Easter Term [2017] UKSC 35 On appeal from: [2015] EWCA Civ 16

JUDGMENT
Gard Marine and Energy Limited (Appellant) v China
National Chartering Company Limited and another
(Respondents)
China National Chartering Company Limited
(Appellant) v Gard Marine and Energy Limited and
another (Respondents)
Daiichi Chuo Kisen Kaisha (Appellant) v Gard Marine
and Energy Limited and another (Respondents)
before
Lord Mance
Lord Clarke
Lord Sumption
Lord Hodge
Lord Toulson
JUDGMENT GIVEN ON
10 May 2017
Heard on 1, 2 and 3 November 2016
Appellant/Respondents
(China National
Chartering Co Ltd)
Respondent/Appellant
(Gard Marine & Energy
Ltd)
Michael Davey QC Mark Howard QC
James M Turner QC
Simon Birt QC
(Instructed by Winter
Scott LLP
)
(Instructed by Ince & Co
LLP
)
Respondent (Daiichi Chuo
Kisen Kaisha)
Dominic Kendrick QC
David Goldstone QC
Gavin Geary
(Instructed by MFB
Solicitors
)
Page 2
LORD CLARKE: (with whom Lord Sumption agrees. Lord Mance, Lord
Hodge and Lord Toulson agree on the first and third issues only)
Introduction
1. This appeal arises out of the grounding of the OCEAN VICTORY (“the
vessel”) in the port of Kashima in Japan on 24 October 2006. She was a Capesize
bulk carrier, built in China in 2005. By a demise charterparty dated 8 June 2005, the
vessel’s owners, Ocean Victory Maritime Inc (“OVM” or “the owners”), chartered
the vessel to Ocean Line Holdings Ltd (“OLH”), which is or was a related company,
on the widely used Barecon 89 as amended. On 2 August 2006, OLH time chartered
the vessel to China National Chartering Co Ltd (“Sinochart”) and on 13 September
2006, Sinochart in turn sub-chartered her to Daiichi Chuo Kisen Kaisha (“Daiichi”
or “the charterers”) for a time charter trip.
2. The demise charterparty and both time charterparties contained an
undertaking (on materially identical terms) to trade the vessel between safe ports.
On 12/13 September 2006, Daiichi (and thus Sinochart) gave the vessel instructions
to load at Saldanha Bay in South Africa and to discharge at Kashima. Between 19
and 21 September she loaded 170,000 tonnes of iron ore. She arrived off Kashima
on 20 October and discharge began that afternoon.
3. The port of Kashima is entered from the sea through the northern end of a
specially constructed channel known as the Kashima Fairway, which runs almost
due north south, and is the only route into and out of the port. The Kashima Fairway
is bounded on one side (to the east) by the South Breakwater and on the other (to
the west) by the land.
4. On 24 October the vessel sought to leave the port during a storm. However,
she allided with the northern end of the South Breakwater and grounded. Shortly
thereafter another Capesize vessel, the ELIDA ACE, also grounded in the Kashima
Fairway while attempting to leave the port. Salvors were engaged but the OCEAN
VICTORY eventually broke in two. Her wreck was subsequently removed in the
course of a lengthy wreck removal operation.
5. Some two years later, on 15 October 2008, Gard Marine & Energy Ltd
(“Gard”), one of the vessel’s hull insurers at the time of her loss, took assignments
of the rights of OLH and OVM in respect of the grounding and total loss of the
vessel. In its capacity as assignee of those rights, Gard subsequently brought a claim
Page 3
against Sinochart (which Sinochart passed on to Daiichi) for damages for breach of
the charterers’ undertaking to trade only between safe ports.
6. On 30 July 2013, Teare J (“the judge”) held that the casualty was caused by
the unsafety of the port in breach of the safe port undertaking in the time charters.
He awarded Gard substantial damages, namely the agreed value of the vessel
(US$88.5m), damages in respect of liability for SCOPIC expenses (US$12m),
damages for wreck removal expenses (US$34.5m) and damages for loss of hire
(US$2.7m).
7. Permission to appeal to the Court of Appeal on certain specific issues was
granted. On 22 January 2015, the Court of Appeal (Longmore, Gloster and Underhill
LJJ) allowed the appeal and set aside the judgment of the judge on the grounds that
the conditions which affected Kashima were an abnormal occurrence and that there
was no breach of the safe port undertaking on the part of the charterers. The Court
of Appeal also held that, in the light of the insurance provisions of the demise
charterparty, OVM and OLH (and Gard as their assignee) were not entitled to claim
in respect of losses covered by the hull insurers. On 20 May 2015 Gard were granted
permission to appeal to this Court.
Issues in the appeal
8. The parties agreed the issues in this appeal as follows.
1. Was there a breach of the safe port undertaking? In particular the
following specific questions were agreed: (1) was the port unsafe within the
meaning of the safe port undertaking, so that the charterers were in breach;
or (2) was there an “abnormal occurrence” within the context of the safe port
undertaking, which was no breach of the undertaking?
2. If there was a breach of the safe port undertaking, do the provisions
for joint insurance in clause 12 of the Barecon 89 form preclude rights of
subrogation of hull insurers and the right of owners to recover in respect of
losses covered by hull insurers against the demise charterer for breach of an
express safe port undertaking.
3. If there was a breach of the safe port undertaking, is Daiichi entitled
to limit its liability for Gard’s losses or any (and, if so, which) of them as
against Sinochart (and Sinochart in turn against Gard) pursuant to section 185
and Schedule 7 article 2(1) of the Merchant Shipping Act 1995?
Page 4
It was agreed that in the event that the appeal succeeds (that is that the answer to
issue 1(1) is “yes”), issues of time-bar and causation should be remitted to the Court
of Appeal. In this judgment I will focus first on the safe port issue.
Safe port – the facts
9. The events which led to the grounding and subsequent loss of the vessel are
summarised in paras 127 and 128 of the judge’s judgment as follows:
“127. The danger facing OCEAN VICTORY was one which
was related to the prevailing characteristics of Kashima. The
danger flowed from two characteristics of the port, the
vulnerability of the Raw Materials Quay to long swell and the
vulnerability of the Kashima Fairway to northerly gales caused
by a local depression. It may well be a rare event for these two
events to occur at the same time but nobody at the port could, I
consider, be surprised if they did. There is no meteorological
reason why they should not occur at the same time. Long waves
were clearly a feature of the port (as they must be of any port
facing the Pacific) and low pressure systems generating gale
force winds cannot, in my judgment, be regarded as abnormal.
I do not consider that the juxtaposition of long waves and a low
pressure system generating gale force winds from the north
amounts to an abnormal occurrence unrelated to the
characteristics of Kashima. Long waves may give rise to a need
for a vessel to leave the port. It may be a matter of chance
whether at that time there is also a low pressure system
generating gale force winds from the north but I am unable to
accept that such winds are so rare that they cannot be said to be
a feature of the port. It is not without significance that the Guide
to Port Entry notes that during periods of northerly swell the
entry channel is fully exposed and that vessels at low speed
generally have difficulty in steering.
128. It may be that the storm which affected the port on 24
October 2006 was one of the most severe storms to have
affected Kashima in terms of severity, speed of deterioration
and duration as suggested by Mr Lynagh’s analysis of its
characteristics. But the relevant characteristics are those which
give rise to the danger, namely the occurrence of long waves
and northerly gales. Neither long waves nor northerly gales can
be described as rare. Even if the concurrent occurrence of those
Page 5
events is a rare event in the history of the port such an event
flows from characteristics or features of the port.”
The principles – abnormal occurrence
10. In the Court of Appeal Longmore LJ (giving the judgment of the court) noted
in para 14 that it was common ground between the parties that, if the damage
sustained by the vessel at Kashima on 24 October 2006 was caused by an “abnormal
occurrence”, then the charterers would not have been in breach of the safe port
warranty. That common ground was based on the classic dictum of Sellers LJ in The
Eastern City [1958] 2 Lloyd’s Rep 127, 131 that:
“A port will not be safe unless, in the relevant period of time,
the particular ship can reach it, use it and return from it without,
in the absence of some abnormal occurrence, being exposed to
danger which cannot be avoided by good navigation and
seamanship …”
11. Longmore LJ added in para 15 that what was in dispute between the parties
on the appeal in relation to this issue was (i) what, as a matter of law, was the correct
test for an abnormal occurrence; (ii) in particular, was the judge correct to hold (in
paras 110, 127-128, 132 and 134 of his judgment) that the combination of two
weather conditions on the casualty date (namely the phenomenon of swell from
“long waves”, which might have forced the vessel to leave the berth, and a very
severe northerly gale which meant that the vessel could not safely exit the port) was
not to be characterised as an abnormal occurrence, notwithstanding that the
coincidence of the two conditions was “rare”, because both conditions were physical
characteristics or attributes of the port; and (iii) on the facts as found by, or
undisputed before, the judge, did the weather conditions on the casualty date amount
to an abnormal occurrence? It is important to note that it was not submitted that the
relevant test could or should be other than that described by Sellers LJ in The Eastern
City. In any event that test has stood the test of time. The question is what is meant
by “an abnormal occurrence”.
12. The Court of Appeal summarised the charterers’ case in para 44 of their
judgment as follows:
“(i) There was no breach of the safe port undertaking. By the
safe port undertaking, the charterers did not assume
responsibility for loss from every foreseeable risk at the port to
which the ship was ordered. They assumed responsibility only
Page 6
for risks which were sufficiently regular or sufficiently
foreseeable to amount to an attribute or feature of the port.
(ii) The prospective nature of the undertaking was material
to the test, because the right way to approach this test was to
imagine a charterer with full knowledge of the port giving the
order on the relevant day. He had to ask himself: ‘will the port
be safe for the ship to reach, use and depart from?’ If he could
say ‘yes’, then, barring some abnormal occurrence, there was
no breach. So a charterer did not assume the risk of loss from
an unusual event which was not characteristic of the port at the
time when the ship should be there. The obligation to give
indemnity for loss from such unusual events lay properly and
legally with the owner’s hull insurers.
(iii) The phrase ‘abnormal occurrence’ was not a term of art.
An occurrence was just an event – something that happened on
a particular time at a particular place in a particular way.
‘Abnormal’ was something well removed from the normal. It
was out of the ordinary course and unexpected. It was
something which the notional charterer would not have in
mind.
(iv) A rare event could not be ‘an attribute’ of a port. It was,
in the language of the cases, an ‘abnormal occurrence’ and so
outside the undertaking. The judge erred in law in holding that
a rare event was a feature of the port.
(v) The judge erroneously held that it did not matter if the
event was rare or unexpected, provided it arose from the
combined occurrence of two or more characteristics or
attributes of the port.
(vi) Words such as ‘characteristic’ or ‘attributes of the port’
were tools to help identify what arose in the ordinary course.
They were not intended to bring events well out of the ordinary
course into the scope of the charterers’ undertaking.
(vii) The judge went wrong by breaking down the question
into components instead of asking one unitary question,
namely: would it be an unexpected event for Capesize vessels
Page 7
calling at Kashima to find it necessary to leave the berth due to
danger from a long-wave swell at the very time when it was
dangerous to transit the Fairway? The judge’s approach was to
consider whether long waves and strong northerly winds from
low pressure storms affecting navigation in the Kashima
Fairway were respectively ‘attributes’ of the port. Having
reached the conclusion that they were ‘attributes’, he wrongly
assumed that it did not matter how rare their combination was.
(viii) On the facts, the combination of the two weather events
(namely long waves and strong northerly winds from low
pressure storms) had never apparently happened in the previous
35 years preceding the instruction to proceed to Kashima.
Accordingly the conditions on 24 October were an abnormal
occurrence for which the charterers were not liable.”
13. It was not in dispute that the question whether the port was unsafe must be
tested as at the moment that the charterers instructed the owners to proceed to it. It
was submitted on behalf of the charterers that the appropriate test was whether a
reasonable shipowner trading the ship for his own account and knowing the relevant
facts would decline to proceed to the nominated port. That is essentially the test set
out in the Court of Appeal’s summary of the charterers’ case in sub-para 44(ii)
above.
14. To my mind the key points in this appeal are to be found in sub-paras 44(i),
(ii), (iii) and (iv). It is important to note that the test is not whether the events which
caused the loss were reasonably foreseeable. Reasonable foreseeability is a well
known test in some parts of the law of tort, notably negligence and remoteness of
damage. The courts could well have adopted such a test but they have not done so.
Instead they have asked whether the relevant event was an abnormal occurrence.
15. What then is meant by “abnormal occurrence”? The question is whether it
has the meaning proposed by the charterers and set out in para 44(iii) quoted above
or the meaning proposed by Gard in para 66 of its case as follows:
“The phrase ‘abnormal occurrence’ in the Court of Appeal’s
judgment took on its own momentum as a term of art or
something to be construed as if in a statute. It is not. The phrase
is not something that appears in the words of the charter. It is a
qualification derived from the authorities intended to assist the
court, and the parties, to work out whether the port was safe
within the contractual clause. It is a description for an
Page 8
occurrence which does not result from the set-up or
characteristics of the port; the set-up and characteristics of the
port (tested at the time the order is given) being the concern of
the safe port undertaking. In other words, ‘an abnormal
occurrence’ is in contradistinction to an occurrence which
results from the set-up or characteristics of the port. The result
of the Court of Appeal’s approach is to widen the category of
‘abnormal occurrences’,so as to include occurrences which do
result from the set-up of the port, and in turn to narrow the
circumstances in which a port will be regarded as ‘unsafe’
(despite the obligation being a strict one).”
16. I would accept the charterers’ submission recorded in sub-para 44(iii) that an
‘abnormal occurrence’ has its ordinary meaning. It is not a term of art. As stated in
that sub-paragraph,
“[a]n occurrence was just an event – something that happened
on a particular time at a particular place in a particular way.
‘Abnormal’ was something well removed from the normal. It
was out of the ordinary course and unexpected. It was
something which the notional charterer would not have in
mind.”
17. We were referred to a number of cases which seem to me to support that
conclusion. Sellers LJ’s famous dictum quoted above was taken from the judgment
of Morris LJ in The Stork [1955] 2 QB 68, 105, which was a time charterparty case.
See also Reardon Smith Line Ltd v Australian Wheat Board (The Houston City)
[1956] AC 266, which was a voyage charterparty case to which essentially the same
principles were applied. In Kodros Shipping Corpn v Empresa Cubana de Fletes
(The Evia) (No 2) [1983] 1 AC 736, which was another time charterparty case, Lord
Diplock said at p 749 that he regarded the nature of the contractual promise by the
charterer in what he called the safe port clause as having been well settled for a
quarter of a century at the very least. He referred specifically to the expression
“abnormal occurrence” used by Sellers LJ in The Eastern City which he said
reflected the previous statement of Morris LJ in The Stork.
18. At p 749H, Lord Diplock referred to the distinction between damage
sustained by a particular vessel in a particular port on a particular occasion caused
by an “abnormal occurrence” and damage resulting from “some normal
characteristic of the particular port at the particular time of year”. He added that
there were dangers that judges of first instance sometimes omitted important
qualifications. He was referring (at p 750A-B) to what he called the heresy that, in
the previous decade or so, had been embraced by judges in the commercial court
Page 9
culminating in that of Mustill J in Transoceanic Petroleum Carriers v Cook
Industries Inc (The Mary Lou) [1981] 2 Lloyd’s Rep 272, 277.
19. Lord Diplock agreed with the judgment of Lord Roskill which had been
prepared with the assistance of Lord Brandon. The heresy to which Lord Diplock
referred was that identified by Lord Roskill at p 757. It arose in the construction of
what he described as these eight words in clause 2 of the relevant time charterparty:
“The vessel to be employed … between … safe ports …” Those were essentially the
same words as in the charterparties in the instant case. The heresy identified by Lord
Diplock was the conclusion of Mustill J in The Mary Lou and, indeed, of Robert
Goff J in The Evia (No 2) that there was an absolute continuing contractual promise
that at no time during her chartered service would the ship find herself in any port
which was or had been unsafe for her: see Lord Roskill at p 756G.
20. On p 757 Lord Roskill gave his reasons for rejecting that approach as a matter
of construction of the charterparty. In particular, he said at p 757D that a charterer
will exercise his contractual right to require the shipowner to carry out his
contractual obligations by giving the shipowner orders to go to a particular port or
place of loading or discharge. He added that it was clearly at the point of time when
that order is given that the contractual promise to the charterer regarding the safety
of that intended port or place must be fulfilled.
21. Lord Roskill then said this at p 757E:
“The charterer’s contractual promise must, I think, relate to the
characteristics of the port or place in question and in my view
means that when the order is given that port or place is
prospectively safe for the ship to get to, stay at, so far as
necessary, and in due course, leave. But if those characteristics
are such as to make that port or place prospectively safe in this
way, I cannot think that if, in spite of them, some unexpected
and abnormal event thereafter suddenly occurs which creates
conditions of unsafety where conditions of safety had
previously existed and as a result the ship is delayed, damaged
or destroyed, that contractual promise extends to making the
charterer liable for any resulting loss or damage, physical or
financial. So to hold would make the charterer the insurer of
such unexpected and abnormal risks which in my view should
properly fall upon the ship’s insurers under the policies of
insurance the effecting of which is the owner’s responsibility
under clause 3 unless, of course, the owner chooses to be his
own insurer in these respects.”
Page 10
22. Having expressed those views as a matter of construction of the charterparty,
Lord Roskill analysed the cases and concluded that they strongly supported the
views he had expressed. The cases included Ogden v Graham (1861) 1 B & S 773
and GW Grace and Co Ltd v General Steam Navigation Co Ltd (The Sussex Oak)
[1950] 2 KB 383 in addition to The Stork, The Houston City and The Eastern City.
He noted in particular the dissenting judgment of Sir Owen Dixon CJ in the High
Court of Australia in The Houston City which was accepted as correct by the Privy
Council.
23. Lord Roskill paid particular attention at p 760 to the reference to “some
abnormal occurrence” by Morris LJ in The Stork, which, as he put it, was the
foundation of the similar view of Sellers LJ in The Eastern City as described above.
Like Lord Diplock, Lord Roskill emphatically adopted that approach as correct. As
I read the remainder of his judgment, in which he refers to a number of other cases,
they are all to substantially the same effect.
24. In all these circumstances, I would accept these three submissions made on
behalf of the charterers arising out of the cases referred to above and, in particular
The Evia (No 2) in the House of Lords. First, the date for judging breach of the safe
port promise is the date of nomination of the port. A safe port promise is not a
continuing warranty. Second, the promise is a prediction about safety when the ship
arrives in the future. These propositions are not in dispute. As I see it, such a promise
necessarily assumes normality; given all of the characteristics, features, systems and
states of affairs which are normal at the port at the particular time when the vessel
should arrive, the question is whether the port is prospectively safe for this particular
ship. If the answer is “yes unless there is an abnormal occurrence”, the promise is
fulfilled. As Robert Goff J said at first instance in The Evia (No 2) [1981] 2 Lloyd’s
Rep 613, 621, “the formulation of a test whether the port is unsafe must assume
normality, and must therefore exclude danger caused by some abnormal
occurrence.”
25. I would further accept the third submission made on behalf of the charterers
that on the authorities to which I have referred, safe port disputes should be
reasonably straightforward. Was the danger alleged an abnormal occurrence, that is
something rare and unexpected, or was it something which was normal for the
particular port for the particular ship’s visit at the particular time of the year?
26. I would also accept the submission that this approach, that is the approach in
The Eastern City as elaborated by Lord Roskill and Lord Diplock in The Evia (No
2), provides a coherent allocation of risk between the various interests as follows.
The owners are responsible for loss caused by a danger which is avoidable by
ordinary good navigation and seamanship by their master and crew. The charterers
are responsible for loss caused by a danger which was or should have been
Page 11
predictable as normal for the particular ship at the particular time when the ship
would be at the nominated port and was not avoidable by ordinary good seamanship.
The owners (and ultimately their hull insurers) are responsible for loss caused by a
danger due to “an abnormal occurrence”. As Lord Roskill put it at p 757E quoted
above, charterers are not insurers of “unexpected and abnormal risks”. On the
contrary, the charterparty terms require owners to take out hull insurance (as they
will invariably do) which is their protection against rare and unexpected events. On
the charterers’ case on the facts, the characteristics of the port were such that the
ship was prospectively safe, but they unexpectedly combined in a critical way such
as to create an exceptional, and apparently unprecedented danger. This was within
the letter and spirit of Lord Roskill’s description of an unexpected and abnormal
risk.
27. It is to my mind important to note the emphasis in the cases upon the meaning
of the expression “abnormal occurrence”. I would accept the charterers’ submission
in para 44(iii) of the Court of Appeal’s judgment that “abnormal” is something well
removed from the normal. It is out of the ordinary course and unexpected. It is
something which the notional charterer or owner would not have in mind.
28. In short, I would accept the charterers’ submission that the first question is
whether a reasonable shipowner in the position of the particular shipowner trading
the ship for his own account and knowing the relevant facts would proceed to the
nominated port. If the answer is “yes unless there is an abnormal occurrence”, the
port is prospectively safe for the particular ship and the promise is fulfilled. In a case
where the vessel suffers loss or damage, a second question arises, namely whether
there was damage caused by an abnormal occurrence as defined above.
Contrast between the approaches of the judge and of the Court of Appeal
29. This contrast can be seen in the judgment of the Court of Appeal. Having
summarised the essential facts at para 48, in para 49 the Court of Appeal described
the core of the judge’s reasoning at paras 127-129 of his judgment thus:
“On analysis his approach appears to have been that, in
deciding whether the casualty resulted from an abnormal
occurrence:-
(i) he did not need to consider the evidence relating
to how ‘rare’ the critical combination of the two
component dangers was, although, without analysing
the evidence in any detail, he was prepared to hold that
Page 12
‘it may well be a rare event for these two events to occur
at the same time’;
(ii) he did not need to consider whether the critical
combination was rare, because ‘[e]ven if the concurrent
occurrence of those events is a rare event in the history
of the port,’ what mattered was that:
(a) separately the two component features of
the critical combination were characteristics or
attributes of the port;
b) looked at separately, neither of the two
component features could be said to occur
‘rarely’; long waves and northerly gale winds
were at least foreseeable in Kashima;
(c) there was no meteorological reason why
the two component features should not occur at
the same time; despite the fact that the storm
which affected Kashima on 24 October 2006 may
have been exceptional in terms of its rapid
development, its duration and its severity, there
was a clear risk of gale force winds from the
northerly quadrant in the Kashima Fairway at the
same time as long waves were affecting the Raw
Materials Quay;
(d) therefore, it was necessarily foreseeable that
at some stage the critical combination would occur
and nobody could be surprised if it did; and
(e) the critical combination was accordingly an
event which ‘flow[ed] from the characteristics or
features of the port’;
(iii) accordingly, in those circumstances the critical
combination could not be said to be an ‘abnormal
occurrence’; in the language of Mustill J (as he then was)
in The Mary Lou …, the critical combination was not
something which ‘could be said, if the whole history of the
Page 13
port were regarded, to have been out of the ordinary’;
again, adopting Mustill J’s words, long waves and
northerly gale winds were ‘events of the type and
magnitude in question [which were] sufficiently regular or
at least foreseeable to say that their occurrence is an
attribute or characteristic of the port’, so as not to amount
to an abnormal occurrence; the critical combination
flowed from those characteristics and therefore could not
be an abnormal occurrence.”
30. In paras 50ff the Court of Appeal embarked upon a critique of the judge’s
approach. In para 50 they set out their conclusion that the judge’s approach was
flawed. They then referred in detail to the speeches of Lord Diplock and Lord
Roskill in The Evia (No 2). In the speech of Lord Roskill they highlighted in bold
the passage at p 757E which I quoted at para 21 above. It is to my mind important
that that passage includes the proposition that where the characteristics of a port
make the port prospectively safe, Lord Roskill did not think that
“if … some unexpected and abnormal event thereafter
suddenly occurs which creates conditions of unsafety where
conditions of safety had previously existed and as a result the
ship is delayed, damaged or destroyed, that contractual promise
extends to making the charterer liable for any resulting loss or
damage, physical or financial.”
He added that so to hold would make the charterer the insurer of such “unexpected
and abnormal risks” which should properly fall on the ship’s insurers.
31. Immediately after quoting an extensive passage from the speech of Lord
Roskill in The Evia (No 2), most of which is quoted above, in para 52 the Court of
Appeal said that its import was clear. They said that charterers do not assume
responsibility for unexpected and abnormal events which occur suddenly and which
create conditions of unsafety after they have given the order to proceed to the
relevant port. They are the responsibility of the ship’s hull insurers (if owners have
insured) or of owners themselves. The Court of Appeal further noted in para 52 that
the concept of safety is necessarily not an absolute one. They did so by reference to
the decision of the Court of Appeal in The Saga Cob [1992] 2 Lloyd’s Rep 545, 551,
where, in the context of political risks, Parker LJ, giving the judgment of the court,
said this:
“In the latter [the safe port warranty case] one is considering
whether the port should be regarded as unsafe by owners,
Page 14
charterers, or masters of vessels. It is accepted that this does
not mean that it is unsafe unless shown to be absolutely safe. It
will not, in circumstances such as the present, be regarded as
unsafe unless the ‘political’ risk is sufficient for a reasonable
shipowner or master to decline to send or sail his vessel there.”
32. In the instant case the Court of Appeal, in my opinion correctly, held (at para
53) that a similarly realistic approach should be adopted to the determination of what
it called the essentially factual question whether the event giving rise to the
particular casualty is to be characterised as an “abnormal occurrence” or as resulting
from some “normal” characteristic of the particular port at the particular time of
year. The Court of Appeal, emphasised the word “normal” in the term “normal
characteristic”, noting that it was used by Lord Diplock when he observed in The
Evia (No 2) at p 749 that:
“… it is not surprising that disputes should arise as to whether
damage sustained by a particular vessel in a particular port on
a particular occasion was caused by an ‘abnormal occurrence’
rather than resulting from some normal characteristic of the
particular port at the particular time of year.”
Importantly in the instant case, the Court of Appeal further observed (also at para
53) that, in what they described as an illuminating passage, in The Saga Cob the
Court of Appeal at pp 550-551 emphasised that the fact that an event (in that case a
guerrilla attack) was theoretically foreseeable did not make it a “normal
characteristic” of the port. They noted that on the facts in The Saga Cob the event
relied upon could not be regarded as other than “an abnormal and unexpected event”.
This approach underlines the fact that foreseeability is not the test of the normality
of an event. The cases show that an abnormal occurrence or event is something that
is unexpected when the vessel arrives at and remains in the port: see eg The Evia
(No 2) per Lord Roskill quoted at paras 21 and 30 above.
33. In para 54 the Court of Appeal placed reliance upon the approach of Mustill
J in The Mary Lou at p 278. They noted that in his description of what constitutes
an abnormal occurrence, Mustill J implicitly recognised the need to approach the
identification of an abnormal occurrence realistically and having regard to whether
the event had occurred sufficiently frequently so as to become a characteristic of the
port. At the end of the same paragraph, having recognised the difficulty of finding
an appropriate turn of phrase, Mustill J said this:
“It may be said that the loss is not recoverable unless events of
the type and magnitude are sufficiently regular or at least
Page 15
foreseeable to say that the risk of their occurrence is an
‘attribute’ or ‘characteristic’ of the port. Or it may be said that
‘abnormal’ or ‘casual events’ do not found a claim.”
34. The Court of Appeal identified a number of respects in which they concluded
that the judge went wrong. They summarised them in para 55 of their judgment:
“First of all he failed to formulate the critical – and unitary –
question which he had to answer: namely, whether the
simultaneous coincidence of the two critical features, viz (a)
such severe swell from long waves that it was dangerous for a
vessel to remain at her berth at the Raw Materials Quay
(because of the risk of damage or mooring break out) and (b)
conditions in the Kashima Fairway being so severe because of
gale force winds from the northerly/north easterly quadrant), as
to make navigation of the Fairway dangerous or impossible for
Capesize vessels, was an abnormal occurrence or a normal
characteristic of the port of Kashima? Or put even more simply,
was it an abnormal occurrence or a normal characteristic of the
port that a vessel might be in danger at her berth at the Raw
Materials Quay but unable at the same time safely to leave
because of navigation dangers in the Kashima Fairway arising
from the combination of long waves and gale force northerly
winds which, in fact, occurred?”
35. The Court of Appeal added in para 56 that, instead of asking the unitary
question directed at establishing the correct characterisation of the critical
combination (abnormal occurrence or normal characteristic of the port), the judge
merely addressed the respective constituent elements of the combination (swell from
long waves making it dangerous for a vessel to remain at the Raw Materials Quay
and gale force winds from the northerly/north-easterly quadrant making navigation
of the Fairway dangerous or impossible for Capesize vessels) separately. He looked
at each component and decided that, viewed on its own, neither could be said to be
rare and both were attributes or characteristics of the port. The Court of Appeal
concluded that that was the wrong approach; what mattered was not the nature of
the individual component dangers that gave rise to the events on 24 October, but the
nature of the event (namely the critical combination of the two) which gave rise, on
the judge’s findings, to the vessel effectively being trapped in port.
36. The Court of Appeal further held in para 57 that the judge was also wrong to
hold that, even if the critical combination was rare, nonetheless it was a
characteristic of the port, for two reasons. The first (as stated in his para 127) was
because, although it might well be rare for these two events to occur at the same
Page 16
time, nobody at the port could be surprised if they did, and there was no
meteorological reason why they should not occur at the same time. The second (as
stated in his para 128) was because, even if the concurrent occurrence of those events
was a rare event in the history of the port, such an event flowed from the
characteristics or features of the port.
37. The Court of Appeal concluded that both reasons were fallacious. In my
opinion they were correct so to hold. As to the first, the Court of Appeal noted in
para 58 that the conclusion that nobody at the port could be surprised that both the
above events occurred at the same time appears to have been based on the idea that,
provided an event is theoretically foreseeable as possibly occurring at the relevant
port, because of the port’s location, then that is enough to qualify the event as a
“characteristic of the port”.
38. The Court of Appeal correctly said at para 58 that the judge appears to have
derived that test from dicta in the judgment of Mustill J in The Mary Lou at p 278,
where (in the passage quoted at para 33 above) he referred to long waves and
northerly gale winds as being “at least foreseeable”. However the Court of Appeal,
in my opinion correctly, held that satisfaction of the test of mere “foreseeability”
was per se clearly not sufficient to turn what the judge himself described as “a rare
event in the history of the port” into a normal characteristic or attribute of the port.
They held that the error made by the judge was to pick up on the words “at least
foreseeable” in his citation from Mustill J’s judgment, and to use minimum
foreseeability, without more, as some sort of litmus test for establishing whether an
event was a characteristic of a port, without having any regard to significant factors
such as the actual evidence relating to the past history of the port, the frequency (if
any) of the event, the degree of foreseeability of the critical combination and the
very severe nature of the storm on the casualty date. The Court of Appeal further
held that in doing so the judge departed from the orthodox and practical approach
of Mustill J in his judgment in The Mary Lou at p 278 and of Lords Diplock and
Roskill in The Evia (No 2), to the question of whether an event was abnormal. Such
an approach necessarily includes an examination of the past history of the port and
of whether, in that evidential context, the event was unexpected. I agree. The Court
of Appeal also noted in para 58 that he took the phrase “at least foreseeable” as used
by Mustill J out of context. I agree with the Court of Appeal that it is clear that, when
the passage is read in context, Mustill J was certainly not suggesting that mere,
theoretical, foreseeability on its own was sufficient. He was not setting up some sort
of alternate test which excluded considerations of questions such as the frequency
of past occurrences of the particular event, or the degree of likelihood that the event
was to occur in the future.
39. Moreover, the Court of Appeal noted at para 59 that, as the Court of Appeal
emphasised in The Saga Cob in the passage cited at para 32 above, one has to look
at the reality of the particular situation in the context of all the evidence, to ascertain
Page 17
whether the particular event was sufficiently likely to occur to have become an
attribute of the port. Otherwise the consequences of a mere foreseeability test lead
to wholly unreal and impractical results. The Court of Appeal focused on these
examples in the instant case: does the mere fact that it is “foreseeable” from the
location of San Francisco that earthquakes may occur in its vicinity, or from the
location of Syracuse, beneath Mount Etna, that there may be volcanic explosions in
its vicinity, predicate that any damage caused to vessels in those ports from such
events, were they to occur in the future, would flow from the “normal characteristics
or attributes” of those ports, and therefore necessarily involve a breach of any safe
port warranty? The answer given by the Court of Appeal was obviously not;
whether, in such circumstances, there would be a breach of the safe port warranty,
or the event would be a characterised as an abnormal occurrence, would necessarily
depend on an evidential evaluation of the particular event giving rise to the damage
and the relevant history of the port.
40. The Court of Appeal was particularly struck (at para 60) by the fact, as they
put it, that the judge provides no evidential basis for his apparent factual conclusion
that “nobody at the port could, I consider, be surprised” if the crucial combination
occurred, or for the conclusion reached earlier in para 110 of the judgment that
“there must have been … a clear risk of gale force winds from the northerly quadrant
in the Kashima Fairway at the same time as long waves were affecting the Raw
Materials Quay”.
41. The final conclusion of the Court of Appeal’s on the first reason advanced by
the judge and referred to by the Court of Appeal in para 57 was set out in para 61 of
their judgment (which must be read in the context of para 60) as follows:
“61. In the light of the evidence to the effect that no vessel in
the port’s history had been dangerously trapped at the Raw
Materials Quay, with a risk of damage or mooring break out, at
the same time as the Kashima Channel was not navigable
because of gale force winds, it is difficult to see how he reached
this conclusion. This may be because he did not adequately
focus evidentially on the particular situation which he had to
consider, namely one where a vessel was effectively trapped,
because the swell from long waves affecting vessels berthed at
the Raw Materials Quay was so severe that it was dangerous
for a vessel to remain there (as opposed to merely a situation
where long waves caused swell and a vessel decided to leave
the Raw Materials Quay) and the Kashima Channel not being
navigable because of gale force winds. It may also be because
he did not give adequate weight to the evidence of Mr Lynagh
(which he gives no cogent reason for rejecting) that the storm
which occurred on 24 October was exceptional in terms of its
Page 18
rapid development, its duration and its severity (see para 48(ix)
above).”
42. As to the second reason advanced by the judge, the Court of Appeal
responded in this way in para 62:
“62. The second reason given by the judge (‘Even if the
concurrent occurrence of those events is a rare event in the
history of the port such an event flows from characteristics or
features of the port’) is, in our view, equally flawed. As we
have already stated in paras 55 and 56 above, what the judge
had to decide was whether ‘the concurrent occurrence of those
events’ (ie the critical combination) was itself a normal
characteristic of the port or an abnormal occurrence. That was
the relevant event which the judge had to characterise. It simply
did not follow, logically or otherwise, from the fact that that
event arose from (or, as the judge said, ‘flow[ed] from’) the
combination of two individual dangers, which he had held were
normal characteristics or attributes of the port, that the
‘concurrent occurrence of those events’ was also a normal
characteristic or attribute of the port.”
43. By way of postscript, I note that on behalf of the owners significant stress
was placed upon the failure of the Kashima port authority to carry out a risk
assessment and put in place a proper safety system to deal with the risk of the two
types of weather conditions referred to by the judge occurring at the same time.
However, while it may be relevant in some cases, the question remains whether the
event (or in this case the combination of natural events) which led to this casualty
was an abnormal and unexpected occurrence or not. For the reasons I have given I
conclude that the Court of Appeal were entitled to reach the decision which they
did.
44. The ultimate conclusion of the Court of Appeal was set out in paras 63 and
64 as follows:
“63. In deciding whether the critical combination was itself a
normal characteristic of the port or an abnormal occurrence,
what the judge should have done was to evaluate the evidence
relating to the past frequency of such an event occurring and
the likelihood of it occurring again. He should have also, in our
view, have taken into account what appears to have been the
unchallenged evidence of Mr Lynagh referred to above relating
Page 19
to the exceptional nature of the storm that affected Kashima on
24 October 2006 in terms of its rapid development, its duration
and its severity. Had he done so, then, on the basis of his own
finding that ‘the concurrent occurrence of those events was
rare’, and on the basis of the evidence which we have
summarised above, there would, in our view, have been only
one conclusion which he could have reached – namely that the
event which occurred on 24 October 2006 was indeed an
abnormal occurrence.
64. For the above reasons we conclude that the conditions
which affected Kashima on 24 October 2006 were an abnormal
occurrence, that there was no breach by the charterers of the
safe port obligation, and accordingly that the appeal should be
allowed on this ground.”
45. I agree with the Court of Appeal.
Conclusion on the safe port issue
46. In my opinion, the Court of Appeal reached the correct conclusions for the
reasons they gave. I initially questioned whether the Court of Appeal should have
interfered with the decision of the judge at first instance. However, in the light of
the submissions made on both sides, I have concluded that this was one of those rare
cases in which the correct conclusion is that the casualty was caused by an abnormal
occurrence as that expression is explained in the cases. I accept the reasoning of the
Court of Appeal and prefer their approach to that of the judge.
47. I would accordingly dismiss the appeal on the safe port issue. I would answer
the questions raised on that issue by the parties and set out in para 8.1(1) and (2) as
follows. The port was not unsafe within the meaning of the safe port undertaking so
that the charterers were not in breach of it. The conditions at the port amounted to
an abnormal occurrence as that expression is understood in the cases.
Joint insurance
48. Issue 2 summarised in para 8 above assumes (contrary to my view) that there
was a breach of the safe port undertaking and asks whether the provisions for
insurance in clause 12 of the Barecon 89 form preclude rights of subrogation of hull
insurers and the right of owners to recover in respect of losses covered by hull
insurers against the demise charterers for breach of the express safe port
Page 20
undertaking. The judge held that the owners and hull insurers were entitled to
recover notwithstanding clause 12, whereas the Court of Appeal held that they were
not. Lord Sumption agrees with the judge whereas Lord Toulson agrees with the
Court of Appeal. I agree with the judge and Lord Sumption, essentially for the
reasons they give. Lord Sumption has set out clauses 12 and 13, which I will not
repeat.
49. I have been particularly struck by these considerations. I agree with the judge
(at para 185) that the demise charterparty must be given the meaning which, having
regard to the background known to both parties, it would reasonably be understood
to bear and that, in circumstances where, in clause 29, the demise charterparty
contains a clear safe port warranty, one would expect any exemption of the demise
charterers from liability in damages for breach of the safe port warranty to be clearly
expressed.
50. In para 190 the judge observed that the charterers relied upon the The Evia
(No 2), in which the question arose whether (as Lord Roskill put it at p 766) the war
risks clause cast upon the owners and their insurers all war risks and thus freed the
charterers from liability for them pursuant to the safe port clause. It was held that
the charterers were freed from any liability that they might otherwise have. Lord
Roskill identified the relevant question as being whether the war risks clause was a
complete code exhaustive of the owners’ rights, which depended upon the
construction of the time charterparty as a whole.
51. As the judge noted at para 191, the charterers adopted that reasoning here and
said that it was applicable to the demise charterparty and, indeed, that it was a
stronger case because, not only did the demise charterers pay for the cost of hull
insurance, but they were also named as joint assureds and, generally speaking, an
insurer cannot exercise rights of subrogation to pursue a claim in the name of one
co-assured against another.
52. The judge rejected that submission in these terms:
“192. Cases decided after The Evia (No 2) have emphasised
that the decision in that case depended upon there being a
clause which, on its true construction, provided an exhaustive
code of the rights and liabilities of the parties; see The
Concordia Fjord [1984] 1 Lloyd’s Reports 385 and The
Chemical Venture [1993] 1 Lloyd’s Reports 508.
Page 21
193. In The Concordia Fjord the arbitrator, Mr MacCrindle
QC, said that he was ‘not aware of any principle exempting the
Charterers from liability for their breaches of contract merely
on the ground that they have directly or indirectly provided the
funds whereby the Owners insured themselves against such
damages.’ Bingham J agreed; see pp 387-388. Thus the mere
fact that the charterer pays for the hull insurance is not enough
to exempt him from liability for breach of his obligations under
the charterparty. There has to be an intention to create an
exhaustive code which determines the parties’ rights and
liabilities by reference to a claim on the insurance policy.
194. If clause 12 of the demise charterparty were such a code
it would apply, not just to a particular issue such as war risks
as was the case in The Evia (No 2), but to all hull, war and P&I
risks. The charterparty contains a clear and express safe port
warranty. If clause 12 were to be construed as an exhaustive
code that clause would be rendered nugatory with regard to
insured risks. It would in effect exempt the demise charterer
from liability for breach of the safe port warranty in exchange
for paying for the hull insurance. For that to be the intention of
the parties there would have to be clear words. Counsel for
Gard emphasised the rule of construction that clear words are
necessary before the court will hold that a contract has taken
away rights or remedies which one of the parties would
otherwise have had; see Gilbert-Ash (Northern) Ltd v Modern
Engineering (Bristol) Ltd [1974] AC 689. This rule of
construction is usually expressed with regard to rights which a
party would have at common law but it must also apply to
valuable rights given by other parts of a contract. Counsel
submitted that there are no such words in clause 12 of the
demise charterparty. The sentences of clause 12(a) and 12(c)
on which particular reliance is placed by Daiichi do not
expressly remove the right to damages for breach of the safe
port warranty. They merely give the demise charterer certain
rights with regard to proceeds of the insurance policy for which
they have paid.”
53. I entirely agree with the judge. It follows from the fact that clause 12 contains
no such express exclusion that any such exemption can only arise by necessary
implication. In short, there is nothing in clause 12 which provides that the demise
charterers have no liability for breach of clause 29 and I see no basis for such a
necessary implication, essentially for the reasons given by the judge and by Lord
Sumption. In particular, it seems to me to be striking that, as the judge observed in
Page 22
para 195, clause 13, which contained an alternative insurance and repairs clause
which not only provided that hull insurance would be paid for by the registered
owner but also expressly stated that the registered owners and/or insurers would not
have any rights of recovery or subrogation against the demise charterers in respect
of insured losses, was deleted from the printed form. Thus the demise charterers
chose not to be bound by clause 13.
54. After a detailed analysis (between paras 196 and the first part of para 198),
the judge said that he did not consider that clause 12 codified the rights and liabilities
of the parties with regard to insured risks. He noted that it provides for the provision
of insurance and who is to pay for it, for the demise charterers to be responsible for
insured repairs and to reimburse themselves from the proceeds of the insurance
policy, for the demise charterers to be responsible for other repairs and for the claims
on a total loss to be paid to the mortgagee for distribution to the registered owners
and demise charterers in accordance with their respective interests. I agree with the
judge that this does not in the required sense codify the rights and liabilities of the
parties with regard to breach of the safe port warranty where the casualty caused by
the breach has given rise to a claim on the insurance.
55. It is true that, as the judge put it in para 199, what clause 12 has, which neither
the clause in The Evia (No 2) nor the clause in The Concordia Fjord (as expressly
noted by Mr MacCrindle QC) had, is a provision that the owners and demise
charterers were to be co-assureds, thereby, on the face of it, bringing into play the
principle that, generally, an insurer cannot exercise rights of subrogation to pursue
a claim in the name of one co-assured against another.
56. After referring (in para 200) to the decision and reasoning of Rix LJ in Tyco
Fire and Integrated Solutions (UK) Ltd v Rolls-Royce Motor Cars Ltd [2008] 2 All
ER (Comm) 584, the judge concluded in para 201 as follows:
“In the present case there was an express safe port warranty by
the demise charterers, there was no code of rights and
obligations in clause 12 with regard to insured losses caused by
a breach of the safe port warranty and there was no express
ouster of the right of subrogation in clause 12. Those features
of the demise charterparty suggest to me that, construing the
charter as a whole, it was intended that the demise charterer
would be liable to the owner for breach of the safe port
warranty, notwithstanding that they were joint assured and
could take the benefit of the insurance in the manner set out in
clause 12.”
Page 23
I agree.
57. For these reasons and those given by Lord Sumption, I would have answered
the question or issue 2 summarised in para 8 above, namely whether the provisions
for insurance in clause 12 of the Barecon 89 form preclude rights of subrogation of
hull insurers and the right of owners to recover in respect of losses covered by hull
insurers against the demise charterers for breach of the express safe port
undertaking, in the negative. I am not persuaded by the judgments of Lord Mance
and Lord Toulson (with whom Lord Hodge agrees) to reach a different conclusion.
Limitation of liability
Introduction
58. Question or issue 3 summarised in para 8 above assumes (contrary to my
view) that there was a breach of the safe port undertaking and asks whether Daiichi
is entitled to limit its liability for Gard’s losses or any (and, if so, which) of them as
against Sinochart (and Sinochart in turn against Gard) pursuant to section 185 and
Schedule 7 article 2(1) of the Merchant Shipping Act 1995, which gave the force of
law to the Convention on Limitation of Liability for Maritime Claims 1976 (“the
1976 Convention”).
59. Like the joint insurance issue, this issue does not arise in the light of our
decision on the safe port issue. However it raises a point of some potential
importance and was fully argued before us. Neither of the courts below considered
it because it was accepted that they were both bound by the decision of the Court of
Appeal in The CMA Djakarta [2004] 1 Lloyd’s Rep 460. In that case Longmore LJ
gave the only substantive judgment, with which Waller and Neuberger LJJ agreed.
60. As formulated on behalf of Daiichi, who were time charterers, the question
at issue is whether Daiichi (hereinafter “the charterers”) can limit their liability for
the loss of the vessel and consequential losses arising out of the loss of the vessel.
The answer to that question depends largely upon whether The CMA Djakarta was
correctly decided in the Court of Appeal. The limitation issue is a short but important
one. It concerns the correct interpretation of the 1976 Convention.
Page 24
61. The Convention provides, so far as relevant, as follows:
“Article 1. Persons entitled to limit liability
1. Shipowners and salvors, as hereinafter defined, may
limit their liability in accordance with the rules of this
Convention for claims set out in article 2.
2. The term ‘shipowner’ shall mean the owner, charterer,
manager or operator of a seagoing ship.
3. …
4. If any claims set out in article 2 are made against any
person for whose act, neglect or default the shipowner or salvor
is responsible, such person shall be entitled to avail himself of
the limitation of liability provided for in this Convention.
5. In this Convention the liability of a shipowner shall
include liability in an action brought against the vessel herself.
Article 2. Claims subject to limitation
1. Subject to articles 3 and 4 the following claims,
whatever the basis of liability may be, shall be subject to
limitation of liability:
(a) claims in respect of loss of life or personal injury
or loss of or damage to property (including damage to
harbour works, basins and waterways and aids to
navigation) occurring on board or in direct connexion
with the operation of the ship or with salvage operations,
and consequential loss resulting therefrom;
(b) …
(c) claims in respect of other loss resulting from
infringement of rights other than contractual rights,
Page 25
occurring in direct connexion with the operation of the
ship or salvage operations;
(d) claims in respect of the raising, removal,
destruction or the rendering harmless of a ship which is
sunk, wrecked, stranded or abandoned, including
anything that is or has been on board such ship;
(e) claims in respect of the removal, destruction or
the rendering harmless of the cargo of the ship;
(f) claims of a person other than the person liable in
respect of measures taken in order to avert or minimize
loss for which the person liable may limit his liability in
accordance with this Convention, and further loss
caused by such measures.
2. Claims set out in paragraph 1 shall be subject to
limitation of liability even if brought by way of recourse or for
indemnity under a contract or otherwise. However, claims set
out under paragraph 1(d), (e) and (f) shall not be subject to
limitation of liability to the extent that they relate to
remuneration under a contract with the person liable.
Article 3. Claims excepted from limitation
The rules of this Convention shall not apply to: (a) claims for
salvage or contribution in general average …”
Both parties rely principally upon article 2.1(a) quoted above. However, as appears
below, reliance is also placed upon articles 6 and 9-11.
62. Although I have included the references to salvors in the above quotation, it
is common ground that those references are irrelevant to the issues in this appeal.
They were included in the Convention in order to depart from the decision of the
House of Lords in The Tojo Maru [1972] AC 242.
63. The critical question for present purposes is whether The CMA Djakarta was
correctly decided in the Court of Appeal. Gard say that it was. They rely upon the
Page 26
fact that it has not been criticised in any case since it was decided. They thus rely
upon the reasoning of the Court of Appeal in The CMA Djakarta. They also rely
upon some at least of the reasoning of David Steel J at first instance in that case,
reported at [2003] 2 Lloyd’s Rep 50 and of Thomas J in The Aegean Sea [1998] 2
Lloyd’s Rep 39. By contrast, the charterers say that both cases were wrongly
decided. I have reached the clear conclusion that the Court of Appeal were correct,
essentially for the reasons they gave.
History of limitation
64. The 1976 Convention had of course been preceded by earlier Conventions,
which David Steel J referred to as part of his historical analysis of the right to limit
liability set out in detail in his judgment at first instance reported in [2003] 2 Lloyd’s
Rep 50. I entirely agree with his analysis (at pp 51-53) and will not repeat it here,
save to note some key points. He referred to the first relevant limitation statute,
which was the Responsibility of Shipowners Act 1733 and then to the Merchant
Shipping Act 1854, the Merchant Shipping Acts Amendment Act 1862 and the
Merchant Shipping Act 1894, which consolidated the earlier legislation.
65. As David Steel J put it, section 503 of the 1894 Act furnished a limit to an
owner’s liability in respect of certain categories of occurrence in these terms:
“503.(1) The owners of a ship, British or foreign, shall not,
where all or any of the following occurrences take place
without their actual fault or privity; (that is to say,) …
(b) Where any damage or loss is caused to any
goods, merchandise, or other things whatsoever on
board the ship;

(d) Where any loss or damage is caused to any other
vessel, or to any goods, merchandise, or other things
whatsoever on board any other vessel, by reason of the
improper navigation of the ship;
be liable to damages beyond the following amounts …”
Page 27
David Steel J added that section 71 of the Merchant Shipping Act 1906 provided
that the expression “owner” would be “deemed to include any charterer to whom
the ship is demised”. He further noted that, even prior to that Act, “owner” was
construed as being inclusive of a demise charterer: The Hopper No 66 [1908] AC
126.
66. The two Conventions to which David Steel J then specifically referred were
the two Conventions for the Unification of Certain Rules relating to the Limitation
of Liability of Seagoing Vessels in 1924 and 1957 (“the 1924 Convention” and the
“1957 Convention” respectively). Although the United Kingdom signed the 1924
Convention, it never became part of English law and was replaced by the 1957
Convention as between states that ratified the 1924 Convention.
67. In para 20 David Steel J noted that the right of limitation was still afforded to
owners for certain occurrences but the categories of occurrence were enlarged by
article 1(1) to include “(a) … loss of, or damage to, any property on board the ship;”
and
“(b) … loss of, or damage to any other property or
infringement of any rights caused by the act, neglect or default
of any person on board the ship for whose act, neglect or default
the owner is responsible or any person not on board the ship for
whose act, neglect or default the owner is responsible: provided
however that in regard to the act, neglect or default of this last
class of person, the owner shall only be entitled to limit his
liability when the act, neglect or default is one which occurs in
the navigation or the management of the ship or in the loading,
carriage or discharge of its cargo …”
In para 21 he added that the range of those entitled to limitation was also enlarged
by article 6(2), which provided:
“… the provisions of this Convention shall apply to the
charterer, manager and operator of the ship, and to the master,
members of the crew and other servants of the owner, charterer,
manager or operator acting in the course of their employment,
in the same way as they apply to an owner himself: provided
that the total limits of liability of the owner and all such other
persons in respect of personal claims and property claims
arising on a distinct occasion, shall not exceed the amounts
determined in accordance with article 3 of this Convention.”
Page 28
68. The impact of the 1957 Convention was enacted in the form of the Merchant
Shipping (Liability of Shipowners and Others) Act 1958 (“the 1958 Act”) by way
of amendment to the 1894 Act. In purported compliance with article 1(1)(b), section
2 prescribed that a new subsection (d) should be substituted in subsection 1 of
section 503 of the 1894 Act as follows:
“(d) Where any loss or damage is caused to any property
(other than any property mentioned in paragraph (b) of this
subsection) or any rights are infringed through the act or
omission of any person (whether on board the ship or not) in
the navigation or management of the ship, or in the loading,
carriage, or discharge of its cargo … or through any other act
or omission of any person on board the ship …”
Further, in purported compliance with article 6(2), section 3 of the 1958 Act
provided:
“(1) The persons whose liability in connection with a ship is
excluded or limited by Part VIII of the Merchant Shipping Act
1894 shall include any charterer and any person interested in or
in possession of the ship, and, in particular, any manager or
operator of the ship.”
The Convention
69. At first instance in The CMA Djakarta David Steel J noted at para 25 that the
Convention introduced radical changes as regards to both the size of the fund and
the circumstances in which the entitlement to limit might be lost. As he put it, in
short, the Convention made available a significantly enhanced fund at what he said
was perceived to be the maximum insurable level, but the entitlement to which could
only be challenged in quite exceptional circumstances: see The Leerort [2001] 2
Lloyd’s Rep 291. He said in para 26 that it was notable that the Convention left
largely untouched the range of persons entitled to limit, although it expressed the
category in somewhat different terms.
70. David Steel J identified the rival contentions of the parties succinctly in paras
28 and 29. It was the charterers’ case, first that, as charterers, they fell squarely
within the category of persons enabled to limit their liability as prescribed by article
1, and secondly that the entire claim for damages arising out of the casualty fell
equally squarely within the category of qualifying claims under article 2. By
contrast, it was the owners’ case that it was clear from the overall context, having
Page 29
regard to the object and purpose of the Convention, that the entitlement to limit was
restricted to those persons identified in article 1(2) whose liability for the qualifying
claim arose qua owner and not otherwise. On the facts, limitation was not available
since it was common ground that no part of the claim arose from the role of the
appellant charterers qua owners. David Steel J essentially accepted the submissions
made on behalf of the charterers and, in doing so, followed the decision and
reasoning of Thomas J in The Aegean Sea.
General approach of the Court of Appeal in the CMA Djarkta
71. Longmore LJ set out his general approach to the Convention in paras 9-11.
He first identified what he concluded were errors made by David Steel and Thomas
JJ. In particular, he did not agree with them that, in order to succeed in limiting their
liability, it was necessary for charterers’ claims to arise from their role qua owners.
I agree that Longmore LJ’s conclusions in that regard were correct for the reasons
he gave and do not need to revisit them.
72. Longmore LJ then cited a number of cases which support the proposition that,
given that the Convention is in its own words incorporated into English law, the task
of the court is to construe the Convention as it stands without any English law
preconceptions. As he put it, the interpretation of international conventions must not
be controlled by domestic principles but by reference to broad and general principles
of construction. He cited a number of well-known cases: Stag Line Ltd v Foscolo,
Mango & Co Ltd [1932] AC 328, 350, James Buchanan & Co Ltd v Babco
Forwarding & Shipping (UK) Ltd [1978] AC 141, 152D-E, Fothergill v Monarch
Airlines Ltd [1981] AC 251, 272E, 282A and 293C, and Morris v KLM Royal Dutch
Airlines [2002] 2 AC 628, 656 at para 78.
73. He added in para 10 that, while it may be difficult to know what are broad
and acceptable principles, some principles are enshrined in articles 31 and 32 of the
Vienna Convention on the Law of Treaties (1969), which was ratified by the United
Kingdom in 1971 and came into force in 1980. Those articles provide:
“ARTICLE 31
General rule of interpretation
1. A treaty shall be interpreted in good faith in accordance
with the ordinary meaning to be given to the terms of the treaty
in their context and in the light of its object and purpose.
Page 30
2. The context for the purpose of the interpretation of a
treaty shall comprise, in addition to the text, including its
preamble and annexes:
(a) any agreement relating to the treaty which was
made between all the parties in connexion with the
conclusion of the treaty;
(b) any instrument which was made by one or more
parties in connexion with the conclusion of the treaty
and accepted by the other parties as an instrument
related to the treaty. …
ARTICLE 32
Supplementary means of interpretation
Recourse may be had to supplementary means of interpretation,
including the preparatory work of the treaty and the
circumstances of its conclusion, in order to confirm the
meaning resulting from the application of article 31, or to
determine the meaning when the interpretation according to
article 31:
(a) leaves the meaning ambiguous or obscure; or
(b) leads to a result which is manifestly absurd or
unreasonable.”
74. Longmore LJ summarised his conclusions derived from articles 31 and 32 in
this way. The duty of a court is to ascertain the ordinary meaning of the words used,
not just in their context but also in the light of the evident object and purpose of the
Convention. The court may then, in order to confirm that ordinary meaning, have
recourse to the travaux préparatoires and the circumstances of the conclusion of the
Convention. The 1957 Convention was signed by the United Kingdom.
75. Like Longmore LJ in para 10, I would regard the existence and terms of a
previous international convention (even if not made between all the same parties) as
one of the circumstances which are part of a conclusion of a new convention but
Page 31
recourse to such earlier convention can only be made once the ordinary meaning has
been ascertained. Such recourse may confirm that ordinary meaning. It may also
sometimes determine that meaning but only when the ordinary meaning makes the
convention ambiguous or obscure or when such ordinary meaning leads to a
manifestly absurd or unreasonable result.
Context, object and purpose
76. In para 11 Longmore LJ set out the object and purpose of the 1976
Convention as agreed between the parties as follows (omitting the reference to the
Tojo Maru). First, the general purpose of owners, charterers, managers and operators
being able to limit their liability was to encourage the provision of international trade
by way of sea-carriage. Second, the main object and purpose of the Convention was
to provide for limits which were higher than those previously available in return for
making it more difficult to “break” the limit. Under the 1894 Act an owner was
entitled to limit his liability if he showed that the casualty occurred without his
“actual fault or privity”. Under the 1976 Convention (and the 1995 Act) the (now
higher) limit is to apply unless it can be shown that the loss resulted from the
personal act or omission of the party relying on the limit “committed with intent to
cause such loss or recklessly with the knowledge that such loss would probably
result”. It is thus particularly difficult to break the limit, but the amount available for
compensation is higher than it was previously. Longmore LJ added that it was not
possible to ascertain with certainty any object or purpose beyond that common
ground. He therefore turned to the ordinary meaning of the Convention, beginning
at the beginning.
Ordinary meaning
77. Leaving the position of salvors on one side, at para 13 Longmore LJ noted in
the context of article 1 that the word “shipowner” was defined as “the owner,
charterer, manager or operator of a seagoing ship”. He then rejected the opinion of
David Steel J and Thomas J that a charterer could only limit his liability if he was
acting in the management or operation of the vessel. In particular he expressed the
view that the mere fact that “charterer” is part of the definition of the word
“shipowner” cannot of itself mean that a charterer, which was “an expression
otherwise unqualified”, has to be acting as if he were a shipowner (ie qua shipowner)
before he can limit his liability. He added:
“To my mind the ordinary meaning of the word ‘charterer’
connotes a charterer acting in his capacity as such, not a
charterer acting in some other capacity.”
Page 32
78. Longmore LJ then said in para 13 that there were two difficulties in the
argument to the contrary. I do not think that it is necessary for me to discuss them
in any detail, since they were not relied upon on behalf of the charterers here. In para
18, to my mind correctly, Longmore LJ said that he would not give any gloss to the
word “charterer” in article 1(2) and that he would give it what seemed to him to be
its ordinary meaning. I agree with that approach.
79. Longmore LJ noted at para 21 that the issue was not resolved by a
consideration of article 1 of the Convention because it was still necessary to
ascertain whether a claim for damage to the ship by reference to which a charterer
seeks to limit his liability is a claim which falls within article 2. In paras 22-24 he
considered loss or damage to the ship under article 2(1)(a), which he correctly held
extends the right to limit, inter alia, to claims in respect of “loss of or damage to
property occurring on board”, which is not apposite to include loss of or damage to
the ship itself since neither the loss of a ship nor damage to a ship can be said to be
loss or damage to property on board. Property on board means something on the
ship and not the ship itself.
80. The question then arises whether this is a claim in respect of loss of or damage
to property “occurring … in direct connexion with the operation of the ship.”
Longmore LJ held in para 23 that the most obvious reason for including this category
of claim is to cater for cases of collision with another ship. Loss or damage to that
other ship (or its cargo) is not “loss of or damage to property … occurring on board”
but is “loss of or damage to property … occurring … in direct connexion with the
operation of the ship”.
81. The critical part of his reasoning is to my mind in the next part of para 23,
where he said that that wording was not apt to cater for a case where the very ship,
by reference to the tonnage of which limitation is to be calculated, is lost or damaged
because the loss envisaged is loss to something other than that ship herself. He added
at para 24 that it was not without interest that in order to describe this category of
claim the framers had used the phrase “occurring … in direct connexion with the
operation of the ship”. That was, he said, virtually the same phrase as that used by
David Steel J to define what he meant by “qua owner”. If one were to postulate the
case of the vessel being in berth when the dangerous cargo exploded and damaged
parts of the harbour, the harbour authority could sue for that damage but one would
expect that the shipowners would be able to limit any liability for that claim. In order
to do so, however, they would have to assert that the loss or damage occurred in
direct connexion with the operation of the ship. The fact that dangerous cargo had
with their permission been loaded on the ship would, one thinks, be enough for that
purpose. But if it would be sufficient for that purpose, it would be odd that a charterer
pursuant to article 1 could not say of his own act in permitting such cargo to be
loaded that it was an act “in direct connexion with the operation of the ship”.
Page 33
Articles 9 to 11 of the Convention
82. Longmore LJ treated these provisions as of some importance in reaching his
conclusion. In my opinion he was correct to do so. In para 25 he noted that Thomas
J set them out in detail in The Aegean Sea and summarised them broadly in this way.
Article 9(1) provides for the claims against (a) the persons mentioned in article 1(2)
(viz owner, charterer, manager or operator) to be aggregated if they arose on distinct
occasions; likewise for claims against (b) the owner of a ship rendering salvage
services and a salvor operating from that ship and (c) a salvor not operating from a
ship. Article 9(2) then deals with passenger claims. Article 10 provides that liability
can be limited without the creation of a fund. Article 11 then provides for the
constitution of a limitation fund when that is, in fact, done; it provides for separate
funds for the “shipowner” category of those entitled to limit and the “salvor”
categories (and for passenger claims) by providing:
“‘A fund constituted by one of the persons mentioned in
paragraph 1(a), (b) or (c) or paragraph 2 of article 9 or his
insurer shall be deemed constituted by all persons mentioned
in paragraph 1(a), (b) or (c) or paragraph 2 respectively.’ Thus
through the references to article 9(1)(a) all those persons
designated as shipowners in article 1(2) of the Convention are
brought together as a single unit for the constitution of the fund.
Thomas J said this (p 49):
‘In my view the combined effect of these articles is
important. As there is provision for a fund for those
categorized as shipowners and that fund is to cover both
charterers and owners, it is difficult to see how
charterers can claim the benefit of limitation through
that fund where a claim is brought against them by
owners. Owners are entitled to the benefit of limitation
for a claim by charterers as that claim is being brought
by charterers not when performing a role in the
operations of the ship or when undertaking the
responsibility of a shipowner, but in a different capacity,
usually through their interest in the cargo being carried.’
While I entirely agree with this passage from The Aegean Sea,
the considerations advanced by the judge to my mind more
effectively support a conclusion that the claims in respect of
which an owner or a charterer can limit do not include claims
for loss or damage to the ship relied on to calculate the limit
Page 34
rather than a conclusion that a charterer can only limit in
respect of operations he does qua owner.”
83. Further, at para 26 Longmore LJ approved Thomas J’s view that, if he was
wrong in his general conclusion that the charterers could only limit when the loss (a
total loss in that case) was caused by an act normally performed by the shipowner,
the claim for the loss of the vessel did not fall within article 2(1)(a) because the loss
of the ship was not “loss of property … occurring … in direct connection with the
operation of the ship”. This in turn was because, as Thomas J put it at (p 51):
“it is the operation of the very ship that must cause the loss of
property; the ship cannot be the object of the wrong.”
84. Longmore LJ added that, similarly in The CMA Djakarta, which was a case
of extensive repair rather than total loss, David Steel J upheld the shipowners’
argument that the vessel cannot be both the victim and the perpetrator and that the
“property” envisaged in the article must be the property of a third party either on
board the vessel (eg cargo) or external to the vessel, for example an SBM. David
Steel J said (at para 52):
“The property damaged cannot be the very same thing as the
operation of which caused the damage.”
I agree with both Thomas J and David Steel J in this respect and conclude that the
ordinary meaning of article 2(1)(a) does not extend the right to limit to a claim for
damage to the vessel by reference to the tonnage of which limitation is to be
calculated.
Confirmation of the ordinary meaning of article 2(1)(a)
85. In para 27 Longmore LJ gave a further reason for his earlier conclusion based
on the ordinary meaning of article 2(1)(a), with which I also agree. I agree that the
effect of giving the words their ordinary meaning is not absurd or unreasonable, nor
is there ambiguity or obscurity. As Longmore LJ pointed out, David Steel J
considered the wording of the 1957 Convention and held that his conclusion derived
support from article 1 of that Convention, which drew an express distinction
between “the ship” and “other property”. Longmore LJ held that it was, if anything,
even clearer than the 1976 Convention on this point. He held that that served to
confirm the proposition since any intention to change the previous agreement so that
damage to the ship itself would be subject to limitation would have been made much
Page 35
more explicitly. I agree, although I do not regard this point as of any great
significance.
86. I should add that, in my opinion, in agreement with David Steel J and the
Court of Appeal in The CMA Djakarta, there is nothing in the travaux préparatoires
which supports any other conclusion. Some reliance was placed upon article 4(1)(iv)
of the Liens and Mortgages Convention (1967), which provides:
“The following claims shall be secured by maritime liens on
the vessel … (iv) claims against the owner, based on tort and
not capable of being based on contract, in respect of loss of or
damage to property occurring, whether on land or on water, in
direct connection with the operation of the vessel …”
No-one suggested that “loss or damage to property” could include loss of or damage
to the very vessel on which the maritime lien was secured. I would accept the
submission made on behalf of Gard that, when similar wording was exported to the
1976 Convention, the expression “loss of or damage to property … in direct
connection with the operation of the ship” was not intended to include loss of or
damage to the very vessel on the basis of whose tonnage limitation was calculated.
Conclusion on limitation
87. For the reasons I have given, which are essentially the same as those of the
Court of Appeal in The CMA Djakarta, I would hold that, if there were a breach of
the safe port warranty, the charterers would not be entitled to limit their liability
under the Convention in accordance with the limitation fund calculated by reference
to the vessel.
CONCLUSIONS
88. On the safe port issue (question 1), I would hold that the port was not unsafe
within the meaning of the safe port undertaking so that the charterers were not in
breach of it. The conditions in the port amounted to an abnormal occurrence as that
expression is understood in the cases.
89. On the joint insurance issue, if (contrary to para 88 above) there were a breach
of the safe port undertaking, I would have answered question or issue 2 summarised
in para 8 above, namely whether the provisions for insurance in clause 12 of the
Barecon 89 form preclude rights of subrogation of hull insurers and the right of
Page 36
owners to recover in respect of losses covered by hull insurers against the demise
charterers for breach of the express safe port undertaking, in the negative. However
I recognise that this is a minority view.
90. On the limitation issue (question 3), again assuming that there were a breach
of the safe port undertaking, I would hold that the charterers were not entitled to
limit their liability in accordance with a limitation fund calculated by reference to
the vessel.
91. It may be appropriate for declarations to be made reflecting the above. The
parties are invited to make submissions on the form of order and on costs within 21
days of the handing down of the judgments.
LORD SUMPTION:
92. I agree that there was no breach of the safe port warranty in this case, for the
reasons given by Lord Clarke, which substantially correspond to those of the Court
of Appeal.
93. On that footing, it was strictly speaking unnecessary for the Court of Appeal
to decide whether, if there had been a breach, there would have been any liability in
damages. For the same reason, it is unnecessary for us to deal with it. Nonetheless,
I propose to do so, because the question is of some general importance and I am not
persuaded that the Court of Appeal answered it correctly. They held that the demise
charterers suffered no loss by the destruction of the ship because, although there was
a corresponding safe port warranty in the demise charter of which they were (on this
hypothesis) in breach, they would have had no liability to pay damages representing
the value of the ship. This is said to be the result of clause 12 of the demise charter,
which provided for the demise charterers to procure insurance for the vessel at their
own expense against marine, war and protection and indemnity risks, for the joint
interests of themselves and the head owners. From this, and from the provisions of
clause 12 relating to the distribution of the insurance proceeds upon a total loss, it is
said to follow that the head owners were obliged to look exclusively to the insurance
proceeds and not to the demise charterer to recover the value of their ship. Therefore
the demise charterers had no liability to pass on to the time charterers as damages
for breach of the safe port warranty, and the insurers, as their assignees, had no
greater right. The same would, on this analysis, have been true if the insurers had
brought a subrogated claim against the time charterers in the name of the demise
charterers. It is accepted that this argument applies only to that part of the loss which
represents the value of the ship. It does not apply to the claim for SCOPIC expenses
(essentially salvage), wreck removal costs or loss of hire, which together accounted
for rather more than a third of the claim.
Page 37
94. It is necessary to draw attention at the outset to the limited basis on which
this issue comes before the court. There are three possible bases on which a demise
charterer might be in a position to claim damages from a subcharterer for the loss of
a ship of which he is the bailee but not the owner:
(1) on the basis that he is himself liable to the head owner under the
demise charter;
(2) on the basis that as a bailee he has a possessory title which entitles him
to recover in his own name, accounting to the head owner for any recovery
exceeding his actual loss: Waters v Monarch Fire and Life Assurance Co
(1856) 5 El & Bl 870; The Winkfield [1902] P 42; and
(3) under what has sometimes been called the principle of transferred loss,
which may permit a contracting party to recover substantial damages for
breach of contract where the loss is foreseeably suffered by a third party and
the latter has no direct claim against the wrongdoer: Linden Gardens Trust
Ltd v Lenesta Sludge Disposals Ltd [1994] 1 AC 85, Alfred McAlpine
Construction Ltd v Panatown Ltd [2001] 1 AC 518. The recovery in such a
case is held on trust for the third party.
For reasons which it is unnecessary to explore, the insurers have confined their case
to basis (1). The claim was argued on that basis in the courts below, and before us
their counsel (Mark Howard QC) confirmed that the appeal would be argued on that
basis alone.
95. The demise charter was on the Barecon 89 form. The form was originally
drafted in 1974 by the Documentary Committee of the Baltic and International
Maritime Council, and revised in 1989. It is said to have become, in one or other of
its variants, the most commonly used form of bareboat charter world-wide. Under
clause 9 of the form, the demise charterers have the usual obligation to maintain the
vessel in good repair and efficient operating condition and to take immediate steps
to have any necessary repairs carried out. The form as printed contains no trading
limits other than clause 5, which simply requires the vessel to be employed in
conformity with the terms of its insurances, including any insurance warranties,
unless the agreement of the insurers is obtained. The insurances are governed by
either clause 12 or clause 13, one of which must be selected. Clause 12, which was
selected in this case, provides (so far as relevant):
Page 38
“12. lnsurance and Repairs
(a) During the Charter period the Vessel shall be kept
insured by the Charterers at their expense against marine, war
and Protection and indemnity risks in such form as the Owners
shall in writing approve, which approval shall not be
unreasonably withheld. Such marine, war and P & I insurances
shall be arranged by the Charterers to protect the interests of
both the Owners and the Charterers and mortgagees (if any),
and the Charterers shall be at liberty to protect under such
insurances the interests of any managers they may appoint. All
insurance policies shall be in the joint names of the Owners and
the Charterers as their interests may appear.
If the Charterers fail to arrange and keep any of the insurances
provided for under the provisions of sub-clause (a) above in the
manner described therein, the Owners shall notify the
Charterers whereupon the Charterers shall rectify the position
within seven running days, failing which Owners shall have the
right to withdraw the Vessel from the service of the Charterers
without prejudice to any claim the Owners may otherwise have
against the Charterers.
The Charterers shall, subject to the approval of the Owners and
the Underwriters, effect all insured repairs and shall undertake
settlement of all costs in connection with such repairs as well
as insured charges, expenses and liabilities (reimbursement to
be secured by the Charterers from the Underwriters) to the
extent of coverage under the insurances herein provided for.
The Charterers also to remain responsible for and to effect
repairs and settlement of costs and expenses incurred thereby
in respect of all other repairs not covered by the insurances
and/or not exceeding any possible franchise(s) or deductibles
provided for in the insurances.

(b) If the conditions of the above insurances permit
additional insurance to be placed by the parties, such cover
shall be limited to the amount for each party set out in Box 28
Page 39
and Box 29, respectively. The Owners or the Charterers as the
case may be shall immediately furnish the other party with
particulars of any additional insurance effected, including
copies of any cover notes or policies and the written consent of
the insurers of any such required insurance in any case where
the consent of such insurers is necessary.
(c) Should the Vessel become an actual, constructive,
compromised or agreed total loss under the insurances required
under sub-clause (a) of clause 12, all insurance payments for
such loss shall be paid to the Mortgagee, if any, in the manner
described in the Deed(s) of Covenant, who shall distribute the
moneys between themselves, the Owners and the Charterers
according to their respective interests. The Charterers
undertake to notify the Owners and the Mortgagee, if any, of
any occurrences in consequence of which the Vessel is likely
to become a Total Loss as defined in this clause.
(d) If the Vessel becomes an actual, constructive,
compromised or agreed total loss under the insurances arranged
by the Charterers in accordance with subclause (a) of this
Clause, this Charter shall terminate as of the date of such loss.
(e) …
(f) For the purpose of insurance coverage against marine
and war risks under the provisions of sub-clause (a) of this
clause, the value of the vessel is the sum indicated in Box 27.”
The alternative insurance clause, clause 13, was intended for short-term demise
charters and envisaged that the demise charterer would become entitled under
existing insurance arrangements made by the head owner. It accordingly provided
for the vessel to be kept insured against marine and war risks by the owner at their
expense under a policy in joint names, and against P & I risks by the charterers at
their expense. For present purposes, however, the most significant difference
between the two clauses consists in the addition in clause 13(a) of an express
provision dealing with the relationship between the liability of insurers and that of
demise charterers. It provided:
“The Owners and/or insurers shall not have any right of
recovery or subrogation against the Charterers on account of
Page 40
loss of or any damage to the vessel or her machinery or
appurtenances covered by such insurance, or on account of
payments made to discharge claims against or liabilities of the
Vessel or the Owners covered by such insurance.”
96. In the present case, the Barecon 89 form was amended by deleting the trading
limits clause (clause 5) and adding at clause 29 a safe port warranty in the following
terms:
“29. Trading Exclusions
Vessel to be employed in lawful trades for the carriage of
lawful merchandises only between good and safe berths, ports
or areas where vessel can safely lie always afloat, always
accessible within IWL except NAABSA in River Plate where
it is customary for similar size or similar dimension vessels to
safely lie aground, specially excluding Abkhazia, Albania,
Angola, Bosnia-Herzegovina, CLS Pacific ports, Democratic
Republic of Congo (formerly Zaire), Eritrea, Israel, North
Korea, Lebanon, Liberia, Libya, Sierra Leone, Somalia, Sri
Lanka, Federal Republic of Yugoslavia, Zimbabwe, in Arabian
Gulf and adjacent waters including the Gulf of Oman North of
24 deg North, any United Nation embargo countries/ports.
Charterers have right to send vessel to the war/warlike zone or
other zones for which additional insurance are levied by
vessel’s war risk insurers. In such event, Charterers are fully
responsible to pay for all additional war risk premium upon
demand by vessel’s underwriters and/or P+I club with all
risks/consequences to be for Charterers’ account.
Charterers shall have right to break IWL in which case
Charterers are fully responsible to pay for all additional
premium upon demand by vessel’s underwriters and/or P+I
Club for breaching IWL with all risks/consequences to be for
Charterers’ account.
Any ice affected port(s) and/or place(s).
No direct sailing between PRC and Taiwan or vice versa.”
Page 41
It is not disputed that if Kashima was an unsafe port there was a breach of clause 29.
97. The provisions which are said to exclude a right to recover the value of the
ship as damages for breach of clause 29 are clauses 12(a) and (c). Clause 12(a)
requires insurance to be in place for the parties’ joint account. Clause 12(c) provides
that “all insurance payments for such loss shall be paid to the Mortgagee, if any, in
the manner described in the Deed(s) of Covenant, who shall distribute the moneys
between themselves, the owners and the charterers according to their respective
interests.” The argument is that the clause as a whole is a complete code governing
financial liability for loss or damage to the ship, and that the words quoted provide
for the relief, and the only relief, available as between the head owner and the demise
charterer for a total loss. To address this argument, I propose to deal first with the
law relating to rights as between co-insured. I shall then consider how, if at all, it
applies to this demise charter.
98. The starting point is the general rule that insurance recoveries are ignored in
the assessment of damages arising from a breach of duty: Bradburn v Great Western
Railway Co (1874) LR 10 Ex 1; Parry v Cleaver [1970] AC 1. This can conveniently
be called the collateral payments exception. It is a departure from the general
principle that collateral benefits are brought into account, and is probably best
regarded as being based on public policy. Insurance recoveries are a benefit which
the injured party has bought in consideration of his premiums, which are intended
to inure to his benefit alone, not that of third party wrongdoers. Moreover, the courts
have traditionally been concerned to preserve the subrogation rights of insurers
against those who are legally responsible for the loss, which are an important part
of the economics of insurance. The effect of the collateral payments exception is
that as between the insured and the wrongdoer who has caused the loss, they are not
treated as making good the former’s loss or as discharging the latter’s liability. The
assumption underlying it is that as far as the wrongdoer is concerned, insurance is
res inter alios acta, ie, loosely translated, none of his business. The rule thus stated
falls to be modified in a case where insurance manifestly is the wrongdoer’s business
because, for example, he is a co-insured and/or the insurance is taken out for his
benefit. The business context in which this has most commonly arisen is the coinsurance of employer, contractor and subcontractors under standard forms of
building contract.
99. It is well established, and common ground between the present parties, that
where it is agreed that the insurance shall inure to the benefit of both parties to the
contract, they cannot claim against each other in respect of an insured loss. Coinsurance is the paradigm case. The principle first appears in the United States, but
was successively adopted in early editions of MacGillivray on Insurance Law, by
the Supreme Court of Canada in Commonwealth Construction Co Ltd v Imperial Oil
Ltd [1978] 1 SCR 317 and by the English courts in a line of cases beginning with
the decision of Lloyd J in Petrofina (UK) Ltd v Magnaload Ltd [1984] QB 127.
Page 42
What is less clear is its juridical basis. Lloyd J was inclined to think that it was based
on the rule against circuity of action, which is difficult to accept given that the
insurer will not be a party to any litigation between the co-insureds. The better view,
which was endorsed by the House of Lords in Co-operative Retail Services Ltd v
Taylor Young Partnership Ltd [2002] 1 WLR 1419, paras 61-65 (Lord Hope), is that
it is an implied term of the contract of insurance and/or of the underlying contract
between the co-insureds pursuant to which their interests were insured. The
implication is necessary because if the co-insureds are both insured against the
relevant loss, the possibility of claims between them is financially irrelevant. It
would be absurd for the insurer to bring a subrogated claim against a co-insured
whom he would be liable to indemnify against having to meet it. It should be noted
that this reasoning is relevant only to the position as between the co-insureds. In all
of the English cases before this one the question arose between the co-insureds and
their insurer. None of them raised the question how the principle about co-insurance
affects claims against a third party wrongdoer who is not himself a co-insured and
is not party to the arrangements between them. There is no necessity to exclude a
claim against him and indeed no reason why either of the co-insureds or their insurer
should wish to do so. It is impossible to identify any contract whose business
efficacy depends upon that result being achieved.
100. As between a co-insured (or his insurer) and a third party wrongdoer, a
different question arises which none of the existing English authorities purports to
answer. The question is this: when we say that one co-insured cannot claim damages
against another for an insured loss, is that because the liability to pay damages is
excluded by the terms of the contract, or is it because as between the co-insureds the
insurer’s payment makes good any loss and thereby satisfies any liability to pay
damages? The significance of this question may be illustrated by a hypothetical case.
Suppose that A and B are engaged in some contractual venture, involving the use of
A’s property. The property is insured in their joint names. It is damaged in breach
of some contractual duty owed to A by B, but the cause of the damage is some act
of B’s agent, X. If the effect of the co-insurance is that B’s liability to pay damages
to A is excluded, then B never had a relevant liability and has suffered no loss which
he can claim over against X. But if its effect is that payment by the insurer makes
good A’s loss as between A and B and thereby satisfies any liability of B, the result
is different. The effect is to exclude the collateral payments exception, so as between
A and B the receipt of the insurance proceeds must be taken into account. However,
the fact that the insurer’s payment has made good the loss as between A and B does
not mean that it has done so as between B and the stranger, X. As between B and X
the insurance is res inter alios acta. Indeed, its normal consequence is that the claim
will survive to be pursued by the subrogated insurers. Either analysis will achieve
the object of the implication, namely to prevent claims between co-insureds. But
they have radically different consequences for claims against third parties. Which is
the correct analysis must depend on the particular terms of the particular contract.
The answer will not necessarily be the same in every case.
Page 43
101. I therefore return to the contractual arrangements between the head owners
and demise charterers of the “OCEAN VICTORY”. We have not seen the actual
policy. What matters, however, is not the actual policy but the policy envisaged in
clause 12 of the demise charter. The relevant insurance is the insurance against
marine risks which is required by clause 12(a). That is an insurance on property. As
far as the demise charterer is concerned, although it is not a liability insurance, he is
treated as having an insured interest in the property as such, because his potential
liability to the head owner as a bailee and time charterer means that he “stands in [a]
legal or equitable relation to the adventure or to any insurable property at risk
therein, in consequence of which he may benefit by the safety or due arrival of
insurable property, or may be prejudiced by its loss, or by damage thereto, or by the
detention thereof, or may incur liability in respect thereof”: see Marine Insurance
Act 1906, section 5(2) (emphasis supplied); cf Arnould, Law of Marine Insurance
and Average, 18th ed (2013), para 11.37-11.38. If the ship is lost or damaged, the
measure of any liability of the demise charterer will be the same as the measure of
the owner’s loss, namely the diminution, partial or total, in the value of the ship.
There is no other basis on which he could be entitled to insure on the same basis as
the owner.
102. When the “OCEAN VICTORY” was lost, the insurers were bound under
clause 12(c) to pay its insured value to the head owner or, rather, to the mortgagee
for the account of the head owner. But the natural legal inference from (i) the fact
that the demise charterer is insured for his interest in the ship, (ii) the implied
prohibition of claims for damages between the co-insured for loss of or damage to
the ship, and (iii) the avoidance of double recovery, is that the insurer’s payment to
the head owner makes good the head owner’s loss not just as between the insurer
and the head owner but as between both of them and the demise charterer. The
demise charterer’s liability under the demise charter for the loss of the ship has not
been excluded. It has been satisfied. It follows that the demise charterer may claim
over against a time charterer who is not party to the insurance or any of the
contractual arrangements connected with it.
103. This may be tested by asking what would happen if the insurer did not pay,
for some reason which did not involve a breach of duty by either co-insured, for
example because the insurer became insolvent after the casualty. The demise
charterer’s obligation to insure ceases upon the termination of the demise charter:
see clause 12(a). And the demise charter terminates upon the total loss of the ship:
see clause 12(d). The result is that there is no effective insurance, no default of the
demise charterer in there being no effective insurance, and no basis on which the
head owner can be supposed to look exclusively to its proceeds to make good a loss
arising from a breach of the safe port warranty. The demise charterer would be
bound to pay damages, not because he was responsible for the lack of insurance but
because he was liable for the destruction of the ship in breach of his contract. This
can only be because it is the payment of the insurance proceeds which discharges
Page 44
the liability of the demise charterer by making good the head owner’s loss. In the
absence of payment, there is no discharge and no bar to a claim between the coinsureds.
104. This does mean that if the insurer, being solvent, delays in paying the claim,
the head owner can require the demise charterer to pay the loss at once. But I do not
regard that as undermining what I consider to be the way that clause 12 works. Quite
apart from the consideration that the contract must be construed on the assumption
that the insurer will perform his obligations, an obligation to pay damages upon a
breach of contract is a routine consequence of the chartered service. It is no more
unthinkable in the case of the demise charter than it is in the case of the time charter,
where no question of co-insurance arises.
105. This analysis derives strong support from a number of other features of this
particular demise charter:
(1) In the Barecon 89 form as printed, the demise charterer’s sole
obligation in relation to the physical condition of the ship was the
maintenance and repairing obligation in clause 9. Breach of that obligation
would not give rise to an insured loss. The only trading warranty was in
clause 5. This did not give rise to an insured loss either, because it did no
more than prohibit the trading of the vessel to places where she would not be
insured. Under the unamended Barecon 89 form, therefore, it is difficult to
envisage circumstances in which an insured loss could arise from a breach of
contract by the demise charterers. The present issue simply could not arise.
The parties are therefore unlikely to have intended to address it. The present
issue arises only because the printed form has been modified by adding in
clause 29 a contractual obligation not to trade the vessel to unsafe ports where
she may suffer an insured loss. In adding clause 29 to the printed form, the
parties must have intended that in relation to loss or damage arising from the
unsafeness of ports, the liability of the charterers and the insurers would
coexist. If, as the Court of Appeal thought, the liability of the charterers did
not extend to damages, the parties must be taken to have included an
elaborate trading warranty which is almost entirely redundant. Its only effect,
on this view of the matter, was to entitle the head owners to protest if they
happened to learn in advance of the demise charterer’s intention to visit an
unsafe port. If the demise charterers persisted (as they were in a position to
do since they were in operational control of the ship and employed the master
and crew), and the ship was damaged or lost, there would be no consequences
in damages. This is not a realistic intention to impute to commercial parties
in the absence of express words to that effect.
Page 45
(2) The only words of clause 12 which are said to have this effect are the
words of clause 12(c) which deal with the distribution of the insurance
proceeds upon a total loss. They seem to me to be irrelevant to the present
issue. They deal with the mechanics of payment of the insurance proceeds
and not the substantive rights of the parties. The proceeds are to be paid in
the first instance to the mortgagee bank, and thereafter to the head owner and
demise charterer according to their respective interests. This provision does
not exclude a right to damages for breach of contract. Its purpose appears to
be (i) to protect the position of the mortgagee, and (ii) to distinguish the
position where there is a partial loss (see the fourth paragraph of clause 12(a))
when the insurance proceeds will go to the demise charterer by way of
indemnity against the cost of repairs.
(3) It is right to add that in relation to war risks clause 29 expressly
provides that “all risks / consequences” of breach are to be for charterers’
account, notwithstanding the obligation under clause 12 to maintain war risks
insurance.
(4) If clause 12(c) contains the decisive language, as the time charterers
have suggested, then it becomes necessary to distinguish between insurance
payments under the hull policy in respect of the loss of the ship and insurance
payments under the P & I insurance representing other elements of the claim
such as SCOPIC and wreck-raising expenses. The latter will be recoverable
as damages in the ordinary way, as well as under the express indemnities at
clauses 17 and 18 of the demise charter. Logically, it also requires a
distinction to be made between the insured value of the ship (in this case,
$70m) and its market value (in this case $88.5m), the difference being
recoverable as damages, as the time charterers accepted before Teare J. It is
very difficult to see why, if the principle underlying clause 12 was that the
parties were to look exclusively to the insurance proceeds for compensation
for a breach of clause 29, they should have intended this arbitrary distinction
between different elements of the loss.
(5) Finally, there are the terms of clause 13(a), which expressly exclude
any right to recover damages in respect of insured loss of or damage to the
ship. If that option is chosen, the result for which the time charterers contend
is achieved by the express words of the contract. I recognise that clause 13 is
designed for a very different kind of chartered service. It is nonetheless a
striking fact that when the draftsman of this contract wished to deal with the
overlap between the liability of the insurers and that of the demise charterers,
he did so in express terms, using language which finds no equivalent in clause
12.
Page 46
(6) Lord Clarke has made some further observations at paras 49 to 57 of
his judgment, with which I agree.
106. In my judgment the Court of Appeal was wrong to hold that the demise
charterers were relieved by the terms of the demise charter of the obligation to pay
damages for the consequences of an order to an unsafe port. I would therefore have
allowed the appeal on this point if it had arisen.
107. On the limitation issue, I agree with Lord Clarke.
LORD MANCE:
108. I agree with the judgment prepared by Lord Clarke on the first and third issues
and with the judgment of Lord Toulson on the second issue in this appeal. In what
follows, I set out some supplementary reasons of my own for agreeing with Lord
Toulson on the second issue.
109. Clauses 12 and 13 are standard clauses of the “‘Barecon 89’ Standard
Bareboat Charter” issued by the Baltic and International Maritime Council
(“BIMCO”). It is relevant to consider how the scheme introduced by clause 12 (or,
where used, clause 13) operates in circumstances where there is no clause 29. A
demise charter involves a bailment on whatever terms may be agreed. Clauses 9 and
12 impose on charterers strict responsibility for having all necessary repairs done.
But a demise dharterer may well cause total loss of the demised vessel in
circumstances constituting a breach of duty or of an express or implied contract
term. Printed clause 5, which was replaced in the present charter by clauses 29 and
30, could itself involve such a breach. Its first and third paragraphs include, for
example, various obligations relating to lawful trading and the carriage of suitable
lawful merchandise.
110. Clause 12 (or, where used, 13) contains a scheme designed to address the
possibility of the vessel requiring repairs or suffering a total loss. First, and
fundamentally, these clauses provide who is to take out marine, war risks and
protection and indemnity (“P&I”) insurance. Clause 12 deals with circumstances
where the charterers are to do both, clause 13 with circumstances in which owners
are to maintain marine and war risks insurance while demise charterers have to
maintain P&I insurance. Secondly, they provide for the nature of the insurance
which is to be maintained. Under clause 12 all the insurances are to be “in such form
as the owners shall in writing approve, which approval shall not be unreasonably
withheld”, whereas clause 13 provides for marine and war risks insurance “under
the form of policy … attached hereto”, while the P&I insurance to be arranged by
Page 47
charterers has again to be “in such form as the owners shall in writing approve,
which approval shall not be unreasonably withheld”. Under both clauses 12 and 13,
the value of the vessel for marine and war risks insurance purposes is to be that
stated in box 27 (here USD 70m), while boxes 28 and 29 are to specify whether
either or both parties may take out additional insurance and if so for how much (each
being, under the present charter, prohibited from so doing).
111. Thirdly, under clause 12 all the insurances are to be arranged to protect “both
the Owners and the Charterers and mortgagees (if any)”and to be “in the joint names
of the Owners and the Charterers as their interests may appear”. Under clause 13,
the marine and war risks insurances are likewise to be “in the joint names of the
Owners and the Charterers as their interests may appear”, but there is also an express
provision that:
“The Owners and/or insurers shall not have any right of …
subrogation against the Charterers on account of loss of or any
damage to the Vessel or her machinery or appurtenances
covered by such insurance, or on account of payments made to
discharge claims against or liabilities of the Vessel or the
Owners covered by such insurance.”
112. Fourthly, both clauses 12 and 13 make the demise charterers responsible for
effecting any repairs, securing reimbursement from underwriters to the extent of the
insurance coverage, but remaining responsible for all repairs not covered by the
insurance and/or falling within any possible insurance franchise or deductibles.
113. Fifthly, both clauses 12 and 13 address the possibility of the vessel becoming
an actual, constructive, compromised or agreed total loss. Clause 12(c) provides that
in this event the marine or war risks insurance payments for such loss “shall be paid
to the Mortgagee (if any), … who shall distribute the moneys between themselves,
the owners and the charterers according to their respective interests”. Clause 13(h)
provides that they “shall be paid to the Owners, who shall distribute the moneys
between themselves and the Charterers according to their respective interests”.
114. The scheme of clause 12 (and 13) is clearly intended to be comprehensive.
Whatever the causes, both repairs and total losses fall to be dealt with in accordance
with its terms, rather than by litigation to establish who might otherwise be
responsible for undertaking them, for bearing the risk of their occurrence or for
making them good. This is reinforced by the provisions for marine and war risks
insurances to be taken out to protect the interests of owners, charterers and any
mortgagees, and to be in the joint names of owners and charterers, as their interests
may appear. It is well established, as Lord Sumption and Lord Toulson both
Page 48
acknowledge, that, where it is agreed that insurance shall inure to the benefit of both
parties to a venture, the parties cannot claim against each other in respect of an
insured loss. This principle is now best viewed as resting on the natural
interpretation of or implication from the contractual arrangements giving rise to such
co-insurance: Co-operative Retail Services Ltd v Taylor Young Partnership Ltd
(“CRS”) [2001] Lloyd’s Ins Law Rep 122 (CA), [2002] 1 WLR 1419, per Lord
Bingham, para 7 (favouring the rationale suggested by Brooke LJ in the Court of
Appeal at para 72) and Lord Hope, paras 61 to 65. It is merely reinforced where, as
here, the principal co-insureds, owners and charterers, are in the same group and
ultimate beneficial ownership. Hull insurance covers losses whether or not it is due
to the fault of any party, and it is, rightly, not suggested that the principle in CRS is
subject to any exception where the loss is due to fault: see also on this point Mark
Rowlands Ltd v Berni Inns Ltd [1986] QB 211, 232G-233B, per Kerr LJ.
115. Repairs were to be undertaken by charterers, regardless whether their cost
was fully covered by insurance. But the scheme of both clauses 12 and 13 was for
marine and war risks insurances in agreed form to be taken out in a fixed amount,
and for no more unless otherwise specifically agreed. At the date of her total loss
the OCEAN VICTORY is said to have been worth some USD 15m more than the
amount for which she was valued for insurance under the demise charter. It is in my
opinion implausible to suggest that, having developed this careful scheme for
specific protection of their respective interests, it should have been intended that
owners should as against charterers, a company in the same group and beneficial
ownership, be able to reopen the scheme by claims of breach, exposing charterers to
paying damages for the hull loss based on a different alleged value to that which
owners and charterers had agreed between themselves. Just as parties must, for
better or worse, accept a valuation agreed under a marine insurance (Marine
Insurance Act 1906, sections 27(1) and (2)), so here the parties to this charter must
be taken to have accepted the value they agreed for insurance purposes as conclusive
as between themselves.
116. In this respect, the schemes of clauses 12 and 13 are in my opinion effectively
mirror images of each other. Clause 13 has an express exclusion of any right of
recovery or subrogation on the part of owners and/or insurers against charterers.
That is explicable in the context of a clause dealing with insurance taken out by
owners. BIMCO in their observations on the Barecon 89 form explain the optional
clause 13 on the basis that “It has been felt that it may be useful to cover also the
possibility which is believed may arise from time to time, that a vessel is bareboat
chartered for a short period, say, four to six months”. BIMCO give the example of
passenger vessels. They continue: “It is believed that it is normal practice that the
Owners carry on with the insurances for their own account”. BIMCO also confirm
that:
Page 49
“The main difference between Clause 12 and Clause 13 is that
in Clause 13 the responsibility for arranging and keeping the
marine and war risks insurances has been shifted back to the
Owners.”
117. There is no suggestion by BIMCO that the responsibilities for repairs and
total loss differ in any other respect as between clauses 12 and 13. There is no reason
to think that clauses 12 and 13 were devised as anything other than two routes to the
same substantive allocation of responsibilities for repairs and total loss, irrespective
of fault. I conclude that the express exclusion of a right of recovery or subrogation
in clause 13 was simply belt and braces in the context of insurances taken out by
owners, and that the reason why no such express term appears in clause 12 was that
it never occurred that there could be such claims in the context of insurances
arranged by charterers to cover their own as well as owners’ interests. It is
inconceivable that the parties intended fundamentally to alter the incidence of risk
by permitting or excluding breach-based claims as between themselves in respect of
a hull loss, depending upon whether it happened to be convenient to continue to use
hull insurances taken out by owners or to rely on fresh insurances taken out by
charterers. The judge’s speculation (para 202) that the risk of insurers’ insolvency
might have motivated such a distinction is unconvincing, where not only is such risk
most unlikely to have been in the forefront of the parties’ minds, but the decision
which party should be responsible for hull insurances has no connection with any
such risk or with the question whether or not breach-based liability should exist.
118. I do not consider that the substitution of printed clause 5 by typed clause 29
can have been intended to, or did, alter this basic scheme. In many respects clause
29 simply makes different provision for the same subject matter as clause 5. Clause
29 overlaps with clause 5, in so far as it requires employment in lawful trades. It
specifies (not entirely felicitously, in so far as it refers to Zimbabwe) certain trading
limits. Clause 5 contemplates that trading limits would be found in Box 19, but adds
that the charterers must furthermore employ the vessel in conformity with the terms
of her insurances. Both clause 29 and clause 5 address additional insurance
premiums, which a vessel’s trading might in some circumstances require to be paid.
Clause 29 has the provision for employment “only between good and safe berths,
ports or areas where vessel can safely lie always afloat” on which Gard relies.
Owners could no doubt object if they learned of intended or actual trading contrary
to clause 29, and a breach by charterers in this respect might have consequences
outside the scope of hull insurance and so outside the scheme of clauses 12 and 13.
But it is, in my opinion, most unlikely that the “safe port” provision in clause 29 can
have been meant to give rise to a system of recourse for loss of the hull, by way of
damages for breach of contract, separate from and potentially counteracting the no
fault scheme of responsibility and insurance recovery for a hull loss introduced by
clause 12.
Page 50
119. Teare J held that the owners, although indemnified by the insurers, had a
subrogated right to claim against the demise charterers damages for breach of clause
29: judgment, paras 203-204. That makes no sense in the context of a co-insurance
such as this, as Mr Recorder Jackson QC pointed out in Hopewell Project
Management Ltd v Ewbank Preece Ltd [1998] 1 Lloyd’s Rep 448, 458, in a passage
approved by Lord Hope in CRS at para 65. It would mean treating the present coinsurance as if it involved two separate and severable insurances, leaving charterers
exposed, without liability cover, to claims for breach of charter or duty brought by
owners in respect of loss of the hull. That is obviously not what anyone contemplated
by clause 12 or clause 13. It is no answer that demise charterers might in turn have
a back to back claim for damages against sub-charterers. They might equally well
not have, eg if they were trading the vessel on their own account, so there was no
sub-charter, or the sub-charter was on different terms or the sub-charterer was not
worth powder and shot. I add, though no-one suggested this as a correct analysis,
that it is inconceivable that anyone contemplated that the co-insurance to be
arranged could give rise to successive payments of the same sum to different parties,
with the second of such payments going to reimburse insurers for the first.
120. In these circumstances, Gard suggests an alternative basis for its claim, that
liability exists in the first instance for any total loss attributable to a breach of the
safe port warranty, but is then discharged to the extent of any recovery under the
marine or war risks insurance taken out under clause 12. This analysis has the twin
results, that (i) owners could after a total loss call upon charterers to pay the value
of the vessel’s hull, even though an insurance claim was on foot, but had not yet
been paid, and (ii) owners could look to charterers in damages for any amount by
which the vessel’s actual value at the date of loss exceeded her insured value, both
before and after insurers had paid the insured value. Both results are in my view
inconsistent with the scheme of clause 12 (or 13), or with any sensible understanding
of its evident purpose to cater comprehensively for responsibility for repairs and
total loss.
121. Still more fundamentally, in the context of the present claim by Gard as
assignees of demise charterers, there is no basis for treating payment of the vessel’s
value under her marine or war risks insurance as “discharging” pro tanto any liability
existing between owners and demise charterers. Both the ordinary marine and the
war risks insurances are property insurances on the vessel’s hull. Payments made
under them go to owners (or their mortgagees) and charterers for their respective
interests in the hull. They cannot be treated as satisfying, at one and the same time,
any liability (if there were any) which charterers had to owners. The contrary
analysis involves the proposition that the parties were prepared to treat the hull
insurance moneys as going in the first instance to charterers, not to the owners or
their mortgagees, and then being passed up to owners. That is clearly not what
happened or would ever happen in fact, and it is only in a counter-factual world and
Page 51
by contradicting the clear intent of clause 12 that it could be treated as if it had
happened.
122. In my opinion, the reason why owners have no claim against charterers for
damages for loss of the hull is not that such a claim exists under clause 29 but is at
some point discharged. It is that, under a co-insurance scheme like the present, it is
understood implicitly that there will be no such claim. This understanding applies,
in my opinion, whether or not the insurance moneys have yet been paid. But, even
if (contrary to my opinion) one were to treat this understanding as biting only upon
payment of the insurance moneys, it still would not arise from or involve the
proposition that some liability of the charterers to owners was “discharged” in the
sense of paid or satisfied by the insurance moneys. The understanding would simply
be that, upon payment of the hull insurance proceeds to those interested in the hull
for their respective interests, no further liability would exist inter se. On that basis,
charterers would still be unable to show that they had or had discharged any liability
to owners, on the basis of which they could pursue a back-to-back claim against subcharterers. But, as I have made clear, in my opinion the implied understanding
arising from the co-insurance scheme is that there would be no liability for the hull
value in the event of a total loss, whether or not the insured value had yet been
disbursed.
123. Lord Sumption raises the question what would happen if an insurer became
insolvent after a loss. This is in my view a remote eventuality which cannot be a
guide to the meaning of clause 12 (or 13). It also raises different considerations of
risk and implications to any which require determination on this appeal. Two views
might be taken. One is that, under a scheme where both parties have agreed to the
particular form of policy, presumably by particular insurers, the risk lies where it
falls. It is to be noted that the risk could impact either party in different
circumstances. It could impact charterers, who undertake an absolute responsibility
for repairs, irrespective of fault, and would be unable to recoup themselves from the
hull insurance as contemplated by clause 12 or 13. It could impact owners and
charterers for their respective interests, in the event of a total loss occurring without
fault. There is no reason to think that the parties intended any different result in the
event of a loss which might be attributed to a breach of clause 29 or any other clause.
But another view might be that the risk lies by implication on the party responsible
for maintaining the insurance during the charter period, even though clauses 12 and
13 provide that the charter terminates as of the date of any total loss. The former
seems the more likely position, but it is unnecessary on this appeal to decide which
applies, since the point cannot assist to determine the present, quite different issue
as to the implications of the insurance scheme when it is effective, which is no doubt
what the parties, and BIMCO, had in mind when using and devising it.
124. Finally, it is submitted that charterers must have some liability towards
owners under clause 29, because otherwise there can be no back-to-back claim down
Page 52
the line under the equivalent clause in the sub-charter between demise charterers
and Sinochart, who in turn cannot pass on liability under the equivalent clause in the
sub-sub-charter between Sinochart and Daiichi, who were and should be ultimately
responsible. The difficulty with this submission is that it has not been tested and I,
for my part, regard it as entirely open. Gard’s case before the Supreme Court has
been put exclusively on the basis that charterers had a liability to owners, which in
turn enables charterers to claim damages down the line. Mr Howard QC, who did
not appear for Gard below, very frankly acknowledged that it had occurred to him,
when he came into the case, that there could have been other bases on which the
claim could have been presented, but that it had been concluded that Gard was
confined to the way in which the case had been argued on its behalf below.
125. Those other bases are (i) that the charterers’ possessory title gave them a
sufficient interest to be able to maintain a claim for the hull loss: compare the
principle in The Winkfield [1902] P 42, whereby a bailee can claim in tort in his own
name, without showing that he has any liability to the head owner, but accounting
to the head owner for any loss exceeding his own loss, and/or (ii) that there are
circumstances in contract where a contracting party can claim substantial damages
for loss of or damage to property, when another person has actually borne such loss
or damage: Dunlop v Lambert (1839) 6 Cl & F 600, The Albazero [1977] AC 774,
846G-847F. More recently, this latter possibility has been more widely recognised,
by giving special treatment to contracts relating to property where loss due to a
breach of the contract will be suffered by a holder of the property other than the
contracting party: see Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd and
St Martins Property Corp Ltd v Sir Robert McAlpine Ltd (“St Martins”) [1994] 1
AC 85 where it was in the contemplation of the parties when the contract was made
that the property, the subject of the contract and the breach, would be transferred to
or occupied by a third party, who would in consequence suffer the loss arising from
its breach: see also Darlington Borough Council v Wiltshier Northern Ltd [1995] 1
WLR 68 and the narrow ground of decision expressed by Lord Browne-Wilkinson
at p 114G-H in St Martins, in which all members of the House joined. In such a
situation, the claimant is seen as suing on behalf of and for the benefit of the injured
third party and is bound to account accordingly: see St Martins, per Lord BrowneWilkinson at p 115A-B and McAlpine Construction Ltd v Panatown Ltd
(“Panatown”) [2001] 1 AC 518, per Lord Clyde, at pp 530E-F and 532D-E. (An
even broader principle was also suggested by Lord Griffiths in St Martins, at p 96F97D and reviewed inconclusively by Lord Browne-Wilkinson at pp 111F-112F as
well as by the members of the House in Panatown, to the effect that a contracting
party might itself have an interest in performance enabling it to claim damages
without proving actual loss.)
126. In the absence of argument, it is not appropriate to reach any conclusion as
to whether or how far either of these principles might have assisted Gard, had it been
open to Gard to rely on them now. Suffice it to say that, since their application has
Page 53
not been tested, I am not prepared to proceed on the basis that charterers must be
recognised as having had liability to owners, in order to be able to claim down the
line against Sinochart and so for Sinochart against Daiichi. If the absence of such a
liability is fatal to a claim by charterers down the line, it must be because neither
alternative basis of claim identified in para 125 is available, and, if neither proved
to be available, that would, presumably, be because the law did not regard this
situation as one where considerations of justice comparable to those reflected in the
reasoning in those cases militated in favour of recognising a right on the part of the
charterers to claim down the line.
LORD HODGE:
127. I also agree with Lord Clarke’s judgment on the first and third issues in this
appeal. On the second issue I agree with the judgments of Lord Toulson and Lord
Mance.
LORD TOULSON:
128. I agree with the judgments of Lord Clarke and the Court of Appeal on the
issue whether there was a breach of the safe port warranty, and I would therefore
dismiss the appeal. I agree also with Lord Clarke on the issue of limitation of liability
and would affirm the decision of the Court of Appeal in The CMA Djakarta [2004]
1 Lloyd’s Rep 460. On the issue as to the effect of the charter’s joint insurance
provisions, I agree with the reasoning and conclusion of the Court of Appeal and
disagree with Lord Clarke and Lord Sumption for reasons explained in the rest of
this judgment.
129. It is a curious feature of this appeal that permission to appeal on the joint
insurance issue was sought on the basis that the Baltic and International Maritime
Council (BIMCO) form of bareboat charter codenamed Barecon 89 is a standard
form of contract in common use and the effect of clause 12 is therefore a matter of
general importance, but the argument turned on the relationship between that clause
and clause 29, which is not part of the standard form.
130. It is nevertheless sensible to begin by considering the effect of clause 12 in
the unamended version of Barecon 89. It is important to understand its basic
structure. The standard terms, set out in Part II of the policy, contain nothing about
safe ports. Clause 5, headed “Trading Limits”, stipulates that the vessel is to be
employed in lawful trades for the carriage of suitable lawful merchandise within
trading limits which may be specified in the schedule which forms Part I of the
Page 54
policy, and that the vessel is not to be employed otherwise than in accordance with
the terms of the insurance which is required to be maintained.
131. The insurance requirements are set out in clause 12 or its alternative, clause
13. Clause 12 is headed “Insurance and Repairs”. It provides:
“(a) During the Charter period the Vessel shall be kept
insured by the Charterers at their expense against marine, war
and Protection and Indemnity risks in such form as the Owners
shall in writing approve, which approval shall not be
unreasonably withheld. Such marine, war and P and I
insurances shall be arranged by the Charterers to protect the
interests of both the Owners and the Charterers and mortgagees
(if any), and the Charterers shall be at liberty to protect under
such insurances the interests of any managers they may
appoint. All insurance policies shall be in the joint names of the
Owners and the Charterers as their interests may appear.
If the Charterers fail to arrange and keep any of the insurances
provided for under the provisions of sub-clause (a) above in the
manner described therein, the Owners shall notify the
Charterers whereupon the Charterers shall rectify the position
within seven running days, failing which Owners shall have the
right to withdraw the Vessel from the service of the Charterers
without prejudice to any claim the Owners may otherwise have
against the Charterers.
The Charterers shall, subject to the approval of the Owners and
the Underwriters, effect all insured repairs and shall undertake
settlement of all costs in connection with such repairs as well
as insured charges, expenses and liabilities (reimbursement to
be secured by the Charterers from the Underwriters) to the
extent of coverage under the insurances herein provided for.
The Charterers also to remain responsible for and to effect
repairs and settlement of costs and expenses incurred thereby
in respect of all other repairs not covered by the insurances
and/or not exceeding any possible franchise(s) or deductibles
provided for in the insurances.
Page 55
All time used for repairs under the provisions of sub-clause (a)
of this Clause and for repairs of latent defects according to
Clause 2 above including any deviation shall count as time on
hire and shall form part of the Charter period.
(b) If the conditions of the above insurances permit
additional insurance to be placed by the parties, such cover
shall be limited to the amount for each party set out in Box 28
and Box 29 respectively. The Owners or the Charterers as the
case may be shall immediately furnish the other party with
particulars of any additional insurance effected, including
copies of any cover notes or policies and the written consent of
the Insurers of any such required insurance in any case where
the consent of such insurers is necessary.
(c) Should the vessel become an actual, constructive,
compromised or agreed total loss under the insurances required
under sub-clause (a) of clause 12, all insurance payments for
such loss shall be paid to the Mortgagee, if any, in the manner
described in the Deed(s) of Covenant, who shall distribute the
moneys between themselves, the Owners and the Charterers
according to their respective interests. The Charterers
undertake to notify the Owners and the Mortgagee, if any, of
any occurrences in consequence of which the Vessel is likely
to become a Total Loss as defined in this Clause.
(d) If the Vessel becomes an actual, constructive,
compromised or agreed total loss under the insurances arranged
by the Charterers in accordance with sub-clause (a) of this
Clause, this Charter shall terminate as of the date of such loss.
(e) The Owners shall upon the request of the Charterers,
promptly execute such documents as may be required to enable
the Charterers to abandon the Vessel to insurers and claim a
constructive total loss.
(f) For the purposes of insurance coverage against marine
and war risks under the provisions of sub-clause (a) of this
Clause, the value of the Vessel is the sum indicated in Box 27.”
Page 56
132. Clause 13 applies in place of clause 12 if the parties so choose in part I of the
policy. In relation to P and I risks during the charter, clause 13 follows the provisions
of clause 12, but in relation to marine and war risks clause 13(a) puts the
responsibility for maintaining cover on the owner. It provides:
“During the Charter period the Vessel shall be kept insured by
the Owners at their expense against marine and war risks under
the form of policy or policies attached hereto. The Owners
and/or insurers shall not have any right of recovery or
subrogation against the Charterers on account of loss of or any
damage to the Vessel or her machinery or appurtenances
covered by such insurance, or on account of payments made to
discharge claims against or liabilities of the Vessel or the
Owners covered by such insurance. All insurance policies shall
be in the joint names of the Owners and the Charterers as their
interests may appear.”
133. The significant feature of bareboat chartering, or chartering by demise, is that
during the period of the charter (in the words of BIMCO’s explanatory notes to
Barecon 89) “the vessel comes in the full possession, at the absolute disposal, and
under the complete control of the bareboat charterers.” The notes add that bareboat
chartering is therefore entirely different from ordinary time chartering when it comes
to the allocation of costs, liabilities and responsibilities.
134. Clause 12 in the unamended form deals comprehensively with the risks of
loss or damage to the vessel and what is to happen in such an event. In summary,
the demise charterer is responsible for arranging and maintaining insurance, in a
form approved by the owner, in the names of both parties for an agreed value; the
charterer is responsible for effecting all insured repairs; the charterer is responsible
for repairs not covered by the insurance, for example, due to deductibles under the
terms of the insurance (or, for that matter, due to use of the vessel outside the terms
of the insurance); and in the case of a total loss covered by the insurance, the clause
provides for the processing of the insurance moneys.
135. BIMCO’s explanation for the optional alternative of clause 13 was that
sometimes a vessel is bareboat chartered for only a short period and it may make
sense for the owners to carry on with the insurances which they are likely to have in
place. Clause 13 therefore provides that the vessel is to be kept insured by the owners
against marine and war risks, and that the owners and their insurers are to have no
right of recovery or subrogation against the charterers on account of loss or damage
covered by such insurance. It would be unnecessary to include equivalent words in
clause 12. It cannot have been the parties’ intention that the charterer’s exposure to
liability should be greater under clause 13, where cover against marine and war risks
Page 57
was to be maintained at the owner’s expense than under clause 12, where it was to
be maintained at the charterer’s expense. Longmore LJ put the point pithily when
he described the exclusion of rights of recovery or subrogation in clause 13 as “a
confirmation rather than a negation of such exclusion in the more usually adopted
clause 12 for the longer term charters when it is the charterers who pay the premium”
(para 88).
136. The critical question then arises as to the effect, in relation to the operation
of clause 12, of the substitution of clause 5 by clause 29, which provides that the
vessel is to be employed in lawful trades for the carriage of lawful merchandises
“only between good and safe berths, ports or areas where the [vessel] can safely lie
always afloat”, etc.
137. On the hypothesis that Kashima was not prospectively a safe port for the
OCEAN VICTORY when Daiichi gave instructions for her to discharge there,
Daiichi was thereby in breach of the safe port undertaking in the time sub-charter
between itself and Sinochart, which in turn was in breach of the equivalent
undertaking in the time charter between itself and the demise charterer, OLH, which
in turn was in breach of clause 29 of the demise charter between itself and the
owners, OVM. The consequence of that breach was that the vessel was lost, but the
demise charterer and the owners were co-insured (as required by clause 12) and the
insurers paid its insured value. One of the insurers, Gard, claims to be entitled to
recover that sum from the time charterers as assignee of the rights of the demise
charterer. The claim therefore depends on the demise charterer being liable to the
owners for that sum by way of damages for breach of clause 29, and thus entitled to
recover the same sum from the time charterer as loss suffered by the demise
charterer.
138. Gard’s case is that the breach of clause 29 caused the loss of the vessel;
therefore the demise charterer was liable to the owners for the vessel’s value, and
the fact that the owners were paid that amount by the insurers is res inter alios acta
as between the demise charterer and the time charterer.
139. The critical question is whether the contractual scheme between the owners
and the demise charterer precluded any claim by the former against the latter for the
insured loss of the vessel. This is a matter of construction. It has become a common
practice in various industries for the parties to provide for specified loss or damage
to be covered by insurance for their mutual benefit, whether caused by one party’s
fault or not, thus avoiding potential litigation between them. The question in each
case is whether the parties are to be taken to have intended to create an insurance
fund which would be the sole avenue for making good the relevant loss or damage,
or whether the existence of the fund co-exists with an independent right of action
for breach of a term of the contract which has caused that loss. Like all questions of
Page 58
construction, it depends on the provisions of the particular contract: see, for
example, Co-operative Retail Services Ltd v Taylor Young Partnership Ltd [2002]
1 WLR 1419.
140. In that case a building owner entered into a standard form of building contract
for the construction of office premises. Under its terms the contractor was required
to take out and maintain a policy in the names of the owner, the contractor and
specialist electrical subcontractors, Hall, for all risks insurance covering loss or
damage to the works from specified perils including fire. Hall entered into a
collateral contract with the owner warranting that it had exercised and would
exercise all reasonable care and skill in the design and execution of the sub-contract
works. A fire occurred causing extensive damage. The owners sued their architects
and mechanical and engineering consultants, who brought third party proceedings
against Hall. This raised the question whether Hall was liable to the owners in
respect of the fire damage, alleged by the third party claimants to have been caused
by Hall’s negligence and breach of warranty. The House of Lords, upholding a
decision by the first instance judge and the Court of Appeal, held that it cannot have
been the parties’ intention that parties who were jointly insured under a contractors’
all risks policy could make claims against one another in respect of damage covered
by the insurance, or that the insurers could make a subrogated claim in the name of
the owners against Hall, and that the court would if necessary hold that there was an
implied term to such effect (which I infer in relation to Hall must logically have
taken effect as an implied term of the collateral contract between itself and the
owners). In so holding the House of Lords approved and applied the reasoning of
Mr Recorder Jackson QC, as he then was, in Hopewell Project Management Ltd v
Ewbank Preece Ltd [1998] 1 Lloyd’s Rep 448, 458, where he described it as
nonsensical if those parties who were jointly insured under a contractors’ all risks
policy would make claims against one another in respect of damage to the contract
works. The implied term presupposes, of course, that the party relying on it has not
by his own conduct prevented recovery of the loss under the policy – a point made
by Jackson J (as he had by then become) in Board of Trustees of the Tate Gallery v
Duffy Construction Ltd [2007] BLR 216.
141. In the present case the Court of Appeal followed the same reasoning in
holding that the proper construction of clause 12 was that there was to be “an
insurance funded result in the event of loss or damage to the vessel by marine risks”
and that, if the demise charterers had been in breach of the safe port clause, they
would have been under no liability to the owners for the amount of the insured loss
because they had made provision for looking to the insurance proceeds for
compensation. It did not consider that the introduction of clause 29 was intended to
alter the way in which clause 12 was to operate.
142. I agree with the Court of Appeal. The demise charter allowed for a subdemise with the owners’ consent, which was not to be unreasonably withheld. The
Page 59
risk existed that the vessel might be directed to an unsafe port, not necessarily by
negligence on anyone’s part, so causing peril to the vessel, but the risk of
consequential damage to the vessel was catered for by the insurance required to be
maintained by the demise charterer in the joint names of itself and the owners. The
commercial purpose of maintaining joint insurance in such circumstances is not only
to provide a fund to make good the loss but to avoid litigation between them, or the
bringing of a subrogation claim in the name of one against the other. I do not accept
that by substituting clause 29 for clause 5 the parties intended to subvert that
purpose. If anything, the present case is stronger in this regard than Co-operative
Retail Services Ltd v Taylor Young Partnership Ltd, because in that case Hall’s
obligation to use skill and care was in a separate contract from the contract between
the owner and the contractor which contained the provisions about insurance
whereas in the present case clauses 12 and 29 are part of a single contract. Gard’s
submission that clause 29 becomes pointless if clause 12 has the effect held by the
Court of Appeal is fallacious. It sets limits on the use of the vessel, breach of which
may give rise to loss, but clause 12 deals with the consequences of loss or damage
to the vessel, regardless of whether it resulted from negligence or other fault of the
demise charterer (or a sub-charterer).
143. Mr Mark Howard QC argued that this interpretation misapprehends the
purpose of clause 12, which in a case of loss caused by a breach of contract by the
charterer does no more than to ensure that the owner’s right to recover damages is
backed by an available fund. This in substance was the argument advanced
unsuccessfully against Hall in Co-operative Retail Services Ltd v Taylor Young
Partnership Ltd. Lord Hope (paras 39 and 40) distinguished between a provision for
insurance which curtails the means of recovering loss whether or not it was caused
by a contracting co-insured’s default, and a provision which backs the other party’s
other obligations with an insuring obligation but leaves the other obligations
enforceable against the other party by other means. He agreed with the judge’s
conclusion that the contractual insurance arrangements meant that if a fire occurred,
the owners were to look to the joint insurance policy to provide the fund for the cost
of restoring and repairing the fire damage rather than, in the judge’s words, indulge
in litigation with each other.
144. In the present case, if one were to ask whether it would have accorded with
the parties’ intentions that on the morning after the loss the owners would have been
entitled to demand immediate payment from the demise charterers, rather than make
a claim on the insurers and wait for it to be settled, my answer would be that they
intended no such thing. The insurance arrangements under clause 12 provided not
only a fund but the avoidance of commercially unnecessary and undesirable disputes
between the co-insured.
145. It does not follow that the demise charterers (or their insurers in their shoes)
necessarily had no available remedy against the time charterers. The court was told
Page 60
that at one stage the insurers intimated a claim analogous to the claim which a bailee
may bring under The Winkfield [1902] P 42. However, the matter was not pursued,
and it would be inappropriate to express a positive view about the likelihood of
success of such a claim (or any alternative), about which the court has heard no
argument.
146. For those reasons I would have upheld the decision of the Court of Appeal
on the recoverability issue, if the demise charterers had been in breach of the safe
port clause. I have had the benefit of reading Lord Mance’s additional reasons, with
which I agree. Like him, I do not think it is necessary to reach a final conclusion
about the position in the case of an insolvent insurer, which it is not reasonable to
suppose was in the minds of the BIMCO drafters of Barecon 89 and ought not to
affect its core interpretation.