The Place that Launched a Thousand Ships1: Some Hellenic Influences on Maritime Law and Commerce
Hellenic Australian Lawyers Association International Conference on Contemporary Legal Issues and the Influence of Hellenism
1 This paper considers some recent United Kingdom Supreme Court cases on safe ports, damages for breach of charterparty and general average that all have some connection with Greece (either through links to Greek shipping commerce or to ancient law): The Ocean Victory, The New Flamenco and The Longchamp. All three cases reflect the clarity of expression and thought which mark that great appellate shipping court.
2 Each, in its own different way, illustrates the importance of a way of thinking about the law and its problems that goes back to Greek philosophy. To this end, the paper focuses on one, in particular, of the cases, not so much to speak of maritime law itself, but to place the cases in a Greek philosophical tradition to make a point about contemporary law, and its shape (shape is an important word), and to illuminate the need to think contextually and holistically about law – a feature so often ignored by Parliamentary law-makers, judicial rule-makers and academic taxonomers.
3 This emphasis on wholeness and shape of the law, and not on analytical theoretical abstraction, is particularly important in respect of maritime law because it is a separate body of law (separate from the common law and equity), distinct in its common sources and derived from international maritime commerce and its relational human activity over centuries.
4 Rhodes, the location for this conference, is a most propitious place to talk about maritime law.
5 Since antiquity, Hellenic peoples have played a role in the development of international maritime law, through their participation in trade throughout the Mediterranean and the world. Rhodes in particular has been influential. In 1759, Lord Mansfield described Rhodian maritime law as "the ancientest laws in the world". Around the eighth or ninth century BC there was in operation here a body of maritime law later referred to as the "Rhodian Code". It is clear that there at least existed an "unwritten body" of customary maritime law that had developed in Rhodes in antiquity. Aspects of this body of law were incorporated into Roman law and were included in the Digest of Justinian that was compiled in 600 AD.
6 During the Byzantine era, around the seventh or eighth century AD, a distinct codification of maritime law known as the "Rhodian Sea Law" was prepared. It was a proper codification of maritime law that differed from the customary "Rhodian code". It can be read in translation today, and had its basis in Roman law. The Rhodian Sea Law was a sophisticated statement of maritime law that later formed a foundation of modern international maritime law.
7 In more recent years, the great Greek shipping magnates have been facilitators of global trade and, through that, maritime law. People such as Onassis and Vergottis (to be discussed later in this paper) became household names. Greece continues to be the largest ship owning country in terms of deadweight tonnage, though invariably flagged elsewhere. The influence of Greek maritime commerce upon English shipping law can also be observed from the names of some of the vessels that have become the subject of famous decisions, for example: The Achilleas; The Evia (No 2); The Eugenia; The Troilus; The Democritos; The Eurymedon; and The Agamemnon. There are many others. It could indeed be said that this was the place that launched a thousand shipping cases.
The Ocean Victory and Safe Ports
8 The first case is The Ocean Victory, which is an important decision on "safe port" obligations. These are obligations, normally imposed by clauses in charterparties, that require a charterer only to give orders to navigate to a port that is "safe". As to whether a port is "safe", a great maritime judge Sellers LJ said the following in 1957 in The Eastern City:
… 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.
9 One can observe the round, human shape to the requirement. It is not linear, but humanly relational, contextual and whole – not defined by reference to parts. Rather, it is a judgment (a practical value judgment) about what a skilled mariner could do with the ship in question absent abnormal human, natural, or seafaring occurrences.
10 The Ocean Victory was a Capesize bulk carrier that grounded, and subsequently broke in two, and was totally lost, while trying to leave the port of Kashima in Japan for safety at sea during a severe storm in 2006. The danger to the vessel arose from a combination of two characteristics of the Port of Kashima – long swell and northerly gales – occurring simultaneously. Neither of these meteorological phenomena were rare occurrences on their own but, as the primary judge noted, the combined occurrence of them both was rare, though it could not be said to be a surprise. The combination was certainly foreseeable. At the time of the loss, the vessel was subject to a demise charter by the owners, a time charter by the demise charterer and a trip time charter by the time charterer. All of the charterparties imposed similarly framed obligations to trade only between safe ports.
11 After the vessel was lost, Gard, as one of the hull insurers and subrogated to and assignee of the rights of the owners and demise charterer, sued the trip time charterer for breach of the safe port obligation. The primary judge found that the obligation had been breached and awarded a sum of over USD 137 million in damages for loss of the vessel and consequential losses. The Court of Appeal overturned that judgment, and Gard appealed to the Supreme Court.
12 In the Supreme Court, the case turned on whether the prevailing weather conditions at Kashima which caused the loss of Ocean Victory were characteristics of the port or whether they were properly characterised as an "abnormal occurrence" within the meaning of the passage in The Eastern City referred to above.
13 The Eastern City had been approved by the House of Lords in The Evia (No 2), the most important modern decision on safe port obligations, in 1983. Evia, or Euboea, is the second largest Greek island, after Crete, to the north and north east of Athens. The vessel that shared its name (Evia) became trapped in Basrah after the outbreak of the Iran-Iraq War in 1980, having been ordered to sail to Basrah by her time charterers. (In a time charter, the charterer has the commercial use and control of the ship that is still in possession of the owner or demise charterer. Having the commercial control and use of the ship, the charterer must only send the ship to ports where she will be safe.) The question was whether Basrah was an unsafe port. Critically, their Lordships clarified that the primary obligation under a safe port clause is a prospective one: the port must be safe at the time the nomination is made. It is a practical judgment, looking forward. Basrah was a safe port at the time Evia was ordered there. The outbreak of hostilities was considered to be an abnormal occurrence and, as the vessel was trapped there, she could not leave. Thus, there was no breach of the safe port obligation.
14 The principles in The Evia (No 2) were accepted in The Ocean Victory. The useful contribution of The Ocean Victory relates to the process of evaluating the evidence in a safe port case, and the making of a judgment as to whether a port is unsafe. For The Ocean Victory, the question was how the weather conditions that caused the loss should be analysed and evaluated, for the purposes of determining whether Kashima was prospectively safe or unsafe.
15 What was said to have made Kashima unsafe was the confluence of long swell and the northerly gales. The Supreme Court, in a judgment on this issue written by Lord Clarke, agreed with the Court of Appeal that the primary judge had erred by considering "… the respective constituent elements of the combination [of weather events] separately … and decided that, viewed on its own, neither could be said to be rare and both were attributes or characteristics of the port", rather than "asking the unitary question directed at establishing the correct characterisation of the critical combination [of the weather events]".
16 The primary judge considered that the combination of swell and northerly gales could not be considered to be an abnormal occurrence as it arose from foreseeable characteristics of the Port, looked at individually. It was also reasonably foreseeable (using that well-known abstraction) that a combination of both could arise. In contrast, Lord Clarke agreed with the Court of Appeal that rather than applying a standard of foreseeability to determine whether these individual weather conditions were nevertheless combined characteristics of the port, the correct approach was instead to:
… look at the reality of the particular situation in the context of all the evidence, to ascertain 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.
17 This statement makes it clear that the question of whether a port is unsafe, and whether something is a characteristic of a port, requires an evaluative judgment based on all the evidence. It is a whole and a practical human commercial evaluation. This is also indicated by Lord Clarke's statement that:
… 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 year?
18 A number of observations can be made about the Supreme Court's approach. First, the Court makes clear that a conclusion as to whether something is a characteristic of a port or an abnormal occurrence necessitates an evaluative judgment based on the whole of the evidence rather than a test of mere foreseeability. Secondly, this judgment should not be unnecessarily complicated. Thirdly, in the case of Ocean Victory, the most important matter was the anterior question of how to characterise the conditions said to make Kashima an unsafe port. The primary judge erred by adopting an artificial and particularised characterisation of these conditions that did not have sufficient to regard to the fact that the prevailing conditions on the day arose from the combination of both long swells and northerly gales. It was thus inappropriate to consider each weather phenomenon separately. The key lesson of the case is that safe port disputes need to be decided having regard to the reality in which the harm to the vessel occurred.
19 This may appear all very straightforward. But the correct approach, particular or whole, abstracted or practically human and relational, was worth USD 137 million.
20 How is this philosophical? Let's see.
21 The early Greek thinkers are considered to be the progenitors of Western philosophy. Bertrand Russell referred to the philosopher Thales, who belonged to the Milesian School of Philosophy that arose in Ionia in the sixth century BC, as "the first philosopher". The Pantheon – if we may use that word – of Greek philosophers can perhaps be classified into the pre-Socratics and the post-Socratics.
22 Pre-Socratic thinkers were concerned, primarily, with attempting to understand the natural world. Their investigations were fundamentally metaphysical. Unlike later Greek philosophers they showed a willingness to trust experience and perception, to seek to accommodate difference and to build a more holistic conception of the world.
23 This pre-Socratic conception of the world is illustrated by Heraclitus, mentioned by the President of Greece in his speech at the Opening of the Conference on Monday. Heraclitus was an Ionian "mystic" who lived around 500 BC.
24 Heraclitus' philosophy can be summarised as attending to experience, rather than pre-conceived ideas about experience. Experience involves the reality of the union of opposites, and reconciling them in the creation of something new.
25 Heraclitus' thinking demonstrates an appreciation of the importance of context and awareness of the implicit. He contended that one must understand the whole of the contradictory aspects that make up existence in order to reach a conclusion about the nature of the world. Through appreciating this wholeness we come to understand the inherent complexity present in the world.
26 Not a long way from a safe port analysis, if one thinks about it. We attend to experience, and the whole human and commercial context with its dangers and opposites. We do not take individual parts or extrapolate with abstracted notions. We look at the phenomenological whole and give a whole, human seafaring answer based on individual experience. This can be seen as contrasted with the perceived need to abstract a form or a rule to allow the mind to categorise in detailed externalised parts an experiential world that is not to be trusted and that is too individualistic to render a reliable rule.
27 Plato's criticism of Heraclitus can be seen in a movement away from experience and perception, and towards abstraction and logic. The abstraction was what was real – the textual abstraction, that forms the tool of particular analysis, clarity and the search for certainty.
28 There is, of course, much more to be said of pre and post-Socratic philosophy and of the riches of Aristotle. But, there is a constant struggle in the law that is reflected in very different ways of looking at or attending to problems – the explicit abstracted rule to define, to grasp, to create so-called certainty, and the whole sometimes contradictory evaluation of the real and the experiential, sometimes (more than we often give credit) containing the implicit and the indefinable.
29 Abstraction has its seductive beauty, just as an appreciation of the whole can have. As was said by William James in 1884:
Beautiful is the flight of conceptual reason through the upper air of truth. No wonder philosophers are dazzled by it still, and no wonder they look with some disdain at the low earth of feeling from which the goddess launched herself aloft. But woe to her if she return not home to its acquaintance … Every crazy wind will take her, and like a fire-balloon at night, she will go out among the stars.
30 What is the lesson of this for the law, and for lawyers? Perhaps, it is that one should not exhaustively rely on analytical, abstracted and de-contextualised methods of thought as the totality of any analysis, as critical and important as such approaches always will be. There will often need to be a synthesis, and a balance: to contextualise the abstracted into experience. Law is about people and human existence, their relationships with each other and the perception and recognition of societal bonds. Law is about power (here the power to send a hugely valuable ship to a port to work) and its control. In its application, law is often non-linear by virtue of the fact that it is necessarily both relational and experiential. The imperative to appreciate wholeness and context was understood by the pre-Socratic Greek philosophers, and by Aristotle. The tension and relationship between the two approaches can be seen in all fields of the law; today I have sought to illustrate it in a minor way in maritime commerce.
The New Flamenco and Damages for Breach of Contract
31 The second recent case is The New Flamenco – the latest in a line of time charterparty cases over the past decade that have dealt with damages for breach of contract. The first was The Golden Victory in 2007. It emphasised the "compensatory principle" established in Robinson v Harman as the basis for the award of damages for breach of contract. That is, a party that suffers a loss due to breach of contract is to be placed in the position it would be in if the contract were performed. Compensation is about actual loss. Thus, in The Golden Victory, it was held that the fact the charter would have been terminated two years early due to the outbreak of the Second Iraq War could be taken into account in reducing the damages for the anticipatory breach; damages were not to be inflexibly calculated at the date of breach where, at the date of assessment, subsequent events were known of that would have led to the termination of the charter. As Lord Sumption said in Bunge, in endorsing The Golden Victory, "it can rarely be thought to justify an award of substantial damages to someone who has not suffered any". The focus is upon awarding damages for actual loss suffered.
32 In 2009 came The Achilleas. Achilleas, the hero of the Trojan War described in the Illiad, was now a she – a bulk carrier. The case deals with remoteness of damage caused by breach of contract. The charterers redelivered the vessel late. Unbeknownst to them, the owners had fixed a lucrative follow-on charter under high market rates prevailing at the time. Due to the late re-delivery, they had to re-negotiate the follow on charter down to a lower rate of hire. The owners sued for the lost hire over the entirety of the subsequent charter, rather than merely claiming for loss at market rates during the overrun of the original charter. The House of Lords did not allow the owners to recover. Lords Hoffman and Hope held that the loss of the lucrative follow-on charter rate was not recoverable, as it was not a loss for which the charterers had assumed responsibility. The losses from the follow-on charter were a different type of risk than those the charters had assumed. The traditional understanding of remoteness was by reference to a rule that losses had to be within the reasonable contemplation of the parties to be recoverable. Arguably, it was clearly foreseeable that such a lucrative contract could be lost on repudiation. The approach adopted by Lord Hoffman in The Achilleas was to re-frame the question of remoteness as being based upon ensuring that the loss is one which the parties have assumed responsibility for. This was a whole, experiential and not abstracted rule-based way of putting it. Whether the loss is within the reasonable contemplation of the parties is an adequate guide in most cases, but it is not sufficient in all cases. Lord Hoffman's speech reflected an approach to remoteness that focused upon the context and experiential reality of the market: from an understanding of how the market worked, an evaluation was made of the fair and just assumption of responsibility.
33 These contractual damages principles are about ensuring there is compensation for the actual loss caused by breach of contract. But what is "actual loss"? It is in fact the first question in the enquiry, and one of characterisation. This leads to the latest of these charter cases: The New Flamenco. It is about mitigation of damages, and what is often described as "avoided loss". The New Flamenco emphasises the importance of properly characterising the loss caused by the breach before applying relevant legal rules. It has been suggested that mitigation in a contractual context is really an inquiry into whether the breach is the legal cause of the steps taken in mitigation by the claimant. This is an evaluative inquiry into whether a benefit arisen out of action taken by a claimant is so closely connected with breach that it is to be taken into account in assessing damages.
34 The New Flamenco was a cruise ship redelivered by its time charterers two years early, at the time of the GFC. There was no available market to re-charter in so the owners decided to sell the vessel for a sum that was USD 17 million more than if the vessel had been sold at the end of the charter and to sue the charterers for two years' lost hire. The arbitrator held the windfall obtained in selling the vessel had to be taken into account. The trial judge reached the contrary conclusion, but was overturned by the Court of Appeal.
35 The approach of Longmore LJ in the Court of Appeal was to have regard to the compensatory principle. He stated a general principle that where a step is taken in mitigation that arises out of the consequences of the breach and was in the ordinary course of business then the benefit obtained is normally to be brought into account unless it is wholly independent of the breach. Longmore LJ placed emphasis on whether there was an available market to charter into, given the prima facie measure of damages for breach of a time charter is the difference between the contract and market rates. Re-chartering is assumed to be a reasonable step in mitigation. Longmore LJ reasoned that this measure did not apply where there was no available market and concluded that the benefit obtained in the sale of the vessel was one that had to be taken into account.
36 The Supreme Court took a different view. Lord Clarke, delivering the judgment of the Court, held that the decrease in the value of the vessel between 2007 and the expected end of the charter was irrelevant. The owners' interest in the vessel's capital value was unrelated to the interest injured by the early redelivery. To be relevant to the assessment of damages, there had to be a sufficiently close causal link between the benefit and the loss. The benefit had to be caused by the breach or an act of mitigation. The benefit obtained by selling the vessel earlier was an independent commercial decision unrelated to the charterparty and it was not legally caused by the repudiation. The loss that the repudiation caused was the loss of two years' hire. This process of characterisation is not a process of definition. It is the placing of events in their human context, and evaluating and characterising the relational advantages or disadvantages to the parties.
37 The sale of the vessel was not, according to Lord Clarke, an act of mitigation as:
The relevant mitigation … is the acquisition of an income stream alternative to the income stream under the original charterparty. The sale of the vessel was not itself an act of mitigation because it was incapable of mitigating the loss of the income stream.
38 The reasoning of the Supreme Court indicates that the focus when assessing damages must be on correctly characterising the loss actually caused by the breach of contract. Selling the vessel did nothing to mitigate the actual loss suffered by the owner as caused by the repudiation of the charterparty. In the absence of a market, the appropriate measure was likely the difference between the charter rate and what could reasonably have been earned under shorter charters. To use the language of Justice Edelman in McGregor on Damages, the sale "… was not a reasonably necessary response to the breach". The change in the value of the vessel, the decision to sell and the value obtained from the sale was independent and distinct to the loss. Sale of the vessel did not mitigate the lost income stream, nor was it legally caused by the repudiation.
39 The importance of how one characterises loss caused by breach of contract and determining whether an act is done in mitigation is illustrated by the decision of the High Court in Clark v Macourt. In that case, the High Court endorsed the compensatory principle. However, the majority and Gageler J reached opposite results in their application of the principle.
40 Clark v Macourt had a peculiar set of facts. Dr Clark had purchased a reproductive health centre, including a quantity of sperm purportedly compliant with the relevant guidelines. The sperm were not compliant, however, and could not be used. Ethical rules prevented any profit being made from the use of the sperm – only costs could be recovered. Dr Clark purchased replacement sperm from the US and recouped most of the cost from her patients. The majority considered that she was entitled to damages for the full "market value" of compliant sperm not delivered and the fact she had recouped most of the cost of obtaining replacement sperm was not relevant as mitigation of loss as the claim was characterised as one for a failure to deliver compliant goods. Gageler J, on the other hand, held that on a proper application of the compensatory principle and given the fact the sperm were only valuable for their use in medical practice the compensable loss was the amount of the cost of the replacement sperm – if any – that Dr Clark had not been able to recoup from patients and so had been unable to mitigate.
41 The facts of Clark v Macourt are worlds away from The New Flamenco, however, they both illustrate how much an award of damages can be altered by how one characterises the loss. The conclusion of Gageler J may be more consistent with The New Flamenco. As Dr Kramer has suggested, the purchase by Dr Clark of replacement sperm and recoupment of the costs of this from patients appears to have been legally caused by the breach. It was a step taken to obtain sperm for use in her practice that she was required to undertake as a result of the breach. It was a step taken in mitigation of her loss, if the loss is characterised as the loss of sperm for use in her practice. On this hypothesis, it would not be correct to characterise the loss as the loss of property with ordinary commercial value given the restrictions upon the permissible use of the sperm. As the interest harmed by breach was that in having sperm available for use in the clinic, the loss was more akin to the loss of hire in The New Flamenco, not the sale of the vessel. It will thus be interesting to see what arguments are made in future cases about the interaction of these cases, and the application of the principles discussed therein.
The Longchamp and the Law of General Average
42 The third case concerns general average. This is a uniquely maritime concept involving the right to payment for contribution to prevent loss in a common maritime adventure. It is the unique "system of maritime law by which sacrifices of property made, and loss and expenditure incurred, as a direct result of actions taken for the purpose of preserving a common maritime adventure from peril are rateably shared between all those whose property is at risk in the adventure." General average has Rhodian origins, and this has been acknowledged over the centuries. In Burton & Co v English & Co, Brett MR described how it arises "…from the old Rhodian laws, and has become incorporated into the law of England as the law of the ocean." The fragment of the original Rhodian code incorporated into Justinian's Digest was about general average (although termed "jettison"), and it was also part of the Rhodian Sea Law. These origins of the doctrine were expressly referred to in The Longchamp. The substance of the rules of general average remain consistent with these Rhodian origins. It was really an early form of marine insurance that has persisted. Much of the modern law of general average is now governed by a set of rules known as the York-Antwerp Rules that are incorporated into shipping contracts.
43 The Longchamp is the first UK Supreme Court decision to deal with a particular rule of the 1974 CMI version of these rules, Rule F. It is also significant as many general average disputes never reach a court room and are instead settled in the trade. The vessel the subject of the case had been hijacked by Somali pirates in January 2009 who demanded a ransom of USD 6 million for its release. The owners managed to negotiate the ransom demand down to a sum of USD 1.85 million. They sought, under the rules of general average, for the cargo owners to make a proportionate contribution toward the ransom payment and, also, toward operating expenses incurred over the three month period while they were negotiating a reduction in the ransom demand. It was accepted that the ransom payment was recoverable under Rule A of the York-Antwerp Rules. Rule A was as follows:
There is a general average act, when, and only when, any extraordinary sacrifice or expenditure is intentionally and reasonably made or incurred for the common safety for the purpose of preserving from peril the property involved in a common maritime adventure.
44 The cargo owners disputed whether the operating expenses incurred while negotiating the reduced ransom were within Rule F, which was as follows:
Any extra expense incurred in place of another expense which would have been allowable as general average shall be deemed to be general average and so allowed without regard to the saving, if any, to other interests, but only up to the amount of general average expenses avoided.
45 The majority of the Supreme Court, in a lead judgment delivered by Lord Neuberger, found that cargo owners were required to contribute proportionally under Rule F to the operating expenses incurred while the vessel was detained, as it enabled a reduced ransom to be negotiated. Such expenses were recoverable to the extent they reduced the ransom amount paid. Lord Neuberger rejected as contrary to the text of Rule F the submission by the cargo owners that was accepted by the Court of Appeal and the average adjusters, that the operating expenses were not an "extra expense" under Rule F as paying a reduced ransom was not an "alternative" course of action. Lord Mance dissented on this point, concluding that Rule F only allowed recovery of expenses incurred as an alternative to the ordinary way of dealing with the situation which in this case was paying a ransom.
46 The majority approach in The Longchamp places emphasis on the text of the Rules. The cargo owners' argument glossed the text of Rule F which, as Lord Sumption said in his concurring speech, was "simplicity itself". The majority concluded that the commercial practice of average adjusters could not dictate the clear meaning of the Rule. Although it is always important to have regard to the practice of the maritime industry in shipping cases, the result reached by the majority is in fact the more commercially attractive one. It seems peculiar that expenses incurred in connection with reducing the ransom would not be within Rule F. The straightforward construction given to Rule F gives greater leeway to parties facing a general average event.
47 Whilst a textual answer was preferred over the views of the practising profession, the result was both intuitively correct and as Lord Sumption said, simplicity itself.
The Contribution of Onassis, Maria Callas and Vergottis
48 Finally, there is one further Greek connection to English shipping law to consider: that of Maria Calogeropoulos (Callas), Aristotle Onassis and Panaghis Vergottis in a case concerning witness credibility, the breakdown of personal relationships and the curious joint acquisition of the bulk carrier Artemison II that went all the way to the House of Lords.
49 The flavour of the case, together with its dramatis personae, was summed up with characteristic style (if here somewhat condescendingly) by Lord Denning in the Court of Appeal:
This is a tale of three friends who fell out. Their quarrel has reached the Courts. It has found its vent in a dispute about shares. Each side says the other is lying. Charges of fraud are freely made. Goodness knows where the truth lies.
The first of the three friends is Mr. Aristotle Onassis. He is 61, a Greek shipowner, as rich as Croesus. He carries on business in Monte Carlo. He has a house in Paris. Another near Athens. He owns the island of Scorpios off Greece, and he has a splendid yacht Christina.
The second is Mr Panaghis Vergottis. He is 77, also a Greek shipowner, a close friend of Mr Onassis for over 30 years. He is rich but nowhere near as rich as Mr. Onassis. He lives at the Ritz Hotel in London, which speaks for itself. Both men were devoted to the third, a lady Mme Maria Calogeropoulos, better known as Maria Callas. She is 43, also a Greek. She is highly gifted as an opera singer and as an actress. She has a world-wide reputation.
50 The facts of the case were as follows. Onassis and Vergottis, in 1964, jointly purchased a bulk carrier, Artemison II for £1.2 million. Each paid half of the cash sum payable for the purchase, and formed a Liberian holding company to own the ship. The documentary evidence indicated that out of 100 shares, Onassis received 50 shares, giving 26 to Maria Callas and 24 to his nephew. Vergottis received 50 shares. It was also clear that Maria Callas had paid £60,000 to the company.
51 Mr Onassis and Mme Callas sued Mr Vergottis. They alleged that there has been an agreement between them under which, in return for the £60,000, Mme Callas was to receive 25 of Mr Vergottis' shares, so that she would end up with majority control of the company. In the alternative, they argued that the sum paid by Mme Callas should be treated as a loan to the company, with an option to call for delivery of the 25 shares. They sued for specific performance of the alleged agreement. The agreement was said to have been struck or amended during negotiation onboard the yacht Christina, on the island of Scorpios, in London at Claridge's, and at Maxim's in Paris. Vergottis denied the contract, and said that the sum paid by Mme Callas was a loan to the company – nothing more.
52 The matter was tried before Roskill J (as he then was) for nine days with an extempore judgment delivered on the tenth. All three gave evidence. Justice Roskill stated that, in the end, the case turned "upon a simple question of personal credibility" – who was telling the truth? His Lordship commented that it was rare "to encounter a [commercial] case where there is so stark a contrast of evidence". Justice Roskill believed Onassis and Mme Callas. She had exercised an option to buy the shares. The case turned on a matter of impression, and Roskill J stated his impressions emphatically:
This is simply a question how the witnesses struck me as they gave evidence. I formed a highly favourable view of Mr. Onassis and of Mme Callas. Their evidence can be criticized in its detail, it was not always consistent; but I could not but take an unfavourable view of Mr. Vergottis as a witness, reluctant as I am to say that of a man of his age and standing. His whole attitude in the witness-box gave me the impression that he had in truth stood upon his rights in this Court, not only that he might see Mme Callas and Mr. Onassis cross-examined at length by his learned Counsel – as was done with the utmost discretion and good taste – as to what the relationship between them might be, but that he might use his opportunity of going into the witness-box to make such venomous remarks about them as he could slip in before he was stopped, either by his own learned Counsel or by Sir Milner or by myself. He made an unfavourable impression upon me, and I have no hesitation in holding that the plaintiffs' story is the true story and the defendant's is the untrue story.
53 Strong stuff. One cannot help thinking (and it should be forgiven as only natural after nine days) that he fell, just a little, for La Divina. The Court of Appeal (after a nine day appeal) disagreed with Roskill J's conclusions. It found that he had erred in making a purported finding that "greed" or avarice motivated Vergottis and in not addressing evidence (especially documentary) that was in Vergottis' favour. One gets the impression that the Court thought Roskill J had fallen for Maria. A majority of the House of Lords (after an eight day appeal) ultimately took a different view on a further appeal and restored the judgment of Roskill J. Their Lordships agreed that this was a case that turned on the judge's assessment of witness credibility at trial and found that the he had not erred. These were matters for the trial judge, and respect had to be given to his impression of the witnesses.
54 The judicial disagreement in this bitterly personal dispute reflects the difficulty in cases that turn on judgments of impression by a trial judge, and in resolving appeals from such assessments. It also reflects an approach to appeals that has passed into memory. The weight to be given to the assessment of impressions made at trial arises from the indefinable and intangible advantages held by a trial judge who is able to observe the witnesses in person and order, contemplate and assess their evidence contemporaneously as it emerges over the course of the trial. These advantages may be "subtle and imprecise". Factual findings made based on impressions of credibility, as occurred in this case, are seen particularly as matters where a trial judge has an advantage. Thus, it has been said that findings based on credit should only be set aside where incontrovertible facts or uncontested testimony show them to be wrong or where they are glaringly improbably or contrary to compelling inferences. But it is important not to transform such a linguistic expression of principle into a rigid, abstract test. There should be no rigid approach as to these matters in cases that turn on the assessment of witness credibility.
55 Lord Pearce recognised this in his dissenting speech in the House of Lords. His Lordship remarked that the "solution" to the appeal was "not to be found in any principle of law" and that there could be no code established to govern when an appellate court should overturn a trial judge or order a new trial. He described the advantages possessed by a trial judge but stated that, although difficult to overturn factual findings, it was not a Sisyphean task. Rather, whether the appellate courts should interfere were "questions of degree". In some cases, the court may be justified in reversing the judgment. In others it may have to order a retrial, depending on the circumstances. Lord Pearce remarked that:
The considerations which I have lightly adumbrated are part of the daily fare of a Court of Appeal in its task of deciding whether there is enough in an appeal to call for its duty to interfere. Reported cases would not greatly help, since each case depends on its particular circumstances.
56 Thus, the question of whether an appellate court should interfere is one that must be resolved in the circumstances of the particular case, having due regard to the nature of the appellate task and the advantages a trial judge has to form an impression of the evidence and, as Lord Pearce described, "absorb the atmosphere of the case".  Of course we have our own sources of guidance in Australia: Warren v Coombes; Abalos v Australian Postal Commission; Fox v Percy; State Rail Authority (NSW) v Earthline Constructions Pty Ltd (in liq) tell us that a formulaic approach is not always very helpful. Such questions require a human judgment, often based on human impressions in the circumstances of the case. It is worth recalling Heraclitus and his emphasis on experience.
57 One cannot leave this case without paying tribute to Eustace Roskill's sense of humour and style. One part of the evidence of Mr Onassis concerning the commercial structure of owning a ship was summarised by Roskill J:
It is worth recording Mr. Onassis's unconscious contribution to English jurisprudence regarding the test for implying a term in a contract. He was asked …
Q.: Did you have a discussion with him about forming a company to own the ship? A.: It goes without saying, even among our teenagers, that when you buy a ship like that, [you are] supposed to form a company which, quite often in the old days, was preferably Panamanian; in the last ten or fifteen year it is preferably Liberian. . . .
That perhaps is a useful alternative to Lord Justice MacKinnon's "officious bystander".
58 One thing all these cases have in common with each other, and with Greek philosophy, is that they show law is as much about shape as it is about lines.
* Chief Justice of the Federal Court of Australia. I acknowledge the assistance of my associate, Samuel Walpole, in the preparation of this paper. This is a revised version of a paper first presented at the Hellenic Australian Lawyers Association International Conference on Contemporary Legal Issues and the Influence of Hellenism, Rhodes, Greece, 12 July 2018.
 With apologies to Helen of Troy.
 Luke v Lyde (1759) 2 Burr 882 at 889; 97 ER 614 at 619.
 William Tetley, Maritime Liens and Claims (Editions Yvon Blais, 2nd ed, 1998) at 7-8; William Tetley, International Maritime and Admiralty Law (Editions Yvon Blais, 2002) at 9; William Tetley, "The General Maritime Law – The Lex Maritima" (1994) 20 Syracuse Journal of International Law and Commerce 105 at 109; James Allsop, "International Commercial Law, Maritime Law and Dispute Resolution: The Place of Australia, New Zealand and the Asia Pacific Region in the Coming Years" (2007) 21 Australian and New Zealand Maritime Law Journal 1 at 3.
 Tetley, 1994, op cit 3 at 109; Tetley, 2002, op cit 3 at 9. See also Robert D Benedict, "The Historical Position of the Rhodian Law" (1909) 18 Yale Law Journal 223; cf Edward Cohen, Ancient Athenian Maritime Courts (Princeton University Press, 2015) at 5.
 Digest 14.2. Book 14 of the Digest is entitled: "The Rhodian Law of Jettison", see Alan Watson (trans), The Digest of Justinian (University of Pennsylvania Press, 1985) vol 1 at 419-422. See also Tetley 1994, op cit 3 at 109; Tetley, 1998, op cit 3 at 7-8; Tetley, 2002, op cit 3 at 9; AR Emmett, "Reception of Roman Law in the Common Law" in JT Gleeson, JA Watson and RCA Higgins (eds), Historical Foundations of Australian Law (Federation Press, 2013) vol 1 at 78.
 Tetley, 1994, op cit 3 at 109; see also David Abulafia, The Great Sea: A Human History of the Mediterranean (Penguin, 2011) at 236.
 Tetley, 2002, op cit 3 at 10-11.
 Walter Ashburner, 'The Rhodian Sea-Law' (Clarendon, 1909).
 Tetley, 1994, op cit 3 at 109.
 Cohen, op cit 4 at 5.
 United Nations Conference on Trade and Development, Review of Maritime Transport 2017 (5 October 2017) <http://unctad.org/en/PublicationsLibrary/rmt2017_en.pdf> at 28.
 Transfield Shipping Inc v Mercator Shipping Inc  UKHL 48;  1 AC 61.
 Kodros Shipping Corporation of Monrovia v Empresa Cubana de Fletes  AC 736.
 Ocean Tramp Tankers Corporation v V/O Sovfracht  2 QB 226.
  1 Lloyd's Rep 467.
 Marbienes Compania Naviera SA v Ferrostaal AG  2 Lloyd's Rep 149.
 New Zealand Shipping Co Ltd v M Satterthwaite & Co Ltd  AC 154 (PC).
 TA Shipping Ltd v Comet Shipping Ltd  1 Lloyd's Rep 675.
 Leeds Shipping Co Ltd v Societe Francaise Bunge  2 Lloyd's Rep 127 at 131.
 Gard Marine and Energy Ltd v China National Chartering Co Ltd  EWHC 2199 (Comm);  1 Lloyd's Rep 59.
 Gard Marine and Energy Ltd v China National Chartering Co Ltd  EWCA Civ 16;  1 Lloyd's Rep 381.
  AC 736 at 749 per Lord Diplock and 760-761 per Lord Roskill.
  AC at 749 per Lord Diplock and 763 per Lord Roskill.
  1 WLR at 1803 - per Lord Clarke.
  1 WLR at 1807 .
  1 WLR at 1805 .
  1 WLR at 1804-1805 .
  1 WLR at 1807-1808 -.
  1 WLR at 1808 , citing The Saga Cob  2 Lloyd's Rep 545.
  1 WLR at 1803 .
 Bertrand Russell, A History of Western Philosophy (Unwin, 1984) 25.
 Russell, op cit 31 at 25; see also Iain McGilchrist, The Master and his Emissary (Yale, 2009) at 266.
 McGilchrist, op cit 32 at 267.
 Russell, op cit 31 at 59.
 Russell, op cit 31 at 59.
 Russell, op cit 31 at 59.
 William James, "The Function of Cognition" (Address to the Aristotelian Society, 1 December 1884).
 (1848) 154 ER 363 at 365.
 Golden Strait Corporation v Nippon Yusen Kubishika Kaisha  UKHL 12; 2 AC 353 at 370  per Lord Bingham; 382  and 383 - per Lord Scott. In Bunge SA v Nidera BV  UKSC 43; 2 Lloyd's Rep 469 at 473 , Lord Sumption stated that "the fundamental principle of the common law of damages is the compensatory principle". See also, in Australia, Haines v Bendall  HCA 15; 172 CLR 60 at 63; Johnson v Perez  HCA 64; 166 CLR 351 at 355-356; Hyundai Merchant Marine Co Ltd v Dartbrook Coal (Sales) Pty Ltd  FCA 1324; 236 ALR 115 at 135 ; Clark v Macourt  HCA 56; 253 CLR 1 at 6 , 11 , 19 , 30 .
  2 Lloyd's Rep at 476 .
 Transfield Shipping Inc v Mercator Shipping Inc  UKHL 48;  1 AC 61.
 Hadley v Baxendale (1854) 156 ER 145 at 147 ; The Heron II  1 AC 350 at 421 per Lord Reid.
 Adam Kramer, The Law of Contract Damages (Bloomsbury, 2nd ed, 2017) at 300 [14-02], referring to Siemens Building Technologies FE Ltd v Supershield Ltd  EWCA Civ 7 at  per Toulson LJ.
 Kramer, op cit 43 at 303-304 [14-13]-[14-17].
  1 AC at 66-67 .
 James Allsop, "Characterisation: its place in contractual analysis and related inquiries" (Paper presented at the Contracts in Commercial Law Conference, Sydney, 18-19 December 2015) at .
 Kramer, op cit 43 at 355 [15-08], 357-358 [15-15]-[15-16].
 Kramer, op cit 43 at 359 [15-20].
 Kramer, op cit 43 at 358 [15-19].
 See  1 All ER (Comm) 1205.
  1 WLR at 2463 , 2466 .
  1 WLR at 2462 .
 The Elena D'Amico  1 Lloyd's Rep 75. In Bunge  2 Lloyd's Rep at 474  Lord Sumption, described this prima facie measure (correctly, I suggest) as "not so much a rule as a technique which is prima facie to be treated as satisfying the general [compensatory] principle".
  1 WLR at 2464 , 2465 , citing British Westinghouse Electric & Manufacturing Co Ltd v Underground Electric Railways Co of London Ltd (No 2)  AC 673.
  1 WLR at 2592 .
  1 WLR at 2592-2593 .
  1 WLR 2593 at -.
  1 WLR 2593 at .
  1 WLR 2593 at .
  1 WLR 2593 at .
 JJ Edelman, McGregor on Damages (Sweet & Maxwell, 20th ed, 2018).
  HCA 56; 253 CLR 1.
 253 CLR at 6  per Hayne J; 11 - per Crennan and Bell JJ; 18-19 - per Gageler J and 30  per Keane J.
 253 CLR at 31-32  per Keane J; 10  per Hayne J; 15  per Crennan and Bell JJ.
 253 CLR at 36  per Keane J
 253 CLR at 19 -,  per Gageler J.
 253 CLR at 22 - per Gageler J.
 Kramer, op cit 43 at 142-143 [4-214].
 The Longchamp  1 Lloyd's Rep at 3  per Lord Neuberger.
 (1883) 12 QBD 218 at 221.
 Digest 14.2 at , where the Digest states that: "The Rhodian law provides that if cargo has been jettisoned in order to lighten a ship, the sacrifice for the common good must be made good by common contribution." See Watson, op cit 5 at 419.
 Ashburner, op cit 8 at ccli-ccxciii.
 See Burton & Co v English & Co (1883) 12 QBD 218 at 221 per Brett MR; The Longchamp  1 Lloyd's Rep at 3  per Lord Neuberger. In addition, Lord Neuberger describes how general average was "first authoritatively discussed" in England in Birkley v Presgrave (1801) 1 East 220 at 228-229. See also NG Hudson and MD Harvey, The York-Antwerp Rules (Lloyd's List, 3rd ed, 2010) at 3-5 [1.04]-[1.10].
 Martin Davies and Anthony Dickey, Shipping Law (Thomson Reuters, 4th ed, 2016) at 762 [18.10]. The substance of these rules, like the common law of general average, can be traced to Rhodian law: AR Emmett, "The Law Merchant: How we came to where we are" in JT Gleeson and RCA Higgins, Constituting Law: Legal Argument and Social Values (Federation Press, 2011) at 222.
 Comité Maritime International.
 Many general average disputes never reach a court room. At  1 Lloyd's Rep 10 , Lord Mance remarked that "general average cases are few and far between". One other recent decision relating to general average is the recent judgment in Australia of McKerracher J in Offshore Marine Services Alliance Pty Ltd v Leighton Contractors Pty Ltd  FCA 333; 252 FCR 574. His Honour held that the obligation to contribute to general average expenses under s 72(3) of the Marine Insurance Act 1909 (Cth) extended only to an owner of the relevant cargo or someone who owed "a contractual obligation to the general average claimant in circumstances governed either by a bill of lading or a general salvage bond":  FCA 333 at -.
  1 Lloyd's Rep 1 at 5-6 - per Lord Neuberger.
  1 Lloyd's Rep 1 at 7 - per Lord Neuberger. If an alternative course of action was required, Lord Neuberger's view was that opting to negotiate a reduced ransom would be sufficient:  1 Lloyd's Rep 1 at 7 .
  1 Lloyd's Rep 1 at 11-12 - per Lord Mance.
  1 Lloyd's Rep 1 at 10  per Lord Sumption.
  1 Lloyd's Rep 1 at 7  per Lord Neuberger. General average contributions are traditionally assessed by "average adjusters": see generally FD Rose, General Average: Law and Practice (LLP, 2nd ed, 2005) Ch 6. The average adjuster in The Longchamp considered that the cargo owners were required to make a contribution under Rule F in respect of the operating expenses. A report from the "Advisory Committee of the Association of Average Adjusters", by majority, concluded that they were not, and the owners subsequently brought their action in the Commercial Court.
 Onassis and Calogeropoulous v Vergottis  1 Lloyd's Rep 607; Onassis and Calogeropoulos v Vergottis  1 Lloyd's Rep 294; Onassis and Calogeropoulous v Vergottis  2 Lloyd's Rep 403.
  1 Lloyd's Rep 294 at 294.
  1 Lloyd's Rep at 613.
  1 Lloyd's Rep at 614.
  1 Lloyd's Rep at 632.
  1 Lloyds Rep at 298, 301, 304.
  2 Lloyd's Rep at 407.
  2 Lloyd's Rep at 415-416 per Viscount Dilhorne; 425 per Lord Morris; 429-430 per Lord Guest. Lords Pearce (at 436) and Wilberforce (at 436) dissented.
 Branir Pty Ltd v Owston Nominees Pty Ltd (No 2)  FCA 1833; 117 FCR 424 at 435-436 ; Stone & Wood Group Pty Ltd v Intellectual Property Development Corporation Pty Ltd  FCAFC 29 at -; State Rail Authority (NSW) v Earthline Constructions Pty Ltd (in liq)  HCA 3; 160 ALR 588 at 619-620 ; Fox v Percy  HCA 22; 214 CLR 118 at 126 .
 Branir 117 FCR at 437 .
 Branir 117 FCR a 436 ; Devries v Australian National Railways Commission  HCA 78; 177 CLR 47 at 479-480 per Deane and Dawson JJ.
 Fox v Percy  HCA 22; 214 CLR 118 at 128  per Gleeson CJ, Gummow and Kirby JJ.
 Fox v Percy  HCA 22; 214 CLR 118 at 128  per Gleeson CJ, Gummow and Kirby JJ.
  2 Lloyd's Rep at 430.
  2 Lloyd's Rep at 430, quoting Benmax v Austin Motor Company Ltd  AC 370 at 377 per Lord Somervell.
  2 Lloyd's Rep at 431.
  2 Lloyd's Rep at 430. Sisyphus being the figure in Greek mythology cursed to try to push a boulder up a hill, only for it fall down as he reaches the top, for eternity.
  2 Lloyd's Rep at 430.
  2 Lloyd's Rep at 430.
  2 Lloyd's Rep at 431.
  2 Lloyd's Rep at 431.
  HCA 9; 142 CLR 531.
  HCA 47; 171 CLR 167.
  HCA 22; 214 CLR 118.
  HCA 3; 160 ALR 588.
  1 Lloyd's Rep at 624.