Mediterranean Salvage & Towage Ltd v Seamar Trading & Commerce Inc

This appeal arises out of a dispute between Mediterranean Salvage & Towage Limited (‘the owners’) and Seamar Trading and Commerce Limited (‘the charterers’) arising out of a voyage charterparty dated 10 January 2003 on an amended Gencon form. The chartered vessel was the Reborn. It is an appeal from an order made by Aikens J (‘the judge’) on 1 August 2008 dismissing an appeal from a final declaratory award on preliminary issues dated 6 November 2007 made by three LMAA arbitrators, namely Bruce Harris, Mark Hamsher and Michael Baker-Harber.

The dispute arose out of damage allegedly sustained by the vessel during loading at Chekka in the Lebanon as a result of her hull being penetrated by a hidden underwater projection at the loading berth nominated by the charterers. The appeal arises out of the answer which the arbitrators gave to the first of four questions which they considered as preliminary issues. They formulated the question in this way: was there to be implied into the charterparty a term that there was an absolute duty upon the charterers to nominate a safe berth. The arbitrators answered that question no.

The judge expressed the question for decision somewhat differently as follows: if a specific load port is named in a voyage charterparty and there are several possible berths within that port to which a vessel could be directed to load by the charterers and there is no express warranty in the charterparty of the ‘safety’ of either the port or the berth to which the vessel is to be directed by the charterers, is the charterparty subject to an implied term that the charterers must nominate a ‘safe’ berth at that load port? The judge observed that his question put the point in slightly more precise terms without altering the nature of the question of law posed. He too answered the question no and accordingly dismissed the appeal. He gave permission to appeal to this court.

The charterparty was for a voyage from Chekka to Algiers with a cargo of 8000 metric tons of cement in bulk, 5 per cent more or less at owners’ option. The charterparty contained a front page with 26 boxes, two pages with amended standard Gencon clauses numbered 1 to 19 and five pages of typed clauses numbered 20 to 42. Box 10 described the “Loading port or place” as “1 BERTH CHEKKA – 27 FT SW PERMISSIBLE DRAFT”. Box 11 described “the Discharging port or place” as “ALGIERS … (SEE CLAUSE 20). By box 20 and clause 19(a) it was agreed that the charterparty was subject to English law and that any dispute arising out of it would be submitted to arbitration in London.

Clause 1 provided, so far as relevant:

“The said Vessel shall … proceed to the loading port(s) or place(s) stated in Box 10 or so near thereto as she may safely get and lie always afloat … and being so loaded the Vessel shall proceed to the discharging port(s) or place(s) stated in Box 11 … or so near thereto as she may safely get and lie always afloat, and there deliver the cargo.”

Clause 20 provided, so far as relevant,

“Owners guarantee and warrant that upon arrival of the vessel to and/or prior its departure from, loading or discharging ports (either in ballast condition prior to loading or laden prior discharging) the vessel including, inter alia the vessel's draft, shall fully comply with all restrictions whatsoever of the said ports (as applicable at relevant time) including their anchorages, berths and approaches and that they have satisfied themselves to their full satisfaction with and about the ports specifications and restrictions prior to entering into this Charter Party.”

It was thus common ground that Chekka was agreed as the load port. It was further common ground that it was for the charterers to nominate the berth at which the vessel was to be loaded and that there was no express warranty of safety in respect of either the port or the berth. It follows that, in order to succeed, it was necessary for the owners to establish that an appropriate term was to be implied into the charterparty as to the safety of the berth. Mr Bailey QC submitted for the owners that the obligation on the charterers to nominate the berth was itself an implied term and that the only question therefore was as to the precise form of the implication.

Although the judge’s formulation of the question might suggest that there were several possible berths which might be nominated at Chekka, we were told on behalf of the charterers by Mr Croall QC that there are only two. Mr Bailey said that there was no finding of the arbitrators to that effect but that he was willing to argue the appeal on that assumption. This might be a relevant factor on the facts of some cases but, in my opinion, it is not critical in this case.

The correct approach to the question when to imply a term into a contract or other instrument, including therefore a charterparty, has recently been considered by Lord Hoffmann, giving the judgment of the Judicial Committee of the Privy Council, which also comprised Lord Rodger, Baroness Hale, Lord Carswell and Lord Brown, in Attorney General of Belize v Belize Telecom Limited [2009] UKPC 11. I predict that his analysis will soon be as much referred to as his approach to the construction of contracts in Investors Compensation Scheme v West Bromwich Building Society [1998] 1 WLR 896 at 912-3. His analysis in the Belize case is extensive: see [16] to [27].

It repays detailed study but for present purposes it is I think sufficient to say that the implication of a term is an exercise in the construction of the contract as a whole: see Trollope & Colls Limited v North West Metropolitan Hospital Board [1973] 1 WLR 601, 609 per Lord Pearson, with whom Lord Guest and Lord Diplock agreed and Equitable Life Assurance Society v Hyman [2002] 1 AC 405, 459, where Lord Steyn said:

“If a term is to be implied, it could only be a term implied from the language of [the instrument] read in its commercial setting.”

See Belize at [19] and [20].

At [17] Lord Hoffmann made the important point that the question of implication arises when the instrument does not expressly provide for what is to happen when some event occurs. The most usual inference in such a case is that nothing is to happen. If the parties had intended something to happen, the instrument would have said so. Otherwise, the express provisions of the instrument are to continue to operate undisturbed. If the event has caused loss to one or other of the parties, the loss lies where it falls. As appears below, that seems to me to be of particular significance here because this is not a case in which each party needs to establish a competing implied term if it is to succeed. Put another way, it is not a two way test. Absent an implied term, the default position is that the owners must bear their own loss.

The charterers do not need to establish an implied term. It is their case that the loss should lie where it falls and that, if the parties had intended the charterers to warrant the safety of the loading berth, they could and would have said so, as is very common in voyage charterparties. Sometimes the printed clauses provide for express safe port and safe berth warranties and sometimes typed clauses do so. The charterers say that whether or not to do so is a matter of choice and there is no warrant for the conclusion that it is necessary to do so in the particular circumstances of this case.

The central part of Lord Hoffmann’s reasoning is from [21] to the first part of [25], where he focused on some of the tests which have historically been used to identify when a term is to be implied into a contract. He said this:

“[21] It follows that in every case in which it is said that some provision ought to be implied in an instrument, the question for the court is whether such a provision would spell out in express words what the instrument, read against the relevant background, would reasonably be understood to mean. It will be noticed from Lord Pearson's speech that this question can be reformulated in various ways which a court may find helpful in providing an answer – the implied term must "go without saying", it must be "necessary to give business efficacy to the contract" and so on – but these are not in the Board's opinion to be treated as different or additional tests. There is only one question: is that what the instrument, read as a whole against the relevant background, would reasonably be understood to mean?

[22] There are dangers in treating these alternative formulations of the question as if they had a life of their own. Take, for example, the question of whether the implied term is "necessary to give business efficacy" to the contract. That formulation serves to underline two important points. The first, conveyed by the use of the word "business", is that in considering what the instrument would have meant to a reasonable person who had knowledge of the relevant background, one assumes the notional reader will take into account the practical consequences of deciding that it means one thing or the other. In the case of an instrument such as a commercial contract, he will consider whether a different construction would frustrate the apparent business purpose of the parties. That was the basis upon which Equitable Life Assurance Society v Hyman … was decided. The second, conveyed by the use of the word "necessary", is that it is not enough for a court to consider that the implied term expresses what it would have been reasonable for the parties to agree to. It must be satisfied that it is what the contract actually means.

[23] The danger lies, however, in detaching the phrase "necessary to give business efficacy" from the basic process of construction of the instrument. It is frequently the case that a contract may work perfectly well in the sense that both parties can perform their express obligations, but the consequences would contradict what a reasonable person would understand the contract to mean. Lord Steyn made this point in the Equitable Life case (at p 459) when he said that in that case an implication was necessary "to give effect to the reasonable expectations of the parties."

[24] The same point had been made many years earlier by Bowen LJ in his well known formulation in The Moorcock (1889) 14 PD 64, 68:

"In business transactions such as this, what the law desires to effect by the implication is to give such business efficacy to the transaction as must have been intended at all events by both parties who are business men"

[25] Likewise, the requirement that the implied term must "go without saying" is no more than another way of saying that, although the instrument does not expressly say so, that is what a reasonable person would understand it to mean. …”

Lord Hoffmann then warned against considering the subjective state of mind of the parties or their representatives and stressed the need for the court to be satisfied that the proposed implication spells out what the contract would reasonably be understood to mean. He then stated the Judicial Committee’s view that the question how the actual parties would have reacted to the proposed amendment was irrelevant and added that it was not necessary for the implied term to be obvious in the sense of being immediately apparent.

Importantly, he concluded his analysis in [26] and [27] by reference to BP Refinery (Westernport) Pty Ltd v Shire of Hastings (1977) 180 CLR 266, 282-283, where Lord Simon of Glaisdale, giving the advice of the majority of the Judicial Committee, said that it was “not … necessary to review exhaustively the authorities on the implication of a term in a contract” but that the following conditions (“which may overlap”) must be satisfied:

“(1)

it must be reasonable and equitable; (2) it must be necessary to give business efficacy to the contract, so that no term will be implied if the contract is effective without it; (3) it must be so obvious that ‘it goes without saying’ (4) it must be capable of clear expression; (5) it must not contradict any express term of the contract”.

Lord Hoffman expressed the Judicial Committee’s opinion thus:

“[27] The Board considers that this list is best regarded, not as series of independent tests which must each be surmounted, but rather as a collection of different ways in which judges have tried to express the central idea that the proposed implied term must spell out what the contract actually means, or in which they have explained why they did not think that it did so. The Board has already discussed the significance of "necessary to give business efficacy" and "goes without saying". As for the other formulations, the fact that the proposed implied term would be inequitable or unreasonable, or contradict what the parties have expressly said, or is incapable of clear expression, are all good reasons for saying that a reasonable man would not have understood that to be what the instrument meant.”

It is thus clear that the various formulations of the test identified by Lord Simon are to be treated as different ways of saying much the same thing. Moreover, as I read Lord Hoffmann’s analysis, although he is emphasising that the process of implication is part of the process of construction of the contract, he is not in any way resiling from the often stated proposition that it must be necessary to imply the proposed term. It is never sufficient that it should be reasonable. This point is clear, for example, from the well-known speech of Lord Wilberforce in Liverpool City Council v Irwin [1977] AC 239, where he rejected at page 253H to 254A the approach of Lord Denning, which was to permit the implication of reasonable terms. He identified two classes of implied term in the case (as here) of a complete, bilateral contract. He said that in a case of established usage the courts are spelling out what both parties know and would, if asked, unhesitatingly agree to be part of the bargain. That is not, in my opinion, this case. Lord Wilberforce added at page 253G:

“In other cases, where there is an apparently complete bargain, the courts are willing to add a term on the ground that without it, the contract will not work – this is the case, if not of The Moorcock … itself on its facts, at least of the doctrine of The Moorcock as usually applied.”

Lord Wilberforce stressed that the test is one of necessity. Is it necessary to make the contract work?

I should also note that, since the end of the argument, Rix LJ has drawn my attention to what he described in Socimer Bank Limited v Standard Bank Limited [2008] EWCA Civ 116, [2008] Bus LR 1304 at [105] as a useful and authoritative modern restatement of the relevant principles by Sir Thomas Bingham MR, giving the judgment of this court, which also comprised Stuart-Smith and Leggatt LJJ, in Phelps Electronique Grand Public SA v British Sky Broadcasting Limited [1995] EMLR 472.

Rix LJ quoted an extensive passage at pages 480 to 482 in Phelps. So I will not do the same but will content myself with these few points, which seem to me to underline the principles stated by Lord Hoffmann but also to stress the importance of the test of necessity. Thus, after saying that both parties accepted the propositions stated by Lord Simon in the BP Refinery case (and quoted by Lord Hoffmann), Sir Thomas Bingham said that they distilled the essence of much learning on implied terms but that their simplicity could be almost misleading. He then said this:

“The courts' usual role in contractual interpretation is, by resolving ambiguities or reconciling apparent inconsistencies, to attribute the true meaning to the language in which the parties themselves have expressed their contract. The implication of contract terms involves a different and altogether more ambitious undertaking: the interpolation of terms to deal with matters for which, ex hypothesi, the parties themselves have made no provision. It is because the implication of terms is potentially so intrusive that the law imposes strict constraints on the exercise of this extraordinary power.”

Reference was then made to cases in which terms are routinely and unquestionably implied, as in the case of a term that a surgeon will exercise all reasonable care and skill. He added:

“But the difficulties increase the further one moves away from these paradigm examples. … It is much more difficult to infer with confidence what the parties must have intended when they have entered into a lengthy and carefully-drafted contract but have omitted to make provision for the matter in issue. Given the rules which restrict evidence of the parties' intention when negotiating a contract, it may well be doubtful whether the omission was the result of the parties' oversight or of their deliberate decision; if the parties appreciate that they are unlikely to agree on what is to happen in a certain not impossible eventuality, they may well choose to leave the matter uncovered in their contract in the hope that the eventuality will not occur.

The question of whether a term is to be implied, and if so what, almost inevitably arises after a crisis has been reached in the performance of the contract. So the court comes to the task with the benefit of hindsight, and it is tempting for the court then to fashion a term which will reflect the merits of the situation as they then appear. Tempting, but wrong. For, as Scrutton LJ said in Reigate v Union Manufacturing Co (Ramsbottom) Limited [1918] 1 KB 592 at 605:

“A term can only be implied if it is necessary in the business sense to give efficacy to the contract; that is, if it is such a term that it can confidently be said that if at the time the contract was being negotiated some one had said to the parties, 'What will happen in such a case', they would both have replied, 'Of course, so and so will happen; we did not trouble to say that; it is too clear'. Unless the court comes to some such conclusion as that, it ought not to imply a term which the parties have not themselves expressed…”

And it is not enough to show that had the parties foreseen the eventuality which in fact occurred they would have wished to make provision for it, unless it can also be shown that one of several possible solutions would without doubt have been preferred: Trollope & Colls … at 609-10, 613-14."

The significance of both Liverpool City Council v Irwin and the Phillips Electronique case is that they both stress the importance of the test of necessity. Is the proposed implied term necessary to make the contract work? That seems to me to be an entirely appropriate question to ask in considering whether a term should be implied on the assumed facts in this case.

In the course of his oral argument Mr Bailey very helpfully reduced to writing a summary of his central submissions, which include the following:

“(1)

The three propositions:

(a)

Unless the charterparty otherwise provides, a shipowner is not obliged to comply with an order to load or discharge at an unsafe place unless that place has been expressly named in the charter party. This is because when charterers reserve to themselves the right to choose from a specified range of unnamed places the place where the vessel is to load or discharge, the exercise of that right necessarily carries with it the implied obligation to choose a safe place.

(b)

On its true construction, this charterparty gave the charterers the right to choose the place for the vessel to load from a range of unnamed places.

(c)

Although the express terms of the charterparty necessarily qualify the scope of the implied obligation in relation to the safety of the berth, the express terms of the charterparty are not inconsistent with an implied obligation on the part of the charterers to choose a loading place which is prospectively safe from risks not affecting the port as a whole or all the berths within it [“…or arising from the specifications or restrictions of the berth”].

(2)

The owners submit that there is a fundamental distinction to be drawn between the nomination or declaration of one place from a series of named places, on the one hand, and the nomination of one of a number of unnamed places from within a specified or limited range on the other. No case (of which the owners are aware) has attempted to analyse the nature of the distinction.”

I would not accept Mr Bailey’s submissions, at any rate in so far as they are said to apply to the facts of this case. While I recognise that there is force in some of his more general submissions and that, somewhat surprisingly, the principles have not been fully worked out in the cases, there is no case that supports his approach to this charterparty and to my mind the appeal fails by reference to the particular terms of the charterparty.

Before considering those particular terms, it is important to note that the cases stress the importance of the particular terms of each charterparty or other contract, just as Lord Hoffmann does in the Belize case. Thus, for example, in Vardinoyannis v The Egyptian General Petroleum Corporation (The Evaggelos Th) [1971] 2 Lloyd’s Rep 200, upon which (even though it was a time charter case) Mr Bailey placed considerable reliance, Donaldson J stressed this point at page 204 in these terms, having expressed his view as to what he would have done if faced with a simple time charter:

“However, I am not faced with a simple charter. Any suggested implied term has to be considered against the general business background to the transaction and the express terms of the charter. Furthermore, if a term is to be implied, it must be such that it would surely have been agreed to both by the reasonable shipowner and the reasonable charterer, not to mention that other immortal abstraction and honorary member of the Baltic Exchange, the maritime officious bystander.”

Thomas J said much the same in Aegean Sea Traders Corporationv Repsol Petroleo SA (The Aegean Sea) [1998] 2 Lloyd’s Rep 39, where he was considering whether, if a party to a bill of lading had a right to nominate the discharging port (which he had held that it did not), a term should be implied into the bill of lading to the effect that the port and place of discharge should be safe. At page 67 he accepted, as is accepted here (and was accepted by the judge), that where the charterparty provides for a choice of a number of unnamed ports (there in Spain), there must be an implied warranty that any port nominated must be one that it is possible for a vessel to reach: see the decision of this court in Aktieselskabet Olivebank v Danske Svovlsyre Fabrick (The Springbank) [1919] 2 KB 162 and Reardon Smith Line Limited v Ministry of Agriculture Fisheries and Food [1962] 2 QB 42 per Willmer LJ at page 110. Thomas J added that beyond that it was an open question whether any term should be implied into a voyage charter where there is a range of unnamed ports and where there is no express warranty of safety in contradistinction to voyage charterparties where there is an express warranty, as in the Norgrain and Asbatankvoy standard forms. He noted that this was an important question because (as he understood to be the position) some standard forms of voyage charterparty do not contain an express warranty of safety.

Thomas J then referred to The Evaggelos Th in support of the proposition that in a time charterparty a warranty of safety would often be implied. In that case Donaldson J had said this at page 204 in a passage relied upon by Mr Bailey:

“For my part, if I were faced with a simple charter which provided that the vessel was only to go to such port or place within a specified range as might be nominated by the charterer and there load a cargo, I should have no hesitation in implying a qualification that the port or place had to be safe. I should make this implication because common sense and business efficacy require it in cases in which the shipowner surrenders to the charterer the right to choose where his ship shall go, and because I think that this is in accordance with the weight of authority.”

Thomas J noted that no decision had yet gone so far in a voyage charterparty case. However he referred to the judgment of Devlin J in Compania Naviera Maropan S/A v Bowater's Pulp and Paper Mills Limited (The Stork) [1955] QB 68 at page 72:

“There must … be an obligation to nominate at least one loading place, and there must be implicit in that some condition about safety to prevent the making of a derisory nomination.”

Morris LJ said much the same in this court: ibid at page 102. Mr Bailey relied upon those statements of principle but it is I think important to note that the court also relied upon the other terms of the charterparty which to my mind made the position absolutely plain.

Based on those dicta, Thomas J noted that in Voyage Charters by Julian Cooke and others (1993), the editors concluded at page 76:

“It is submitted that the court would imply a warranty as to safety and, a fortiori, the charterer will be under an obligation not to nominate a port which is impossible.”

He added:

“Although there is much to be said for this view, I do not think that one can conclude in general that a term as to safety will always be implied into voyage charter-parties where there is an unspecified range of ports. The issue as to whether a term should be implied as to safety and the extent of the obligation may turn on the specific terms of the charter.”

I agree. This is the same point as that made by Donaldson J, namely that there are no absolute rules but that much depends upon the terms of the particular charterparty.

Mr Bailey referred us to the most recent edition of Voyage Charters, which is the third edition, 2007. At paragraph 5.5 the authors describe the principal ways used in practice for identifying the port or ports: by identifying a named port, such as “Rotterdam”, or by stipulating a number of named ports from which a choice of one or two is to be made, such as “Amsterdam/Rotterdam/Antwerp” or by stipulating a number of unnamed ports within a specified range, from which a choice is likewise to be made, such as “Bordeaux/Hamburg range”. The authors add that in each of those cases, it is implied, if not express, that the charterers will also nominate a berth or place within the port. They also add that it is common to specify a warranty of safety, such as “1/2 safe berth 1/2 safe ports Bordeaux/Hamburg range” and that, even without such a clause the charters are both entitled and bound to designate a berth. I would agree that that accurately describes the general position.

As to safety, the authors discuss various types of problem in some detail between paragraphs 5.31 and 5.110. In this appeal we are concerned with only some of the problems discussed there. At paragraphs 5.31 and 5.32 they note the dicta in The Stork referred to above and the approach of Thomas J in The Aegean Sea. In particular they note Thomas J’s view that whether or not there is an implied warranty of safety will depend upon the normal contractual rules for the implication of terms. I agree. It follows, as I see it, that the test is one of necessity.

The authors add that whether such a warranty will be implied will be influenced greatly by the degree of liberty which the charterers enjoy under the terms of the charter to choose the port or place where the ship is to load or discharge. They conclude paragraph 5.32 by saying this:

“In principle, the more extensive that liberty, the greater the necessity to imply a warranty: conversely, the more specific the information given in the charter to the owner about the intended port or place, the more reasonable it is to conclude that he has satisfied himself as to its safety, or that he is prepared to take the risk that it is unsafe.”

I would accept that as a correct general approach, for which the authors cite Reardon Smith Line limited v Australian Wheat Board (The Houston City) [1954] 2 Lloyd’s Rep 148 and Atkins International HA v Islamic Republic of Iran Shipping Lines (The AJP Priti) [1987] 2 Lloyd’s Rep 37. The reference to The Houston City is I think intended to be a reference to the dissenting judgment of Dixon CJ in the High Court of Australia, especially at page 153, which was subsequently approved by the Privy Council at [1956] AC 266.

At paragraphs 5.34 to 5.36 the authors of Voyage Charters discuss the three types of situation they had identified earlier and conclude that, where there is one or more named ports or berths without an express warranty of safety, no such warranty is to be implied. They again rely upon Dixon CJ and upon The AJP Priti (see below). Mr Bailey relied in particular upon the judgment of Dixon CJ, which included this:

“When the charterer is prepared at the time of taking the charter to specify the place where the cargo will be available or the place where or the place at which he desires it to be delivered, the shipowner must take the responsibility of ascertaining whether he can safely berth his ship there or will take the risk of doing so. If he agrees upon the place, then, subject to excepted perils, his liability to have his ship there is definite.”

Dixon CJ compared that situation with the situation where there is a range of ports or places, although it is relevant to have in mind that in The Houston City the charterparty provided that the vessel should proceed to “one or two safe ports Western Australia” and there load a cargo.

The owners’ case is in essence that, since no berth was named in the charterparty and there was more than one berth at Chekka and, since in these circumstances it was for the charterers to nominate the loading berth, it was necessary to imply a term that the berth was safe. I can well understand that in some cases that might be so and, indeed at paragraph 5.37 the authors of Voyage Charters draw attention to Thomas J’s view that there was much for the view that it is necessary to imply a term that a nominated port will be safe. At paragraph 5.38 the authors quote the passage from the judgment of Donaldson J in The Evaggelos Th which I have set out above and say that, although, as Thomas J pointed out in The Aegean Sea, no decision had gone so far as to equate voyage charters with voyage charters in this respect, the reasoning of Donaldson J seemed equally applicable to a voyage charterparty, at any rate unless the range of unspecified ports is very limited. In the 21^st edition of Scrutton on Charterparties, 2008, at page 119, the editors say that if the charterparty, whether time or voyage, provides for the nomination of a port or berth, a warranty that the port or berth is safe will probably be implied but they add that this is not invariable and may depend upon the specific terms of the charterparty. They cite The Aegean Sea. For my part, I would not apply the reasoning directly from a time charterparty to a voyage charterparty, especially where, as here, the danger at the berth was (it appears) unascertainable by either the owners or the charterers and the question is simply which party has to bear the risk or, put another way, how the risks should be apportioned. As I see it, all will depend upon the circumstances and, in particular whether, having regard to the terms of the particular charterparty, it is necessary to imply the warranty.

In these circumstances, I would not accept the bald proposition stated in Mr Bailey’s proposition (1)(a). I would accept his proposition (1)(b) but not his proposition (1)(c). I would accept proposition (2), subject to this. Although it is not necessary to determine the question on the facts of this case, I am not persuaded that the distinction is between a charterparty with named ports or places and one with unnamed ports. As at present advised I do not see why the ports have to be specifically named. If they were described as 1/2 ports on a particular river, such that they could be readily identified by looking at an atlas, it is to my mind difficult to see why that would not be the equivalent of naming them.

I recognise that a distinction may be drawn between named or identified ports or places on the one hand and un-named or unidentified ports or places on the other. However, as I see it, the relevant principles are not so prescriptive as set out in Mr Bailey’s proposition (3) set out above.

In the case where there is an express warranty of safety of the port but not the berth but the charterers are left to nominate the berth, the authors of Voyage Charters say at paragraph 5.42 that, if the port to be nominated must be safe, it follows that the berth must impliedly be safe. I would accept that proposition on the footing that it follows from Sellers LJ’s description of a safe port in Leeds Shipping Co Ltd v Société Française Bunge (The Eastern City) [1958] 2 Lloyd’s Rep 127. He said at page 133:

“If it were said 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, it would probably meet all circumstances as a broad statement of the law.”

A safe port must thus be prospectively safe for the vessel to use, which of course involves use of a loading berth. It is thus implicit in the express warranty of safety of the port. To the extent that Mr Bailey did not accept that analysis, I would not accept his submissions.

In the context of the present case, it appears to me that the decision of this court in The APJ Priti is of some considerable importance. It does not follow from the mere fact that the charterers are under a duty to nominate the berth that they warrant that the berth is safe. The facts of this case demonstrate that. Mr Bailey correctly accepted that, if there is such an implied term, it must be expressed in narrow terms. He correctly accepted that there was no express warranty that the port was safe and that, if there was an implied safe berth warranty, it was not a safe berth warranty of any width.

This can be seen from a consideration of The AJP Priti and of the particular terms of the charterparty in the instant case. In The AJP Priti the charterparty was in an amended Gencon Form which provided for the carriage of a cargo of bagged urea from Damman to 1/2 safe berths Bandar Abbas, 1/2 safe berths Bandar Bushire, 1/2 safe berths Bandar Khomeini in charterers’ option. It further provided that the vessel “shall proceed to the discharging port or place … as ordered or so near thereto as she may safely get and lie always afloat …”. It was accepted that there was no implied warranty that the load port was safe because there were three named ports. In any event Parker LJ said at page 40 that a claim based on an implied warranty could not have succeeded for no such warranty could be necessary to give business efficacy to the charterparty or on the official bystander or any other test.

Bingham LJ summarised the same point in this way at page 42:

“There is no ground for implying a warranty that the port declared was prospectively safe because the omission of an express warranty may well have been deliberate, because such an implied term is not necessary for the business efficacy of the charter and because such an implied term would at best lie uneasily beside the express terms of the charter.”

That case was concerned with risks arising out of the Iran/Iraq war and the terms of the charterparty were of course different from this but here too the omission of an express safe port warranty may well have been deliberate. This charterparty provided for loading at one named port and Mr Bailey did not argue that the charterers impliedly warranted that the port of Chekka was safe. Any such argument would have been doomed to failure.

If there is an implied safe berth warranty here but not a safe port warranty, as there was expressly in The APJ Priti, the next question is what is meant in this context by a safe berth. This question was expressly considered by Bingham LJ at page 42. In doing so, he adverted to Sellers LJ’s description of a safe port referred to above and that the same principle should apply to safe berths, subject to two qualifications. The first qualification was that, since the charterers had not promised that the port would be safe, the vessel’s passage to and from the berth would not include the passage to and from the port. The second qualification was that the charterers’ promise should be understood as limited to a promise that the berth or berths nominated would be prospectively safe from risks not affecting the port as a whole or all the berths at it. He accepted the charterers’ contention that the relevant safety (or unsafety) must be particular to the berth nominated and not general to the port as a whole or all the berths in it. It is not necessary to discuss in this case how that principle applies in a particular case. Much may depend upon the configuration of the particular port. However, in the present case Mr Bailey correctly accepted that, if there is an implied term, its nature must be affected by the express terms of the contract and, in particular clauses 1 and 20 quoted above.

As to clause 1, he accepted (in my opinion correctly) that the fact that the word “safely” in the printed Gencon terms was struck out by agreement was relevant to the true construction of the charterparty. It is the only place in the printed terms that there is any reference to safety. The fact that it was struck out seems to me to be a pointer to the fact that the parties did not intend there to be any express term requiring the port or berth to be safe. It is, however, no more than a pointer. More important is that Mr Bailey correctly accepted that the fact that there is no warranty that the port be safe, when read with clause 1, leads to the conclusion that the charterers were not warranting that the channel to the berth was safe. He submitted however, that a warranty should be implied that the berth itself was safe.

Initially he submitted that any unsafety of the berth would be a breach of the warranty but, in the course of his submissions he accepted that such a warranty did not have sufficient regard to clause 20. This is reflected in the words in square brackets at the end of his proposition (1)(c) set out above. He thus correctly recognised that, if the unsafety was a feature of the specifications or restrictions of the berth, the express terms of clause 20 negatived an implied term with regard to them. This was on the footing that, if the parties had made specific provision for them, there was no sensible basis upon which it could be said that it was necessary to imply a term that the charterers warranted that the berth was safe in this regard or, put another way, that the charterers warranted that the berth was not unsafe by reason of its specifications or restrictions.

However, Mr Bailey submitted that the unsafety here was not by reason of the specifications or restrictions of the berth but by reason of a hidden and unknown underwater projection. It followed, he said, that the warranty which is to be implied was not contradicted by clause 20. I would accept Mr Bailey’s submission that the reference in clause 20 to specifications and restrictions is to advertised (or at least known) specifications and restrictions and not to hidden and unknown features of the berth. However, as Bingham LJ put it in The AJP Priti at page 42 in the passage quoted above, albeit in connection with a possible safe port warranty, such an implied term would at best lie uneasily beside the express terms of the charterparty. Like the judge at [31], I would go further and hold that it would be inconsistent with them.

Mr Croall QC submitted on behalf of the charterers that the basis of the suggested implied term is that it was for the charterers to nominate the loading berth and that it was therefore reasonably to be expected by the parties that the charterers would ascertain the safety of the berth before nominating it because the owners would not know what berth the charterers were going to nominate. That is indeed the essential basis of the alleged implied term, as appears from the owners’ central propositions referred to above. Mr Croall submitted however that, whatever the position would be in the ordinary case, such an implied term would be inconsistent with clause 20.

I would accept that submission. By clause 20 the owners guaranteed and warranted that upon arrival at the loading port, which was Chekka, the vessel complied with all restrictions of the port, including its berths and approaches and that they had satisfied themselves to their full satisfaction with and about Chekka’s specifications and restrictions before entering into the charterparty. The expression “restrictions” in the penultimate line naturally includes restrictions of the berths, as expressly provided earlier in the clause. Thus the owners were willing to give those warranties in relation to the berths at Chekka. In order to do so it may well be that they in fact checked the berths at Chekka, which would have involved their identifying the possible berths at Chekka. In any event I would agree with the view expressed by the judge at [28] that box 10 and clause 20 read together indicated that the owners agreed that they would either investigate Chekka, that is the berths at Chekka, or take the risk of any dangers getting to whatever berth was nominated, using it (ie loading at it) and departing from it.

I would further agree with the view expressed by the judge at [29] that the only choice the charterers had was to choose and nominate the loading berth. Once it was nominated it was treated as having been written into the charterparty at the outset. In all the circumstances the owners were accepting that the charterers had the right to elect a berth of their choice, with only one limitation, namely that a vessel of 27 feet salt water draft could berth there. A nomination of a berth which could not accommodate such a vessel would be a bad one because it would be an “impossible berth”.

Finally I would agree with the views expressed by the judge at [30]. They were these. When clauses 1 and 20 are read together, the owners undertook that the vessel would proceed to the nominated berth at Chekka or so near thereto as she may get and lie afloat and load the cargo and that that obligation was not contingent on the vessel’s safety. This case is quite unlike those charterparties where the charterers have the right to nominate a port out of a range of unnamed ports.

For all these reasons, I would agree with the arbitrators and the judge that it was not necessary to imply the suggested term in order to make the contract work. Moreover, if one asks the question identified by Lord Hoffmann at [21] of the Belize case as the only question to be asked, namely whether the charterparty could reasonably be understood to mean, when read against the relevant background, that the charterers warranted the safety of the berth at Chekka from risks not affecting the port as a whole or all the berths in it or arising from the specifications and restrictions of the berth, the answer is in my opinion no. The question is simply whether the charterers agreed to take the risk of unsafety at the berth from hidden dangers and the answer is no.

For these reasons and the reasons given by Rix and Carnwath LJJ, whose judgments I have seen in draft, I would dismiss the appeal.

Lord Mansfield is supposed to have said, in effect, that it is easier for a judge to get the answer right than it is for him to give sound reasons for his correct answer. In a sense, this is one of those areas of the law which puts me in mind of that dictum.

It is, I think, relatively easy to say in this case, along with the three experienced maritime arbitrators and the experienced commercial judge, that the charterers did not here implicitly warrant the safety of the berth at Chekka to which they directed the vessel. After all, the charterers did not expressly warrant the safety of the port or berth; the port was a named, accepted and agreed port; it is hard to think that it does not follow that the (two assumed) berths in the port are also to be regarded as berths of which the owners had agreed to accept the risk; and for good measure the owners had, by the special clause 20, expressly agreed to familiarise themselves with all aspects of the port, including its berths, as well as relevant restrictions and specifications. To cap this, there is no single case which Mr Bailey has been able to refer to us, for all his industry, in which the safety of a berth at a named port, whose safety has not itself been expressly warranted, has been implicitly warranted. Whether or not the, albeit limited, implied warranty of safety for which Mr Bailey now contends is actually inconsistent with clause 20, or whether, as I am inclined to think, in Sir Thomas Bingham’s insightful phrase in The APJ Priti at 42, “such an implied term would at best lie uneasily beside the express terms of the charter”, I would reject the owners’ submission. Moreover, I would do so whether or not an essential requirement of the implication of a term is that it is necessary, and whether or not Lord Hoffmann’s new formulation of the implied term test, that the term contended for is what the instrument “must mean” to the “reasonable addressee” (para 18 of Attorney-General of Belize), is the ultimate test. In all this, I agree with the judgment of Sir Anthony Clarke MR, which I gratefully adopt in full.

What is more mysterious, however, is an aspect of the argument which Mr Bailey addressed which relates, not to the time of the misfortune, but to the time of the order to berth, or at any rate of the berth’s nomination. He submits that if a berth is in fact unsafe at the time of its nomination, then a vessel is not obliged to proceed there. That is because a charterer is not entitled to nominate an unsafe or impossible berth. It ought to follow, therefore, he submits, that it is an unlawful breach of contract to nominate such a berth. If, however, this is so, then it must follow that there must be an implicit obligation not to nominate an unsafe berth, and that that must be a general obligation which is to be implied in every case, save where an express term is plainly inconsistent, or, which has come to be regarded as its equivalent, where the owner has already agreed to go where he is ordered. The latter reasoning typically operates in the case of a port which is named in the charter, but even in such a case the charterer retains an option as to the berth, and the agreed nomination of a berth written into the charter must be a rare event. Therefore, there is always, or nearly always, at least an implied, even if not an express, warranty as to the safety of the berth. In the present case, because of the naming of the port and the inclusion of clause 20, the standard implied term had to be limited to make way for the charter’s express provisions, so that the implication was only that the berth should not be unsafe because of features which did not apply to the port as a whole and did not fall within the specifications or restrictions referred to in that clause. Subject to those limitations, however, the implied term contended for was only that which was necessary and standard in every charter, voyage or time charter alike.

In support of that reasoning Mr Bailey referred to dicta in the cases which attached the implication of safety to the right of the charterer to choose where to send the vessel. Thus in Reardon Smith Line Ltd v. Australian Wheat Board [1954] 2 Lloyd’s Rep 148 (The Houston City), Sir Owen Dixon CJ in the High Court of Australia stated in his dissenting judgment at 153:

“If he [the owner] agrees upon the place, then, subject to excepted perils, his liability to have his ship there is definite. But where the charterer cannot specify the place of loading or discharge at the time of the charter, then, subject to excepted perils, the shipowner must agree to submit his ship to the charterer’s orders.”

That dissenting judgment was preferred (to the judgments of the majority in the High Court of Australia) by Devlin J and the court of appeal in Compania Naviera Maropan S/A v. Bowaters Lloyd and Pulp and Paper Mills Ld [1955] 2 QB 68 (The Stork): and in due course when The Houston City reached the Privy Council, it was Devlin J’s analysis in The Stork which was cited with approval at [1956] AC 266, 282/3. However, both The Houston City and The Stork concerned charters with express port and/or berth warranties, and the issue with which those cases primarily dealt was the different issue (for our present purposes) of whether a master who entered port and/or berth despite misgivings as to their safety had done an act breaking the chain of causation. Therefore, what was said there about the rationale for any warranty, express or implied, was said in the context of express warranties. See, for instance, the quotations from Sir Owen Dixon’s judgment in The Stork at 91/92, which do not include the passage relied on by Mr Bailey.

Nevertheless, some dicta, particularly in The Stork, went a little wider. This was because in that case there was a subsidiary submission, bound up with the primary argument, that, although the charter contained an express warranty of safety of the berth (in fact a “loading place” out at anchorage, where the vessel had been damaged), there had been no express warranty of safety of the port: so that, when the vessel entered the port it did so of the owner’s own doing and without further obligation. This argument arose on the following clause, which provided that the vessel was to proceed to –

“two approved loading places as ordered in Newfoundland on the east coast, or so near thereunto as she may safely get, and there load, always afloat, or safe aground…Charterers have the right to order the ship to load at two safe berths or loading places…”

There was an issue as to whether “two approved loading places” involved a warranty of safety. Devlin J distinguished between that phrase, which he thought was the equivalent of a reference to the “port place”, and the later phrase “two safe berths or loading places”, which he regarded as the equivalent of a reference to the “berth place” (at 72). In the end, Devlin J regarded the former reference as encompassing a warranty of safety, as a matter of pure construction (at 73). He said:

“I think that the parties would probably have inserted the word “safe” in what I have called the “port place,” if they had not used the word “approved.”…Thus its use, I think, takes for granted that the approved place will be safe. Whether the use of the expression by itself would comprehend a warranty of safety I need not decide, since I am satisfied that this charterparty, construed as a whole in the way that I have already indicated, is to be read as if it referred to an approved and safe loading place as ordered in Newfoundland on the east coast.”

Therefore it was a “safe port” charter.

It was in the context of that discussion that Devlin J had said this (at 72):

“There must be an obligation, express or implied, to nominate a loading place of some sort; if the ship is not told where to load she cannot earn her freight, and so, by failing to nominate or by nominating an obviously unsafe place the charterers would defeat the object of the charterparty without being liable for damages. There must therefore be an obligation to nominate at least one loading place, and there must be implicit in that some condition about safety to prevent the making of a derisory nomination. The obligation on the ship to proceed to a loading place “or so near thereunto as she may safely get” suggests strongly that the loading place itself must be safe. And when one finds the obligation on the charterers to nominate a loading place of some sort amplified by an express right to nominate one or two safe loading places, it does not require much effort of construction to conclude that the loading place which must be nominated must be safe.”

All that, however, is in the context of “loading place” qua port; and in the context of an express provision for a safe berth (loading place qua berth). Given that the whole of the east coast of Newfoundland was open for nomination of the loading places qua port, what Devlin J said there in my judgment takes the matter no further. In any event, by a process of construction, this was an express safe port charter, and above all without any question an express safe berth charter.

In the court of appeal, the argument on construction appears to have metamorphosed. Singleton LJ discusses a submission that it is only if two loading places qua berths were nominated that they are warranted safe, and not if only one is nominated (at 86). He seems puzzled by that submission, as am I. He rejected it. Hodson LJ agreed with Devlin J’s approach, albeit he described the warranty of safety of the loading places qua ports as “a necessary implication” (at 96). Morris LJ seems to have thought that there was an implied term requiring the nomination of at least one loading place “together with an express or implied term that the place to which the ship would be directed would be safe”, but that even in the absence of the express requirement of safety there would have been an implied warranty (at 102). That is hardly surprising, but it is all a long way from our case. Morris LJ also said that an order to proceed to a berth known to be unsafe would be an unlawful order which would not command obedience, but that there was no requirement to test the validity of an apparently valid order (at 103). However, if an order was known to be unlawful, “The owners must not throw their ship away” (at 104).

However, none of this assists in a case such as ours, where there is no express warranty of safety of either port or berth. Moreover, even in the absence of any warranty of safety, it may well be a breach of contract to order the vessel to a port or berth known to be unsafe, or obviously unsafe: because such an order is to an impossible port or berth, and would prevent the owners earning their freight. On that hypothesis there would be no need for an implication of safety.

In The Evaggelos Th [1971] 2 Lloyd’s Rep 200 there was an express warranty that the vessel could “always lie safely afloat” at any nominated berth. The vessel was hit by rocket fire at Suez. Donaldson J held that the express warranty related only to marine risks and was therefore irrelevant to the war risk which had occurred. However, the owners also put their case on an implied term that the vessel would only be employed between safe places. The charter concerned was a time charter with a wide trading range. Donaldson J was prepared to imply a term (but only in a limited form, ie in a form which was less wide ranging than an express term of safety) –

“because common sense and business efficacy require it in cases in which the shipowner surrenders to the charterer the right to choose where his ship shall go” (at 204).

In the event, however, there was no breach. It seems to me that that dictum is of little assistance in a voyage charter to a named port, especially where marine risks are concerned. Moreover, in Kodros Shipping Corporation of Monrovia v. Empresa Cubana de Fletes (The Evia (No 2)) [1983] 1 AC 736 at 761, Lord Roskill cautioned about accepting the reasoning (although he did not doubt the result) in The Evaggelos Th.

In The Aegean Sea [1998] 2 Lloyd’s Rep 39 the vessel grounded on rocks while moving between the roads and a berth at the discharge port. A question arose as to whether a warranty as to the safety of a port was to be implied into a bill of lading which it was suggested allowed nomination from a range of ports. There was no case where such an implication had been made in a bill of lading. The discussion by Thomas J was all obiter, since he held that there was no right to nominate. On the assumption of such a right, however, he held that there was no implication of safety. Even if one assumed a right to nominate, that involved no implied warranty of safety (at 68):

“Although it would be necessary to imply a term that the port nominated be one at which it was possible for the vessel to be able to discharge the cargo she had loaded, the implication of a term as to safety of that port would create a very onerous liability on those who became subject to the liabilities under the bill of lading” (emphasis added).

In this context Mr Bailey’s submission relating to the question of safety as it occurs at the time of nomination raises in my judgment a non sequitur. If a charterer has warranted the safety of a berth, then of course an owner can consider at the time of nomination whether he is willing to accept the nomination or refuse it on the ground of lack of safety. However, if the charterer has not warranted the safety of a berth, but it is obvious that it is unsafe or (perhaps) the owner has good reason to doubt its safety, he may possibly be entitled on the ground of other provisions or considerations of the contract between the parties to refuse to proceed. In either case, of course, the owner takes the risk of being able to establish a legitimate basis for his refusal. Thus it might be that what entitles the owner to refuse to proceed is that an order to go to an unsafe berth is an illegitimate order not because of any implied warranty of safety but because of an obligation not to nominate an impossible port (or berth) which, if persisted in, might amount to a repudiation, or at least to a frustration of the contract. A similar problem might arise if a valid nomination of a port or berth (even if warranted safe) becomes unsafe as a result of some abnormal occurrence after nomination (a problem left open in The Evia (No 2) at 749, 764/5). In such a case it is an open question whether a voyage charter is frustrated, or whether the charterer must renominate, or whether the problem is to be resolved by other terms of the contract: see the helpful discussion in Cooke etc, Voyage Charters, 3^rd ed, 2007, at paras 5.49/5.53. The position in this respect under a voyage and time charter may well not be identical.

Ultimately, Mr Bailey’s submission assumed that any right to choose on the part of a charterer, even so limited a choice as a right to choose between a small number of berths at a named port to which the owner has agreed to go and as to which there is no warranty of safety, must or at any rate prima facie ought to bring with it an implied warranty of safety. However, no case has gone so far, and in my judgment the proposition is not correct. The first and most basic understanding must be that an owner’s vessel is at his risk, subject to further agreement. That is why an express warranty of port and/or berth is an absolutely standard provision, in order to reverse the basic incidence of risk. Such an express warranty would not be necessary if Mr Bailey’s basic assumption were correct.

In this connection, although there is much about the learning of safe ports and berths which crosses the boundaries between voyage and time charters, nevertheless a typical time charter is a potentially different arrangement, because in such a case there is usually a wide trading range and the understanding is that the vessel is under the charterer’s orders on the basis of an employment and indemnity clause. Even so, a warranty is frequently included in such a time charter to the effect that the vessel is only to trade “between safe ports and safe places” (see NYPE clause 5, or Baltime clause 2). Thus an owner could refuse to proceed to a port which was unsafe, even though ordered to go there.

The appearance of express terms as to safety of ports, berths or places, so frequently found in charters of all kinds, even in time charters where the natural background is one in which the vessel’s trading is left broadly open to the charterer’s choice, emphasises to my mind that the basic risk to the integrity of his vessel is upon its owner. If he wishes to shift that risk on to his charterer, the warranty of safety is the time-honoured way of doing so. In the present case, the owner has chosen not to extract such a warranty, but on the contrary has agreed to the port and to taking the risk of its restrictions and limitations, as well as those of its anchorages, berths and approaches. In such a case it seems to me that if there is some unforeseen risk which lies outside those terms, it does not lie with the charterer, but with the owner.

In sum, there is no authority which extends any implied warranty of safety to a voyage charterer’s choice of berth in a port which is not itself warranted safe. It appears to be accepted that a warranty of safety as to a port will encompass a warranty of safety as to its berths (Voyage Charters at para 5.42, Time Charters, 6^th ed, 2008, at para 10.33). It seems to me that the corollary also applies: where there is no warranty of safety as to the port, there is unlikely to be any warranty of safety as to its berths, in the absence of an express warranty. Whether that is always the case, as for instance in a vast port such as Rotterdam, it is not necessary to decide. But in the present case, it seems to me to be impossible to imply the residual term for which the owners contend.

I agree that the appeal should be dismissed. As the Master of the Rolls has pointed out (paras 10-11), the charterers do not need to prove anything. The owners have to establish an implied term whereby the charterers impliedly undertook responsibility for even unforeseeable risks associated with the nominated berth. Otherwise the damage lies where it falls. The owners have not in my view shown that the implication of such a term is necessary, nor is their case strengthened by substituting any of the other formulations of the test discussed in the cases. Nor does any of the cases relied on go that far. Furthermore, for the reasons discussed by the Master of the Rolls (paras 38-44), I agree that any such implication is negated by the express terms of the contract.