Doctrine of Impossibility in Contracts


This essay will consist in an attempt to analyse the doctrine of impossibility and its operation in relation to contracts. It will look closely at both the concept of initial impossibility arising from a common mistake on the part of both parties as to the state of things before the contract was agreed and the concept of subsequent impossibility and frustration. The latter deals with a situation whether the parties enter into agreement on terms both express and implied and then a supervening event renders the performance of that agreement radically different from that which was envisaged by both parties at the outset. These themes will be discussed in greater detail in the first section and will run throughout the work. The essay will examine the concept of objective and subjective impossibility, and the rules relating to discharge of contractual obligations and allocation of risk. It will look at the situation when either the subject matter or a thing essential for performance is destroyed or unavailable, either partially or completely. It will then look at how the death or supervening incapacity of a party will affect a personal contract. Towards the latter part of the essay, it will discuss the problems that arise when a method of performance becomes impossible or a particular source becomes unavailable. It will conclude by looking at the effect of delay and temporary impossibility on a contract.

The concept of impossibility

The concept of impossibility in contract law can be split into two distinct categories. There are the cases where the parties never actually reach a true agreement because they are mistaken as to some element of the contract before the contract is concluded and the cases where the contract becomes impossible to perform subsequent to the agreement having been reached. Generally speaking, in the first instance, the contract is void ab initio and in the second, an otherwise valid contract is brought to an end from the point when the impossibility arises. A basic example to illustrate the difference would be a contract for the sale of a car. If unknown to the parties, the car had blown up 5 minutes before the contract was signed the contract would be void ab initio, whereas if the car blew up 5 minutes after the contract was signed, the contract would be valid, but brought to an end by the fact that its subject matter no longer existed. Essentially the courts are implying into the contract a condition precedent that the subject matter exists and is capable of transfer. This concept of implied condition precedent has been regarded with considerable scepticism among commentators in light of the traditional common law view that the courts should neither make nor amend a bargain.

The main problem arises when dealing with the first type of impossibility. It is not always entirely clear how the courts will formulate the implied condition precedent. Smith and Thomas suggest three possibilities:

  1. A impliedly promised B that the thing existed.
  2. A impliedly promised B that he had taken reasonable care to ascertain that the thing existed.
  3. A and B proceeded on the common assumption, for which neither was more responsible than the other, that the thing existed and its existence was a condition precedent of the contract.[1]

Which of these options it will be, depends largely on the relative means of knowledge of the parties and whether one is relying on the other. This will be discussed at length through the course of the work.

It also may be that on proper construction of the contract either, or both of the parties have made absolute promises. In that event, the courts will not excuse non-performance for either type of impossibility. There are also cases where the contract has not become entirely physically or legally impossible, but an event has occurred which “strikes at the base of the contract so as to frustrate its purpose.”[2] This is commonly referred to as frustration and it operates as a form of subsequent impossibility.

Objective and Subjective Impossibility

The contract will have to be objectively impossible to perform before it is held to be void. The case of Thornborow v Whitacre (1705) 2 Ld Raym 1164 held that a party cannot escape liability on the grounds of impossibility purely relating to his individual ability or circumstances. Neither will he be discharged from his obligations simply because he finds the contract particularly difficult or onerous to perform:

“It is not hardship or inconvenience or material loss itself which calls the principle of frustration into play”[3]

Subsequent impossibility will similarly not excuse the parties from performance if it was brought about by the conduct of one of the parties. The case of Southern Foundries (1926) Ltd v Shirlaw [1940] AC 701 held at 717 per Lord Atkin:

“…conduct of either promisor or promisee which can be said to amount to himself of his own motion, bringing about the impossibility of performance is in itself a breach.”

Clearly, any impossibility that can be attributed to either party will be considered a breach of contract and the defaulting party will become liable in damages in the usual way. Where the impossibility brought about by one of the parties existed at the time of the contract he is likely to be held to have warranted possible performance of the contract and held to be in breach of that warranty.

As discussed above it is sometimes possible for the courts to hold that a party made an absolute promise and therefore accepted the risk of the fact that the contract might be impossible to perform. Whether a contract is considered to be absolute will be a matter of objective construction of the terms of the contract. If the contract is held to be absolute, the party will be held to his performance whether or not the impossibility is his fault or not. In the case Paradine v Jane (1647) Aleyn 26 a lessee was held liable to pay rent even though he had been evicted from the property by armed forces during the civil war. A lease is a type of contract that is commonly regarded as being objectively absolute without reference to the subjective intentions of the parties.

Overall the contract must be objectively impossible to perform, the subjective views of the parties as to their circumstances and their personal ability to perform the contract will not usually be taken into account. Similarly, if a party is active in bringing about the impossibility the contract will not be seen as objectively impossible, but as having been breached. Conversely, some contracts will be held to be objectively absolute and the subjective intentions of the parties in forming the contract and their level of fault in bringing about the impossibility of performance will not be relevant.

Destruction of the Subject Matter

In the case of Taylor v Caldwell (1863) 3 B & S 826 the claimants granted the defendants the use of a music hall and gardens for a series of music concerts. After the contract had been concluded, but before the concerts had begun the music hall was destroyed by fire and the concerts could no longer be held there. The claimants argued that the defendants were in breach of the contract for failing to provide the music hall and sought to recover £58, which they had spent on advertising the concerts. The courts however held that the contract had become impossible to perform and was therefore frustrated. Both parties were therefore released from their obligations under the contract. In coming to this conclusion Blackburn J referred to the dicta of Pothier[4] stating that:

“The debtor is freed from obligation when the ting has perished, neither by his act nor his neglect and before he is in default, unless by some stipulation he has taken on himself the risk of the particular misfortune which has occurred.”

He recognises that the civil law is not binding on English Courts, but states that it is a useful indicator of the principles on which the law is grounded.

Blackburn J also refers to a line of authority involving bailment. For example the case of Williams v Lloyd W.Jones 179 the claimant had delivered a horse to the defendant on the condition that it be returned on request. Without fault on the part of the defendant, the horse became sick and died and was therefore not able to be returned on the request of the claimant. It was held that bailee was discharged from his promise by the fact that the horse had died. Blackburn J stated that it was a settled principle of English law that in contracts for loans of chattels or bailments, if the promise of the bailee or borrower to return the goods becomes impossible because the goods have perished through no fault of his own, the bailee is excused from this promise. It is noted that in none of the cases relating to bailment was it expressly agreed that the destruction of the subject matter would release either party from their obligation, “the excuse is by law implied” [5]

This principle established in Taylor and subsequent cases[6] is now contained in section 7 of the Sale of Goods Act 1979

“Where there is an agreement to sell specific goods and subsequently the goods, without any fault on the part of the seller or buyer, perish before the risk passes to the buyer, the agreement is avoided.”

Partial Destruction of the Subject Matter

It is interesting to note that the contract in Taylor was for the use of ‘Surrey Music Hall and Gardens’. It was therefore only part of the subject matter that was destroyed by the fire; the gardens were still in tact. However, it was held that the destruction of the music hall rendered performance of the contract impossible. This implies that when part of the subject matter is destroyed the courts will investigate the purpose of the contract. If the part that is destroyed renders that purpose impossible the contract will be held to have been frustrated by its destruction.

Discharge and Rules Governing Risk

As discussed above a contract, which is the subject of a mistake made by both parties prior to its formation that makes performance impossible, will be void ab initio. This is not the case if the impossibility arises after the formation of the contract, i.e. the contract is frustrated. In that event, the contract is said to be discharged from the time when the frustrating event arose. The parties are discharged from any future performance without having to elect that that will be the case.[7] Where the core of the contract is the happening of some future event and that event is cancelled the time of frustration will be the time when the cancellation is announced. In the case of Krell v Henry [1903] 2 KB 740 the defendant hired a flat on Pall Mall to watch the coronation procession of Edward VII, though this purpose was not expressed in the contract. The procession was cancelled before the formation of the contract, but the announcement was not made until after the contract had been agreed.

If the contract is severable, it may be that only part of the contract is frustrated and the other parts remain in force. It seems that even when an entire contract of sale is held to be discharged because it has become impossible to deliver some of the goods, the buyer can ‘waive’ this and demand delivery of the rest of the goods. This was the case in HR & S Sainsbury Ltd v Street [1972] 3 All ER 1127. Supervening events may also make the suspend the contract without actually discharging it.[8] Temporary impossibility will be discussed in greater detail in a later section. Furthermore, illegality may frustrate a minor obligation without discharging the entire contract. The implications of things like this for a potential doctrine of partial frustration will also be discussed later.

The essay will now go on to look at who should bear the risk and hence the loss of a frustrating event. For a long time it was thought that the losses resulting from the frustration of a contract should lie where they fell. This led to the conclusion that any money paid before the frustrating event occurred was irrecoverable and conversely any money already due under the contract for services provided was enforceable. In the case of Chandler v Webster [1904] 1 KB 493, CA, a room was hired to view the coronation procession, the price being payable immediately. When the procession was cancelled, £100 had been paid on account. It was held that the contract was frustrated thereby releasing the parties from further performance, but leaving promises performable before the frustrating event still standing. On the other side a party who had only partially performed the contract could not recover anything for his services even when he had conferred a benefit on the other side.[9] It is possible however that a party who, after a frustrating event, takes reasonable steps to protect the other party’s interest will be entitled to recover remuneration for his expenditure on a restitutionary quantum meruit basis. This was the case in Société Franco-Tunisienne d’Armement v Sidermar SpA [1961] 2 QB 278[10].

Until 1942 it was also considered that there could be no recovery for total failure of consideration. This was on the basis that up until the point of frustration the party who had paid any money had the benefit of a executory contractual promise and that was consideration enough[11]. However, in the case of Fibrosa Spolka Akcyjna v Fairbairn Lawson Combe Barbour Ltd [1943] AC 32 the House of Lords held that a party could recover where there had been a total failure of consideration. This was an improvement on the Chandler position discussed above, but two principle defects in the law remained. The first was that the principle only applied when there was a total failure of consideration; where there was a partial failure the claimant could not recover anything.[12] The second defect was that the payee could not set off any expenditure that he had incurred in the performance of his side of the contract. These defects were rectified by section 1(2) of the Law Reform (Frustrated Contracts) Act 1943. The subsection states:

“All sums paid or payable to any party in pursuance of the contract before the time when the parties were so discharged (in this Act referred to as “the time of discharge”) shall, in the case of sums so paid, be recoverable from him as money received by him for the use of the party by whom the sums were paid, and, in the case of sums so payable, cease to be so payable:

Provided that, if the party to whom the sums were so paid or payable incurred expenses before the time of discharge in, or for the purpose of, the performance of the contract, the court may, if it considers it just to do so having regard to all the circumstances of the case, allow him to retain or, as the case may be, recover the whole or any part of the sums so paid or payable, not being an amount in excess of the expenses so incurred.”

This deals with the defects in the common law by stating that monies paid before the frustrating event are recoverable, sums payable prior to the time of discharge cease to be payable and the payee is entitled to set off expenses reasonably incurred in their performance of the contract. Goff and Jones note that whilst the Act does deal in outline with the deficiencies of the common law it does not completely resolve the issues.[13] For example, the Act does not say what principles the court ought to employ to decide how much the payee is entitled to set off. In the case Gamerco SA v ICM/Fair Warning Agency Ltd [1995] 1 WLR 1226 Garland J felt that the court’s task was to:

“…do justice in a situation which the parties had neither contemplated nor provided for, and to mitigate the possible harshness of allowing all loss to lie where it has fallen.”

Section 1(2) does permit the payee to recover or retain more than he has been paid up to the tine of frustration. I.e. for expenses incurred in expectation of future payment. They may be able to recover such expenditure under section 1(3), which takes effect when one party has conferred a valuable benefit on the other party (other than money) before the time of discharge. In that event, he will be able to recover a just sum, which shall not exceed the value of the benefit conferred.

Robert Goff J held in the case of BP v Hunt [1979] 1 WLR 783 that there were two steps to assessing a claim under section 1(3), the first was identifying and valuing the benefit conferred. Goff J held that usually the benefit would be the end product of any services. In some contracts the services were the end product themselves, for example, a contract for the transportation of goods. He held that if the end product is destroyed by the frustrating event then no benefit is conferred because the other party does not have the product either. This interpretation has been heavily criticised as failing to give effect to the intention of the Act.[14] This section of the Act was intended to mitigate against the harsh consequences of the common law rule of ‘entire obligations’. In the case of Appleby and Myers (1876) LR 2 CP 651 the claimants contracted to make machinery in the defendants factory and to maintain the machinery for two years. Payment was upon completion of the work. After part of the machinery had been erected, a fire destroyed the whole factory and all the machinery. The claimants could not recover anything, as they had not completed the work. Goff J’s interpretation of section 1(3) would lead to the same result. However, this interpretation has also been adopted in the Commonwealth[15]. It does appear to accord closely with the wording of section 1(3), which draws a distinction between the performance by on party and the benefit conferred on the other. This implies that the claimant must actually have received the benefit of any performance on the part of the defendant before the defendant can recover or retain any money.

The second step Goff J laid down was the measurement of a ‘just sum’. Contractual allocation of risk will of course be a factor. Goff J thought that it ought to be as much as is necessary to prevent the unjust enrichment of the other party. This approach was rejected by the Court of Appeal in the same case, who simply held that it was in the almost unrestricted discretion of the trial judge. In conclusion, the Act is sadly deficient in its guidance as to the allocation of risk and loss between the parties to a contract that has been discharged for frustration.

It is possible for the parties to allocate the risks contractually. This is one of the main reasons that the courts have kept a tight reign on the doctrine of frustration. Parties are expected to be able to foresee the possibility of dramatic price increases and the outbreak of labour disputes etc. Contracts therefore regularly include clauses which allocate the risk of such an unforeseen event occurring. One common example is a ‘force majeure clause’. In the case of Channel Island Ferries Ltd v Sealink UK Ltd [1988] 1 Lloyd’s Rep 323 the relevant clause stated:

“A party shall not be liable in the event of non-fulfilment of any obligation arising under this contract by reason of Act of God, disease, strikes, Lock-Outs, fire and any accident or incident of any nature beyond the control of the relevant party.”

The advantages of such clauses are that they provide a degree of certainty and the parties can agree to a wider range of circumstances than are currently available under the doctrine of frustration. For example, an unexpected increase in prices is not considered to be a frustrating event,[16] but it is common in a commercial contract to see a force majeur clause containing provision for ‘abnormal increase in prices and wages.’ It also allows the parties to determine their future relationship. The frustration doctrine discharges the contract regardless of the wishes of the parties, but they can provide for a continuing, adapted relationship if they so wish.

Unavailability of the Subject Matter

Where both parties are mistaken as to the availability of the subject matter at the time of the contract, this may be sufficiently fundamental to avoid the contract. The leading case on this issue is that of Courturier v Hastie (1856) 5 HLC 637 in which the parties entered into a contract for the sale of a cargo of corn, which was believed to be in transit from Salonica to England. Unknown to both the parties, the corn’s quality had deteriorated to such an extent that the master had sold it. The House of Lords held that the matter turned on the construction of the contract concluding that:

“The contract plainly imports that there was something which was to be sold at the time of the contract, and something to be purchased, no such thing existing,… judgment should be given for the defendants.”[17]

The exact legal basis for importing this term has been the subject of some debate among commentators and will be discussed briefly now. The draftsmen of section 6 of the Sale of Goods Act 1979 appear to have interpreted the decision as stating that a mistake as to the existence of the subject matter of the contract inevitably renders it void:

6 Goods which have perished

Where there is a contract for the sale of specific goods, and the goods without the knowledge of the seller have perished at the time when the contract is made, the contract is void.

The court in Couturier did not however mention the word mistake; they based their reasoning on the construction of the contract and the fact that there was a total failure of consideration on the part of the sellers.

Lord Denning applied a different interpretation in the case of Solle v Butcher [1950] 1 KB 671 at 691 in which he held that there was an implied condition precedent that the contract was capable of performance. He reasoned that in Couturier the parties had proceeded on the assumption that the goods were capable of being sold, when in fact they were no longer available for sale. Lord Denning’s interpretation does seem to give effect to the most likely intention of the parties. However, in the absence of a clear intention to release each other from the agreement if the subject matter is not available, it is not clear when Lord Denning is suggesting a term of this nature should be implied into the contract.

The third interpretation is that whether or not the contract will be void, depends on the its construction. This was the interpretation put on Couturier by the High Court of Australia in the case of McRae v Commonwealth Disposals Commission 84 C.L.R. 377. The defendants invited tenders for the purchase of an oil tanker described as lying on the Jourmand Reef off Papua, together with its contents, which were stated to be oil. The Claimants won the tender and spent a considerable amount of money modifying a vessel for the salvage work. In a bizarre turn of events it was later discovered that no such tanker had ever existed. The court held that:

“The only proper construction of the contract is that it included a promise by the commission that there was a tanker in the position specified.”

On that construction the Commission had assumed the risk of the tanker not existing. They distinguished Couturier, holding that this was not a case in which both parties had entered the contract on a common assumption. The Commission had assumed the existence of the tanker, but the buyers had only relied on their assertion. In policy terms there can be little doubt that the approach taken in McRae is a sound one and one which ought to be followed by the English courts, but its is somewhat difficult to reconcile with section 6 of the Sale of Goods Act. There is the possible argument that McRae does not fall under section 6 because the tanker had never existed and therefore could not have ‘perished’. This distinction does seem somewhat artificial and not within the intention of the court in McRae.

If the subject matter becomes unavailable after the contract has been concluded this may also render the contract frustrated for impossibility. For example in the case of Bank Line Ltd v Arthur Capel & Co [1919] AC 435 a charterparty was held to be frustrated when the ship was requisitioned and so unavailable to the charterer. Temporary unavailability may also suffice, but this will be discussed later.

Destruction or Unavailability of a Thing Essential for Performance

Lord Atkin in the case of Bell v Lever Brothers Ltd [[1932] A.C. 161, discussed the circumstances in which one might wish to imply a condition into the contract. He states that a condition derives its efficacy from the consent of the parties, express or implied. He supposes a possible term:

“Unless the facts are or are not of a particular nature, or unless an event has or has not happened, the contract is not to take effect.”

If there are express words in the contract such as ‘a foundation essential to the existence’, there need not be any further enquiry, but when there are no such words the court must investigate the circumstances of the agreement to see whether any such condition can be implied. Lord Atkin uses the example of the hire of a professional vocalist whose continued health would be essential to the performance of the contract.

The case of Krell v Henry [1903] 2 KB 740 has been discussed earlier. For present purposes it can be described in the following terms: The contract was for the hire of a room on Pall Mall to watch the coronation procession of Edward VII. The subject matter of the contract was the room and that was still in tact. However, the purpose of the contract was to watch the procession and without the procession the contract was not capable of full performance. Vaughn Williams LJ refers in his judgment to the case of Nickoll v Ashton [1901] 2 K.B, which is authority for the proposition:

“ English Law applies the principle not only to cases where performance of the contract becomes impossible by the cessation of existence of the thing which is the subject matter of the contract, but also to cases where the event which renders the contract incapable of performance is the cessation or non existence of an express condition the continued existence of which is necessary for the fulfilment of the contract, and essential to its performance.”

This concept was extended in Krell to include a situation in which that particular set of circumstances (the viewing of the coronation) was not expressly mentioned in the contract. The contract in Krell was, however a strange one; the room was only hired out by the day, not the night, and the purpose for the contract on both sides was the viewing of the coronation. It is clear that the particular set of circumstances must have been in the contemplation of the parties and one that they both realised was necessary for the full performance of the contract. There is some dispute surrounding the Krell case. Cheshire and Fifoot point out that the cancellation was probably not in the contemplation of the parties, but with regard to the proposition that the buyer should be discharged from his obligation to pay on cancellation:

“It is incompatible with the character of a hard bargainer to say that the owner of the room would have agreed to this proposal if it had been put to him during negotiations.”[18]

It is more likely that the owner would have told the hirer that that was a risk he would have to take. It seems somewhat unreasonable to import to the seller a state of mind which he may well not have been in had he thought about it. McElroy and Williams, on the other hand say that the contract was impliedly for the hire of “rooms to view the procession”, the fact that there was no procession therefore amounted to a complete failure of consideration on the part of the owner of the rooms, discharging the hirer from his obligation to pay.[19]

The circumstances in which Krell will apply are extremely limited. The set of circumstances, which the parties assume to be continuing, must be the common foundation of the contract. In the case of Herne Bay Steamboat Co v Hutton [1903] 2 KB 683 the claimant hired a ship from the defendant to watch the naval review and for a day’s cruise around the fleet. After the contract, the naval review was cancelled owing to the same illness of Edward VII, but the contract was held not to have been frustrated. This is thought to be because the hirer could still see the fleet and the boat had not been hired out by the owner for the specific purpose of seeing the Naval Review. This meant that seeing the Naval Review was not the common purpose of the contract and its cancellation was not therefore a frustrating event. Thus interpreted, Krell can be seen as a very narrow decision and as indeed been distinguished in more recent cases.[20]

The Death of a Person Essential to Performance

In the case of Galloway v Galloway (1914) 30 TLR 531 the defendant thought that his first wife had died and married the claimant. The defendant and claimant subsequently separated and entered into a deed of separation under which the defendant agreed to pay the claimant a weekly sum in maintenance. The defendant then discovered that his firs

Bill Carlson