other theory of causation, such as "but for" causation.
102. I also conclude that the proviso does not apply to proximate causes other than the named peril which the assured establishes as providing cover under the clause. ITC Clause 6.2 is concerned with a number of separate specified perils and the proviso must be construed in that context. It is a qualification of the cover granted in relation to the named perils. The clause provides extended cover over and above that which would exist for the other insured perils identified in ITC Clause 6.1. The proviso in ITC Clause 6.2 is, therefore, most naturally read as confined to want of due diligence in relation to the specific insured perils with which the clause is concerned. If it is confined to those perils, it is most naturally to be read as confined to the particular peril upon which the assured relies as affording cover, in this case crew negligence. The effect of the Underwriters' argument is to treat the proviso as if, in the context of the clause, it were a freestanding exception of cover wherever loss or damage is caused by want of due diligence on the part of the assured, owners or managers. Clear words would be needed to achieve that result because "it is a basic principle of the law of insurance that negligence by the assured which causes a loss insured under the policy is no defence to the insurer in the absence of some special provision to the contrary": per Kerr J in Global Tankers Inc v Amercoat Europa NV [1977] 1 Lloyd's Rep 61, 66. The proviso is not expressed as a general exception from cover; on the contrary, it is expressed as a qualification to the cover for the specified insured perils.
103. The Underwriters contend that their construction is supported by the proviso being expressed to apply to loss and damage, as opposed to the enumerated causes of the loss or damage. They contend that the words used in the proviso do not refer back to the acts or occurrences enumerated in Clauses 6.2.1 to 6.2.5 but to the loss and damage caused by such acts or occurrences. In my view this is to mistake the operative antecedent expression to which "such" refers. The words "such loss and damage" mean not simply the loss and damage for which the claim is made, but refer back to the whole expression "loss of or damage to the subject-matter insured caused by [the enumerated perils]".
104. The Underwriters' construction would have surprising consequences. Where the two proximate causes were a Clause 6.2 peril and an unrelated cause resulting from want of due diligence by the assured, owners or managers, the proviso would exclude cover. But the same would not apply if a proximate cause were a Clause 6.1 peril. The proviso would not have any application if, for example, one of the proximate causes were perils of the seas, as the Underwriters accept. This has no logical or commercial justification, and is not supported by the historical development of the clauses: see Arnould - The Law of Marine Insurance (18^th Ed) at 23-04, 23-05.

Issue 5: Did the loss result from want of due diligence by the Owners or managers in relation to the bilge alarms?

105. On my previous findings this issue does not arise because (1) there is cover under ITC Clause 6.1 for a loss by perils of the seas, and (2) there is cover for crew negligence and contractors' negligence under the Inchmaree clauses, to which these questions are irrelevant. I state my findings briefly in case I am wrong in both conclusions.
106. There was a forward bilge alarm which was situated at the aft end of the bowthruster space beside the manhole cover into the duct keel. It is now common ground that it did not go off prior to the casualty. The alarm float was of a type used to stop and start submersible pumps. Following the casualty it was tested and found to be working by the Dutch Shipping Inspectorate. This would have been by a bucket test which tested the buoyancy of the float and the effectiveness of the activation mechanism; but since the vessel was without power, and had suffered the flooding in the engine room, it would not have been possible to test whether the electrical connections were in order and so whether it would have sounded the alarm in the engine room. The evidence of the Engineer to Mr Billowes, the Admiralty Manager in the firm of Ince & Co LLP instructed by the Underwriters, when interviewed shortly after the casualty, was that he tested the bilge alarms every two weeks, with the last test having been carried out in the second week of January, when the alarms were found to be working. Although he was not called as a witness, there is no substantial ground for doubting this evidence. The alarms were inspected and tested by Class every year and passed each year. The most likely explanation for the forward bilge alarm failing to sound prior to the casualty is that it was working, but that the float was situated too high by comparison with the open cable gland. The forward bilge alarm appears to have had an activation level between 77 and 84 cms above the floor plating which was about the same height as the top of the cable gland. Accordingly the water could have flowed into the duct keel tunnel without reaching a sufficient height in the bowthruster room to activate the alarm.
107. The failure of the forward bilge alarm to sound does not assist the Underwriters for two reasons. First it was not a proximate cause of the loss. But for the deficiency in the watertight bulkhead, the alarm would have sounded; it was the open cable gland, however, not the absence of a forward bilge alarm, which was the real and efficient cause of the water being allowed to continue its ingress and reach the engine room via the duct keel tunnel. Secondly, there was no want of due diligence on the part of the owners in relation to the alarm. It was common ground between the experts that SOLAS does not provide for any minimum height for the bilge alarm in the bowthruster space. Had there been a watertight bulkhead, as the owners and crew were entitled to expect, there would have been nothing dangerous or untoward about the amount of water which would have to be in the bowthruster room before the alarm was triggered. The experts agreed that a superintendent would not be expected to measure or check the heights of the alarms or their type. No adverse comment had been made about the height of the alarm by any of the Class surveyors who had carried out inspections since 2001. In these circumstances the Underwriters' expert, Mr Tanner, agreed that Owners were not negligent in relation to the height of the alarm. The Underwriters' case on want of due diligence was focussed on an allegation that there were inadequate systems to ensure that the bilge alarms were working. In relation to the forward bilge alarm, this allegation falls away with my finding that it was working.
108. As for the bilge alarm in the engine room, it is not necessary to rehearse the evidence touching on the dispute as to whether it sounded, including detailed calculations and submissions on rates of ingress, and engine room heights and volumes. It too was tested by the Dutch Shipping Inspectorate shortly after the casualty and found to be working. My conclusion on the evidence is that it did sound shortly before 2100, although the condition of the float may have meant that the water was above the level of the alarm, 95mm above the bottom shell plating, before the float achieved sufficient buoyancy to activate the alarm. But even if I had concluded that it had not been working, such conclusion would not have assisted the Underwriters because again its failure to sound would not have been a proximate cause of the loss. The highest the Underwriters were able to put their case was that had it sounded, it would have alerted the crew to the presence of water in the engine room at a stage a little earlier than when the Engineer observed the water shortly before 2100. But the Underwriters were unable to contend that in those circumstances the crew would have been in a better position to deal with the ingress or would have acted differently. As it is, the allegation that there was a want of due diligence in relation to testing that alarms were working also falls away with my finding that the alarm went off.

Issue 6: s. 39(5) Was the loss caused by the unseaworthiness of the vessel to which the Owners were privy?

109. I have not overlooked the Underwriters' allegation that the Owners' want of due diligence in relation to the engine room pumping system engages the proviso to the Inchmaree clauses, although on my previous findings this issue does not arise. It is convenient to address it as Issue 7 after first addressing the Underwriters' defence under s. 39(5), which is also concerned with the engine room pumping system and alleges privity against the Owners.

The law

110. Section 39(5) Marine Insurance Act 1906 provides that where, with the privity of the assured, the ship is sent to sea in an unseaworthy state, the insurer is not liable for any loss attributable to unseaworthiness. The "privity of the assured" means with the assured's personal knowledge and consent. The assured must (i) know the facts constituting unseaworthiness and (ii) realise that those facts render the ship unseaworthy. Knowledge for the purposes of section 39(5) includes "blind eye knowledge", consisting of a suspicion that the vessel might be unseaworthy combined with a conscious decision not to inquire for fear of confirming that suspicion. See Manifest Shipping Co Ltd v Uni-Polaris Insurance Co Ltd (The Star Sea) [2003] 1 AC 469 at [3], [23]-[26], [113]-[116]; The Eurysthenes [1976] 2 Lloyd's Rep 171, at 179, 184, 188.
111. It is common ground that the state of mind of each of Chris Kornet, Gertjan Kornet and Kees Parel is to be treated as the knowledge of the Owners for these purposes.

The pumps and piping system

112. There were three relevant pumps available in the engine room to deal with the ingress, namely a port and starboard general service pump ("GS pumps") and a ballast pump. There was also a main engine seawater cooling pump which was capable of being used as an emergency bilge pump through its own emergency bilge suction, but throughout the casualty this was being used to service the main engine, and no criticism was made of the crew's failure to use it to deal with the ingress.
113. The GS pumps were electrically driven and each had a rating of 74 m³ per hour against a head of 20m. They were generally used to pump ballast, pressurise the fire main and deck wash system and to drive the eductor, rather than to pump the bilges, but they could be used to pump the bilges in an emergency through two suctions, one in the forward part of the engine room and one in the aft. When doing so, the Starboard GS pump, drawing through the starboard manifold, would normally be connected to the aft bilge suction, and the Port GS pump, connected through the port manifold, would draw through the forward bilge suction. This was the arrangement used by the crew on the night in question. The piping system also allowed for each GS pump to be connected by crossover lines to draw on the other bilge suction. On the night the crew did not attempt to use the crossover to use the Port GS pump to draw on the aft bilge suction or vice versa.
114. The ballast pump was hydraulically driven and so could work when submerged under water. Its power pack was located in the generator room one deck above the engine room floor plating. It had a rating of either 250 or 300 m³ per hour (the evidence was unclear as to which). The piping system allowed it to be connected to the bilge piping so as to draw on either of the bilge suctions. When so connected it would have a smaller capacity by reason of the narrower diameter of the bilge piping. The experts agreed that its capacity would be reduced by about 50%.
115. There was agreement between the experts that the rate of ingress through the crack in the emergency fire pump would have been about 6 m³ per hour and that from the strainer lid between about 44 m³ and 61 m³ per hour. It is readily apparent that the pumping system in the engine room should have been able to cope with an ingress of no more than 67 m³ per hour.
116. On the night in question the crew sought to deal with the ingress by using the Port and Starboard GS Pumps connected to the fore and aft suctions respectively. When these did not appear to be coping with the ingress, an attempt was made to use a submersible pump, but this was unsuccessful. The crew did not attempt to swap the connection between the GS Pumps and the suctions using the cross over line, and it was not ultimately suggested by Underwriters that they should have done so, or that it would have assisted. Blockages in the crossover lines are therefore only relevant as background to the issues surrounding the use of the ballast pump.
117. The crew did not attempt to use the ballast pump to connect with the bilge suctions. Instead it was decided to cut a hole in a line on the suction side of the pump so as to allow the pump to draw water directly through the hole rather than from the bilge suctions. This decision was taken following discussions on the telephone between the Master and the Kornet brothers and Mr Parel, who were on a management team building session at a training establishment in Bosch en Duin when first alerted to the casualty by the Master.
118. The DSI report recorded that the pumps had later been inspected by the shipyard and found to be in good working order. As to the lines, valves and suctions it found:

(1) There were blockages in the bilge crossover lines:

(a) The connection between the starboard suction manifold of the Starboard GS Pump and the main bilge system was completely blocked; a photograph of the cross section of the piping illustrated the finding. There was an issue as to whether this meant it was totally blocked along its whole length or simply that there was a total blockage at some part of it. I find that the latter is more likely: where the report refers to a blockage along the length of a pipe it says so, as it did for the partial blockage in (b) below; and it would not have been possible to examine the extent of the full blockage without taking numerous cross sections of the piping.

(b) There was a partial blockage along the entire length of the crossover pipe between the port and starboard main bilge systems.

(2) The mesh filters in the baskets at the fore and aft bilge suctions restricted flow for two reasons. First they were clogged with dirt and paint flakes, which was explicable as normal debris from the vessel which accumulated in the filters as a result of pumping during the casualty. The Engineer's evidence in his statement was that he had recently cleaned the filters, although it appears from the DSI report that his evidence to the Inspectorate was that he had not opened the filters "during his term". On either view, it was Underwriters' case that the filters were clean before the casualty. Secondly these mesh filters were not positioned vertically away from the suction aperture of the bilge pipe, but were curved so as to line the basket and lie directly against the aperture. This arrangement restricted the flow of water through them to that which could pass through the area equivalent to the diameter of the pipe rather than the full area of the mesh. It is not clear when such filters were installed; the Underwriters did not contend that the Owners or Managers were at fault in this respect.

(3) The main bilge valve on the port side suction manifold of the Port GS pump had too long a spindle. The result was that when fully opened the valve restricted the capacity of the passage through it by 69%. It was not suggested that the Owners were at fault in relation to this defect or that they should have been aware of it.

(4) A mop head was found to have become jammed in the hole cut in the ballast suction line. The crew were unaware of this on the night in question.

119. The result of these conditions was as follows:

(1) The Starboard GS pump was only partially effective in drawing through the aft bilge suction to which it was connected because of the filter arrangement and filter blockage in the aft suction.

(2) The Port GS pump was only partially effective in drawing through the forward bilge suction to which it was connected because of (a) the 69% reduction in capacity caused by the overlong spindle and (b) the filter arrangement and filter blockage in the forward suction.

(3) Had the crew sought to use either GS pump on the other bilge suction, they would have been unable to do so because of blockages in the pipes. They did not, however, seek to do so, and no criticism is made of their failure to do so.

(4) Both GS pumps were electrically driven and incapable of operating under water. They were located a little above the engine room floor plates. Accordingly when the water reached that level, they were incapable of operating.

(5) The ballast pump was capable of being connected to the bilge suctions and the blockages in the crossover lines would not have prevented it from drawing on the bilge suctions, had the crew chosen to do so, although the partial blockage in the cross over line would have restricted flow from the forward bilge suction. The Underwriters suggested that the total blockage would have prevented any suction from the aft bilge suction, but I reject that submission. It rested upon two false assumptions, one being that the blockage extended along the full length of the pipe and the other being that that blockage intersected with, and blocked, the line between the ballast pump and the ballast suction valve at the starboard manifold.

120. By the conclusion of the hearing the Underwriters' case was simplified as amounting to the following. The total and partial blockages in the bilge lines rendered the Vessel unseaworthy; Gertjan Kornet was aware of the blockages and that they rendered the vessel unseaworthy; that was why he requested the ballast line to be cut rather than connecting the ballast pump to the bilge system; had the ballast pump been connected to the bilge system it would have controlled the ingress and the Vessel would not have suffered any loss and damage above the level of the floor plates.
121. I reject that case for three reasons. First, so far as the managers were concerned, and in particular Gertjan Kornet, the reason for cutting the ballast line was nothing to do with the bilge piping, but was to create a new source of suction against the background that the GS pumps working at full capacity through the existing suctions were not coping with an unknown rate of ingress. Secondly, I am not persuaded that the managers were aware of any blockage in the bilge piping system. Thirdly, I am not persuaded that if the ballast pump had been applied to the bilge system the additional capacity through the obstructed suctions would have coped with the ingress.

The decision to cut the ballast line

122. A considerable amount of time was spent with the witnesses in examining who first suggested cutting the ballast line and whether the managers instructed the Master to do so. To a large extent it was directed to an aspect of the "fraudulent device" issue which was not pleaded and was ultimately not pursued by Underwriters. It is impossible to tell whether the idea was first mooted by the Master or by the managers in the course of the conversation. Whoever first raised it, I regard it as probable that the Master would have sought the assurance of the managers that such a step could be taken because it was unusual. He would have sought authority from the managers. On the other hand the evidence from the managers' side that the Master was the man on the spot who had to assess what was happening and make the ultimate decision as to what to do was persuasive. I find that the Master was authorised to cut the ballast pipe, but not instructed to do so; he retained the responsibility for the ultimate decision in the fast moving circumstances he was facing.
123. Gertjan Kornet explained in his cross examination the thinking behind a decision to cut the ballast line, rather than connect the ballast pump to the bilge system. He said that they were faced with a situation in which the two service pumps were working fully through the bilge suctions, but were unable to cope with the ingress. The water was assumed to be coming in through the engine room and the source of the ingress had not been identified. They did not know the rate of ingress but it appeared to be substantial. An urgent decision needed to be made to try to deal with this unknown rate of ingress against that background. The full capacity on the existing suctions was not coping with the ingress, so what seemed to be needed was another source of suction rather than any additional pumping capacity applied to the existing suctions.
124. I found this explanation coherent and convincing. It accords with the inherent probabilities and I accept it. The cutting of the ballast line does not give rise to the inference that the owners knew of blockages in the piping system. The reasons for cutting the line were unconnected with the piping: they were to create a new point of suction.

Managers' alleged knowledge of blockages

125. The engine room pumps had not previously been used for pumping water from the engine room bilges. The engine room bilges were routinely emptied using a vacuum pumping truck to pump out the bilges and carry away the oily water. The Vessel would have been in breach of the International Convention for the Prevention of Pollution from Ships ("MARPOL") had the engine room pumps been used in this manner, except in an emergency. Accordingly there was no reason for the managers to have discovered any problem in the bilge pipes from any such operation.
126. The Underwriters relied upon two aspects of the evidence to support their case on knowledge. Each concerned the Vessel's use of the pumps to discharge ballast from the holds. The first is a repair list from 15 December 2007 which records the following exchange:

Crew: "There are still problems with pumping water out in the hold. We are not able or hardly able to pump water out of the hold in the aft of the ship. This is life-threatening and remains so."

Managers: "UNDERSTAND. IF THERE IS TIME THEN WE WILL TRY TO REPAIR THIS. BUT IT REMAINED THE CASE IN BOLNES THAT IT IS EXTREMELY DIFFICULT TO DO THIS AND WAS AGREED TO DO THIS TEMPORARILY WITH THE SUBMERSIBLE PUMP."

Crew: "At the last inspection from nnpc [the P & I Club] it was assumed by the inspector that this was working. He had tested this, we had had a problem. The bilge pumping gear fore was rectified with the emergency fire pump. This works properly, but aft ….???? When pumping out with a small pump, it is accompanied by a shot of oil. So all valves need to be tackled thoroughly.

Managers: "WE LOOKED AT THIS IN BOLNES AND AGREED THAT THE LADS WOULD DO THIS ON BOARD DURING THE JOURNEY.

Crew: "An injector has been brought on board, but has never been fitted, due to lack of time and lack of fitters in Bolnes."

Managers: "IT WAS AGREED IN BOLNES TO TEMPORARILY POSTPONE THIS."

Crew: "This point remains a hot item !!!!!!! Portside sump sometimes work, but starboard side bilge sump does not work !!!

Managers: "GET HENK TO HAVE A LOOK AT THIS HERE"

127. The Underwriters suggested that the reference to the P & I Club surveyor's inspection should be read as meaning "[If] He had tested this, we [would have] had a problem". This is not borne out by the Surveyor's report of his inspection carried out three days earlier on 12 December 2007. The Surveyor's report recorded:

"Hold bilge pump system

Tested while washing the hold and found to be in good working condition. The floors of the sumps were contaminated with old cargo/rust."

128. This suggests that the problem was not with the pipes, but with obstruction to the suctions in the sump. Gertjan Kornet testified that he could not remember precisely when those problems were rectified, but he knew that the whole bilge system was working in 2010. This is confirmed by the fact that on 24 April 2009 the Engineer was overhauling valves for the hold suctions, as is evidenced by a log entry. This suggests that the hold bilge system was working in 2009, as it would have made no sense for the Engineer to replace the valves if he believed that the bilge pipes were blocked so as to prevent the hold bilges being pumped.
129. The second aspect relied on by the Underwriters was the regular use of submersible pumps in the hold to pump out hold washings. Gertjan Kornet explained why there was nothing suspicious about the use of a submersible pump. The Vessel regularly carried cargoes of stone. When the holds were washed down, the hold bilges would fill with gravelly water which the crew did not want to clog up the bilge system. To avoid blockages and damage, the crew used the submersible pump instead. I accept that explanation. It was the one given by the Chief Officer and Engineer to a P&I surveyor on 14 January 2010 and recorded in the surveyor's report. Mr Barker's opinion was that it was not particularly unorthodox or unusual when discharging cargoes such as stone or cement, which owners would be extremely keen to keep out of their bilge system. During an inspection which took place since the casualty, Mr Tanner observed that a submersible pump was still being used, and he accepted that the bilge pipes were unlikely to have remained blocked since the casualty because Class would have ensured that they were checked and renewed as necessary.
130. There was some unclear evidence from the Master which I do not think greatly assists on this issue. The notes taken by Mr Billowes of his interviews following the casualty included the following being attributed to the Master: "We also have a portable pump – capacity about 50 metres cubed per hour. Electric. Armoured tube for suction was in hold. We use it if bilge is blocked." When the Master was cross examined about this note, he confirmed that it was his recollection that if the bilge was blocked then the portable pump would be used, but he rejected the suggestion that it was used because the starboard GS pump was unable to pump the holds because of blocked lines, and he said that he did not have any knowledge of a blockage on this Vessel. His command of English was less than perfect, and I would not treat him as accepting any more than that pumps on the vessels on which he served would be used if bilges were blocked and that was what he understood to be their general purpose. I do not regard his evidence as suggesting that a portable pump was used on the DC Merwestone because he knew the bilges were blocked on that vessel, and he denied any such knowledge.
131. I find that no one at the Managers, including Gertjan Kornet, was aware of the blockages in the bilge lines.

Causation: additional capacity sufficient if applied through the suctions?

132. The Underwriters' case on causation was advanced by the following steps (i) the maximum rate of ingress into the engine room was 67 m³ per hour; (ii) the amount not being coped with by the GS pumps was a maximum of about 32m³ per hour; (iii) the additional capacity of the ballast pump would have enabled at least an additional 32m³ per hour to be drawn through the bilge suctions despite the arrangement and clogging of the filters. Step (i) was common ground between the experts, but steps (ii) and (iii) were not, and bear closer examination.
133. The figure of 32m³ in step (ii) is derived by taking the level of water visible in a photograph taken at 2225 as indicating a level of 2.3m above the shell plating. By taking the height which the water had reached when the GS pumps were started 85 minutes earlier, at about 2100, a calculation is made of the increase in volume over that period, which will indicate the rate of net increase which is not being pumped out by the GS pumps. Assuming the level at 2100 to be the level of the bilge alarm, 95cm above the shell plating, the net increase in 85 minutes would be about 44-48m³, equivalent to an hourly rate of net ingress of the order of 32m³. At a maximum rate of gross ingress into the vessel of 67m³ (step (i)), this would mean the GS pumps were coping with 35m³ per hour, leaving the ballast pump to have had to cope with 32m³ per hour. If the level of the water at 2100 was above the bilge alarm level, the amounts being coped with by the GS pumps would be greater, and the required additional capacity of the ballast pump lower. At 1.5m the net rate of ingress would be 20m³ per hour, and at 1.8m it would be 12m³ per hour. The calculation of volumes from heights was not agreed, nor can one be sure from the conflicting evidence from the crew what the levels were at 2100. Nevertheless I am satisfied that the level was not lower than 95cms at 2100 and probably a little higher. I therefore accept that Underwriters have established step (ii): the net rate of ingress with which the ballast pump would have had to cope was no more than about 32m³ per hour, although not likely to have been substantially less.
134. Step (iii) relies upon an answer given in re examination by Mr Tanner on the assumption that the required additional capacity achievable by the ballast pump when connected to the bilge system was 32m³. It was not however supported by any calculation or reasoning. It is wholly dependent on the extent of clogging of the filters, which Mr Tanner took to be about one third, on the basis of counting up the total number of holes in the mesh which appeared to be clogged from the photos of the filter meshes taken after the casualty. This unscientific method was rejected by Mr Barker as a reliable method of estimation, to my mind convincingly. The holes which would have mattered were those adjacent to the aperture in the suction pipe, but the location of the aperture is not apparent from the photos. Moreover the debris seen in those photos at the bottom of the basket would have been likely to have been drawn against the mesh filter under suction and so have impeded flow to a greater extent than that caused by the blockage evident in the photos. The extent of the blockage during the casualty is simply a matter of speculation. If the gross rate of ingress was 50m³ per hour and the net ingress 32m³, the GS pumps were only coping with 18m³ per hour, despite a rated capacity of almost 97m³ per hour (taking account of the 69% reduction in the Port GS pump caused by the overlong spindle). The filters would be reducing the rated capacity to an efficiency of less than 20%; if the rated capacity of the ballast pump operating through the bilge lines was 125m³ (or even 150m³), it could not have coped with 32m³ per hour if the filters were reducing its efficiency to less than 20%. Continued pumping by the ballast pump would have tended to increase the amount of debris in the filters. On the evidence it remains a matter of speculation whether the ballast pump attached to the bilge system would have coped with the ingress given the arrangement and state of the suction filters.
135. Accordingly I do not consider there is any reliable basis in the evidence for reaching the conclusions for which Underwriters contend in step (iii). Underwriters have therefore failed to discharge the burden of proving that had the ballast pump been connected to the bilge system the rate of ingress would have been controlled and the casualty averted.

Issue 7: Did the loss result from want of due diligence by the Owners or managers in relation to the engine room pumping system?

136. I return to the question whether there was a want of due diligence on the part of the owners in relation to the bilge line blockages for the purposes of the proviso to the Inchmaree clauses. This issue does not arise on my previous findings because (1) there is cover under ITC Clause 6.1 for a loss by perils of the seas, and (2) there is cover for crew negligence and contractors' negligence under the Inchmaree clauses, to which these questions are irrelevant and (3) the blockages were not causative of loss, still less a proximate cause. It is sufficient therefore simply to state my conclusion that had it been relevant, I would not have found any want of due diligence on the part of the Owners or managers in this respect.

Issue 8: Quantum

137. The Owners' claim is in the total sum of €3,241,310.60, being (i) €3,257,650.00 for the Particular Average due to it and (ii) €8,660.60 in respect of its General Average contribution, both figures approved by the independent Average Adjuster in his adjustment, less the policy deductible of €25,000.00. The only difference between the parties is whether it was reasonable to replace the Vessel's main engine, which was damaged beyond repair, with a new engine, as opposed to a reconditioned engine.
138. Section 69 of the Marine Insurance Act 1906 provides in relevant part as follows:

"Where a ship is damaged, but is not totally lost, the measure of indemnity, subject to any express provision in the policy, is as follows:—

(1)Where the ship has been repaired, the assured is entitled to the reasonable cost of the repairs, less the customary deductions, but not exceeding the sum insured in respect of any one casualty …"

139. Clause 14 of the Institute Time Clauses – Hulls 1/10/83 further provided "Claims payable without deduction new for old." The Owners argued that this meant that the Insured is entitled to the reasonable cost of replacing the old engine with a new one, and that the deduction for which Underwriters contend is contrary to the express terms of the policy, since it would amount to a new-for-old deduction. That is not the purpose of the clause. It provides that there shall be no deduction for betterment if a repair involves replacing an old part with a new one. But the measure of indemnity dictates an inquiry into whether the cost of repairs is reasonable, and there remains the question whether it was reasonable to repair the vessel by replacement of the damaged part by a new part as opposed to a second hand part.
140. The quantification of the difference in cost between the reconditioned and the new engine was estimated by Mr Tanner at €540,000. A spreadsheet prepared on behalf of Underwriters using figures from the draft Adjustment (which contemplated a reconditioned engine) and final Adjustment (which provides for the new one) gives a difference of €565,686.63.
141. I find that the cost of installing a new engine was reasonable. If the Owners had chosen to install a reconditioned engine, they would have incurred significant extra costs and encountered practical difficulties. In this regard, it was the uncontradicted evidence of Chris Kornet that:

(1) Steps had been taken to reduce the output of the original engine from 1,324 kW to 749 kW so as to allow the Vessel to sail with a Periodically Unattended Machinery Space ("UMS Notation") in accordance with the DSI minimum safe manning requirements.

(2) The DSI and Bureau Veritas told the Owners that it would be impossible to implement a reconditioned engine of the same type in 2010 and to obtain a UMS Notation, due to a change in the regulations applicable to the Vessel. In particular, the Vessel would only have been able to retain its UMS Notation if the alarms on the reconditioned engine and gearbox had been modified. This would not have been possible because the manufacturer of the gearbox and thrust bearings was no longer in business.

(3) Had the Owners installed a reconditioned engine, it would therefore have been necessary to sail with a chief engineer, a second engineer and probably a third engineer or an oiler. A reconditioned engine would not have amounted to a like-for-like replacement, because it would have exposed the Owners to additional labour expenses of about €82,000 per year, a figure which the Underwriters did not challenge. By contrast, the new engine allowed the Vessel to retain its UMS Notation and allowed the Owners to trade the Vessel in the same way and with the same manning as before the casualty.

(4) As the independent Average Adjuster, A.C.G. Schoutens, stated

"A second hand engine of the same type was available, however, it was not permitted to reduce the output anymore to 750 HP to allow the vessel to trade with an unmanned engine room… Consequently this second-hand engine was no reasonable option for the Assured. The Assured opted for a new main engine so that they could continue to trade with an unmanned engine room. This seems to be a reasonable decision that, given the circumstances, would have been taken by any prudent uninsured Shipowner too."

142. The Underwriters argued that this justification was not in the Owners' mind when they chose a new engine, and the evidence supports their position. But this is irrelevant. Whether the cost of repairs is reasonable is an objective test and the cost of an objectively reasonable repair is recoverable whatever the motives of the assured in taking that course.
143. Accordingly the Owners have a valid claim for €3,241,310.60 under the policy unless it is forfeit by their having employed a fraudulent device in pursuit of the claim.

Issue 9: Have the Owners forfeited the claim by reason of fraudulent devices employed in supporting it?

144. The Underwriters' case is that Chris Kornet deliberately or recklessly gave a false narrative of the casualty in (1) a letter to the Underwriters' solicitors, Ince & Co, dated 21 April 2010 and (2) a letter to the First Defendant ("Gerling") as lead underwriter dated 27 July 2010 and (3) a preliminary report of TMC dated 27 January 2011 sent to the Underwriters. The main thrust of the allegation is that he falsely claimed that the bilge alarm had gone off at about noon on 28 January 2010; that the alarm had been ignored because it was attributed to the rolling of the vessel in heavy weather; and that he had been told both these things by the Master or crew. It is alleged that he did this because he had been advised of the Inchmaree proviso, and understood a need to distance the Owners themselves from any fault in relation to the casualty, and was therefore keen to explain the quantity of water reaching the engine room by a narrative which involved the bilge alarms working but being ignored by the crew. A further allegation of falsehood in the 21 April 2010 letter, relating to who had suggested or instructed the cutting of the ballast pipe, was advanced at the hearing, but had not been pleaded; in the face of objection during final speeches, Mr Jacobs QC made clear that he did not pursue it as a separate fraudulent device under this head, but only as relevant to Chris Kornet's credibility.

The law

145. There is a long line of authority which stands for a rule of law, applicable even when there is no express clause in the policy, to the effect that an insured who has made a fraudulent claim forfeits any lesser claim which he could properly have made. It originates from the early 19^th century (see Levy v Baillie (1831) 7 Bing 349, Goulstone v Royal Insurance Co (1866) 1 F&F 276 and Britton v Royal Insurance Co (1866) 4 F&F 905), and has been confirmed and applied at appellate level up to the present day: see Lek v Matthews (1927) 29 Ll. L. Rep 141 (HL); Orakpo v Barclays Insurance Services [1995] Lloyd's Rep IR 443 (CA); Galloway v Guardian Royal Exchange (UK) Ltd [1999] Lloyd's Rep IR 209 (CA); The Star Sea (HL); Direct Line Insurance v Khan [2002] Lloyd's Rep IR 364 (CA); Agapitos v Agnew [2003] QB 556 (CA); Axa General Insurance Ltd v Gottlieb [2005] All ER 163 (CA).
146. More recently the fraudulent claims rule has been extended by the decision of the Court of Appeal in Agapitos v Agnew [2003] QB 556 to cases in which the assured has deployed in support of a wholly valid claim some fraudulent means or device to advance the claim. Strictly speaking this aspect of the decision in Agapitos v Agnew was obiter, because it was held that the rule, whether or not extended, could not apply after the commencement of litigation in the light of the House of Lords decision in The Star Sea. Whilst there is room for argument whether such extension is justified (see for example Lord Hobhouse's observations in The Star Sea at [71] on the controversial decision of Hirst J in The Litsion Pride [1985] 1 Lloyd's Rep 437), the question was fully considered in the judgment of Mance LJ at [19]-[46], in a judgment concurred in by Brooke LJ and Park J. Mance LJ's judgment was described by Park J as a comprehensive and scholarly exposition, with which I would respectfully agree. The extension has been recognised by the Supreme Court in Summers v Fairclough Homes [2012] 1 WLR 2004 at [29] and applied by the Privy Council in Stemson v AMP General Insurance (NZ) Ltd [2006] Lloyd's Rep IR 252, and recognised or applied in a number of first instance decisions, including Eagle Star Insurance Co Ltd v Games Video Co (GVC) SA (The Game Boy) [2004] 1 Lloyds 238, Joseph Fielding Properties (Blackpool) Ltd v Aviva Insurance Ltd [2011] Lloyd's Rep IR 238, and Aviva Insurance Ltd v Brown [2012] Lloyd's Rep IR 211. In those circumstances it would not be right for a judge of first instance to decline to apply the extension, however much he might regret that the submissions of counsel for the insurers in Agapitos v Agnew prevailed. Mr Karia QC reserved his right to challenge the extension in a higher court.
147. The juridical basis for the rule has caused some difficulty. Its origin was linked to the continuing duty of good faith in insurance contracts, for which the remedy provided by s. 17 of the Marine Insurance Act 1906 is avoidance of the policy; and the rule was thought to exist under, or by analogy with, s. 17. It has also been treated as deriving from an implied term of the contract (see Orakpo), a view rejected by the House of Lords in The Star Sea (see per Lord Hobhouse at [62]). Recent authority, in which Lord Mance has played a leading role, favours an analysis which treats it as a sui generis principle of the common law resulting in the forfeiture of the claim to which the fraud relates: see Agapitos v Agnew at [45(g)] and Axa v Gottlieb at [18]-[20], [22], [31]-[32].
148. The justification for the rule and its extension, as articulated by Lord Hobhouse in The Star Sea at [62], is that: "The fraudulent assured must not be allowed to think: if the fraud is successful, then I will gain; if it is unsuccessful, I will lose nothing". It reflects a policy of the law to discourage the making of fraudulent claims: Agapitos v Agnew at [14] per Mance LJ and Galloway v Guardian Royal Exchange per Lord Woolf MR at 213 and Millet LJ at 214. In Axa General Insurance Ltd v Gottlieb Mance LJ described it at [31] as "deliberately designed to operate in a draconian and deterrent fashion" and said: "……the policy of the rule is to discourage any feeling that the genuine part of a claim can be regarded as safe - and that any fraud will lead at best to an unjustified bonus and at worst, in probability, to no more than a refusal to pay a sum which was never insured in the first place."
149. Two aspects of the rule and its extension merit closer examination in the present case. The first is as to the state of mind which is sufficient to make the claim or device "fraudulent". The second is what may be termed materiality, that is to say the relationship which the fraudulent means or device must bear to the valid claim.

State of mind

150. In Agapitos v Agnew Mance LJ said at [30] that "A fraudulent device is used if the insured believes that he has suffered the loss claimed, but seeks to improve or embellish the facts surrounding the claim, by some lie". He did not expand upon the concept of fraud in this context.
151. The state of mind necessary to establish fraud for the purposes of the tort of deceit is to be found in the well known passage in the speech of Lord Herschell in Derry v Peek (1889) 14 AC 337 at 374:

"I think the authorities establish the following propositions: First, in order to sustain an action of deceit, there must be proof of fraud, and nothing short of that will suffice. Secondly, fraud is proved when it is shewn that a false representation has been made (1) knowingly, or (2) without belief in its truth, or (3) recklessly, careless whether it be true or false. Although I have treated the second and third as distinct cases, I think the third is but an instance of the second, for one who makes a statement under such circumstances can have no real belief in the truth of what he states. To prevent a false statement being fraudulent, there must, I think, always be an honest belief in its truth. And this probably covers the whole ground, for one who knowingly alleges that which is false, has obviously no such honest belief. Thirdly, if fraud be proved, the motive of the person guilty of it is immaterial. It matters not that there was no intention to cheat or injure the person to whom the statement was made."

152. That a reckless untruth, of the kind described in Lord Herschell's third category, is sufficient to amount to fraud for the purpose of the fraudulent claims rule, is apparent from the speech of Lord Sumner in Lek v Matthews at 145 and the decision of Saville J in Bucks Printing Press Ltd v Prudential Assurance Co [1994] 3 Re LR 219.
153. Mr Karia QC argued that it is also necessary for the insurer to prove dishonesty, and in particular [1] that the assured was dishonest by the ordinary standards of reasonable and honest people and [2] that the assured himself realised that by those standards his conduct was dishonest. This submission was taken from a submission made in Brown v Aviva which Eder J recorded at [67] as having been accepted on behalf of the insurer. This combined test originates in the speeches of Lord Hutton and Lord Hoffmann in Twinsectra Ltd v Yardley [2002] 2 AC 164, where it was formulated in the context of constructive trusts imposed for dishonest assistance in a breach of trust. The test of dishonesty in that context was reconsidered and explained by the Privy Council in Barlow Clowes Ltd v Eurocrest [2006] 1 WLR 1476 in terms which represent English law (Abu Rahmah v Abacha [2007] 1 All ER (Comm) 827, at [66] – [70]). The subjective element requires the assister to know facts which make his conduct, objectively viewed, dishonest, but there is no subjective element of conscious dishonesty; the test for dishonesty does not require that the assister considers that he is acting dishonestly.
154. In these circumstances, where the fraudulent device consists of a representation, there is little room for any distinction between the test of fraud provided for in Derry v Peek and the test of dishonesty which applies to dishonest assistance in a breach of trust. If any of the three limbs of the Derry v Peek test are fulfilled, the statement will have been made without an honest belief in its truth. That is what would be regarded as dishonest for the objective limb of the test in Barlow Clowes, irrespective of whether the representor consciously believes himself to be acting dishonestly. In this context "in effect, recklessness is a species of dishonest knowledge, for in both cases there is an absence of belief in truth": per Rix LJ in The Kriti Palm [2007] 1 Lloyd's Rep 555 at [257]. Where, as in this case, the fraudulent device alleged is a statement, the appropriate test for fraud is that set out in Derry v Peek. In particular conscious dishonesty is not a separate element of the test, any more than it is for dishonest assistance constructive trust liability.
155. This is not to diminish the burden which lies upon insurers in seeking to establish fraud. The standard of proof is the balance of probabilities, but the cogency and strength of the evidence required to prove fraud is heightened by the seriousness of the allegation: Re H [1996] AC 563, 586. The court will be astute not to water down the requirement of fraud into something akin to negligence, even gross negligence. When considering whether there has been recklessness as to the truth of a statement, not caring whether it be true or false, "not caring" does not mean not taking care; it means indifference to the truth, the moral obloquy of which consists in a wilful disregard of the importance of truth: see per Bowen LJ in Angus v Clifford [1891] 2 Ch 449, 471.

Materiality

156. In the cases concerned with the fraudulent claims rule itself, the preponderance of authority is that the fraudulent part of the claim must be substantial in the sense of being not insubstantial or immaterial or de minimis: Goulstone's case ("wilfully false in any substantial respect" per Pollock C.B. at 279); Lek v Matthews ("anything not so unsubstantial as to make the maxim de minimis applicable" per Lord Sumner at 145); Orakpo ("fraudulent to a substantial extent" per Sir Roger Parker at 452); Galloway (not "immaterial", "substantial" per Lord Woolf MR at 213, 214).
157. This is not a high threshold. In Galloway a householder made a claim on his contents policy for loss arising out of a burglary. In addition to a valid contents claim for £16,133.94 the assured presented a fraudulent claim of £2,000 for a computer, supported by a bogus invoice in respect of its purchase. This was sufficiently material to vitiate the valid claim. Millett LJ, with whom Mummery LJ agreed, expressed the view that assuming, without deciding, that the rule was limited to claims which were substantially fraudulent or fraudulent to a substantial degree, that was a question to be addressed by reference only to the fraudulent element, not by reference to the proportion which the fraudulent element bore to the valid element of the claim. A fraudulent claim for £2,000, viewed in isolation, was sufficiently serious to be stigmatised as a breach of good faith so as to engage the rule. In Direct Line Insurance v Khan [2002] Lloyd's Rep IR 364, a husband and wife brought a claim for property and contents damage following a fire to their home which was properly quantified at £61,342, and a fraudulent claim for rental of alternative accommodation for £8,257. The court accepted a concession that the rental claim was "sufficiently substantial" to taint the whole claim and make it irrecoverable (see Arden LJ at [12]). If the approach of Millett LJ in Galloway be right, a fraudulent element of £2,000 (and quite possibly considerably less) is sufficiently substantial to vitiate a marine insurance claim of €3 million or more.
158. In the context of the extension of the rule to fraudulent devices, Mance LJ conducted an analysis in Agapitos v Agnew which emphasised the potential need to draw a distinction between a test of materiality, which would apply to a breach of the duty of good faith under s. 17 Marine Insurance Act 1906, and the test applicable to fraudulent claims or devices. His analysis deserves extensive citation:

"31. The authorities also indicate that there are differences between, on the one hand, a fraudulent claim to recover a non-existent or exaggerated loss and, on the other, a breach of the duty of good faith under section 17. Rix J said in the Royal Boskalis case, at p 599:"

"upon my understanding of the nature of a fraudulent claim, there is no additional test of materiality or, to put the same point perhaps in another way, the test of materiality is built into the concept of a fraudulent claim…"

32. This observation merits some further examination. I start by noting an aspect of this court's decision in Galloway's case [1999] Lloyd's Rep IR 209. The claim there made, following a burglary, for some £18,143 consisted in the main of genuine loss, but as to £2,000 involved the alleged loss of a non-existent computer. The court agreed with the judge that the whole claim was forfeit. Lord Woolf MR, at p 213, explained references in Orakpo's case [1995] LRLR 443 to the need for "substantial" fraud as intended to exclude fraud which could be regarded as "immaterial" (or, in Viscount Sumner's words in Lek v Mathews 29 LlL Rep 141, 145, so "unsubstantial as to [be] de minimis"). In this context the right approach was to look at the size of the non-existent loss alone and not to draw some comparison between it and the size of the genuine claim. Millett LJ [1999] LRLR 209, 215 suggested that the right approach was to consider whether the making of the claim was "sufficiently serious to justify stigmatising it as a breach of [the insured's] duty of good faith so as to avoid the policy". This assumes that the remedy of avoidance is available in this context. Whether it is available was not in issue in either Orakpo's or Galloway's case (as Lord Hobhouse observed in The Star Sea [2003] 1 AC 469, 501, para 67), and is a matter which, I suggest, merits further examination before the common law commits itself.

33. Secondly, in relation to Rix J's observation in the Royal Boskalis case [1997] LRLR 523 to the extent that loss claimed is non-existent, the claim will fail anyway and the fraud is clearly material in so far as it amounted to an attempt to recover for non-existent loss. But the real bite of the fraudulent claim rule is to forfeit even the genuine part of any claim; and the fraud by definition is not material in any ordinary sense to the genuine part. Thus, it is sufficient for the rule to apply that the fraud occurs in making a claim and relates to a part of the claim which, when viewed discretely, is not itself immaterial or "unsubstantial".

……………

"36. What relationship need there then be between any fraud and the claim if the fraudulent claim rule is to apply? And need the fraud have any effect on insurers' conduct? Speaking here of a claim for a loss known to be non-existent or exaggerated, the answers seem clear. Nothing further is necessary. The application of the rule flows from the fact that a fraudulent claim of this nature has been made. Whether insurers are misled or not is in this context beside the point. The principle only arises for consideration where they have not been misled into paying or settling the claim, and its application could not sensibly depend upon proof that they were temporarily misled. The only further requirement is that the part of the claim which is non-existent or exaggerated should not itself be immaterial or unsubstantial: see paragraphs 32-33 above. …..

37. What is the position where there is use of a fraudulent device designed to promote a claim? I would see no reason for requiring proof of actual inducement here, any more than there is in the context of a fraudulent claim for non-existent or exaggerated loss. As to any further requirement of "materiality", if one were to adopt in this context the test identified in the Royal Boskalis case [1997] LRLR 523 and The Mercandian Continent 572 [2001] 2 Lloyd's Rep 563, then, as I have said, the effect is, in most cases, tantamount to saying that the use of a fraudulent device carries no sanction. It is irrelevant (unless it succeeds, which only the insured will then know). On the basis (which the cases show and I would endorse) that the policy behind the fraudulent claim rule remains as powerful today as ever, there is, in my view, force in Mr Popplewell's submission that it either applies, or should be matched by an equivalent rule, in the case of use of a fraudulent device to promote a claim—even though at the end of a trial it may be shown that the claim was all along in all other respects valid. The fraud must of course be directly related to and intended to promote the claim (unlike the deceit in The Mercandian Continent). Whenever that is so, the usual reason for the use of a fraudulent device will have been concern by the insured about prospects of success and a desire to improve them by presenting the claim on a false factual basis. If one does use in this context the language of materiality, what is material at the claims stage depends on the facts then known and the strengths and weaknesses of the case as they may then appear. It seems irrelevant to measure materiality against what may be known at some future date, after a trial. The object of a lie is to deceive. The deceit may never be discovered. The case may then be fought on a false premise, or the lie may lead to a favourable settlement before trial. Does the fact that the lie happens to be detected or unravelled before a settlement or during a trial make it immaterial at the time when it was told? In my opinion, not. Materiality should take into account the different appreciation of the prospects, which a lie is usually intended to induce on insurers' side, and the different understanding of the facts which it is intended to induce on the part of a judge at trial.

38. The view could, in this situation, be taken that, where fraudulent devices or means have been used to promote a claim, that by itself is sufficient to justify the application of the sanction of forfeiture. The insured's own perception of the value of the lie would suffice. Probably, however, some limited objective element is also required. The requirement, where a claim includes a non-existent or exaggerated element of loss, that that element must be not immaterial, "unsubstantial" or insignificant in itself offers a parallel. In the context of use of a fraudulent device or means, one can contemplate the possibility of an obviously irrelevant lie—one which, whatever the insured may have thought, could not sensibly have had any significant impact on any insurer or judge. Tentatively, I would suggest that the courts should only apply the fraudulent claim rule to the use of fraudulent devices or means which would, if believed, have tended, objectively but prior to any final determination at trial of the parties' rights, to yield a not insignificant improvement in the insured's prospects—whether they be prospects of obtaining a settlement, or a better settlement, or of winning at trial. Courts are used enough to considering prospects, e g when assessing damages for failure by a solicitor to issue a claim form within a limitation period."

………….

"45. What then is the appropriate approach for the law to adopt in relation to the use of a fraudulent device to promote a claim, which may (or may not) prove at trial to be otherwise good, but in relation to which the insured feels it expedient to tell lies to improve his prospects of a settlement or at trial? The common law rule relating to cases of no or exaggerated loss arises from a perception of appropriate policy and jurisprudence on the part of our 19th century predecessors, which time has done nothing to alter. The proper approach to the use of fraudulent devices or means is much freer from authority. It is, as a result, our duty to form our own perception of the proper ambit or any extension of the common law rule. In the present imperfect state of the law, fettered as it is by section 17, my tentative view of an acceptable solution would be: (a) to recognise that the fraudulent claim rule applies as much to the fraudulent maintenance of an initially honest claim as to a claim which the insured knows from the outset to be exaggerated; (b) to treat the use of a fraudulent device as a sub-species of making a fraudulent claim—at least as regards forfeiture of the claim itself in relation to which the fraudulent device or means is used (the fraudulent claim rule may have a prospective aspect in respect of future, and perhaps current, claims, but it is unnecessary to consider that aspect or its application to cases of use of fraudulent devices); (c) to treat as relevant for this purpose any lie, directly related to the claim to which the fraudulent device relates, which is intended to improve the insured's prospects of obtaining a settlement or winning the case, and which would, if believed, tend, objectively, prior to any final determination at trial of the parties' rights, to yield a not insignificant improvement in the insured's prospects—whether they be prospects of obtaining a settlement, or a better settlement, or of winning at trial; and (d) to treat the common law rules governing the making of a fraudulent claim (including the use of fraudulent device) as falling outside the scope of section 17 (as advocated, though more generally, by Howard N Bennett in the article to which I have already referred in paragraph 36). On this basis no question of avoidance ab initio would arise."

159. The test "tentatively" proposed, therefore, is that fraudulent means or devices are sufficient to vitiate a valid claim if:

(1) they are directly related to the claim and intended to promote the claim; and

(2) the fraudulent means or devices would, if believed, tend, objectively and prior to any final determination at trial of the parties' rights, to yield a not insignificant improvement in the insured's prospects, whether they be prospects of obtaining a settlement, or a better settlement, or of winning at trial. In this context "not insignificant" has the same connotation as "not insubstantial", "not immaterial", "not de minimis", which is the test for the fraudulent claims rule, of which the rule as to fraudulent means and devices is to be treated as a subspecies.

160. The logic and juridical pedigree of such an analysis is, with the very greatest respect, powerful. It was adopted and applied by the Privy Council in Stemson v AMP; by Blair J in Sharron's Bakery (who as a result of a concession treated the test under a clause applying to claims "in any respect fraudulent" as replicating the common law); and by Eder J in Brown v Aviva. Nevertheless the low and relatively inflexible threshold which this test of materiality imposes is one which I find in a number of respects unsatisfactory.
161. The policy considerations which apply to fraudulent claims take on a very different aspect in the context of fraudulent means or devices which are deployed to support a wholly valid claim. The policy of the rule as applied to fraudulent claims is to deter an assured from seeking to recover more than his entitlement without risk of penalty; this is explicit in the rationale given by Lord Hobhouse in The Star Sea at [62] and Mance LJ at [31] of Axa v Gottlieb. In the context of the extension of the rule to fraudulent means or devices deployed in support of a valid claim, the assured is seeking to recover no more than his entitlement, albeit employing underhand means in doing so. It is true that he is still seeking to gain to some extent by his fraud: he intends to persuade the insurer to pay his valid claim more promptly than the insurer otherwise might and without recourse to litigation, or to make a more generous offer of settlement than otherwise might be forthcoming. But he is not seeking to gain any more than that which the court has subsequently been able to determine is his contractual entitlement.
162. A different justification for the fraudulent claims principle which has been suggested is that it lies in the asymmetry of information which also underpins the duty of disclosure at the time of making the contract of insurance. This is the explanation for the rule in the majority judgments of Hoffmann LJ and Sir Roger Parker in Orakpo at 450 and 452: a duty of good faith arises at the time of making a contract of insurance because the material facts are peculiarly within the knowledge of the assured; the same is generally true in relation to the casualty, and the continued existence of the duty of good faith at the time the claim is presented and pursued is justified by this imbalance; insurers have to be able to trust the assured to put forward a claim in good faith. But this asymmetry of information at the claims stage is a commonplace of many civil claims. Outside the insurance context, the quantification of a claimant's damages will often depend upon facts and circumstances which are peculiarly within the knowledge of the claimant. I find this rationale for the rule, with respect, insufficient to explain its peculiar and exclusive application to insurance claims. It is not the justification which has been adopted in the subsequent authorities, including in particular The Star Sea, Agapitos v Agnew and Axa v Gottlieb.
163. The assured's contractual entitlement is an entitlement to be indemnified, that is to say to be held harmless against the suffering of the loss in the first place: see Firma C-Trade S.A. v Newcastle Protection and Indemnity (The Fanti and The Padre Island) [1991] 2 AC 1, 35). Non payment by the insurer, even during a reasonable period for investigation, is a breach of the insurer's obligation to indemnify. His failure to pay may well cause the assured to suffer consequential loss; but the assured cannot recover for losses caused by the insurer's wrongful refusal to pay a valid claim: see Ventouris v Mountain (The Italia Express No 2) [1992] 2 Lloyd's Rep 281. In this respect the law disadvantages the assured, and is widely regarded as unfair and in need of reform. In seeking prompt payment of a valid claim, an assured is not only seeking that to which he is entitled, but may often be seeking to avert consequential losses which are being caused by the insurer's wrongful refusal to pay the claim. In the marine insurance context these may be very substantial. The irrecoverability of losses caused by the insurer's failure to pay a valid claim bears on the current problem in two ways. First the degree of culpability which attaches to an assured who uses a fraudulent device to seek prompt payment of a valid claim, in order to avert further irrecoverable losses which are being caused by non payment, may lie at the lower end of the scale, especially when set in the context of the fact that the non payment of the claim is a failure by the insurer to do that which is the very essence of the contract of indemnity, namely to hold the assured harmless from the consequences of the casualty. This will especially be so if the insurers' conduct is unreasonable. Secondly, if the fraudulent devices rule deprives the assured of his claim, it leaves him to bear losses comprising not only the amount insured, but also losses suffered by reason of non payment of his valid claim, which may potentially be very substantial. These may be very harsh consequences to visit upon an assured whose culpability is at the lower end of the scale.
164. All fraud rightly attracts condemnation. In the Law Commission's recent second consultation paper on "Insurance Contract Law: Post Contract Duties and other issues", it is recorded at paragraphs 6.1 and 6.6 that fraudulent insurance claims are a serious and expensive problem. The Association of British Insurers (ABI) reported that in 2010 insurers uncovered 133,000 fraudulent claims; and that the value of these claims totalled £919 million or 5% of the value of all claims made on its members that year. According to an ABI news release dated 28 July 2011, insurance fraud is said to cost the UK economy £2 billion every year. If these figures are accurate, they fully justify the widespread judicial comments that insurance fraud is prevalent and a scourge. But it may be appropriate to sound a note of caution. I would venture to doubt whether a significant element of the problem reflected in these figures is represented by wholly valid claims which are supported by a fraudulent device.
165. Whilst any fraud is to be condemned as reprehensible and is to be discouraged, the degree of culpability and the force of the condemnation must take their colour from the differing circumstances of each case. Not all fraud attracts the the same moral obloquy, as is recognised in the sentencing practice applied to criminal offences involving dishonesty and fraud. The assured who burns down his house for the insurance proceeds, or makes a claim for a fictitious loss attracts greater condemnation than the householder who includes, within a property and contents claim of £500,000, a claim for £2,000 for a computer he has genuinely lost, but believing that the insurers will not pay without some proof of purchase, supports it by presenting an invoice which is for a similar computer which has not been lost, in circumstances where he knows it is not the right invoice but can't be bothered to search for the right one. Those cases where fraud is employed by an assured in pursuit of a wholly genuine and valid claim may come towards the lower end of the scale of culpability. The conduct of the insurer in declining the claim may be unreasonable. Insurers may be seen by assureds, whether legitimately or not, as "messing them about" in delaying or declining the payment of a valid claim. In the marine context, assureds often pay large premiums, and the insurance is taken out with the purpose, and legitimate expectation, that the cover will enable them to cope with the serious financial impact of a casualty: delay in payment of a claim can cause real financial hardship. That was the position of the Owners in this case and Chris Kornet's frustration at the continuing and serious adverse consequences for the Owners resulting from the Underwriters' delay in accepting the claim, however reasonable the Underwriters' conduct, is an understandable human reaction.
166. As formulated in Agapitos v Agnew, the materiality test is not concerned with the protection of the particular insurer who is the victim of the fraudulent device. There is no requirement that the insurer should have been deceived by the lie, or that the lie should have played any part in his consideration of whether or when to pay the claim. It is the assured's attempt to deceive which is sufficient to attract the penalty of forfeiture of the valid claim. Moreover the logic of the test is that the attempt to deceive, once committed, is irremediable. A correction or retraction would be ineffective. The assured who in a fit of exasperation tells a lie but, having calmed down, corrects it the following day, would still forfeit his claim: see Stemson at [34].
167. The test is therefore capable of operating to visit disproportionately harsh and unjust consequences upon an assured in favour of an undeserving insurer.
168. Such potential consequences fall to be judged against the background of two anomalous aspects of the fraudulent devices rule. The first is that the rule of law under consideration is one imposed upon the parties as an incident of their contract notwithstanding that the parties have not agreed it. Many non marine policies contain a term which provides for forfeiture of the claim in the event of fraudulent claims or means: see for example the clause in Sharron's Bakery at [75]. Insurers are free to include such a term in their policies if the assured will agree to it. In the absence of such a clause, the parties have not agreed that any fraudulent device will forfeit a valid claim, either expressly or by a process of implication or interpretation of their agreement. The fraudulent claims rule appears to have had its origin in clauses inserted in fire policies until the law took a turn in the early 19^th century of treating such clauses as no more than a reflection of the position at law. Whether that was a wrong turn is not now open to debate, for the fraudulent claims rule is established at the highest level. But the scope of the fraudulent devices rule is less well established, and the question under consideration is as to the consequences which the law should visit on parties to a contract who could have provided that a valid claim should be forfeited if tainted by a fraudulent device used in support of it, but have chosen not to do so. That would suggest that the approach should be to interfere with the parties' bargain no further than the policy of the law strictly requires.
169. The second anomalous aspect of the rule is that whilst any fraud is reprehensible and is to be discouraged, it is not normally the function of the civil law to provide such deterrence. The fraudulent claims rule in insurance is a form of penal non damages which, so far as I am aware, has no parallel elsewhere in the common law. Yet deliberate exaggeration of claims, for example in the context of personal injuries, occurs regularly and does not attract the sanction of loss of that part of the claim which is valid: see for example the observations of Lord Clarke in Summers v Fairclough Homes at [32], of Park J in Agapitos v Agnew at [58], and of Smith LJ in Ul-Haq v Shah [2010] 1 WLR at [17]. Fraudulent claiming is not a problem peculiar to claims by assureds under contracts of insurance. His Honour Judge Hawkesworth QC (quoted in Ul-Haq at [13]) has said the following in relation to road traffic accidents, where the claim is in tort and does not arise under a contract of insurance, and the defendant is, nominally at least, not an insurer:

"Unhappily such fraudulent claims are now legion. They occupy the court time of District Judges and Circuit Judges in West Yorkshire literally week in and week out. My own judicial experience reflects, I have no doubt, that of many of my brethren throughout the country. Just about every variant of a fraudulent claim comes before the court, including deliberately staged collisions, damage caused to vehicles which have never been in collision at all, claims deriving from the most trivial touching of vehicles, and claims in which a driver will assert that his car was carrying other members of his family including his children, when in fact none were present but all of whom have reported to a hospital or their General Practitioner that they have been injured, and who are then able to produce an apparently independent expert's report confirming the fact of such injury. The cost to the insurance industry and to other honest policy holders must be very substantial. In addition, and of more relevance to these proceedings, the cost in court time in trying such cases is very high, with the added knock-on effect of casting suspicion onto many genuine claims so that claimants are put to proof of their legitimate and genuine claims for compensation when in other circumstances they might not have been called upon to do so."

170. These are all examples of fraudulent claims which are in whole or in part invalid, rather than valid claims supported by a fraudulent device. Yet the common law makes no provision for the forfeiture of the valid element of such claims outside the insurance context, let alone those which are wholly valid. If the anomalous rule is to be extended to fraudulent devices used in support of valid claims, it is to my mind important that it should not itself be allowed to be used as an instrument of injustice.
171. As a result of all these considerations, I would be strongly attracted to a materiality test which permitted the court to look at whether it was just and proportionate to deprive the assured of his substantive rights, taking into account all the circumstances of the case. The blunt instrument of a relatively inflexible test of materiality, reminiscent of the old latin tag "fraus omnia corrumpit", must surely be capable of yielding to a more proportionate response, which can meet the varying circumstances of each case. Such a flexible approach has been introduced in Australian law in section 56 of the Insurance Contracts Act 1984 (which however does not apply to marine insurance) which allows a court, in relation to claims which are partly false and fraudulent, to order payment of an amount which it thinks just and convenient where forfeiture of the part of the claim which is valid would be harsh and unfair.
172. Support for such an approach can be found in the decision of the Supreme Court in Summers v Fairclough Homes [2012] 1 WLR 2004. In that case the claimant suffered an accident at work and sued his employers. The claim was defended by the employer's liability insurers who were the real party at interest. The claimant obtained judgment on liability with damages to be assessed. For the purposes of the quantum hearing the claimant signed a witness statement describing the debilitating effect of his injuries and the extent to which they had rendered him unable to work. His initial schedule of loss claimed over £800,000. The employer's liability insurers discovered by means of covert surveillance that this was fraudulent and involved a gross exaggeration of his injuries, and sought to strike out the claim as an abuse of process. The trial judge held that the exaggeration of the injuries, and the consequent claims for loss advanced in successive schedules of loss, were dishonest and that he had deliberately lied as to his ability to work. The judge assessed the value of the valid claim at £88,716.76. He declined to strike out that claim, on the grounds that he was bound by the decisions of the Court of Appeal in Ul-Haq v Shah and Widlake v BAA Ltd [2009] EWCA 1256 which decided that there was no power to do so. The Supreme Court overruled those decisions and held that there was power to strike out claims as an abuse where they were fraudulent in any respect, but declined to do so in respect of Mr Summers' claim. There was no appeal against the amount which the trial judge had assessed as being the extent of his valid claim.
173. Lord Clarke, giving the judgment of the Court, recorded at [24] that it was accepted on behalf of the claimant that in making the false statements of truth and in presenting a dishonest case as to the effect of his injuries and on quantum, the claimant was guilty of a serious abuse of process. At [33] he stated:

"We have reached the conclusion that notwithstanding the decision and clear reasoning of the Court of Appeal in Ul-Haq, the court does have jurisdiction to strike out a statement of case under CPR 4.3(2) for abuse of process even after the trial of an action in circumstances where the court has been able to make a proper assessment of both liability and quantum. However we further conclude, for many of the reasons given by the Court of Appeal, that, as a matter of principle, it should only do so in very exceptional circumstances."

174. Having repeated at [36] that the court would only strike out a claim as an abuse at the end of a trial in very exceptional circumstances, Lord Clarke went on at [43]-[44] to approve paragraph 72 of the judgment of the Court of Appeal in Masood v Zahoor [2010] 1 WLR 746, to the effect that it would only do so where the abusive conduct was such that the claimant had forfeited his right to have his claim determined, which would be "a very rare case". At [46]-[49] he referred to a claimant's rights under Article 6 of the European Convention on Human Rights requiring the court to examine the circumstances of the case scrupulously in order to ensure that to strike out the claim would be a proportionate means of achieving the aim of controlling the process of the court and deciding cases justly. At [49] he said:

"The draconian step of striking a claim out is always a last resort, a fortiori where to do so would deprive the claimant of a substantive right to which the court has held he was entitled after a fair trial. It is very difficult indeed to think of circumstances in which such a conclusion would be proportionate. Such circumstances might, however, include a case where there had been a massive attempt to deceive the court but the award of damages would be very small."

175. At [61] he concluded that

"The test in every case must be what is just and proportionate. It seems to us that it will only be in the very exceptional case that it will be just and proportionate for the court to strike out an action after a trial."

176. I find this decision instructive in the context of fraudulent means or devices deployed by an assured prior to the commencement of litigation. The fraudulent claims rule itself, and its extension to fraudulent devices, is restricted to the period prior to the commencement of litigation (see Summers v Fairclough Homes at [29], The Star Sea at [77] and Agapitos v Agnew at [47]-[53]). But the policy of deterrence of fraud, which is said to underpin the rule, is not diminished at the point of time when the assured brings the claim before the court. On the contrary, one might have thought that the policy of deterrence was at its strongest after the commencement of litigation, when such fraud amounts not only to an attempt to deceive insurers but additionally to an attempt to manipulate the court's process. If in the latter context an assured is not to be deprived of his substantive rights save in the very rare and exceptional circumstances where that is a just and proportionate sanction for his fraudulent conduct, it is difficult to see a justification for applying a lower threshold to the same conduct in presenting his claim prior to the commencement of litigation.
177. In these circumstances, and with the very greatest diffidence and respect, I would hesitate to follow the tentative proposals suggested by Mance LJ in Agapitos v Agnew. My own view would be that if the law is to extend the draconian effect of an anomalous rule, applicable only to insurance claims, and then only prior to the commencement of litigation, to striking down wholly valid claims, the policy of the law should be to require at least a sufficiently close connection between the fraudulent device and the valid claim to make it just and proportionate that the valid claim should be forfeit. The law does not provide in this context that the end always justifies the means; but nor should it say that any dishonest means which are more than de minimis should deprive a litigant of his just ends. What will be just and proportionate will depend upon the circumstances of each case, which may vary considerably.
178. Can such a conclusion be reconciled with the decision of the Privy Council in Stemson v AMP General ? I do not believe that it can. In that case an assured brought a claim for property damage arising from his house burning down. It was common ground that the fire was arson and the main issue was whether the assured was party to the arson. The trial judge held that he was, and the decision was upheld by the Court of Appeal of New Zealand and the Privy Council. The trial judge held that the claim also failed on the additional ground that it was supported by a fraudulent device. The assured had made a dishonest statement to the insurer's claims investigator that he had not previously attempted to sell the property or even considered putting it on the market. The trial judge found that he had been to see a real estate agent and asked him to put the property on the market, in a serious attempt to sell the property, shortly before the fire. He described this as a knowingly false statement "of very distinct significance in the context of the investigation into this fire", and held that this provided a separate ground entitling the insurer to avoid liability under the policy quite independently of the arson.
179. Lord Mance, giving the judgment which represented the unanimous opinion of the Board on this point, upheld this ground. He recorded at [35] that counsel for the assured did not challenge any of the statements of principle which, as Mance LJ, he had set out in Agapitos v Agnew and concluded at [36]:

"On that basis, the Board has no hesitation in upholding the judge's conclusion that, quite apart from any question of arson, the respondent was entitled to reject the appellant's insurance claim (as it did) on the ground that the appellant had sought to promote it in May 1992 by lying to the respondent about the position, and his state of mind, regarding any attempt to sell the house prior to the fire. The materiality of such matters to the respondent's investigation and evaluation of this insurance claim is not challenged and is, as the judge indicated, obvious."

180. In that case the assured's claim was wholly fraudulent, so that the Board was not considering the point in the context of a valid claim. Nevertheless the unspoken assumption underlying the conclusion that it was an additional ground for declining the claim is that it would have been sufficient to defeat the claim if the householder had been the innocent victim of a malicious arsonist. In treating the test as one of the materiality to the insurer's investigation and evaluation of the claim, the Board were applying the test proposed by Mance LJ in Agapitos v Agnew.
181. Decisions of the Privy Council are not binding on me, but they command the very greatest respect. The decision in Stemson was reached without any argument on this point, but was reached unanimously and without hesitation. The reasoning of Mance LJ in Agapitos v Agnew, whilst strictly speaking obiter and so not binding on me, was a full and considered analysis after argument on the point, and was agreed with by the full court. By contrast, I have not had full argument on the point. In those circumstances I shall apply the test of materiality proposed by Mance LJ in Agapitos v Agnew and ignore my own tentative inclination to apply a test of what is just and proportionate.

Investigation of the casualty and presentation of the claim

182. On 29 January 2010 Gertjan Kornet attended the Vessel at Gdynia in the immediate aftermath of the casualty, as did Han Gravendeel of Doldrums, the surveyors appointed on behalf of the Underwriters. On 30 January 2010 the crew signed a joint "declaration" pursuant to a request from Gertjan Kornet that they talk together and prepare a joint statement. The declaration dealt only with the narrative of the casualty, and started its account of events with the bilge alarm being heard by the Engineer at 2058.
183. Water was pumped from the Vessel between 30 January and 2 February, and diving inspections took place on 30 and 31 January. Arrangements were made to discharge the cargo, which was completed on 20 February, and the Vessel moved to drydock between 22 and 24 February, where surveys and inspections were undertaken by Gertjan Kornet and Mr Gravendeel amongst others. On 26 February the Vessel left Gdynia under tow for Bremerhaven, where she arrived on 4 March for permanent repairs.
184. On 24 February Gertjan Kornet prepared a report headed (in translation) "EXPLANATION OF THE CAUSE OF THE LEAKAGE DC MERWESTONE" which was emailed to Chris Kornet and to Carins, the Owners' Dutch broker. The report concluded that the initial suspicion that the stern screw was the source of ingress was misplaced; and that the sealing and pressure testing of compartments through which the leak might have come pointed to the bowthruster room as the source; but that the question which "everyone was wondering" was whether that could have caused the engine room to fill so quickly. It was in this context that Gertjan Kornet recorded his discussions with the crew, which referred to the bilge alarm going off at 2115 but did not address whether an alarm had gone off earlier in the day. The report was based on what were described as some "fairly critical questions" asked of the crew who remained on board, which included the Master and Engineer.
185. The claim was presented to Gerling through Carins, the producing broker, and Jardine Lloyd Thompson ("JLT"), the placing broker. At the request of Gerling, and with the cooperation of the Owners, arrangements were made for the Underwriters' Admiralty Manager, Mr Billowes of Ince & Co, to attend the Vessel and to interview the crew. He attended the Vessel on 2 April 2010 and interviewed the Masters (Capt Loosman and Capt Lilipaly), the Chief Officer and the Second Officer in Holland on 1, 6 and 7 April 2010, and the Engineer in Manila on 12-13 April 2010. Gertjan Kornet and Kees Parel appear to have sat in on at least part of the interviews at Werkendam, but the Owners did not have a representative present throughout and were not provided with notes or statements of the evidence taken.
186. On 16 April 2010 Mr Billowes sent an email to Werner Schurink at Carins asking him to ask the Owners to provide 36 categories of documents and in addition "Owner's explanation for the cause of the ingress, the spread of the ingress into the engine room and the reason the ingress could not be controlled by the ship's pumps." It was forwarded by Mr Schurink to Chris and Gertjan Kornet, Ton Schootens the adjuster, and Han Gravendeel of Doldrums, the Underwriters' surveyor, with the suggestion that they all work on it "full steam ahead".
187. The response came in a letter dated 21 April 2010 signed by Chris Kornet which was forwarded to Mr Billowes by Carins on 23 April 2010. It included the following passages:

"After further internal investigation due to the problems onboard DC Merwestone we will inform you as follows.

Facts

Floating visual by crew 28 January 2010 around 21.00 hrs.

Bilge alarm from bow thrusters and engine room have the same alarm point so nobody can see exact which area gives alarm.

First alarm was going off around noontime due to rolling vessel (weather SSE 7) vessel rolling. No investigation in bow thruster room.

When bilge alarm activate during rolling nobody has controlled this due the rolling.

…………….

Due to the weather circumstances the first bilge alarm has go gone of off [sic] in the morning. If there was leakage or due to rolling we are not sure.

………..

Conclusion

1. starting time of leakages is around 13.00 hrs on 28 of January"

188. On 1 June 2010 the two case handlers at Ince & Co sent an email to Mr Schurink stating that they had reviewed the documentation and information provided by Owners, including the letter dated 21 April. Ince & Co sought clarification in the form of questions on 13 topics. Question 2 was as follows:

"(2)(a) What evidence do Owners rely on that the bilge alarm operated around noon on 28 January?

(b) Who heard it and when?

(c) Who acknowledged it?

(d) Who took the decision not to investigate it?"

189. The response came in a letter to Gerling signed by Chris Kornet dated 27 July 2010 giving the following answers:

"2(a) Owners were told by the vessel's captain that the bilge alarm operated around noon on 28 January.

2(b) Owners believe it was the watchkeeper/Master on the bridge and the Engineer in the engine room. They thought the alarm was routine. For further information, Owners suggest Underwriters refer to their own interview notes.

2(c) Owners do not know. They suggest that Underwriters refer to their own interview notes.

2(d) Owners assume it was the Master. It was in any event not possible to go forward. Temperatures were minus 10 to minus 20, and the vessel was shipping heavy seas and water over the fore deck. Furthermore, Underwriters have interviewed the crew members."

190. The Report of Captain Barker of TMC dated 27 January was forwarded through the broking channel to JLT on 22 February 2011. At paragraph 3.3 it included the following as part of the narrative of the casualty:

"Whilst at sea, in SSE'ly Beaufort force 7 winds, at about noon on 28^th January the bilge alarm sounded. There is a bilge alarm fitted in the Bow Thruster Room and the Engine Room however the alarm is connected to a single buzzer/warning light in the Engine Room and on the Bridge. If the alarm sounds it is necessary to verify the bilge levels by visual inspection. The vessel was rolling, and due to the adverse weather (which would have been just aft of the port beam), the alarm was accepted without anyone going forward to inspect the Bow Thruster Compartment. The crew's experience was that the alarm is sometimes triggered when the vessel was rolling and they were not unduly concerned. The bilge warning light would have remained on."

191. It will be recalled that by the time of the trial the Owners no longer contended that an alarm had gone off before 2058 on the 28 January 2010.

Preparation of the 21 April and 27 July letters and TMC report

192. As indicated above, Chris Kornet received by email from Carins, on the evening of Friday 16 April 2010, Ince & Co's request for the Owners' explanation for the cause of the ingress, the spread of the ingress into the engine room and the reason the ingress could not be controlled by the ship's pumps. On the morning of Tuesday 20 April at 1053, he sent his draft response to Mr Schurink, which contained the relevant passages about the noon alarm in materially identical form to those in the letter as sent on 23 April.
193. Chris Kornet's evidence in cross examination was that before preparing that draft, he spoke to the Master on the telephone for 5 to 10 minutes, on 20 April, and that the Master told him during that conversation that the alarm had gone off at around noon, although not that it had been ignored as attributable to the Vessel rolling. The watchkeeping arrangements on the Vessel were for watches from 1-5, 5-9 and 9-1. The Master took the 0900-1300 watch and so would have been on watch at around noon. Chris Kornet accepted in his evidence that the explanation for not investigating the alarm was speculation; it was based on the Master's confirmation that there had been an alarm and what he, Chris Kornet, thought the most likely explanation based on his own understanding of the conditions. His understanding that the weather at noon on 28 January 2010 was SSE 7, as recorded in the 21 April 2010 letter, originated in the deck log, which records that as the wind speed and direction on that day, together with rough sea and moderate swell. That entry is for a 0800-1200 watch. For a noon to 1600 watch the relevant entry is wind N 3, sea slight, swell low. Chris Kornet was cross examined on the basis that the log showed that the weather at noon was slight seas and a low swell, but that is not the obvious reading of the log. The log was on a pre printed form which assumed normal watch changeover times of noon/midnight, 4 o'clock and 8 o'clock, whereas on the Vessel the watches changed at 1 o'clock, 5 o'clock and 9 o'clock. Although one might expect the wind and sea state recorded to pertain to the beginning of the watch, the entries record the vessel's position at the end, rather than the beginning, of the watch, so that the 0800-1200 entry has the GPS position at noon. The 0800-1200 entry records the pilot coming on board at Klaipeda at 0800, casting off lines at 0950, and the pilot disembarking at 1050, from which it would have appeared that at the beginning of the watch the Vessel was still in Klaipeda and not subject to rough seas and a moderate swell which were recorded against that watch. Even if that is not the correct reading of the log, judging from the photocopies in the bundles, which do not have the entries on the same page as the time to which they relate, it would be an easy mistake to make to treat that as the noon weather and sea state. I accept that when preparing his responses, Chris Kornet believed the weather at noon to have been SSE7 with rough sea and moderate swell as a result of his reading of the deck log entry, whether that was a correct reading or a misreading, or a misremembering of the document.
194. Chris Kornet's draft was promptly passed on by Mr Schurink to Mr Gravendeel with a request to consider whether the theory of ingress was plausible from a technical aspect. Mr Gravendeel responded later that day with a calculation that the cable duct openings were equivalent to a six inch diameter pipe, and other points not here relevant, which were passed to Chris Kornet for preparation of his final draft. On 21 April Mr Gravendeel said that he would take a critical look at one or two things when he was on board the Vessel the following day at Bremerhaven in relation to the filter lid; he did so and reverted on 23 April. There was no input or change to the passages dealing with a noon alarm prior to it being sent, save for an apparent intention to correct "go of" to "gone off" which suffered a word processing mishap of addition rather than substitution. The 21 April letter, as sent on 23 April, was drafted by Chris Kornet on his own, not in conjunction with his brother Gertjan. The question about the filter lid was dealt with separately in a letter dated 23 April.
195. When the 1 June email was received by Mr Schurink, it was promptly forwarded to Chris Kornet and Gertjan Kornet the same day, with Mr Schurink's comment that he "could only see one really annoying question, namely 10A". That was a question asking who at K&Z had authorised the cutting of the ballast line.
196. Gertjan Kornet prepared a response for internal consideration the same day which answered Questions 2(a) to (d) compendiously as follows:

"No crew member report this. But according the last ism report it was working so theoretical it have to be working. Also at that time. We have to be aware the report which was send by owners was a report about the situation which most probably happened, not the one who happens. Looking to the weather forecast and wind direction the vessel was rolling at that time, so from experience we have the idea what has happened. No crew member has report this."

197. This response was forwarded to Chris Kornet that afternoon. Chris Kornet's reaction was "Questions, but when is my question going to be answered" meaning when were the Underwriters going to make a decision on payment of the claim. He did not at that stage, or indeed at any subsequent stage, take up with Gertjan Kornet the latter's assertion that no crew member had reported a noon alarm. He provided his own additional comments to Mr Schurink on 2 June without addressing question 2.
198. At 1248 on 2 June Mr Schurink coordinated the answers into a single document and sent it to Chris and Gertjan Kornet emphasising that it was a very important document and that "you as shipowners have to back it completely". It answered question 2 as follows:

"2a. According to the last ISM report it was working properly

2b. no crew member reported this to us

2c. nobody

2d. we cannot say if this decision was made at all as no comment was made on this alarm."

199. This draft was also forwarded to Mr Gravendeeel for his advice from a technical aspect, which resulted in a slight change to the answer to question 6. When Mr Schurink forwarded a revised version to both Chris and Gertjan Kornet on 3 June he left the answers to questions 2(b)(c) and (d) blank. His covering message recorded that Chris Kornet was going to give him the correct answers and commented "after all on 21/4 we wrote "alarm was sounded at noon"". This drew attention to the difficulty in reconciling what had been drafted as a result of Gertjan's Kornet's response with what Chis Kornet had said in the 21 April letter.
200. Chris Kornet's response was to draft a form of wording which he sent by email to the Master at 1515 on the same day, 3 June, in the following terms (correcting some of the typographical errors and linguistic and syntactical infelicities in the original):

"In the morning a bilge alarm went off do you know roughly at what time and what did you do and why? If I say the following can you go along with it think about what you stated earlier. [this part was in Dutch, the remainder was in English]

Please be aware the report we sent was based on brief information from various persons. Based on the information we sent the message as what could happen based on this info

When vessel starts rolling after leaving Klaipeda we had bilge alarm. This is a combined alarm between bow thruster room and engine room. We check the engine room but this looks in normal condition. Due to the rolling we take it as a fact bilge alarm was from bow thruster room and activated due to rolling.

Alarm sound is on the bridge and in the engine room so watch keeping person should have heard this

Acknowledgement has to be done in the engineroom

We can not give a clear answer on the last questions. It should be motorman or officer of watch."

201. In his second witness statement which he verified in evidence, Chris Kornet said that he had convinced himself at around this time that there had been an alarm earlier in the day, and he had spoken to the Master "on several occasions" to say that he could not understand how there could not have been an alarm; and that eventually the Master said something like "I guess there must have been an earlier alarm". Chris Kornet's statement said "I took that to mean that he himself had heard the alarm."
202. The passage in the email asking the Master to "think about what you stated earlier" is not a clear pointer to Chris Kornet already having had a conversation with the Master on this topic by that time: it might be an exhortation to ensure that what the Master confirmed was consistent with what he had said when interviewed by Mr Billowes of Ince & Co. It does not suggest that in any earlier conversation the Master had confirmed a noon alarm: it asserts a morning alarm and asks the Master what time it went off.
203. A little under four hours later, on the evening of 3 June, Chris Kornet sent an email to Carins saying that he had checked the information set out with the Master, and setting out answers in materially identical terms to the text of what had been sent to the Master earlier in the afternoon to see if he could "go along with it". Chris Kornet's evidence was that he spoke to the Master in the interim and the Master confirmed he was happy with the wording, although that was reconstruction rather than recollection because he freely accepted that he could no longer recall the specific conversation.
204. In the Master's evidence to me, the Master accepted that the alarm had not gone off in the morning; he also appeared to accept in cross examination that he had never had a recollection at any stage that it had. However I formed the impression that he did not really understand the difference between these two concepts. He was asked whether he had told Chris Kornet that it might have happened but that he had no recollection, to which he replied "Yes, something along those lines". He rejected the suggestion that Chris Kornet was very insistent on this account of events. The inherent probabilities are that Chris Kornet did have a further communication with the Master following the form of wording sent to him for approval; there would have been little point in sending it to him unless a response was insisted upon. Chris Kornet had copied in Mr Schurink with his email to the Master before he had a response and said "we will confirm". I conclude that there was such a conversation in which the Master confirmed that he would support the account set out in Chris Kornet's email. My assessment of what is most likely to have happened is that in this conversation, Chris Kornet made clear to the Master his view that the alarm must have gone off around noon, and that the latter ultimately conceded that there must have been such an alarm, which Chris Kornet took to be confirmation from the Master that he must have heard it because it was during his watch. The Master gave Chris Kornet confirmation that he would support the account sent to him in Chris Kornet's email, which included the explanation for ignoring the alarm as attributable to the Vessel rolling.
205. On 7 June Mr Schurink adapted the answers he had been sent into a further draft and sent it back to Gertjan and Chris Kornet with the following rubric:

"The further edited version. Read your corrected item two (a/b/c/d) very carefully again! If I see things from the point of view of the insurers, I would still have a question about why the bow thruster room was not checked? Possible reply could be weather on board too bad and therefore dangerous?????"

206. In fact the account which Chris Kornet had sent to Mr Schurink confirming what he had checked with the Master did include a reference to the strong wind and temperatures of minus 20, but Mr Schurink had not included it in his redraft. Chris Kornet replied a week later on 14 June:

"We as owners have asked the crew why they have not checked the bow thrusters.

Answer weather was very bad se 7/8 see log book. Vessel was rolling water covers deck and hatches tempeture -20 degrees, so it was not safe for crew to go outside"

207. There is no suggestion that there was any new enquiry of the crew which prompted this response. Chris Kornet's response was then added to the current draft as part of answer 2(d).
208. There is no documentation in the brokers file or from the Owners which reveals substantial discussion about the terms of the answers over the following six weeks. When the final version was signed by Chris Kornet on 27 July 2010 and sent to the Underwriters, the answers to question 2 had changed to the form I have set out above. These included

(1) confirmation that the Master had said that he had heard the noon alarm;

(2) confirmation that the Owners believed that it had been heard also by the Engineer;

(3) confirmation that it had not been investigated because they thought it was routine; and

(4) reference to the notes of interviews taken by Ince & Co on behalf of the Underwriters; this latter reference suggests that Chris Kornet believed that his account would be at least consistent with, if not corroborated by, what the crew had previously said during those interviews.

209. It is not clear what source of information was used by Captain Barker in compiling the TMC report of 27 January 2011, and the question was not explored at the hearing. The account set out in the report identified, as is usual, the factual basis on which he was expressing his conclusions. In this context it treated as fact the sounding of the noon alarm, and gave the rolling of the Vessel as the reason for the crew's failure to investigate it. Chris Kornet could not remember whether the report had been sent to him, but accepted it was possible. In my view it is likely. By sending the report in support of the claim, the Owners were putting those matters forward as their explanation of what had happened on 28 January 2010, and I infer that Chris Kornet was aware that that was taking place.

Chris Kornet's state of mind

210. There are two relevant features which form important context. The first is that as a result of legal advice from Carins, Chris Kornet believed, when preparing his responses in the 21 April and 27 July letters, and sending the TMC report, that it would assist the claim if he minimised any opportunity for attributing fault to the Owners, rather than the crew, in relation to the cause of the casualty. That is apparent from the following:

(1) In an email of 26 March 2010, sent in the context of the forthcoming interviews of the crew by the Underwriters' solicitors, Mr Schurink advised Chris and Gertjan Kornet as follows:

"This email is strictly confidential, please do not forward.

….. Important; nobody can say "the insured is to blame" (in the sense of the shipowner, let's say "Werkendam"). That would unleash a discussion about the final sentence in 6.2.5 of the policy conditions, namely;

Provided such loss or damage has not resulted from want of due diligence by the Assured, Owners or Managers.

On the other hand, it will probably not come to that; after all, we've got nothing to hide.

However, the men (crew) must not screw things up at the expense of the office. Admit mistakes. Everybody makes mistakes once in a while, especially if there is a bit of panic, and errors by the crew (crew's negligence) is covered."

Chris Kornet accepted in evidence that he regarded this as legal advice, and that he understood that it was to the effect that the Owners and their office should be distanced from the casualty.

(2) The advice was reinforced by a further email from Carins on 16 April in the context of the request in Ince's email for an account of the cause of ingress:

"We have let the list settle. I assume we can answer everything. We have to be careful with the first two questions. On the one hand, we have to give an "adequate", satisfactory answer, on the other hand, the owner does not have to be aware of (important) defects …. "

(3) In similar vein was an email of 21 April from Mr Schurink to Chris Kornet, in which the broker addressed the other question asked by Ince & Co in their email of 16 April, which requested an explanation of the fire pump filter cover being found to be loose. Mr Schurink emphasised that it didn't matter what the answer was (so long as it was the truth) if it was a mistake by someone other than the owners.

(4) On 24 May 2010 Mr Whaley of JLT summarised his assessment of the legal position in an email which was forwarded to Chris Kornet in the following terms:

"Further to our exchange of emails over the weekend, I have reviewed the Doldrums report once again and based on their conclusions the cause of loss is Frost dmg due to the extreme weather conditions experienced in Northern Europe in Jan 2010, exacerbated by the actions of the crew once the Ingress of water commenced. We have the benefit of the Additional Perils cls, which basically provides all risks cover with specific exclusions. Unless Gerling/Ince can demonstrate that Owners did not act with due diligence and knowingly allowed the vessel to put to sea in an unseaworthy condition or the damage is due to wear and tear, then there is a claim under the policy, lack of due diligence is a very difficult defence to run under English law, unless of course the vessel was in breach of Class/ISM etc."

(5) In the light of these exchanges Chris Kornet would have understood why Mr Schurink described Ince's question 10A, asking who at K&Z had authorised cutting the ballast line as "annoying"; it was because it focussed on the involvement of the Owners.

(6) On 7 June 2010 the Owners instructed solicitors, Holman Fenwick Willan LLP. It is to be inferred that they gave advice in relation to the policy terms.

211. The second element of context is that Chris Kornet became increasingly frustrated that the Underwriters were not paying the claim. By the time the 21 April letter was finalised and sent on 23 April, the yard at Bremerhaven had indicated that they would not release the Vessel without some payment, or security, for the substantial cost of repairs. Chris Kornet was anxious for a swift decision from the Underwriters to make a payment on account of the claim, to enable the Vessel to be released. As the months passed and the brokers pressed for a decision, those on the Owners' side became increasingly convinced that the delay was not promising and that the Underwriters would require further information. In the week following receipt of Ince & Co's 1 June questions, Chris Kornet's frustration with the Underwriters' failure to pay the claim had become more pronounced. The Vessel was still in the yard, and the Owners needed the insurance proceeds to provide a bank guarantee for €1.2m which the yard was demanding for release of the Vessel. In his evidence Chris Kornet repeatedly emphasised that his main preoccupation in the week he received the 1 June questions was to get the Vessel released from the yard. He wanted to be paid by the insurers, not to have to answer further questions. His frustration was palpable in his evidence, and was exemplified in an answer he gave in cross examination, without any intended humour or irony, that he regretted that on his computer "there was only one button which was not there, and that was the button F*** off I want the money". At the end of that week at the beginning of June he managed to secure more favourable terms on which the yard would release the Vessel, but his formulation of the answers to question 2 took place in that week against the background of his frustration with the Underwriters.
212. I cannot accept Chris Kornet's evidence that he spoke to the Master on 20 April before preparing his letter dated 21 April, or that he was told by the Master then that the alarm had gone off at around noon. There is no reference to this conversation in his second witness statement, which was provided pursuant to an order to respond to the allegation of fraudulent presentation of the claim in the Underwriters' Re-Amended Defence, and was expressly relied upon in support of the Owners' pleaded case. The contents of this witness statement were confirmed in his evidence in chief. The statement deals in detail with the circumstances leading to his letter dated 21 April without any reference to any conversation with the Master on 20 April. It refers instead to conversations with the Master in June, which is consistent with and corroborated by the documents identified above. If there had been a conversation on 20 April it is most improbable that he would not have told his brother Gertjan, who could not then have drafted the initial response in the terms he did on 1 June saying that no one had reported a noon alarm; and he would have been bound to have corrected what would have been an important and obvious mistake by Gertjan Kornet, when he received that draft; whereas he made no adverse comment on it in his responses on 1 and 2 June, and went back to the Master on 3 June with an account which needed the Master's confirmation. Chris Kornet's evidence of a 20 April conversation came unheralded for the first time in cross-examination and after the Master had been cross-examined and released, so that he could not be asked about it. But the Master's evidence casts some further doubt on it. The Master was on holiday in Indonesia for four weeks after his interview with Mr Billowes on 6 April, and when on holiday was not in contact with his office or at best "hardly ever". Moreover if the Master had stated on 20 April that there had been an alarm at noon, it is improbable that there would not have been a discussion about whether it was investigated and to what it was attributed, yet Chris Kornet says he was told nothing on this occasion by the Master about the crew's reaction to the alarm.
213. In these circumstances my assessment of Chris Kornet's state of mind at each stage is as follows.
214. When he prepared the 21 April letter, his principal focus was on addressing a source and rate of ingress which was realistic. In doing so he reached a conclusion that the water in the bowthruster room was probably sufficient to trigger the alarm at around noon. He was conscious that that required an explanation as to whether the alarm had gone off at around noon, and if so why it had not been investigated. He also believed that it would not assist Owners' case if the alarm had failed to trigger, because it would point to the defective condition of the Vessel rather than crew negligence, and that an explanation that the alarm had gone off, and been ignored, would be preferable to one which involved the alarm not sounding. He had had no evidence from the crew that an alarm had gone off around noon, or been ignored, or as to what the reason was for ignoring it. He genuinely believed that if the alarm had gone off, it would probably have been ignored as a result of the weather conditions he believed the Vessel was encountering; ignoring alarms being triggered by rolling, with little water present, was from his experience a plausible explanation for ignoring the alarms. As he accepted in evidence, this was an explanation he himself hit upon as a matter of speculation, rather than based on anything the crew had told him. His account was given to fit his theory of ingress without making any attempt to check whether it was supported by anything the crew had said about the alarm going off. He suspected that his account might not be supported by the crew, because he knew that Gertjan Kornet had asked "critical questions" of the crew in preparing his report, and he had not been told by his brother or anyone else that any of the crew had heard an alarm earlier in the day. He genuinely believed that his account of the noon alarm and the crew ignoring it was a realistic explanation of events, but he was reckless whether it was supported by the crew's recollection because he did not want the absence of confirmation from the crew to get in the way of an explanation which involved no fault on the part of the Owners or managers. He was indifferent to whether the Master or any other crew member had previously said that the alarm had gone off.
215. Mr Karia QC submitted that a finding of fraud was inconsistent with the explanations being copied to Mr Gravendeel, the Underwriters' surveyor. But the latter's involvement was expressed to be one of considering whether the explanation for the ingress was plausible from a technical point of view. He would not be expected to check or consider the Owners' account of what the crew said about a noon alarm. His involvement gives no reason to doubt that Chris Kornet's state of mind was as I have found.
216. When he put forward the answers in the 27 July letter, Chris Kornet's state of mind was as follows. When the specific questions were posed in paragraph 2 of Ince & Co's email of 1 June, he recognised that the account he had given had not been based on evidence from the crew but that the 21 April letter had given the impression that it had. He was therefore anxious to answer the question by suggesting that it was supported by what had been said by a member of the crew, so as not to appear to have said anything misleading in the letter. To this end he asked the Master whether he could live with an account which involved the Master having heard the alarm at noon, and the crew having ignored it as attributable to the rolling of the vessel. He received confirmation from the Master that he could support that account of events. He took this to be confirmation by the Master that he had heard a noon alarm. Accordingly his confirmation that the Master had said that he had heard the noon alarm was true; and he honestly believed that it had been heard also by the Engineer and that it had not been investigated because they thought it was routine. Although he did not check the latter aspect with the Engineer or the crew, he was not reckless as to its truth: it had been confirmed by the Master's confirmation that he could support the account set out in the email Chris Kornet sent to him. Chris Kornet believed the answers given in the 27 July letter to be true, and that his answers would be at least consistent with, if not corroborated by, what the crew had previously said during Mr Billowes' interviews.
217. For similar reasons the account given in the TMC report was believed by Chris Kornet to be true.
218. In these circumstances the Underwriters' fraudulent device defence only falls to be considered by reference to the 21 April letter.
219. I should record that I derived no assistance from the considerable debate about the passages in the 21 April letter, and subsequent documents, dealing with the decision to cut the ballast line, whether it was the Owners or the crew who first suggested it, and whether the Owners authorised or instructed it. The nuances deployed in interpreting the language used by the Owners in the letter, and in evidence, were more sophisticated than the circumstances warranted. English was not the first language of Chris or Gertjan Kornet. I acquit the Owners of any dishonesty or intention to mislead in those passages.

Was the 21 April letter a "fraudulent device"?

220. The letter addressed in some detail Chris Kornet's theory as to the source and volume of water ingress at various stages, concluding that the source of ingress was a combination of the emergency fire pump and possibly ingress of ballast water. In many respects it was self evidently his theory as to the course and cause of the casualty, rather than recitation of underlying evidence or raw data. The cross examination proceeded on the footing that there was a clear dichotomy between a theory as to what might have happened on the one hand, and a factual account of what had happened on the other. But in my judgment no such clear distinction can be drawn. It is a common experience of marine casualties that what happened is a matter of contention and debate based on numerous indicia, of which human recollection is only a part, to be considered together with scientific and engineering expertise and the inherent probabilities informed by maritime experience. The indicia are often not all of a piece. There is no clear boundary between fact and theory.
221. On the other hand the relevant passage in the letter came under the heading "Facts", and was said to be "after further internal investigation", which in relation to whether an alarm went off, and if so why it was ignored, would reasonably have been taken to include inquiries of the crew. Read as a whole, the letter contained an assertion that one or more members of the crew had claimed to have heard the alarm going off at about noon and had given an explanation that the alarm had not been investigated because attributed to the rolling of the Vessel. Making all due allowance for the fact that English is not Chris Kornet's first language, in which he is proficient but not completely fluent, this must have been the impression the letter was intended to give. I have little doubt that this was how he intended and understood it. That was why when subsequently faced with Ince & Co's question 2, and recognising that the 21 April letter was not consistent with the answers his brother had drafted, he approached the Master for confirmation of the account; and that is why in his evidence to me he felt the need to invent a conversation with the Master on 20 April. In this respect the letter was false and misleading. In this respect Chris Kornet had no grounds to believe it was true, and was reckless whether or not it was true. It was an untruth told recklessly in support of the claim.
222. Mr Karia QC submitted that the letter of 21 April was not written with a view to improving the Owner's prospects of settlement, but merely reflected Chris Kornet's attempt to explain the rate of water ingress. That does not adequately reflect the circumstances in which it was written. Ince & Co were investigating the casualty on behalf of Underwriters and asked for the information in their email of 16 April "in order to complete our report to underwriters". Prior to receipt of the email, the Owners had been pressing Gerling through the broking chain for payment of the claim. In an email of 23 April Mr Schurink confirmed to Chris Kornet that the letter was to go to Ince & Co so that they could complete their report to underwriters. It was clear from the questions asked by Ince & Co in their email of 16 April that the Underwriters had concerns as to the underlying condition of the vessel, the assured's knowledge as to the underlying condition, and why the ingress of an apparently small quantity of water could have given rise to such a major casualty. Chris Kornet was keen to secure a payment on account of the claim, at least, in order to ensure the release of the vessel from the yard after the repairs which were shortly to be completed. I have no doubt that the letter was intended by him to promote the claim in the hope of a prompt settlement, and that the purported factual account about the noon alarm was part of that promotion.
223. The false statement was directly related to the claim and intended to promote the claim. Does it meet the "limited objective element" of the test of materiality that, if believed, it would have tended at that stage to yield a not insignificant improvement in the Owner's prospects of getting the claim paid, bearing in mind that in this context "not insignificant" has the same connotation as "not insubstantial", "not immaterial", "not de minimis"? The answer must be yes. Insurers investigating a casualty of this nature would understandably be sceptical of how debilitative flooding of the engine room could have resulted from a relatively small leak in the bowthruster room, and an explanation for the failure of the Vessel's alarms to prevent such a result would be a not insignificant factor in an insurer's assessment of the validity of the claim, including in particular consideration of the application of the Inchmaree proviso. Mr Karia QC did not advance any argument to the contrary on behalf of the Owners.
224. It follows that the Underwriters' defence of fraudulent device succeeds, and that the Owners' claim, which is otherwise valid, is for that reason forfeit.
225. I have reached this conclusion with regret. In a scale of culpability which may attach to fraudulent conduct relating to the making of claims, this was at the low end. It was a reckless untruth, not a carefully planned deceit. It was told on one occasion, not persisted in at the trial. It was told in support of a theory about the events surrounding the casualty which Chris Kornet genuinely believed to be a plausible explanation. The reckless untruth was put forward against the background of having made the crew available for interview by the Underwriters' solicitor, who had had the opportunity to make his own inquiries of the crew on the topic. To be deprived of a valid claim of some €3.2 million as a result of such reckless untruth is, in my view, a disproportionately harsh sanction.

Amendment Application

226. At the conclusion of the hearing the Underwriters applied to amend their defence. I refused the application and indicated that I would give my reasons for doing so together with my judgment. I have set them out in a separate judgment.

Conclusion

227. The Owners' claim fails.