The following essay discusses the history and rationale underpinning the case of Sumpter v Hedges, a classic case in the field of quasi-contract and thus now restitution. It is useful to those seeking to gain an understanding of this rather difficult area of law and the problems inherent in it.(This essay was previously published by myself on a different blog some while back, simply because I was a little bit proud of it. I republish it here by way of introduction to my new blog)
Quasi-Contract and Restitution: The Case of Sumpter v Hedges
The case of Sumpter v Hedges
1 is notable for the quasi-contractual principle it enunciated that 'the circumstances must be such as to give an option to the defendant to take or not to take the benefit of the work done' for an action for a quantum meruit to be successful. It is an exemplar of what has become known as the principle of free acceptance. Yet despite its adoption in subsequent case law as good authority, Sumpter and the principle of free acceptance raise considerable conceptual problems in terms of unjust enrichment. How well does the null hypothesis of a defendant being able to reject a benefit sit with the concepts of benefit, fault, or even under the rubric of unjust enrichment? How much authority should be given to a case decided within the ambit of quasi-contract and, given the historical forms of pleading, was it decided correctly then? This essay will address these questions through an analysis of Sumpter. It will reconsider the decision in Sumpter and determine whether or not it is correct in light of the law at the time and our understanding of it. It will argue the decision in Sumpter is correct but not on the basis of free acceptance; that the forms of pleading distorted the view of the rationale underpinning the restitutionary principles present within quasi-contract and that modern understanding of the law of restitution can serve as illustrative in so contending.
Quasi-Contract
The theory of quasi-contract was of an obligation created by an implied promise, whether it was for a sum of money or a service to be rendered. It was a wholly fictional promise which could arise out of a genuine one or the surrounding circumstances and framed as such within the historical forms of pleading which required a plaintiff to sue in indebitatus assumpsit for money had and received, and/or quantum meruit.
2 As Ibbetson outlines, an action in indebitatus assumpsit for money had and received 'came to be used as a contractual action in the early years of the sixteenth century.
3 It was an action analogous in purpose to the recovery of a debt, and based on 'a promise made subsequent to the creation of the debt and logically separate from it.
4' Its purpose was to simply allow a plaintiff to use a separate form of action from assumpsit, as being formally an action on the case, it could not be used 'if an action of debt was available in the same cause of action.'
5 It created no new liability nor was any further consideration required and, based as it was on the subsequent promise, 'did not in any way depend on the debt which lay at the root of it having grown out of any contract.'
6 It arose from the circumstances which rendered the retention of money by a defendant as being unjust.
7 During the course of its development in the sixteenth century, the courts came to require no proof of this promise and by the seventeenth it could be said that indebitatus assumpsit rested on an implied promise, though one that was understood as wholly fictional.
8 Quantum meruit, deriving from the Latin “what one has earned”, was just that. It was neither a debt, nor resting upon a fictional promise. Rather it was based on an actual agreement to pay a reasonable sum for services performed.
9 It was a request made by a defendant which could be directly inferred from circumstances, and one which could be said to rest on a true implied contract.
10 As Winfield notes, these cases were based on true contracts, in which case, it was quite right to assert they are firmly rooted in many of the doctrines and principles of contract law.
11
The developments of these two conceptually distinct, though related actions over the following centuries led to them both becoming incorporated into what we have come to know as quasi-contract. The first was that a plaintiff was allowed to recover on a quantum meruit basis in an indebitatus assumpsit action where an agreement to pay a fixed sum could not be ascertained.
12 The second were a minority of cases not easily justifiable as being decided on the basis of an implied contract between the parties.
13 As Barton notes, if a plaintiff who, contracting to do one thing does another, seeks a quantum meruit for the work done, it cannot arise from a promise implied from the circumstances, but rather 'upon the defendant's obligation, arising from the equity of the case rather than from his consent, to pay for the benefit which he has received.'
14 Ibbetson identifies two conclusions which subsequently arose, the first being that an action for a quantum meruit was one in indebitatus assumpsit, rather than that damages in the latter could be assessed on a quantum meruit where there was found to be a promise, but not one with a fixed sum.
15 The second, that all actions to recover a quantum meruit were based on an implied contract rather than most being decided on this basis while others on restitutionary grounds.
16 The union which occurred between both is obvious when considering that plaintiffs could sue on an indebitatus assumpsit count for a quantum meruit, and the courts now had to consider what would otherwise have been cases on a quantum meruit count, within the form of action of indebitatus assumpsit and which would become known as quasi-contract.
The conceptual issues arising from this fusion of both actions into quasi-contract led to a variety of questions concerning its principles and rationales which were never sufficiently addressed. Hunter and Carter note the issues concerning the very nature of an implied promise by asking whether it was genuine, though tacit, or one imposed by the circumstances?
17 These very questions harken back to the distinctions between the old forms of indebitatus assumpsit and quantum meruit discussed. Indeed, the development of the principles within quasi-contract and their application within Sumpter v Hedges illustrate these tensions.
Acceptance
The concept of free acceptance was defined by Birks as being that acceptance 'occurs where a recipient knows that a benefit is being offered to him non-gratuitously and where he, having the opportunity to reject, elects to accept,'
18 a view which is in accordance with the definition of a quasi-contract advance by Latham CJ in
Steele v Tardiani.
19 Acceptance is assessed subjectively and serves as a basis for unjust enrichment on subjective acceptance and it being unjust to accept a benefit with knowledge it was not gratuitous. It is the subjective aspect of free acceptance though, which has had a deleterious impact on plaintiffs' claims. Sumper is the classic example, where the English Court of Appeal upheld the trial judge's ruling the builder could only recover for the value of the loose materials but not the work unpaid for, nor on a quantum meruit for the work performed. Collins LJ stated the view of the court that where the work was done on land, and there is no option provided to the defendant of accepting or rejecting the benefit, the recourse must be had to other facts to 'ground the inference of a new contract.
20 There are two primary justifications as to why free acceptance ruled the quasi-contractual world. The first is it served to protect defendants from claims as to an objective but unrequested benefit being conferred upon them which they could be compelled to repay. This recalls the rationale of indebitatus assumpsit discussed above, that considered whether it was just in the circumstances for a plaintiff to be compensated. The second goes to the characterisation of quasi-contract as an agreement arising between the parties which, but for the failure or non-compliance with some other requirement, formed a binding promise between them. It was an implied contract by operation of law recognising a subjective agreement had arisen. It reflects the rationale underlying the old quantum meruit count: the ability to reject served as proof of acceptance in circumstances where often times explicit acceptance, which could have given rise to an actual contract, was not present.
While Sumpter correctly stated the principle of free acceptance, it is debatable as to whether the principle was correctly applied or rather the result of an historical anachronism borne out of pleading within the form of indebitatus assumpsit and the merging of principles which took place via a quantum meruit being claimed under it. This contention is supported by the judgements as to there being no circumstances which could have given rise to a new contract, and neither could any claim be brought concerning work done under the abandoned contract. Chitty LJ invoked the rule in
Cutter v Powell21 that a plaintiff could not sue on a quantum meruit for work performed if there is a subsisting contract which set out the renumeration on terms for that work. The rationale underlying the rule is that, if I have contracted with you to perform a service on particular terms, then I cannot claim you should pay me for that same work on different terms without showing an agreement to that effect. In Sumpter then, the plaintiff would have been required to show that, after the contract had been abandoned, a new agreement arose whereupon the defendant accepted the incomplete work on the terms of providing reasonable renumeration for it. Failing the null hypothesis of not being free to reject, such an agreement was impossible to infer. As the abandoned contract was on different terms, it could not show acceptance of the work even though logically the construction of whole buildings would, by necessity, require the partial construction of buildings along the way to completion.
22 In a sense, the decision asks us to adopt a kind of fiction: that under an entire contract, a building does not exist until it is finished, as though where one minute there is no building the next a whole one appears. Conceptually, this reflects the idea of quasi-contract expressed above which was based upon there being an actual agreement; not one arising by operation of law but rather simply one recognised by law. Therefore, within the rationale underpinning the old indebitatus assumpsit count it is necessary to ask whether it is, in reality, just and correct for a defendant to retain an objective benefit where a request has been made under a contract which would obviate any concerns as to injustice arising by forcing a defendant to pay for that benefit.
Benefit
Benefit, in quasi-contract may be either subjective or incontrovertible (regarding to money, being no reasonable person could deny a benefit).
23 The subjective test of benefit is one predicated on that of free acceptance, and has served as an efficient way for courts to circumnavigate the question of whether a defendant has been enriched. If something has been requested by a defendant it must be of benefit to them, it is not for the court to inquire. Recognising outside of money that courts only had recourse to a narrow, subjective test which could result in a variety of injustices, the case of
Craven-Ellis v Canons Ltd24 has come to be recognised as broadening the concept of incontrovertible benefit, the rationale underlying which went towards a consideration of the circumstances. Greer LJ held the defendant's argument was invalid where it sought to assert it should not be liable to pay for services which were a necessary expenditure; they 'would have had to get some other agent' to perform them.
25 The issue arose again in the case of
Greenwood v Bennett26 where Lord Denning MR based his view on the objective benefit the defendant received and the absence of risk assumed by the plaintiff. He considered whether a defendant had been benefited by repairs done to the plaintiff's stolen car (of which he was completely innocent) so as to enable the plaintiff to have a direct cause of action against the defendant. Central to his reasoning was that the plaintiff did not assume a risk he would be deprived of the fruits of his labour, and despite not being free to reject the work, a greater injustice would result if the plaintiff were not entitled to recover, especially if doing so resulted in no injustice being done to the defendant, in terms of burdening them with a prohibitive cost. Neither decision rests on acceptance, but a combination of benefit and fault, the former assessed on an objective basis. In this sense, the reasoning in
Craven-Ellis and of Lord Denning MR in
Greenwood, illustrate a different formulation of assessing unjust enrichment than in Sumpter, and is more consistent with the “Bargain-for-Benefit” test advanced by Burrows. Burrows, rejecting Birks' free acceptance as the basis for assessing benefit, proffering that evidence of actual receipt of an objective benefit–being something realisable in money–was required.
27 Once that objective benefit was found, any recourse to subjective devaluation could be rebutted if it was shown the defendant 'bargained-for' the objective benefit `thereby manifesting a positive desire, and willingness to pay for it'.
28 On this formulation, while the factual basis of the claim rests in a failed or terminated contract, the rationale is different to one requiring there be evidence of a new, separate agreement. Of note though, is absent request in
Craven-Ellis and
Greenwood there is no role to play for subjective devaluation or a bargain for the benefit. However, this is resolvable by recourse to the rationale underlying free acceptance. Free acceptance is not simply a ground for unjust enrichment, but a test. In Birks' formulation, it prevents recourse to subjective devaluation for, as noted above, if the defendant has accepted then it must be beneficial and it would be unjust for the defendant to deny enrichment. In other words, in looking towards whether something is a benefit a test is to see if it is subjectively valuable, but logically this cannot exclude objective value, only operate where it is in doubt. Returning to
Sumpter, a mistake was made in holding that absent subjective benefit, there is no benefit. This cannot marry in a system which recognises there are incontrovertible benefits, even if limited to money. It illustrates a failing of the application of the law in the context of asking whether it is unjust for a defendant to retain a benefit in the circumstances. However, in the context of a contractual agreement, such rationale makes perfect sense if one were to remove from consideration the whole context of the original building contract. Absent any law of restitution, all the court need ask is whether a contract arose between the parties that was unenforceable for not complying with some prescribed requirement. If so, then it is a valid quasi-contract. If not; if no such agreement had been made–no offer, no acceptance–then how could the defendant be liable? If though, we turn to an analysis based upon benefit, that request or acceptance goes to the gratuity of the work (or risk taken), then we are left with a benefit which cannot be subjectively devalued. Returning to
Craven-Ellis and
Greenwood, there were measurable objective or incontrovertible benefits. No recourse was necessary to free acceptance, as benefit was not in doubt and therefore did not require assessment. In terms of subjective devaluation, the fact it was incontrovertible addresses this also, in which case, subjective devaluation is not synonymous with free acceptance, but a requirement which free acceptance is also a test for. For example, I cannot subjectively devalue money, services I require, a car I wanted repaired, or building work I wanted performed. Outside the context of a contract, it would be contrary to the rationale of indebitatus assumpsit to contend, I did not want this building because I wanted more of it done. The subjective devaluation of free acceptance would only serve to contextualise the benefit received as to the renumeration due, but not to its existence as being owed.
Fault
There is an issue in
Sumpter as to what role did fault play? Collins LJ noted the case may have had a different result had the plaintiff 'merely broken his contract in some way so as not to give the defendant the right to treat him as having abandoned' it. It is difficult to follow why some lesser breach would have given the plaintiff a right to a quantum meruit in light of the preceding discussion. However it does raise an issue of fault and there are cases involving where a plaintiff not in breach has recovered on a quantum meruit, yet free acceptance could not apply. The first is
Planche v Colburn29 where Tindal CJ awarded the value of the work under a contract to author a manuscript, even though the publisher had not received any part of it, as it had been abandoned by the publisher and no new contract had arisen which allowed the defendant to require production of the work for a different use. The plaintiff was successful even though the defendant had received no benefit and thus, could neither accept nor reject. There was no new special contract, nor was the original one still in existence and open. The rationale advanced by Jones in terms of benefit was that 'The publisher had gained a benefit, for which he had to pay a reasonable sum, because he had bargained for and obtained the writing of the manuscript'.
30 Birks has explained the result in terms of 'limited acceptance' whereby sufficient acts of part performance would prevent a defendant from having recourse to subjective devaluation when, through the actions of the defendant the plaintiff is prevented from performing the whole agreement.
31 The second case of
De Barnardy v Harding32 concerned an agreement to advertise the sale of tickets for seating to view a funeral. After the plaintiff had incurred expenses, but before any tickets were sold, the defendant requested them not to sell any tickets as he would do so himself. Accordingly, the plaintiff forwarded all applicants for tickets to him. The plaintiff sent a bill to the defendant for all of the work done and expenses incurred. The defendant paid for the expenses but refused to pay for the work.
Planche v Colburn was stated as authority for the decision 'the defendant having refused to perform his part of the contract, the plaintiff was entitled to treat it as rescinded, and sue on a quantum meruit.' Though there was no question of fault relating to the original contract, having been abandoned by mutual consent, there was in relation to the new agreement insofar as it was implied the plaintiff should recover for the work done as it was at the request of the defendant the old agreement be abandoned.
The applicability of fault is clarified by reference to US jurisprudence. Hunter and Carter, assessing whether the contract should set a ceiling on the amount of recovery under a claim for restitution, discuss the US majority position it should not.
33 Inherent in these decisions is the notion that 'because the breaching party forfeits [their] rights under the contract by reason of the breach' they should not have recourse to the price in limiting a plaintiff's recovery. Why on principle which party breaches should determine the level of recovery is problematic. The parties have agreed to price in the context of their risk allocation. To ignore that would be to ignore the agreement regardless of fault in its breach. Furthermore, the point is often raised in regard to unprofitable agreements of a plaintiff suing in restitution in order to obtain a higher sum. Indeed, on Birks' 'limited acceptance' test there is no reason why, if a plaintiff can claim under part performance in restitution, they cannot claim for full performance. It is also difficult to see in
Sumpter how the defendant being in breach could have constituted acceptance of the work if that acceptance rested upon a new agreement and, putting to one side any contractual actions which might have accrued, the defendant would still have not been free to reject the work. An explanation must relate to consideration of unjust enrichment which countenances a concept of fault. Had the agreement in
Planche been found to be void or unenforceable contractually, and if the author refused to hand in the manuscript, then should he have still recovered on a quantum meruit? According to the court there was a benefit conferred and therefore why should it not, in principle, be remunerated? There was also no benefit to the defendant in
De Barnardy and no acceptance of the work. The judgements indicate the sole reason why a claim could not be brought under the contract was that it only provided for a share of profits and not work done. As no profits were realised and the work done under the contract went with it, where is the implied agreement? If however, these decisions are viewed through the prism of fault, there arises a coherent explanation. It is best akin to the equitable maxim that “he who seeks equity must do equity” and there is little reason why such principles should not be looked at in explaining decisions based on an action that, as the old indebitatus assumpsit action for money had and received was formulated, asks the question of whether it would be just for the defendant to retain the benefit. In
Sumpter, as it was not the defendant who abandoned the contract, and as they were not free to accept the work, was it unjust for them to retain it unpaid? No, they did not force the plaintiff to abandon the work, as neither would someone receiving a gratuitous benefit be accused of forcing the other to offer it as such. There is a case on the preceding discussion under acceptance and benefit that the plaintiff should still be entitled to some form of quantum meruit considering the justness of the situation, but that is not to say any restitution made should not be done without the plaintiff's abandonment in mind. That being said, it is important to note that quasi-contract does have within it a fault element. It would be troubling if any principle which has the term “unjust” within it did not.
Conclusion
The historical development of indebitatus assumpsit for money had and received and quantum meruit led to a theory of quasi-contract that contained cases explainable by references to implied, but actual agreements, and those which were properly resitutionary. The result are cases such as
Planche,
De Barnardy,
Craven-Ellis, and
Sumpter. The ill-ease with which these cases sit with one another is obvious and the cause is this melding. If this is not evidence of the application of inconsistent principles, it is difficult to conceive of what is. Examining those principles in
Sumpter have illustrated this inconsistency, and the explanation proffered here has sought to reconcile them with historical and doctrinal bases. If
Sumpter is correct then the other cases are wrong, yet courts have continued to apply them. The converse could be said but for the fact the principles stated in
Sumpter were entirely correct; it was their subsequent application which was flawed and it is on this basis it should be understood. There is some evidence that it is, with the courts moving towards an objective assessment of benefit which must, by necessity, put free acceptance into its proper theoretical context. Likewise, the movement of the law towards a remedial model will allow a greater role for the concept of fault in unjust enrichment separate from contract. That being said, this essay has shown that such developments will find an amiable basis within those decisions made in the era of quasi-contract and are therefore ones which will continue to serve it well in the future development of the law of restitution.
1[1898] 1 QB 673.
2D. Ibbetson, 'Implied Contracts and Restitution: History in the High Court of Australia' (1988) 8 Oxford J. Legal Stud. 313.
3Id at 314.
4Ibid.
5Ibid.
6Ibid.
7NSW Law Reform Commission, Community Law Reform Program: Eleventh Report - Restitution Of Benefits Conferred Under Mistake Of Law Report No 53 (1987) [
http://www.lawlink.nsw.gov.au/lrc.nsf/pages/R53CHP2].
8Ibbetson at 314.
9Ibid.
10Id at 315.
11P.H. Winfield, The Province of the Law of Tort (1931) at 159 cited in Ibid.
12Ibbetson at 316.
13Id at 316.
14J.L. Barton, 'Contract and Quantum Meruit: The Antecedents of Cutter v Powel', (1987) 8 Jnl Legal History 48 at 58.
15Ibbetson at 316.
16Ibid.
17H.O. Hunter and J.W. Carter, 'Quantum Meruit and Building Contracts – Part I' (1989) 2 JCL 95 at 96.
18P. Birks, An Introduction to the Law of Restitution, (1989) at 265; J.G. Tolhurst and J.W. Carter, 'Acceptance of Benefit as a Basis for Restitution' (2002) 18 JCL 52 at 53.
19(1946) 72 CLR 386 at 393.
20Id at [676].
21(1795) 101 ER 573.
22P. Birks, `In Defence of Free Acceptance' in A Burrows (ed), Essays on the Law of Restitution (1991) at 140; P. Birks, An Introduction to the Law of Restitution (1989) at 233.
23J.W. Carter, E. Peden & G. Tolhurst, Contract Law in Australia (5th ed, 2007) at [38-10]-[38-13].
24[1936] KB 403.
25Id at 412.
26 [1973] 1 Q.B. 195 (C.A.).
27A. Burrows, `Free Acceptance and the Law of Restitution' (1988) 104 LQR 576.
28A. Burrows, The Law of Restitution (1993) at 12.
29(131) 8 Bing 14, 131 ER 305, 5 Carr & P 58, 172 ER 876.
30G. Jones, 'Claims Arising Out of Contracts Which Do Not Materialize' (1979-1980) 18 U. W. Ontario L. Rev. 447 at 458.
31P. Birks, (1989) at 126-7; J.W. Carter, 'Discharged Contracts: Claims for Restitution' (1997) 11 JCL 130 at 140.
32(1853) 8 Ex 822, 155 ER 1586.
33H.O. Hunter & J.W. Carter, 'Quantum Meruit and Building Contracts–Part II' (1989) 2 JCL 189 at 191.
