English law is much used in international shipping and commerce, and London is one of the major centres of maritime dispute resolution, whether by litigation or arbitration, (English courts have also exercised considerable control over arbitrators, and this has brought in litigation even on arbitrated disputes. The amount of such control has been much reduced by statutes: the most recent is the Arbitration Act 1996) in the world.
Sea transport has to be seen, however, also against the background of insurance: the risks are so great that all involved are likely to insure, and one of the main objects of shipping contracts is to allocate risks, so that parties know those for which they are responsible and which they should insure (See further J Gilman and others (eds), Arnould’s Law of Marine Insurance and Average (18th edn, 2013); H Bennett, The Law of Marine Insurance (2nd edn, 2006); FD Rose, Marine Insurance: Law and Practice (2nd edn, 2012). Thus the cargo owner may insure the risks to his property with a cargo underwriter (as a shipowner may insure the risks to his property with a hull underwriter); but shipowners commonly insure most of their liability to cargo (and for other matters, e.g. personal injury and oil pollution.) with a mutual insurance association (traditionally called a ‘Protection and Indemnity Association’, or ‘P & I Club’). Separate insurance may also be effected against the risks that the cargo owner has paid, or the shipowner has not earned, freight for a voyage that is not completed.
The picture is further complicated where a casualty occurs, requiring services to redeem the situation. Liability to pay a ‘salvage’ reward to a salvor extricating property from danger rests on the owner of the interest salved (whether ship, cargo or freight) – though, e.g., a cargo salvee may seek recourse from a shipowner whose fault caused the casualty (see further FD Rose, Kennedy & Rose: The Law of Salvage (8th edn, 2013). But in cases of a common danger a single interest may incur a loss in order to preserve the common adventure. The classical example is where an item of cargo is jettisoned in order to lighten the vessel; the usual modern example is where the shipowner incurs extraordinary expense to complete the adventure. This is a case of ‘general average’ (see further J Cooke and R Cornah (eds), Lowndes & Rudolf: General Average and the York-Antwerp Rules (13th edn, 2008); FD Rose, General Average: Law and Practice (2nd edn, 2005) requiring all who benefit from that loss to contribute to it – albeit both the person incurring the initial loss and the contributories may insure such risks. All this means that much litigation in this area is often technical, designed to clarify points of risk allocation, and supported or brought by insurers on both sides.
Traditionally, there are two types of contract available to a person who wishes to make use of a ship. The first is the charterparty. The second is the bill of lading contract. In relation to a particular vessel or voyage, charterparties and bill of lading contracts may exist independently of each other or simultaneously.
However, most charterparties are concluded contemplating the carriage of one or many more shipments of goods either by virtue of the charterparty itself or under contracts made consequent upon the charterparty. Thus, a ship may be chartered for a particular voyage to someone who has a complete cargo, usually of a bulk commodity, which will in effect fill the whole carrying capacity of the ship. Many forms of chartering in this context involve use of the ‘tramp ship market’, a market of vessels which hold themselves available for such arrangements; such an arrangement can be called shipping on ‘tramping terms’. Alternatively, a person may charter a ship with a view to making the ship available (in accordance with the terms of the charterparty) for the carriage of separate consignments for many consignors. The arrangements for their carriage will generally utilize the second main type of contract, the bill of lading contract. The phrase ‘bill of lading’ (regularly shortened to ‘B/L’) is an ancient form of ‘bill of loading’; the document may be given a different name in other languages (e.g., connaissement, conocimiento de embarque).
It is a general rule that parties’ relationships are governed by the particular terms upon which they have contracted rather than the apparent form of their agreement. There can be sub-charters, sub-sub-charters and so forth. There are other variants on chartering. Thus there can be charters (or sub-charters) of particular holds of a vessel or of space. There is in the container trade a recently developed institution called ‘slot chartering’, under which the (so-called) charterer contracts for a certain number of container ‘slots’, i.e. contracts to use and/or pay for, and the carrier contracts to provide, the space occupied by so many containers on a particular voyage or on several voyages, or for periods otherwise determined. These only come within the traditional description of charterparties in a very limited sense (The charterer has no control over the general use of the ship. However, the Court of Appeal has held a slot charter to be a charterparty for the purposes of Admiralty Jurisdiction: The Tychy  2 Lloyd’s Rep 11, CA.). Finally, there can be long-term contracts of affreightment, under which an operator contracts to provide ships to lift cargo within certain periods from certain places, without its being initially settled which ships will be used and when. Such contracts contain provisions for narrowing down to specific vessels at appropriate times, and usually provide for the form of charterparty to be used for the actual voyages.
A charterparty (often shortened to ‘charter’) is a contract whereby (subject to the reservations made above) the carrying capacity of a ship is by contract placed by the owner (or sometimes demise charterer) at the disposal of another person, referred to as the ‘charterer’. Charterparties are of two basic types, the voyage charter and the time charter.
The voyage charterparty is ‘a contract to carry specified goods on a defined voyage or voyages, the remuneration of the shipowner being a freight calculated according to the quantity of cargo loaded or carried, or sometimes a lump sum freight’ (Scrutton, para. 4.017). This is the older type of charterparty and dates from times when only the shipowner could take the responsibility for estimating the risks and likely duration of the voyage and what he needed to charge for it. The destination of the voyage is specified, though the charterer may have discretion to determine or change it within limits. Voyage charters are made on standard contract forms, but their long history means that there is considerable scope for the application of general principle.
In substance the voyage is undertaken for a fixed price. The risk of delay affecting the profitability of the carriage contract is therefore on the shipowner, and within that price he normally allows fixed free time for loading and unloading, called ‘laytime’ (lying time), ‘laydays’ or other variants. The calculation of this – it’s starting point, whether it can be interrupted and for how long it runs – gives rise to disputes which, because of the great significance of delay in maritime operations, involve in their resolution technical questions of interpretation of wording and, in their result, large sums of money. If the charterer delays the ship beyond laytime he is in breach of contract and liable in damages, but it is normal to agree in advance the rate at (and time during) which these are payable, under the title ‘demurrage’ (French demeure, delay money). This is by modern standards normally classified as liquidated (or agreed) damages for breach of contract by detaining the ship beyond laytime: that is to say, the charterer is not entitled to keep the ship on demurrage, but is paying damages for doing so (In theory a stipulation for demurrage could be a penalty, and relief granted against it accordingly. But the sum involved is more likely to undercompensate than overcompensate, as high demurrage rates could be a weak selling point. Some standard forms also fix ‘dispatch money’, a bonus for quick loading or unloading, as a fraction of demurrage, this again discourages high demurrage rates. For a dispute turning on such factors see the famous case of Suisse Atlantique Cie d’Armement Maritime SA v NV Rotterdamsche Kolen Centrale  1 AC 361, HL, which additionally stands as a leading English decision on a much more general issue relating to unfair contract terms). Demurrage disputes – as to what sorts of losses are covered by demurrage, when demurrage starts and finishes, whether on the terms of the contract it can be interrupted and so forth – are likewise technical and involve large sums of money. Once laytime and time on demurrage have ceased, damages for continuing delay are payable on normal common law principles.
Another way in which the charterer may cause the shipowner unjustifiable loss of anticipated profit is, when the freight is calculated by quantity or size of the cargo carried, by not utilizing all the cargo space or not shipping the cargo contracted for. Damages for failure to do the former are equivalent to the freight which would have been earned, and are called ‘dead freight’. Here too, complex disputes can arise as to whether indications of cargo carrying capacity in the charterparty (or sometimes elsewhere) are contractual promises by the charterer to load that quantity or fill that space, or simply a general indication of the sort of capacity involved. (The same problem can arise in reverse, where the ship proves unable to carry the anticipated quantity.)
A voyage, charterparty is a contract for services (locatio conductio operarum), to be rendered by the shipowner, his equipment and employees. It is definitely not a hire of the ship (locatio conductio rei), for the ship remains under the control of the shipowner and master.
Time charterparties are a later invention and date from the times when sail was giving way to steam and ship movements were becoming more predictable. ‘The shipowner agrees…to render services for a named period by his master and crew to carry goods put on board the ship by or on behalf of the time charterer’14 (Scrutton, para. 4.017) in return for payment of ‘hire’. The shipowner is said to ‘deliver’ the ship to the charterer, at the beginning of the period in which its services are to be made available; and, at the end of that period, the ship should be free of commitments and ‘redelivered’ to the shipowner by the charterer. However, such (re)delivery is metaphorical, as the shipowner and his master remain in control of the vessel throughout. The charterer can give commercial (but not navigational) (the difference may prove highly controversial, for instance if the charter specifies a route: see The Hill Harmony  1 AC 638) orders to the shipowner within specified geographical limits (which may be anything from very narrow to ‘worldwide’) during the charter period. The arrangement is essentially one whereby the shipowner makes available the services of his ship on an agreed basis, within which the charterer arranges for the carriage of cargo (normally cargo belonging to third parties). The contract normally provides for the circumstances in which the charterer can give the shipowner orders as to the ‘employment’ of the vessel (in particular, requiring him to sign bill of lading contracts), in return for which he is obliged to indemnify the shipowner for losses resulting from obedience to such orders (This includes the obligation to pay the shipowner for caring for the cargo after the shipowner has lawfully exercised his right to withdraw the vessel (11.43-11.45): see ENE Kos 1 Ltd v Petroleo Brasileiro SA (The Kos)  UKSC 17,  2 AC 164). Whether third parties contract for the carriage of their goods with the charterer or the shipowner, the freight payable by cargo owners is ultimately the source of both the shipowner’s and the charterer’s income and profit, so the charterparty will need to provide for how the financial arrangements are made between its parties. Thus, to ensure that the charterer does not collect freight from cargo owners without paying the hire due under the charterparty, the charterparty may provide that the shipowner has a ‘lien’ on freights payable under bill of lading contracts (Such liens can be problematic. See Western Bulk Shipowning III A/S v Carbofer Mariti eTrading APS (The Western Moscow)  EWHC 1224 (Comm),  2 Lloyd’s Rep 163).
A time charterparty is similar to a voyage charterparty except I the method of calculation of payment and hence the allocation of risk of delay during the carriage. The shipowner is paid by time so will usually have no objection to being delayed by the charterer’s orders or failure to give them. Hence a counterpart to the laytime and demurrage regime is required. This is normally provided by an ‘off-hire clause’, under which no hire is payable while the ship is (in whole or in part, depending on the wording of the clause) in effect not working for the charterer.
The time at which the charter is due to expire often gives rise to disputes. On the one hand, the voyages that the charterer can arrange at the end of the charter period may well not fit precisely within it, and the charter will be reluctant to pay for unused time, but the shipowner will wish to earn the maximum possible under the contract. On the other hand, if a voyage is prone to overrun, there may be dispute as to whether the shipowner must proceed on it and, if he does, whether remuneration is payable at the contract rate or (more likely) as damages for breach of contract, particularly when market rates have increased (see Hyundai Merchant Marine Co v Geduri Chartering Co Ltd (The Peonia)  1 Lloyd’s Rep 100). In general, the ship must be redelivered either within a reasonable period after the end of the charter period or in accordance with contractually agreed tolerance. If the ship is redelivered early, freight would normally be payable until the end of the charter period, unless a refusal by the shipowner to take it back was totally unreasonable (see Isabella Shipowner SA v Shagang Shipping Co Ltd (The Aquafaith)  WEHC 1077 (Comm),  2 Lloyd’s Rep 61, interpreting the decision in White & Carter (Councils) Ltd v McGregor  AC 413, HL as to an innocent party’s entitlement to continue performance despite the other party’s repudiatory breach of contract).
There can also be mixed time and voyage charterparties, where the ship is contracted to go to a particular port, but paid on a time basis – perhaps, e.g., because the shipowner is uncertain as to the time likely to be required at the destination port. These, often called ‘trip charters’, are time charters as regards payment; within the voyage route the charterer can give orders (Ocean Tramp Tankers Corp v VIO Sovfracht (The Eugenia)  2 QB 226 CA) and there may be special provisions for delivery and redelivery; but the ship cannot be ordered off the normal route for the voyage (see Temple SS Co Ltd v VIO Sovfracht (1945) 79 LI L R 1).
This is a totally different arrangement, as the ancient English word ‘demise’, formerly used in a property context, indicates. It is not a contract for services at all but rather a contract for the hire of the ship itself (location conductio rei). The charterer has possession of the ship, employs the master and crew, makes contracts of carriage (whether by charterparty or bill of lading) for it, and for most purposes is the equivalent of owner of it during the duration of the charterparty. Such contracts are not contracts for the carriage of goods by sea and are used for different purposes, such as operating ships without capital expenditure, loan financing and various devices connected with the changing of flags. There are nowadays standard forms of demise charter, though there are older cases on the question whether a particular contractual arrangement constituted a demise or a time charter (e.g. Baumwoll Manufactur von Carl Scheibler v Furness  AC 8, HL) which would nowadays more usually be clear.
The general principle in this area is that there is complete freedom of contract. Although international conventions concerning charterparty terms have been mooted, they have not yet come to fruition. The perception is that charterers and shipowners are on a more or less equal bargaining footing: indeed, in contrast with the normal situation under bills of lading, some charterers, such as governmental import or export agencies, may be in a stronger position than shipowning companies.
The basic common law rule then suffered a not very clearly traced metathesis into certain basic common law principles of interpretation of the contract terms used. These remain valid to this day (see Paterson SS Ltd v Canadian Wheat Producers Ltd  AC 538, 544-545). They are as follows:
Two further duties apply to the voyage:
It is often said that there is a further implied duty: to contribute in general average (see 11.06. The obligation is generally imposed by law and has in consequence become a contractually implied term: see a book review by Sir Christopher Staughton (1998) 114 LQR 677).
Contract terms in charterparties and bills of lading are interpreted against the background that these principles are assumed not to have been excluded unless clearly, and there are many dramatic examples of this (e.g. Tattersall v National SS co (1884) 12 QBD 297). But, if they are clearly excluded, freedom of contract applies in both types of contract and the exempting term is valid: unlike the position in some countries, there is no rule of law making terms excluding these duties (e.g. excluding liability for negligence) ineffective on grounds of public policy. Indeed, this was one of the factors leading to the adoption of the Hague Rules, which regulate the liability of the carrier under bill of lading contracts. The Unfair Terms in Consumer Contracts Directive (Operative under the Unfair Terms in Consumer Contracts Regulations 1999, SI 1999/2083) is obviously not relevant, as these are not consumer contracts; and the only general law in the United Kingdom on unfair contract terms, the Unfair Contract Terms Act 1977 is specifically excluded from the vast majority of maritime contracts (Unfair Contract Terms Act 1977, Sch 1 excludes from the Act contracts for the carriage of goods by sea except (i) in respect of exclusion or restriction of liability for death or personal injury resulting from negligence of (ii) in favour of a person dealing as consumer) as from international supply contracts (Unfair Contract Terms Act 1977, s 26).
Implied duties are here less prominent, but in general a charterer must: have cargo ready (a separate duty, but usually subsumed into the laytime and demurrage obligations); load a full cargo of the merchandise specified; and not without agreement of the owner ship dangerous goods (a topic the subject of much case law, in particular on who takes the risk as to goods not known to either side to be dangerous (see The Giannis NK  AC 605 (a case on the Hague Rules, but a leading case on this topic in general)). These duties are strict, in the absence of other indication.
The questions of who pays for and who takes the risks involved in (these are not necessarily the same: the burdens could lie on different parties. The standard term ‘FIO’ (‘Free in and out’) indicates that the freight agreed does not cover the cost of loading or unloading. But the question who takes the responsibility for these functions (as opposed to pays for them) requires further study of the contract terms. See The Jordan II  UKHL 49;  1 WLR 1363) loading or unloading operations turn on the individual contract. The starting point is often said to be that the charterer lifts the goods to the ship’s rail and the shipowner takes them from that point, though it is also said that the duty prima facie rests with the owner (see The Jordan II, n 43, at ). In any case, this is, however, virtually always modified by contract terms, usage or custom of the port (Pyrene Co Ltd v Scindia Navigation Co Ltd  2 QB 402). The time at which the risk passes from seller to buyer under a contract of sale of the goods need not be the same as the time when the shipowner assumes responsibility for the goods: the former is more likely to be the moment of crossing the ship’s rail (which seems to have been the case in Pyrene Co Ltd v Scindia Navigation Co Ltd  2 QB 402) which is thought to be a notion still familiar to merchants (However, when risk passes is more complex where the International Chamber of Commerce’s INCOTERMS 2010 are used).
To determine when the breach is sufficiently serious to entitle the innocent party to treat the contract as discharged, it can be said that three techniques are available, two of them at least stemming from maritime law.
The first technique looks to the term (or promise) of the contract broken and asks whether it is a term in respect of which the parties are to be regarded as having agreed that there should be exact compliance (see Photo Production Ltd v Securicor Transport Ltd  AC 827, 849, HL per Lord Diplock, who was largely responsible for the development of this branch of the law). If so, where there has been no such compliance the innocent party may terminate the contract, at least unless the other party can put the matter right within any express or implied time limit, and sue for damages also. Such a term is called a ‘warranty’. This is an established but curious use of the term ‘condition’: the term is really a promise rather than a condition. What is the condition is in fact the making good of the promise, which is a condition precedent to the other party’s duty to proceed.
The second technique looks to the nature and consequences of the breach. Its classic formulation, derived from the judgment of Diplock LJ in Hong Kong Fir Shipping Co Ltd v Kawasaki Kisen Kaisha Ltd ( 2 QP 26), a seminal case for the general law, but actually on the seaworthiness term, asks whether the effect of the breach is to deprive the innocent party of ‘substantially the whole benefit’ of the contract (see  2 QB 26, 70-71. In the Photo Production case, n 54, Lord Diplock called this a ‘fundamental breach’ (see 849), which invites confusion with older cases using this notion in connection with unfair exclusion clauses: the Suisse Atlantique case  1 AC 361 concerns this usage). This is more adverse to termination than the ‘condition’ technique: ‘deprive of substantially the whole benefit’ is not the same as ‘substantially deprive of the benefit’. Its most common application is to cases of delay, where it is asked whether (e.g.) to make the charterer load in the circumstances created by the breach would be to hold him to a different contract from that which he had made (Freeman v Taylor (1831) 8 Bing 124, 131 ER 348). Such a delay is often called a ‘frustrating delay’, and the test is the same as that in the doctrine of frustration, which releases both parties by a change of circumstances attributable to neither.
The third technique, quite close to the second, asks whether one party has behaved towards the other in such a way as to indicate repudiation or renunciation of his obligations under the contract (See Withers v Reynolds (1831) 2 B & Ad 882, 109 ER 1370 (a case on sale of goods)). Here the breach need not be of condition, nor its consequences in themselves so drastic as are required by the Hong Kong Fir test referred to above, but the situation must be sufficient to raise grave doubts in the innocent party as to the likelihood of the contract’s being properly performed over a period. Such reasoning is found in maritime cases, but more often in cases of instalment sales (see Sale of Goods Act 1979, 8 31).
The inevitable uncertainties of invoking the general law on termination (except where there is no doubt that the term concerned ranks as a condition), and the serious consequences of making the wrong decision, lead to the presence in charterparties of clauses designed to enable one party to avoid uncertainty and terminate regardless of whether the general law permits this. Of these, two, one in favour of the charterer and one in favour of the shipowner, deserve explanation here, since they impinge on and interact with general contractual principles.
This can be found in voyage and in time charterparties, and in general permits the charterer to cancel the contract if the ship does not arrive for loading by a certain time. The same objective could be partially achieved by stipulating that the time of sailing on the approach voyage, or of tender for loading, is a condition of the contract (e.g. Glaholm v Hays (1841) 2 Man & G 257, 133 ER 743). The cancelling clause is however wider in several respects and cannot be regarded simply as a condition expressed in other wording (against which there is no objection in principle). First, it applies even where the late arrival is excused by an excepted peril, i.e. is not a breach of contract at all: in this sense it may enable the charterer to act without reference to the doctrine of frustration (Smith v Dart & Son (1884) 14 QBD 105). Secondly, however, it does not displace the general law as to discharge by breach (Hong Kong Fir Shipping Co Ltd v Kawasaki Kisen Kaisha Ltd  2 QB 26 CA) and frustration (Bank Line Ltd v Arthur Capel & Co  AC 435), with the result that, even if the clause is not exercised, a later frustrating delay would entitle the innocent party to terminate the contract if there was a breach, or, if there was not, the contract could be frustrated. Finally, it may only be exercised exactly according to its terms; under many forms of the clause the charterer may and even must wait until the ship arrives, even if it is certain to be late, before deciding whether to cancel. By the same token, however, a cancellation outside the terms of the clause, e.g., too soon, (The Madeleine  2 Lloyd’s Rep 224) may be ineffective, even if late arrival is at that point inevitable. It will be seen therefore that, as elsewhere in English shipping law, the parties are held strictly to the terms of their contract, and in general no argument is possible that the right was exercised from what may be argued to have been inappropriate motives.
This is peculiar to the circumstances of time charterparties. It provides (with many variants of detail) that, if the hire is not paid by or before a certain time, the shipowner may withdraw the vessel, i.e. terminate the charter. This again protects the shipowner from the general law as to discharge by breach, which does not easily apply to late payment. For example, in land transactions it is often said that the time payment is not ‘of the essence’ (see also Sale of Goods Act 1979, s 10. cf. however the words of Lord Wilberforce as to time clauses in mercantile contracts, quoted in text to n 57). Again, the power can only be exercised in exact accordance with its terms, but again the motive for doing so is not relevant. So it can sometimes, depending on its formulation, be exercised by a shipowner who has already received late payment (unless this constitutes a waiver, which it is unlikely to do since banks, to whom such payment is made, may have no authority to waive contract terms) (see on both points Mardorf Peach & Co v Attica Sea Carriers Corp (The Laconia)  AC 850, HL). On the other hand it cannot be exercised before the time stipulated for its exercise, even if it is clear that timely payment can no longer be made (Afovos Shipping Co SA v Pagnan & Fratelli (The Afovos)  1 WLR 195, though the speech of Lord Diplock seeks to solve the case on the wider and more doubtful ground that the clause constitutes a condition, and there can be no anticipatory breach of condition: ‘Anticipatory breach is but a species of the genus repudiation and applies only to fundamental breach’ (at 203)).
Unlike late arrival, late payment will rarely be excused by any contract term, and the withdrawal clause is therefore almost bound to be exercised in respect of such a breach of contract. It is arguable, therefore, that unlike the cancelling clause it is to be regarded as a condition of the contract expressed in other words: interpretation is a matter for the court, and there is no objection to a condition not specifically expressed as such. It seems clear however that this is not so, though a small addition to the wording could give it this effect (as, in the context of hire purchase, in Lombard North Central plc v Butterworth  QB 527, CA (‘punctual payment [of] which shall be of the essence’). The difference in practice is this. If the withdrawal clause is a condition, the shipowner who withdraws the vessel is exercising his right to treat the contract as discharged under the general law and can sue, not only for accrued unpaid instalments, but also for loss of the profit contemplated from the rest of the charterparty, as arising from breach of contract, damages being available at common law in addition to termination. If it is not, no such damages are recoverable, the shipowner having himself chosen to cause the ending of the contract: all that is recoverable is the instalments already due but unpaid (Financings Ltd v Baldock  2 QB 104). It seems clear that the latter represents the present position in English law. The result is that the shipowner can only sue for full damages for loss of the remainder of the contract if the late payment constituted a repudiatory breach on general principles (as in Leslie Shipping Co v Welstead  3 KB 420).
Deviation (See in general FMB Reynolds, ‘The Implementation of Private Law Conventions in English Law’ in The Butterworth Lectures 1990-91 (1992).
A somewhat unusual doctrine operates in English Law in connection with deviation from the contract route. It applies both to voyage charterparties and to bills of lading, including bills of lading governed by the Hague Rules (Stag Line Ltd v Foscolo Mango & Co Ltd  AC 328, HL. There seems no reason why the doctrine should not apply under the Hague-Visby Rules also: even though these have by virtue of their enacting statute the ‘force of law’, they only apply to contracts, and if the contract is displaced they would be displaced also. However, under the Rotterdam rules, deviation does not displace defences, except in the case of deliberate fault). This route must first be ascertained from the terms of the contract and commercial and navigational practice. It need not be a direct geographical route; and especially in bill of lading contracts, which are likely to involve a ship which calls at several ports, it is unlikely to be so. Most contracts will in fact contain a clause purporting to permit what would otherwise be a deviation in very wide terms, frequently ending with wording such as ‘and all of the above shall be deemed to be part of the contract voyage’. Even with such wording these clauses, called ‘liberty clauses’, have been subjected to very extreme techniques of interpretation contra proferentem: they have been read to be consistent with the contract voyage and not as justifying a ship departing from the normal route (e.g. Glynn v Margetson & Co  AC 351, HL (starting off in wrong direction)).
The contract route being ascertained, a deliberate (a negligent departure from the route could be negligence in navigation, an excepted peril in most contracts) going off it is a deviation unless specially justified by the requirements of preserving the safety of the adventure, or saving life (A deviation to save the property of others, eg for the purpose of earning a salvage award, would not be justified unless permitted by the terms of the contract – which it often is, and is in most bill of lading contracts by virtue of the Hague Rules and their variants: see Article IV.4). The accepted results of this in English law at present appears to be that the shipowner, from the moment the deviation (for a different approach, that the contract is displaced from the beginning, see the authoritative but in this case extremely doubtful view of Scrutton, para 12.014, fn 31. (In an authoritative work of this sort, pregnant footnotes such as this can be of great forensic importance)) loses the benefit of all excepted perils and is remitted to the common law position of a common carrier, the common law exceptions also being displaced, with the result that he is liable unless he can establish that the loss would necessarily have occurred even though he had not deviated (Davis v Garrett (1830) Bing 716, 130 ER 1456. This in effect limits the operative perils to inherent vice).
The doctrine is said to be a reaction to the insurance position, under which when a ship went off the normal route it and its cargo were uninsured: hence the carrier, by doing this took on the insurer’s liability (he was even sometimes called an insurer, though this was to use the word in a different sense from its modern meaning) and became absolutely liable – but subject to the exception regarding events which would have occurred anyway. This could explain the displacement of excepted perils, but not of demurrage, general average and freight provisions.
The doctrine seems historically to be an amalgamation of rules of: proof regarding bailment (a bailee who stores goods in an unauthorized place is liable unless he can establish that the loss must have occurred anyway) (Lilley v Doubleday (1881) 7 QDB 510); interpretation (excepted perils are not intended to apply where the goods are subjected to risks totally different from those contemplated) (Gibaud v Great Eastern Rly Co  2 KB 426); insurance (an insurance policy is avoided if the ship deviates) (the leading case on this point, Joseph Thorley Ltd v Orchis SS Co Ltd  1 KB 660, invokes, however, the rules on displacement of insurance by unseaworthiness, which, since the seaworthiness duty applies at the time of sailing, applies from that point. This has been a source of confusion. In the case itself the difference does not appear to have mattered, since the ship took a wrong route immediately on leaving port); and discharge of contract by breach. This last explanation was accepted in the most authoritative case on the doctrine, the decision of the House of Lords in Hain SS Co Ltd v Tate & Lyle Ltd (see n 96). The main issue in this case was however whether a deviation could be waived: on the basis that it was a breach of condition it was held that it could. But the doctrine of discharge by breach has been further developed since that case, and it is now accepted that, where the innocent party chooses to terminate the contract, it comes to an end at that moment only and not ab initio, nor from the time of breach (see Photo Production Ltd v Securicor Transport Ltd  AC 827, HL). Since few cargo owners are aware of the deviation when it occurred (though it was so in Hain v Tate & Lyle: indeed this is what gave rise to the principal issue, which concerned waiver), this means that the discharge by breach explanation does not account for substantial features of the doctrine, in particular the displacement of the contract from the moment of deviation, which would in most cases have to be retrospective.
It is important to know in a general way how the bill comes to be issued. Plainly practices must vary enormously over the world. But one may start with the proposition that a shipper or his agent may make a reservation for goods to be carried. If so, an issue may arise as to whether this is contractual, i.e. binds the carrier to carry and (though not necessarily) the shipper to ship (see Scancarriers A/S v Aotearoa International Ltd (The Barranduna and the Tarago)  2 Lloyd’s Rep 419, PC). It may or may not be. Depending on the facts, the goods may then be sent down, and at the latest a contract of carriage is (normally) formed when they are loaded on the ship. After they have been loaded (and often after the ship has sailed) the carrier issues a bill of lading (the bill may have been drafted by the shipper and issued by the carrier after he has confirmed it; and there is usually more than one copy) to the shipper, usually signed by or on behalf of the master, who is an employee of the owner. The bill of lading being issued after the goods are loaded, it is in theory a record of a contract previously made; hence, if it is not in accord with prior contractual arrangements, they are not the bill in theory prevail (The Ardennes  1 KB 55). But for most practical purposes the bill of lading constitutes the contract; and it certainly does when in the hands of a third party transferee.