Oxford University Comparative Law Forum

ISSN 1743-8713

Last updated: 18/05/16

Valid HTML 4.0!

Oxford University Comparative Law Forum

Unjust Enrichment Understood as Absence of Basis: a Critical Evaluation with Lessons from Canada

Chris D. L. Hunt*

(2009) Oxford U Comparative L Forum 6 at ouclf.iuscomp.org | How to cite this article

Table of contents


In England, the cause of action for unjust enrichment requires the claimant to satisfy four elements:

  1. That the defendant has been enriched;
  2. that the enrichment was at the claimant's expense;
  3. that the enrichment was unjust; and
  4. that there is no recognized defence.1

There is a vigorous debate among legal scholars about how best to characterize the third of these elements. Broadly speaking, there are two approaches to the 'unjust' question: The civilian approach, which, having regard to the claimant's purpose in transferring a benefit, enquires whether there is a valid legal basis justifying the defendant's retention of an enrichment; and the common law approach, which considers whether the claimant can demonstrate a positive ground for the reversal of the defendant's enrichment.

In his final Book, Professor Peter Birks drew on a series of ultra vires interest rate 'swaps' cases as evidence that English law had in fact reoriented from the traditional common law approach to a civilian model.2 While this view has not been endorsed by the majority of academics,3 and is inconsistent with subsequent judicial dicta,4 the question as to whether English law should in the future embrace a civilian model remains a "live" one.5

Much has been written about the merits and demerits of shifting to a civilian analysis. By and large, this valuable body of scholarship has evaluated the question from a theoretical perspective by considering the hypothetical application of a civilian model operating in England. This paper enters the debate from a different perspective: It analyses the jurisprudence in Canada, subsequent to that jurisdiction's shift from the traditional common law approach to a civilian one. Canada's experience with its civilian 'juristic reasons' model thus provides valuable and concrete lessons for jurists as to the wisdom of considering any similar shift in England.

This paper has three parts. In section I, I set out the competing models of unjust enrichment. In section II, I briefly review some of the academic literature to demonstrate that the supposed economy of Birks' 'absence of basis' civilian model is superficial in some respects and dangerous in others. This is an important point, since the primary argument in favor of a civilian approach is its alleged economy and elegance when compared with the traditional unjust factors. I then discuss three 'structural' problems inherent in absence of basis which contribute to confusion and inevitable error. In section III, I consider the law of unjust enrichment in Canada, demonstrating that the 'structural' problems identified in section II have manifested in the Canadian courts confusing the jurisprudence. The Canadian experience thus serves as a cautionary tale about the dangers a common law system faces when moving from the unjust factors approach to a civilian model. Throughout this paper I argue that English law should not abandon the current approach to unjust enrichment by adopting an absence of basis model. Such a fundamental reorientation should not be undertaken unless strong justifications are offered. In the short term at least, this shift must create uncertainty, increase litigation costs, contribute to errors and injustice6 and undermine the security of transactions.7 Any justification must of necessity point to serious inadequacies in the existing model or real advantages in the new model. Below, I argue that neither of these criteria is satisfied.

I. Two Legal Frameworks: Unjust Factors and Absence of Basis

As mentioned, there are, broadly speaking, two approaches to the 'unjust' question: the civilian approach, which, having regard to the claimant's purpose in transferring a benefit, enquires whether there is a valid legal basis justifying the defendant's retention of an enrichment; and the common law approach, which considers whether the claimant can demonstrate a positive ground for the reversal of the defendant's enrichment.

Birks explains the basic analytical structure of the civilian conception of 'unjust' as follows:

[Civilian models] begin from the proposition that every enrichment at another's expense either has an explanation known to the law or has not. Enrichments received with the purpose of discharging an obligation or, if without obligation, to achieve some other objective as for instance the making of a gift, the satisfaction of a condition, or the coming into being of a new contract. The outcomes succeeding, the enrichment is sufficiently explained. An enrichment which turns out to have no such explanation is inexplicable and cannot be retained. The recipient is not entitled to it.8

The common law approach, in contrast, requires the claimant to establish an 'unjust factor' which serves to invalidate the transfer and justify restitution. The main grounds of restitution under the unjust factors approach (the 'traditional approach') were identified in Moses v Macferlan by Lord Mansfield:

[This action] lies for money paid by mistake; or upon a consideration which happens to fail; or for money got through imposition (express or implied) or extortion; or oppression; or an undue advantage taken of the plaintiff's situation, contrary to laws made for the protection of persons under those circumstances.9

Through rigorous analysis of this and other early cases, Birks developed what became the 'orthodox view' of the English law of unjust enrichment, in which the notion of injustice derives from various unjust factors which he grouped under three heads: claimant-oriented, defendant-oriented and policy- oriented. Claimant-oriented grounds of restitution concern those unjust factors that, when viewed from the claimant's perspective, reveal either an absence of intention (theft), vitiated intention (mistake), or qualified intention (failure of consideration). Defendant oriented grounds of restitution concern those unjust factors that, when viewed from the defendant's perspective, demonstrate some form of exploitation such as undue influence, unconscionable conduct, or free acceptance. Policy-oriented grounds of restitution are grounded in the principle that public policy considers the benefit was transferred in an unjust circumstance and should be returned.10

In his final book, Birks rejected the unjust factors approach that he had pioneered and, drawing on principle and precedent, argued that an essentially civilian model is to be preferred.11 Birks' new 'absence of basis' approach is framed with elegant simplicity. He writes:

A single proposition covers every case: an enrichment at the expense of another is unjust when it is received without explanatory basis.12

If this sounds too good to be true, it is; absence of basis involves more than a single proposition. Birks explained absence of basis by grouping unjust enrichments into two categories: participatory and non-participatory. The former was then subdivided into voluntary and obligatory enrichments. Non-participatory enrichments (where the claimant has no control over the transfer) are prima facie unjust as there is no explanatory basis.13 Obligatory enrichments are transfers designed to discharge an obligation. If there is indeed a valid obligation which is discharged, then there is an explanatory basis. If not, then prima facie restitution should be awarded.14 Voluntary enrichments are transfers made without obligation but in order to achieve some outcome. If this purpose is achieved, there is an explanatory basis for the defendant's retention. If it is not achieved, then prima facie restitution should be awarded.15

Of course, stated thus absence of basis is conclusory: it tells us that if the basis upon which the enrichment was transferred fails (or does not exist in the first place), then retention of said benefit is prima facie unjust. Absence of basis does not, however, explain why the basis fails. For this we still need to consider the traditional unjust factors. Birks explained the interaction between the traditional approach and absence of basis by analogy to a pyramid which he argued provided a "limited reconciliation"16 between the two:

A pyramid can be constructed in which, at the base, the particular unjust factors such as mistake, pressure, and undue influence become reasons why, higher up, there is no basis for the defendant's acquisition, which is then the master reason why, higher up still, the enrichment is unjust and must be surrendered.17

Despite Birks' assertion that a "limited reconciliation" was possible, it must be emphasized that the "profound"18 differences between these two legal frameworks means a choice between them must be made. Indeed, Birks himself acknowledged that absence of basis could not be regarded as simply another unjust factor since it cuts across the entire list of factors and so, logically, they must stand under, not beside, absence of basis.19 Elsewhere Birks asserted that because unjust factors and absence of basis are "two entirely different methods" of determining whether enrichment is unjust, these approaches "cannot be mixed or merged". 20

As mentioned, unjust factors are reasons for reversing enrichments, whereas absence of basis inquires whether there are valid reasons for the recipient to retain them. Smith notes the significant implications that follow from the very different orientations of these approaches:

There is a tremendously important issue here. On it turns the whole orientation of the law of unjust enrichment. A list of reasons for reversing enrichments will look very different from a list of reasons for keeping them. 21

These differences reflect the very different historical and conceptual approaches of common and civilian legal systems more generally. McInnes observes:

Unjust factors traditionally were employed within the common law and, true to that system's basic orientation, inductively operate from the bottom. The specific reasons for restitution evolved, piecemeal and over a prolonged period, on the basis of practical experience [...]
Juristic reasons, in contrast, traditionally operated within civilian jurisdictions and, true to that system's basic orientation, deductively operate from the top down. At root, there is only ever one reason for restitution: a transfer occurred without legal basis.22

Thus, Edelman concludes that because the civilian and common law approaches are "antithetical inquiries", and because "methodology is important to clarity of the law", it is not sensible to have both approaches operating simultaneously.23 From a practical perspective, this must be right. We must choose between these two frameworks, for any mixing of them threatens chaos,24 as is evident from the mixed approach adopted in Canada.25

That a choice must be made follows also from the reality that absence of basis and the traditional approach do not always produce the same results. Before discussing four examples of differing results, it is worth first observing that the two approaches will typically lead to the same conclusion in the following general circumstances:26

  • Payments made under liability mistakes of fact;
  • Payments made under void contracts;
  • Benefits conferred in anticipation of a contract that does not materialise;
  • Payments consequent to an ultra vires demand by a public authority;
  • Compulsory discharge of another's liability.

Having said this, it must be emphasized that absence of basis is broader than the traditional unjust factors. Unless strong defences are crafted, its logic will arguably result in restitution in the following general circumstances that were foreclosed under the traditional approach:

  • Incidental benefits conferred on the defendant while the claimant pursues his own self interest;
  • officious unrequested improvers of another's property;
  • payments made under protest while doubting liability (which does not exist);
  • mispredictions regarding the future.27

Below, I briefly discuss each example.

(i) Incidental Benefits

Incidental benefits arise where the claimant, in pursuit of his own self interest, happens to enrich the defendant. Classic examples here are where C improves his own land which results in an increase in value to D's land, or where C heats his home and the escaping heat warms D's apartment located above. Normatively, there should be no question of restitution in these examples.28 This result is reached easily under the traditional approach as there is simply no reason, no unjust factor, permitting recovery of an enrichment conferred voluntarily.29 Logically, however, absence of basis would regard D's retention of these incidental benefits 'unjust' (as there is no legal ground justifying retention by D)30; and to avoid recovery it must either adopt artificial reasoning31 or specific defences32.

(ii) Unrequested Improvers of Another's Property

Whether a claimant who improves another's property without being requested to do so can claim a restitutionary remedy from the owner under the traditional approach is not straightforward. If the claimant is aware: (i) that the property improved is not his, (ii) that there has been no request for improvement by the owner, (iii) and that no obligation exists to improve it, then he cannot be said to be labouring under a mistake.33 Subject to what is said below, there is no unjust factor, and therefore no unjust enrichment. And properly so: an officious improver should not be able to recover in these circumstances as he is a risk- taker, hoping to be compensated but willing to take the chance he may not be.34 However, a claimant in this circumstance could still point to the unjust factor 'free acceptance' to ground a claim if: (i) the defendant had an opportunity to reject the benefit before it was provided, (ii) the defendant knew the claimant expected to be compensated and (iii) the defendant failed to avail himself of the opportunity to reject the improvement at the time it was performed.35 The defendant's conscious receipt of a benefit conferred by the claimant in these circumstances makes his retention of any enrichment unjust. Additionally, if the claimant was in fact mistaken (i.e. thinking he was in fact the owner of the property), then he could potentially obtain a restitutionary remedy so long as he is able to establish an enrichment to the defendant and the latter is not able to 'subjectively devalue' the service.36 The point, for our purposes, is to witness the nuances under the traditional approach; it draws distinctions depending on the knowledge and behaviour of both parties when fashioning a restitutionary response.37 The same cannot be said for a civilian model. Under that approach, each of these situations is treated the same, for in each case the determinative question is whether a valid legal ground exists to justify the defendant's retention of enrichments conferred at the claimant's expense. In each case, there is no legal ground. Retention of the benefit is prima facie unjust.

(iii) Doubting Payor of Non-Existent Debt

Under the traditional approach, a claimant who, in order to avoid a contractual penalty, makes a payment under protest in response to a defendant's bona fide request, but at the time doubts the money is actually due, cannot recover in the event that it turns out the money was not in fact due (unless compulsion can be established).38 Again, this is rationalized either on the basis that no unjust factor exists (doubt being different from mistake) or that the claimant is a risk-taker.39 In contrast, under the absence of basis approach restitution would prima facie be allowed as there is no legal basis for the defendant's retention of the enrichment.40

(iv) Mispredictions

Several authors have illustrated a further inconsistency. Under the traditional approach, the law draws a fundamental distinction between a mistake as to a present fact and a misprediction as to the future. The former, if causal of a transfer, is an unjust factor that can ground recovery. The latter, however, does not constitute a mistake and will not provide for recovery.41 Again, the distinction here is based on sound policy: "The claimant who has made the misprediction bears the risk as to future events occurring".42 Under absence of basis, this distinction is lost: Mispredictions as to future events would result in recovery.43

II. The Superficial Elegance of Absence of Basis

The most common argument in favour of a civilian model is based on its supposed elegance and economy when compared to the unjust factors approach.44 The civilian approach, it is said, has a "surgical simplicity", for the question is "always the same": What, if any, was the basis of the enrichment?45 The unjust factors approach, in contrast, is said to be untidy for it involves needless duplication.46

Several authors have demonstrated, quite convincingly, that these criticisms leveled against the traditional approach are unwarranted and that the alleged elegance of the civilian approach is superficial.

(i) Unnecessary Duplication

An oft-cited example of the inelegance of the traditional approach is that the analysis often requires unnecessary duplication. In the case of mistake, for instance, the court must first inquire whether the mistake is sufficiently fundamental to invalidate the contract (in contract law) and then inquire whether the mistake was causative (in unjust enrichment law) in order to obtain a restitutionary remedy. The civilian approach is said to be cleaner, for once mistake vitiates the basis of the transfer, restitution follows automatically.

Chen-Wishart rejects this argument as a false contrast. It looks to the form rather than the substance of each approach, because (i) under either model a claimant must first establish the vitiating factor in contract law before restitution can be awarded, and (ii) since mistake as a vitiating factor is narrower in contract law than it is as an unjust factor in unjust enrichment, "satisfaction of the former will automatically satisfy the later". As a result:

Nothing is saved by the apparently more economical German approach [which is exemplary of an absence of basis model] and nothing is added by the apparent duplication of the English approach.47

Furthermore, arguments based on the apparent elegance and purity of the German system, owing to its single rule, are, as Edelman observes, "highly artificial" for they simply push difficult questions into other areas of the law - areas, to be sure, where they do not belong:

...[I]f absence of basis... existed only from the time the contract no longer existed there would be a large gap in the [English] law... [For instance], in what part of private law should the rules governing undue influence be classified? As Birks said, a power to avoid a contract is 'not granted by contract and cannot be said to arise from [its] own effect...[it] arise[s] from the enrichment rendered unjust by the invalidity of the contract under which it was transferred'.48

This is an important point. Under the German Civil Code, unjust enrichment plays a very limited and defined role within a larger, fully integrated "master plan" for the whole of private law.49 Unjust enrichment is not so confined in England's common law system; it mingles in other areas, which means "adopting an absence of basis approach cannot be confined to the unjust enrichment taxonomy".50 Adjacent areas of law will need to accommodate the change, something many commentators fear threatens to be messy.51

Sheehan highlights another aspect of German law which further illustrates that it is not as elegant as may first appear, making the comparative inelegance of unjust factors ring hallow. Because the logic of legal grounds reasoning permits restitution of the doubting payor's discharge of a non-existent obligation,52 German law creates a specific defence (that the payor knew the money was not due) which operates, much like risk-taking in the common law, to deny recovery.53 Sheehan correctly observes that "the defence of knowledge brings in many of the same questions as the unjust factor of mistake under a different guise", which results, inevitably, in the unjust factors asserting themselves into the legal grounds analysis thereby undermining the model's supposed economy.54

(ii) Difficulties Removed or Submerged

Edelman's point, above, is relevant here also. A codified system, like Germany's, can remove difficult questions (about factors vitiating a contract, for instance) from the law of unjust enrichment by pushing them to other areas (such as contract law). Unjust enrichment law thus ostensibly becomes a simpler and smaller subject; but arguments as to its elegance are artificial because these difficult questions must still be answered, just in a different place.55

The Birksian absence of basis scheme, however, is more problematic than a codified civilian system, because these difficult questions, which must still be answered,56 are not pushed out of the law of unjust enrichment through sharp 'codal' distinctions; instead they lurk, submerged, at the base of his pyramid. Virgo puts it best when he writes:

...Birks' pyramidal analysis of the unjust factors...suggests something elegant when in fact it is unwieldy. The analogy with a pyramid is misplaced. The better analogy is with an ice-berg, where nine-tenths of the object is hidden below the surface.57

The point, for our purposes here, is to notice that (i) not only is absence of basis not in fact as elegant as it seems since it must still have regard to unjust factors, but (ii) that it may also be dangerous, since difficult questions are obscured rather than confronted openly. Burrows illustrates this latter point with reference to gifts, arguing that submerging complicated questions about the circumstances in which gifts should be invalid "...does not help a judge faced with deciding... whether (for instance) there should... be restitution of a gift which the donor wishes to recover because he has made it mistakenly believing that the recipient was poor or shared his political view".58

Virgo makes a similar point in the context of risk-taking and its relationship to the unjust factor of mistake, which both operate at the base of Birks' pyramid.59 This creates potential for confusion, since, by virtue of being submerged, rather than addressed head on, the risk-taking issue can be easily missed.60 The need to consider risk-taking and its relationship to mistake also undermines the ostensible elegance of absence of basis.61

(iii) Artificial Reasoning

The ostensible elegance of the Birksian scheme is further undermined by its resort in places to artificial modes of reasoning. Two examples illustrate the point.

As we saw above, Birks employed the concept of 'grudging' gift to explain denying restitution for incidental benefits conferred while the plaintiff pursues his own self interest. This rationale is problematic for three reasons. First, on its face, it does not accord either with the current law of gifts62 or with the lay person's understanding of a gift: In the escaping heat example, how can one be said to make a gift if they are not aware of the transfer, cannot control it, or do not desire it?63 Certainty of the law is not advanced by such artificial concepts. The second point follows from the first: Because the notion of gift is non-intuitive and problematically wide, applying the concept requires inelegant mental gymnastics. As Barker explains, with reference to the escaping heat example (where 'B' is the claimant):

[Birks'] explanation...constructs as a reason against liability ('B intended a gift') what is actually the absence of a positive reason in favour of it (B's consent to the provision of a benefit did not fail). Not only is this confusing, it seems a protracted way of getting the common law back to its own, very simple starting point.64

Third, from a taxonomical perspective, it is curious that gift is employed under both of Birks' primary categories of division (participatory and non-participatory enrichments).65 Such wide deployment of the concept must surely rob it of any precise legal scope and threatens to undermine its practical application. This is not a trivial quarrel, for Birks relies on this problematic notion of gift throughout his text to rein in the pyramid's expansionary momentum.

Several authors have also argued that Birks' approach to voidable and terminable contracts is "unnatural".66 Birks drew a distinction between contracts which are void (a nullity), and those that were merely voidable or terminable in the future.67 Regarding the latter, he asserted that once a contract was terminable or voidable by one party (due to a repudiatory breach or misrepresentation inducing the contract, respectively), then, at that point, the basis of any transferred benefit was absent, despite the reality that the claimant had not yet exercised his power to terminate or avoid the contract.68 Virgo highlights the artificiality of this reasoning when he writes: "But surely the contract remains effective until terminated or rescinded and constitutes a continuing basis for the defendant's receipt of the enrichment".69 An additional artificiality emerges: As Birks himself expressly recognized,70 this logic works only one way, meaning that it is only from the victim's standpoint that there is an absence of basis. That there can simultaneously be an absence of basis and a presence of basis, depending on the party viewing it, is highly artificial and further undermines the pyramid's ostensible elegance.71

(iv) Three 'Structural' Problems with Absence of Basis Borne out in Canada

It is my purpose here to outline in brief three 'structural' issues related to the absence of basis model. I will then consider, in section III of this paper, the extent to which these three issues have become problematic in the Canadian jurisprudence, rendering it confused and leading to inevitable error.

Role of Unjust Factors Problematic

As we saw above, the unjust factors lurk at the base of Birks' pyramid. The claimant's intention in transferring a benefit also remains relevant.72 Thus, we saw that in the case of voluntary enrichments, we must consider, first, what purpose the claimant sought to achieve and then, second, whether it was in fact achieved. If such purpose is achieved, there is an explanatory basis; if not, then prima facie unjust enrichment is established.73 However, in determining whether the claimant's purpose was achieved and in explaining why in the circumstances it may not have been, we will usually have to consider the traditional unjust factors.74 Above, I argued that this reasoning undermines the pyramid's ostensible elegance and may be dangerous because it obscures, rather than openly confronts, difficult issues.

An additional problem arises - one that has, as will be discussed below, posed particular trouble for the juristic reasons test in Canada: the interaction between unjust factors and absence of basis often confuses the analysis, for in some cases courts award restitution on the ground that there is no basis75, while in other cases the analysis suggests the reason for restitution is actually the presence of an unjust factor.76 In this latter group of cases, employing the phrase 'absence of juristic reason' is simply an expression of the conclusion reached under the traditional approach.77 As a result, at times it appears both approaches (civilian and traditional) are operating simultaneously. But, as discussed in section I, above, this cannot be. The two approaches are inconsistent and a choice between them must be made. Arguably, this confusion, resulting from simultaneous application of the two approaches, is an inevitable product of adopting in a common law system an absence of legal grounds analysis which still draws upon the explanatory power of unjust factors. Indeed, it is this confusion between the two approaches that prompts Tang to observe that absence of basis reasoning represents an "unsatisfactory half-way house between a civilian model of unjustified enrichment and the unjust factors approach".78

Nature of Basis Unclear

A second structural problem is that the nature of basis is unclear. It is obvious, of course, that the viability of any scheme based on absence of legal grounds depends, fundamentally, on clarity as to what those grounds consist of.79 Yet Birks was not able to draft a comprehensive list of bases. This represents a major structural deficiency in the pyramid's construction,80 which undermines its supposed elegance and taxonomical purity.

The lack of clarity as to basis is exacerbated by the non-intuitive, abstract manner of its expression.81 Burrows observes that the pyramid's intuitive appeal contrasts unfavourably with the traditional approach:

[A] virtue of the common law as opposed to civil law [is] that most of the unjust factors are readily explicable to non-lawyers...[T]he 'I did not mean to give' scheme of unjust factors, broken down into 'no consent', 'vitiated consent', and 'qualified consent', is beautifully simple and clear. In contrast, an absence of basis approach is not readily understandable... because it immediately entails a legalistic explanation in order to understand what is meant by 'basis' and, in turn, 'invalidity'.82

Uncertainty of basis is further undermined by the manner in which Birks employs the concept. This was illustrated above with reference to 'grudging gifts' and to the operation of basis in the context of voidable contracts. It is further evidenced by Birks' treatment of natural obligations,83 which he argued afforded the defendant a moral defence.84 Several problems can be identified with this. First, the scope and operation of a natural obligations defence is unclear. Second, it is unclear why, conceptually, natural obligations should be regarded as defences and not bases in their own right. Indeed, Dannemann observes that while classifying natural obligations as defences makes sense under the traditional approach, under absence of basis they are better understood as legal grounds justifying retention of enrichments.85 He draws a convincing analogy to the logic of gifts:

Just as the doctrine of consideration is directed against the enforcement of a gratuitous promise but not against its voluntary execution, so the rules on limitation of claims [for instance] are directed against the enforcement of obligations after a certain time has passed, not against the voluntary execution of such obligations. In both situations the unenforceable obligation provides the basis which allows the recipient to keep the enrichment.86

Of course, whether natural obligations are regarded as defences or bases has practical implications for the onus of proof. This is especially significant because (i) not only is the list of natural obligations unclear87 but, furthermore, (ii) there is no agreement on how to evaluate whether a putative natural obligation should operate to block a claim.88

Thus, if natural obligations are regarded as bases, which seems plausible, then a claimant is faced with the difficult task of identifying all possible natural obligations, and showing why none apply - a task that approaches the impossible given the two problems just identified. Conversely, if natural obligations are treated not as bases but as defences, the conceptual clarity as to what constitutes a basis under Birks' pyramid is further undermined, given the logical parallel between natural obligations (which Birks treats as not being bases)89 and gifts (which he says are bases).

As I discuss in section III, below, these aspects of uncertainty of basis have posed considerable difficulty for the juristic reasons test in Canada.

It is arguable that, on a more general level, this lack of clarity as to basis is inevitable in any common law system due to their shared history and mode of reasoning. As mentioned above, common law reasoning develops incrementally, from the bottom up, rather than deductively, from the top down. This is reflected in the logic of the unjust factors; but absence of basis calls for an analytical orientation unlike that applied to any other area of private law. It is thus not surprising the concept of juristic reason has proved difficult for common lawyers and judges to apply in Canada, and we should expect no easier transition if this foreign mode of reasoning is adopted in England.

Furthermore, the task of listing all possible bases is inherently more difficult, and less certain, in a common law as opposed to civilian system due to its non-codified nature and tendency to expand through analogy. The situation in Canada would seem to confirm this. There, the unjust factors approach (founded on English jurisprudence) has been supplanted by a civilian framework. Despite the shift, the common law foundation continues to assert itself in the new test, as is evident by the Supreme Court's insistence that new categories of juristic reasons (i.e. legal grounds or bases) may be established as further cases "refine" the "evolving" test.90 Thus, Birks' failure to identify every legal basis appears to be an inevitable feature of legal grounds reasoning operating in a common law jurisdiction.

Expansionary Tendencies Shifting Work to Defences

A third structural problem can be identified. The ostensible simplicity of absence of basis creates an expansionary momentum. As has been observed, with reference to the absence of legal grounds approach in Germany:

[T]he main dilemma with general provisions 'is that with them one tends to get more than one has bargained for: their wording will often cover more than it should'.91

Indeed, this observation is especially relevant to Birks' pyramid given its lack of clarity as to basis and artificial reasoning (discussed above). Two other features of the absence of basis militate in favour of inevitable expansion: Its civilian orientation that looks to reasons justifying retention, rather than reversal, of benefits received; and the abstract expression of absence of basis itself. According to Barker:

The potential price of this powerful combination of abstraction with a reversed justificatory orientation is the overexpansion of restitutionary rights.92

Birks was aware of this expansionary potential and attempted to control it with "vigorous defences"93 and through the creation of artificial bases (such as "grudging-gifts"). We have already explored the problems associated with the latter device. But what of Birks' defences? Before considering these, two preliminary points must be made. First, as the analysis is shifted toward defences they obviously assume an increasingly important gate-keeping function within the overall unjust enrichment action.94 As a result, their failure to operate effectively threatens to undermine commercial certainty.95 Second, it is important to note that crafting defences is no simple task. According to Krebs, it took German jurists and academics nearly fifty years to work out a patchwork of defences and even now there remains "considerable controversy" over their operation in certain contexts.96 The non-codal nature of England's common law system, coupled with its piecemeal bottom-up evolution, suggests that the task of developing a comprehensive list of defences would be a great deal more difficult here than in Germany. These observations alone are strong recommendations against shifting to an absence of basis approach.

Two broad categories of defences employed by Birks are of particular concern. The first is what he calls stultification:

If without good reason an action in unjust enrichment were to contradict a considered position elsewhere, the unreasoned contradiction would stultify the law. This danger is averted by raising a defence to a claim in unjust enrichment which would otherwise be available.97

Stultification is problematic for the same reasons Birks' basis is: (i) it is hopelessly vague and abstract, (ii) is much wider than the orthodox defences under the traditional approach,98 and (iii) it represents a non-exhaustive super-category without each example identified.99

As mentioned above, Birks also relied on the concept of natural obligations to provide defences. In addition to the problems highlighted above, the efficacy of natural obligations is further undermined by its apparent inconsistent interaction with the concept of stultification. For example, we saw above that Birks endorsed Lord Mansfield's list of natural obligations for use as defences, which includes contracts during infancy.100 This means that minors cannot recover enrichments transferred to adults during infancy even when the minor subsequently avoids the contract. However, it is difficult to see why this rule does not undermine (i.e. stultify) the protective policy animating the rule that minors may avoid contracts for non-necessities.101 It is here that the concept of stultification devolves into conceptual anarchy. On the one hand, it would seem that not allowing the minor to recover would stultify the protective policies allowing him to avoid the contract; yet, on the other hand, according to Birks:

The law of unjust enrichment would itself be stultified if the criteria which normally identify an unjust enrichment were allowed...to compel restitution of an enrichment which was not unjust. The claimant cannot say that the money was not due if, behind the technicalities of the law, there was still a moral obligation to pay.102

Thus, if we allow restitution we stultify the law of unjust enrichment; but if we provide a natural obligations defence to the adult, we stultify the policies protecting minors. Moreover, Birks' subsequent assertion that minors must make restitution to adults seems to further undermine these protective policies.103

III. Absence of Juristic Reason in Canada: a Cautionary Tale

In section I, I argued that a choice must be made between absence of basis and the traditional approach. One legal system cannot employ both approaches simultaneously without risking chaos, since they are 'antithetical inquires' that employ different reasoning and in some cases yield different results. In section II, we saw that the supposed economy and elegance of absence of basis is superficial in some respects and dangerous in others. We also considered three 'structural' problems inherent in the absence of basis model which I argued confuse the analysis and will likely cause judicial error.

In this section, I consider the law of unjust enrichment in Canada, which provides a cautionary tale about the dangers a common law system faces when moving from the traditional approach to a civilian model. While many of the problems plaguing the Canadian jurisprudence stem from its uniquely framed 'juristic reasons' test, and are due also to the inconsistent application of same, I will argue, with reference primarily to the three 'structural' issues identified above, that these difficulties are, on a more general level, likely to emerge whenever a civilian model is transplanted suddenly into a common law system.

(i) A Short History of a Troubled Jurisprudence

Historically, Canada followed the traditional unjust factors approach. Then, inexplicably, Dickson J., in the Supreme Court of Canada, expressed in Rathwell v Rathwell the third element of the test as requiring proof of "absence of any juristic reason for the enrichment".104 This phrase had not appeared previously in the body of unjust enrichment jurisprudence or in the academic literature.105 Initially, this reformulation was of little practical consequence; few jurists noticed the phrase, and fewer still believed it was intended to transform the law of unjust enrichment.106 However, two years later, in Pettkus v Becker, Dickson J. expressed the requisite elements of unjust enrichment as consisting of:

...[a]n enrichment, a corresponding deprivation and absence of juristic reason for the enrichment... The common law has never been willing to compensate a plaintiff on the sole basis that his actions have benefited another... It must, in addition, be evident that the retention of the benefit would be "unjust" in the circumstances.107

Again, commentators were not sure what to make of this. As McInnes observes, with reference to the above formulation:

...[It] appears to pull in both directions. The traditional common law theory is reflected in the Court's insistence that there must be...something unjust in the defendant's retention of the benefit. A civilian analysis is suggested by the reference to the absence of any juristic reason...108

Also, confusingly, a juxtaposition of Rathwell and Pettkus revealed contrasting lessons: The examples of juristic reasons in Rathwell ('contract or disposition of law') are reasons justifying retention of benefits, whereas in Pettkus, despite the reference to 'juristic reasons', liability was in fact imposed pursuant to the unjust factor free acceptance, which is a reason for reversing enrichment.109

Two lines of authority subsequently developed. The majority of cases continued to follow the unjust factors approach (either ignoring the concept of juristic reasons entirely or simply invoking the label after following the traditional approach), while others employed a truly civilian model.110 This confused blending of approaches occasionally developed even within the same case, as different members of the Supreme Court followed different conceptual paths - although without commentating that these approaches were not the same.111 A consequence of all this was not only doctrinal incoherence112, but also the emergence of a list of juristic reasons, developed ad hoc from the bottom up,113 without any organizing logic as would exist in a codified civilian system.

In an influential article published in 2000, Smith reviewed this troubled jurisprudence and concluded that it was "totally unclear as a matter of authority" whether Canadian law was "oriented towards reasons for keeping enrichments...or whether it remains, despite some terminological confusion, committed to the view that the plaintiff must show a reason for reversing... [enrichments]".114

The simultaneous application of both approaches created an intolerable situation. Certainty in the law was undermined as litigants were unable to anticipate with any degree of confidence which approach would be followed and therefore how to frame their arguments.115 We can speculate that transactional costs must have increased as complex legal memoranda would be required for hitherto straightforward cases. Counsel would be less inclined to settle before trial and more likely to advise disappointed litigants to appeal unfavourable judgments.

(ii) The Purported Clean Up

The Supreme Court addressed this inconsistency in Garland v Consumers' Gas Co.116 The unanimous Court reformulated the 'unjust' question as follows:

First, the plaintiff must show that no juristic reason from an established category exists to deny recovery. By closing the list of categories that the plaintiff must canvass in order to show an absence of juristic reason,...[the] objection that it require[s] proof of a negative is answered. The established categories that can constitute juristic reasons include a contract ..., a disposition of law ..., a donative intent ..., and other valid common law, equitable or statutory obligations...
If there is no juristic reason from an established category, then the plaintiff has made out a prima facie case...
The prima facie case is rebuttable, however, where the defendant can show there is another reason to deny recovery. [T]here is a de facto burden of proof placed on the defendant to show the reason why the enrichment should be retained. This stage of the analysis thus provides for a category of residual defence...
As part of the defendant's attempt to rebut, courts should have regard to two factors: the reasonable expectations of the parties, and public policy considerations. It may be that when these factors are considered, the court will find that a new category of juristic reason is established...117

(iii) The Continuing Mess

Despite the Court's effort to clarify the test, it continues to suffer from structural deficiencies similar to Birks' absence of basis model: (i) Uncertainty of basis; (ii) unclear role for unjust factors; and (iii) expansionary tendencies shifting work to vague defences. Furthermore, the reasoning employed is no more elegant than that followed under the traditional approach. These issues will be discussed in turn.

Juristic Reasons

Most fundamentally, notwithstanding a concerted effort to "construe...[the test] in a manner that is responsive to...criticism",118 the concept of juristic reason remains hopelessly vague. The Court's efforts to clarify it by closing the list of juristic reasons are unsatisfactory. First, this goal is obviously undermined as the list is non-exhaustive.119 Second, the list is uncertain as the examples given are themselves vague since they encompass both legal and non-legal grounds. This is confusing: While the label 'juristic' implies 'legal', the phrase 'donative intent' is not necessarily so, either ostensibly120 or in the manner it is applied.121 Thus, as in Birks' model, we have a lack of clarity as to basis, for the operational concept (juristic reason) is open-ended, encompass both legal and non- legal grounds, and is expressed at a high level of abstraction.

Arguably, the first of these problems is inevitable in any common law system due to its non-codal nature and incremental evolution over time. It is simply not possible to close the list of bases. Indeed, Garland recognizes this by making express provision for the recognition of new juristic reasons.122 There is potential for chaos here, since in Canada new juristic reasons are created by reference to public policy and the parties' reasonable expectations. The latter consideration, which threatens inappropriately to introduce an expectation interest in the law of unjust enrichment,123 is not part of Birks' model. But policy considerations are relevant to Birks' scheme at the defence stage.124 It is here that a common danger presents itself: New bases for retention will be snuck through the back door as trial judges justify retention of enrichments by reference to the unworkable defence of 'policy'.125 That Birks regards policy not as a basis but as a defence is of little practical significance - as a matter of precedent, once a 'policy' defence is recognized in one case, it will likely operate as a basis (in effect, if not in name) for retention of enrichments in future cases.126

The second problem identified above (uncertainty as to legal expression of basis) may also be inevitable. For instance, both schemes (Birks and Canada) utilize the concept of gift to control the expansionary momentum of legal grounds reasoning. But, if gift were applied in a technical legal sense it would be too narrow to effectively serve this purpose. As a result, both schemes take an elastic, non-legal approach to the concept, applying it, for instance, to deny recovery for transfers of services.127 Reliance on these unnatural concepts undermines the intuitive coherence and logical elegance of legal grounds reasoning and contributes to uncertainty as to what constitutes a valid basis.

A further problem has emerged, one foreseen by Virgo: Submerging the issue of risk-taking means it can easily be missed.128 This point is related to the concept of gift in both schemes. Because both models use gift in an unnaturally wide manner to control the momentum of legal grounds, the concept stands so close to risk-taking there is a danger judges will employ it instead129. This is problematic, since the concepts are not equivalent in scope.

Consider Peter v Beblow, in which restitution was awarded for the provision of domestic services by a wife to her partner over a twenty year period. The Supreme Court rejected a juristic reason defence of gift on the basis that, factually, "[t]he central element of a gift at law - intentional giving to another without expectation of remuneration - is simply not present".130 By focusing on gift, the Court erred. Had risk- taking been the issue, the Court could have denied recovery because although the wife provided services in the hope and expectation of sharing in the family property, she was willing to run the risk that a proprietary right might not arise and is therefore deemed to have acted officiously.131 Thus, we see that by inadvertently replacing risk- taking with the more stringent standard of donative intent, the Supreme Court has unwittingly expanded the scope of recovery.

This is a major criticism of the Canadian approach, but one that could also apply to Birks' model. If judges under Birks' scheme applied 'gift' in a legal sense (intention to transfer property without expectation of remuneration) in place of risk- taking, then, logically, this same expanded scope of recovery would result. Neither escaping heat nor the busker's services would constitute legal gifts. Thus, we see a common problem between these models: The concept of gift is required to rein in the legal grounds reasoning, but it is an unsatisfactory substitute for risk- taking. Applied properly (meaning legally) and it (gift) will not prevent recovery, but applied in the Birksian 'grudging' manner and it devolves into fictitious nonsense and undermines the pyramid's reasoning.

A solution would be to take a narrow legal view of gift and bring risk-taking clearly to the fore as a bar in its own right.132 But this begs the question: why chance the above confusion when risk- taking already operates in this suggested manner under the traditional approach?

An additional problem has emerged. As commentators predicted, the abstract, non-intuitive concept of juristic reason has been difficult to apply in practice. Consider the following two cases.

In Campbell v Campbell133 the Ontario Court of Appeal suggested that a 'moral obligation' was a juristic reason justifying the retention of benefits a mother received in the form of improvements to her farm performed by her sons. While Birks would not likely have endorsed this suggestion,134 judges can easily fall into this mode of reasoning when applying absence of basis, given its foreign orientation and abstract manner of expression. Indeed, even Birks, famous for taxonomical rigour, himself twice refers to natural obligations as "moral" when discussing their continued relevance under absence of basis.135 The potential for chaos is amplified when we recall that natural obligations could logically function as either defences or bases in their own right,136 and that there is no agreement among commentators as to the continued relevance of Lord Mansfield's proposed list of natural obligations nor is this list necessarily closed.

In Pacific National Investments v Victoria (City) 137 the Supreme Court and the British Columbia Court of Appeal took very different views as to what constituted a juristic reason. PNI, a commercial land developer, contracted with Victoria to develop the City's harbour. The City granted PNI residential zoning rights to the land; in exchange, PNI agreed it would spend $1.08 million to develop a portion of this land as park land for public use. In the next municipal election, officials hostile to the development were elected to City Council. They downgraded the zoning permission granted to PNI, rendering the development less profitable. However, PNI had already developed much of the park land. PNI initially commenced an action for breach of contract, which was dismissed on the basis that since City Council cannot bind a subsequent Council on matters of zoning, the initial contract with the first council was ultra vires its statutory capacity. This decision, removing the contract, cleared the way for the unjust enrichment claim.

PNI next sued in unjust enrichment to recover the money it expended on these improvements. The Court of Appeal denied recovery, on the basis that the legislative capacity to downgrade the zoning permission provided a juristic reason. Southin J.A., for a unanimous Court of Appeal, held:

[T]he juristic reason for what the [City] did... is that the Legislature had conferred upon it the power to do the act of downzoning. The bylaw is of the same force and effect as if it had been enacted by the Legislature itself and provides a complete answer to any and all claims arising out of it.138

The unanimous Supreme Court took a very different view. Citing the above passage, Binnie J. held:

With respect, this argument presupposes that the claim for unjust enrichment "arose" out of the down-zoning. However, the claim for unjust enrichment does not depend on the down-zoning. It depends on the fact that the City has obtained $1.08 million worth of extra works and improvements at [PNI's] expense to which, after ... [the ultra vires determination], the City has no legitimate entitlement. 139

It is difficult to imagine such confusion arising under the traditional approach, where the focus of PNI's claim would be either failure of consideration or mistake as to the City's legal capacity. But, as predicted by commentators, the focus on juristic reasons confuses the analysis and leads to inevitable error. Given the extreme cost of conducting an appeal to the Supreme Court, and the paucity of cases it grants leave to hear, such errors will typically go uncorrected, causing serious injustice to the parties.

Unclear Role for Unjust Factors

Although the Supreme Court did not sketch a pyramid, the unjust factors are nonetheless as relevant to Garland as to Birks' absence of basis; in most cases they must be considered to explain why a purported juristic reason does not exist.140 This has created confusion in the subsequent jurisprudence, which suggests a collective judicial cognitive dissonance. As judges charge ahead with the civilian orientation they continue to look backwards to derive explanatory power from the traditional approach. In Chambers' words: "The shift from unjust factors to absence of basis has been difficult. Courts seem unsure what to do with the pre-Garland authorities".141 As a result, the unjust factors inevitably bubble up to the surface, resulting in a hybrid of civilian principles lying atop, at a higher level of abstraction, what continues to be a common law claim. This has resulted in what McInnes describes as "the restitutionary equivalent of Frankenstein's monster" which is "doomed to disaster".142

Consider the following examples, which happen to be the only times the Garland approach has been considered by the Supreme Court. In PNI,143 (facts summarised above), the Supreme Court began on a strong footing. It affirmed the Garland approach,144 and then observed, correctly, that the City's previous appeal, which resulted in the contract being declared ultra vires, "knocked out of contention the juristic reason (the contractual provision)" justifying retention of the benefits the City received to the park land.145 Following the logic of absence of basis, this should have ended the matter (subject to defences). However, Binnie J., writing for the Court, then said this:

[T]he trial judge found that the ultra vires arrangements rested on a common mistake. Both the City and [PNI] assumed the City had the legal authority to make zoning commitments the City did not possess. The finding of common mistake is important to the appellant's claim...If there had been just the ultra vires transaction without the added element of common mistake, it would have been a different case and the outcome would not necessarily be the same.146

Mistake, of course, is an unjust factor. It has no role to play in the absence of juristic reasons inquiry - except to explain, in certain cases, although not in PNI itself, why there is no juristic reason. Once the basis of the transfer is removed, the presence of an unjust factor is irrelevant. The reference to mistake is, in this respect and at this stage in the analysis, inconsistent with the Garland framework. It is also inconsistent with the application of the Garland test in Garland itself.

Garland involved a claim in unjust enrichment to recover moneys paid to the defendant gas company. The company charged Garland late penalty fees for several years under authority of provincial legislation ('Ontario Energy Board Orders'). The OEB Orders were declared invalid for violating criminal interest rate provisions in the federal Criminal Code. By virtue of the constitutional doctrine of federal paramountcy, the provincial OEB Orders were deemed ultra vires.147 Regarding the issue of unjust enrichment, the Court in Garland held that since the juristic reason upon which these late penalties were collected was no longer legally valid, then, subject to defences, restitution should follow. As McInnes observes, although there were several potential unjust factors the Court could have used to justify restitution (mistake as to validity of the authority and failure of consideration, for instance148), the Court did not in any way rely on these.

Garland and PNI are analogous in that plaintiffs in each case expended money pursuant to an obligation that, by virtue of the doctrine of ultra vires, did not exist. In Garland, the ultra vires nature of the transaction was itself a complete justification for reversing the enrichment, but in PNI the Court emphasized the need for the additional unjust factor of mistake. The Court's different treatment of the same legal question (only months apart) reveals the confusion wrought by the application of the new approach to unjust enrichment. Indeed, this confusion may be inevitable, when an unfamiliar civilian framework lacking explanatory power is transposed upon a common law foundation.149

The Supreme Court's next consideration of juristic reasons is similarly troubling. Kingstreet Investments Ltd v New Brunswick150 involved an action in unjust enrichment for the recovery of taxes which were declared ultra vires the provincial legislature on constitutional grounds.151 Although applying the Garland framework to a case of undue taxes should be "extremely straightforward",152 Bastarache J., writing for the Court, dealt with the case under separate constitutional principles on the grounds that an "unjust enrichment framework is inappropriate".153 As Smith notes, this is an especially bizarre statement since "[m]any of the Supreme Court of Canada's most important and formative decisions on the law of unjust enrichment have involved claims against governments...to recover taxes".154

That aside, three other statements of Bastarache J. merit closer attention here, for they bear on the practical (un)workability of the Garland approach, notwithstanding the Court's refusal to actually apply it in Kingstreet. First, as part of his justification for why unjust enrichment is an inappropriate framework for ultra vires tax cases, Bastarache J. sets out the Garland framework in brief and then observes that "[it] is, as one can see, very complex".155 Second, Bastarache J., like Binnie J. in PNI, alludes to the continued relevance of an unjust factor. In obiter, while discussing the abolition of the distinction between mistake of fact and mistake of law, Bastarache J. states: "There can be no doubt that the ordinary principles of unjust enrichment continue to apply in cases of payments made pursuant to mistake of law".156 Moreover, when discussing the relevance of unjust enrichment principles for recovery of payments made under protest in the non-governmental context, Bastarache J. states: "[M]y view is that courts should insist on proof of compulsion in fact".157

Compulsion, of course, is an unjust factor. It has no place in the civilian framework set out in Garland.158 Likewise, the reference to 'ordinary principles' applying to mistake of law confusingly draws upon pre-Garland unjust factors. What are we to make of these statements? Some commentators have suggested the Court is having second thoughts about Garland,159 but this is unlikely - if the Court wanted to revise Garland, surely it would have said so expressly. The more likely, if unfortunate, explanation is that the Court is confused by its own test,160 finding it easier to avoid its application even at the cost of creating a novel, and doctrinally questionable, principle of constitutional restitution.161

Indeed, the Supreme Court is not the only appellate court having difficulty transitioning from the traditional approach to the civilian model. Consider, for example, the New Brunswick Court of Appeal's reasoning in the same Kingstreet Investments Ltd v New Brunswick162 case, before the matter was appealed further to the Supreme Court. The majority held the defences of change of position and passing on were available for claims based on mistake but not for those grounded on compulsion. But such distinctions, based on unjust factors, should not be relevant post Garland where absence of juristic reason is the basis for liability. As Chambers notes:

There may well be a good justification for treating those two [above] situations differently, but [post Garland] it will have to depend on something other than the unjust factor, such as the defendant's knowledge of the absence of basis for the enrichment.163

All of this confusion, I suggest, is inevitable when we consider the powerful combination of forces inherent in absence of basis: (i) a sudden 180 degree inversion of the legal paradigm, from reasons for reversing to reasons for retaining enrichments that (ii) expresses organisational concepts at high levels of abstraction, but (iii) still draws on the previous law for explanatory power, (iv) requires deductive 'top down' reasoning unfamiliar to the common lawyer, and (v) does not benefit from a carefully integrated, fully defined set of principles by virtue of its operation in a non-codal legal system. This surely is a recipe for inevitable error.

The Supreme Court's most recent restitutionary foray is perhaps the most perplexing example of its tendency to draw upon unjust factors. BMP Global Distribution Ltd v Bank of Nova Scotia164 is a case of complicated facts. What follows is a simplified version concentrating on the Court's reliance on the unjust factor mistake of fact. BMP entered an agreement with a rogue, Mr. Newman, whereby Newman was to pay $1,200,000 for the right to distribute BMP's crockery in the United States. Newman paid BMP $900,000, in the form of a cheque, as a first installment. The cheque turned out to be a forgery. It was drawn on an account held by an unrelated entity (First National) from an account First National held at the Royal Bank of Canada (RBC). RBC drew the money from First National's account and forwarded the funds to an account held by BMP at the Bank of Nova Scotia (BNS). Upon discovery of the forgery, RBC requested BNS to freeze the funds in BMP's account.

Deschamps J., writing for a unanimous Supreme Court, noted that the question as to whether BNS had a right (vis a vis BMP) to restrain these funds and return them to RBC turned, in the first instance, on whether RBC was entitled to recover the funds it transferred pursuant to the forged cheque.165 Deschamps J. answered this question in the affirmative: RBC's prima facie right derived from a "straightforward"166 application of the test for "mistake of fact" articulated by Goff J. (as he then was) in Barclays Bank Ltd v W.J. Simms & Cooke (Southern) Ltd.167

What is troubling here is not this conclusion (which is undoubtedly correct under the traditional approach, and would also be reached easily under an absence of basis/juristic reasons analysis), but rather the method. The unanimous Supreme Court examines the case using an unjust factor without once mentioning Garland or juristic reasons. Whether by accident or design, the unfortunate (albeit implicit) effect of BMP Global Distribution is to return Canada's jurisprudence to its troubled pre-Garland state, wherein both approaches to unjust enrichment apparently operate interchangeably. Compounding the misfortune, the Court failed to provide any guidance as to which model operates under which circumstances. Absent some guidance from the Court, how is the conscientious lawyer to proceed in future cases? As a matter of precedent, similar cases should be plead according to Deschamps J.'s unequivocal, clear and exclusive reliance on "the doctrine of mistake of fact".168 Logically, however, "[u]nless the Supreme Court... expressly indicates an intention to reverse or confine Garland,... it would be wrong to assume that the new juristic reason model of restitutionary relief covers anything less than the whole of its natural territory", which must include mistaken payments as the "paradigm [example] of unjust enrichment".169

Shifting Work to Problematic Defences

We have seen that legal grounds reasoning is expansionary by nature. It requires comprehensive defences to limit its overly broad scope of recovery. Common law systems have difficulty here: By virtue of their non-codal nature, they cannot draft fully integrated and comprehensive lists of specific defences at the front end. True to their basic orientation, they must develop incrementally over time. This poses obvious problems for adopting absence of basis: Expanded recovery appears, in the short term at least, inevitable.170

Ironically, even if a high court were to draw up a comprehensive list of defences at the front end (similar to that operating under the German Code), its very existence would go some way toward undermining the simplicity and purity of absence of basis, which is the major argument in favour of its adoption. How is an absence of basis model buttressed by a hotch potch list of fact specific defences any more elegant, pure or simple than the traditional approach with its eclectic list of unjust factors? The untidiness, it seems, is simply moved from the front door to the back.

To rein in the expansionary momentum both Garland171 and Birks,172 create a flexible super-category of defence based on 'policy' considerations. This is the only solution for a common law system, but it is very problematic.

First, and most fundamentally, 'policy' as articulated by Garland and Birks is hopelessly vague. Such an ambiguous concept, unless tightly defined, must be difficult to apply consistently. The ambiguity is exacerbated by history: Under the traditional approach, 'policy' favouring restitution is an unjust factor173, whereas under juristic reasons and absence of basis, policy favouring retention is a defence.174 This has created confusion in the Canadian courts; and may very well be inevitable, since, as we have seen, the unjust factors continue to inform legal grounds reasoning, meaning the traditional jurisprudence remains relevant.

In Garland the Supreme Court was clear that 'policy' favouring retention of benefits is a residual defence, not a ground of restitution.175 However, a mere three pages later, the Court, while considering said policy defence, looks also to policies in favour of restitution:

[T]he overriding public policy consideration in this case is the fact that the [late penalty fees] were collected in contravention of the Criminal Code. As a matter of public policy, a criminal should not be permitted to keep the proceeds of his crime...[A]llowing Consumers' Gas to retain the [penalty fees]...would let... [it] profit from a crime and benefit from its own wrongdoing.176

Similarly, in PNI, the Court, under stage two of Garland, again looks to policies supporting restitution:

[It is not]... good public policy to have municipalities making development commitments, then not only have them turn around and attack those commitments as illegal and beyond their own powers, but allow them to scoop a financial windfall at the expense of those who contracted with them in good faith. This is precisely the sort of unfairness that the doctrine of unjust enrichment is intended to address.177

This confusion is compounded in Kingstreet. As part of his case for why unjust enrichment is an inappropriate framework for obtaining recovery of ultra vires taxes, Bastarache J., for the Court, states:

The Garland approach is, as one can see, very complex; it requires that courts look only to proper policy considerations. By proper policy considerations, I mean only those that have traditionally informed the development of restitutionary law. [emphasis original] 178

Exacerbating the ambiguity, Bastarache J. gives two examples of public policy to illustrate the point. The first concerns a policy defence raised by La Forest J. in Air Canada v British Columbia,179 to the effect that protection of the treasury should immunize the government from claims for restitution of unconstitutional taxes. The second example concerns the policy identified in Garland - that criminals should not be permitted to profit from their crimes.180 According to Bastarache J., the former is not an appropriate consideration under stage two of Garland, but the latter is. But this cannot be correct. As we saw, under the second stage of the Garland test, policies serve as residual defences, not as grounds of restitution. Consequently, whether one ultimately accepts La Forest J.'s proposed defence in Air Canada, it cannot be doubted that it is precisely the type of consideration relevant under the second stage of the juristic reasons analysis. Conversely, as we saw above, the principle that criminals should not be permitted to profit from their crimes is a policy in favour of recovery - it most emphatically does not fall to be considered as a residual defence.

The above survey demonstrates the danger of incorporating a vague policy defence in an absence of basis scheme that must of necessity continue to draw on the traditional unjust factors for explanatory power. It seems inevitable that in such a model judges will forget that 'policy' now plays the opposite role it once did. The result is not only doctrinal confusion and uncertainty in the law, but also unfairness to litigants. The latter point has a very practical implication: Knowing which policies are relevant (i.e. those favouring recovery versus retention of enrichments) influences the manner in which the parties frame their pleadings and discharge their respective burdens.181 Poorly decided cases frustrate focused legal argument in subsequent cases and contribute to further uncertainty in the law. A vicious cycle ensues.

A final problem with this vague 'policy' defence threatens to be the most significant: In applying stage two of Garland the Supreme Court has now taken 'policy' to mean 'fairness'. Consider Bastarache J.'s justification in Kingstreet for rejecting La Forest J.'s proposed defence in Air Canada:

Considerations related to preserving the public purse do not properly fall within the second branch of the juristic reasons analysis, which is more concerned with broad principles of fairness.182

Not only are such statements hopelessly vague, but they threaten to undermine the modern respectability of unjust enrichment that Birks spent his academic career striving to achieve. If such reasoning were adopted in England, the following dicta of Lord Goff may soon become more relevant to students of legal history than to scholars of unjust enrichment:

The recovery of money in restitution is not, as a general rule, a matter of discretion for the court. A claim to recover money at common law is made out as a matter of right; and even though the underlying principle of recovery is the principle of unjust enrichment, nevertheless, where recovery is denied, it is denied on the basis of legal principle.183

IV. Summary and Conclusion

In section I, I contrasted the traditional approach to unjust enrichment with Birks' absence of basis model. I argued that a choice must be made between these models, for they are fundamentally different orientations that employ different reasoning and in some cases yield different results. In section II, I examined the corpus of academic literature to demonstrate that the supposed economy and elegance of absence of basis is superficial in some respects and dangerous in others. I argued that absence of basis contains three 'structural' problems (unclear role for unjust factors; nature of basis unclear; and expansionary tendencies shifting work to defences) which contribute to confusion and inevitable error. In section III, I considered the law of unjust enrichment in Canada, illustrating that it provides a cautionary lesson about the dangers a common law system faces when moving from the traditional approach to a civilian model. I demonstrated, in particular, that the 'structural' problems inherent in absence of basis have manifested in the Canadian jurisprudence, causing uncertainty and confusion.

Clearly, the Canadian position is unsatisfactory. After dramatically reorienting the law toward juristic reasons in Garland, the Supreme Court's subsequent forays have made perplexing, and inconsistent, references to the traditional approach. Consequently, it is not clear whether, or to what extent, the pre-Garland jurisprudence remains relevant under the juristic reasons framework. We should expect no easier transition in England if absence of basis were adopted in that jurisdiction.

In light of the above, we must conclude that adopting absence of basis in England would be a grave mistake. Arguments in favour of its adoption are not persuasive, and criticisms of the traditional approach are not compelling. McInnes suggests that if courts choose to reorient the law, "they are obliged to devise a system that is at least as good as ... [the one] they reject".184 I would take this further. Change for change's sake cannot be countenanced, because any transition will undermine certainty and destabilise the security of commercial transactions. Errors are inevitable as courts grapple with the new approach. In a common law system, details will have to be "hammered out on the anvil of concrete cases".185 Such fundamental reorientation of the law is not justified unless the move promises significant advantages. Absence of basis does not. It is an inferior model. Moreover, even if we ignore its deficiencies, and assume it is as good as the traditional approach, the fundamental reality is that absence of basis is "not about helping judges to do things they want to do but need theoretical help with. It is simply about applying [arguably] better labels to well understood results".186 This cannot justify reorienting the law. Thus, the strongest argument against absence of basis is a practical one: Adopting it is "the juridical equivalent of starting with a greenfield site, [requiring] one part of a legal system [to] turn its back on its historical and conceptual roots..."187 while inevitably suffering the growing pains such a shift necessarily entails. Such a move must not be undertaken without good reason. It is also a recipe for disaster, as the Canadian experience confirms.


* B.A (Mount Allison University), LL.B (University of Manitoba), LL.M (Cambridge), Ph.D Candidate in law (Gonville & Caius College, Cambridge). I would like to thank Graham Virgo for his insights on an earlier draft of this paper. I would also like to thank Gerhard Dannemann and an anonymous reviewer for their helpful comments.

1 Banque Financiere de la Cite v Parc (Battersea) Ltd [1999] 1 AC 221 at 227 per Lord Steyn and at 234 per Lord Hoffmann.

2 Birks, Unjust Enrichment (2d.), (Oxford: Oxford University Press, 2005), Ch. 5.

3 Most leading academics do not accept that England has already moved to a civilian framework (see A. Burrows, 'Absence of Basis: The New Birksian Scheme', in Mapping the Law: Essays in Memory of Peter Birks, Andrew Burrows and Lord Rodger of Earlsferry (eds.), (Oxford: Oxford University Press, 2006), Ch. 2; G. Virgo, The Principles of the Law of Restitution (2d) (Oxford: Oxford University Press, 2006) 128; R. Stevens, 'The New Birksian approach to Unjust Enrichment' (2004) 12 RLR 260 at 271; J. Edelman, 'The Meaning of 'Unjust' in the English Law of Unjust Enrichment', (2006) European Rev Private L, 326.

4 See Deutsche Morgan Grenfell v Inland Revenue Commissioners [2006] UKHL 49, in which each judge in the House approached the unjust enrichment action by looking for a positive ground of restitution, and none adopted a civilian analysis.

5 G. Virgo, 'Demolishing the Pyramid-the Presence of Basis and Risk-Taking in the Law of Unjust Enrichment', (2008) unpublished draft paper, p.2, noting that in Deutsche Morgan Grenfell v Inland Revenue Commissioners Lord Walker (para. 158) did recognise that it might be appropriate to consider adopting Birks's civilian absence of basis approach in the United Kingdom in the future.

6 M. McInnes, 'Making Sense of Juristic Reasons: Unjust Enrichment After Garland v. Consumers' Gas', (2004) 52 Alta. L Rev 400.

7 Virgo, Principles (n 3) 129.

8 Birks (n 2) 103. The approach codified in German Law provides:
"He who obtains something through someone else's performance or in another way at his expense without legal cause, is obligated to make restitution to the other. This obligation arises also where the legal cause later disappears or where the result contemplated by the legal transaction is not attained.' See Section 812(1) of the German Civil Code (Bürgerliches Gesetzbuch, abbreviated 'BGB'), translated by T. Krebs, Restitution at the Crossroads: A Comparative Study (Cavendish: London, 2001) Ch. 11. An English Translation of the BGB is available at: http://www.gesetze-im-internet.de/englisch_bgb/englisch_bgb.html#BGBengl_000P812 An English language compilation of BGB provisions specifically related to restitution is available at: http://www.iuscomp.org/gla/statutes/BGBrest.htm It should be noted that civilian systems are not monolithic; differences exist among them.

9 Moses v Macferlan (1760) 2 Burr 1005; 97 ER 676.

10 For a summary of Birks' three categories of unjust factors, see Virgo, Principles, (n 3) 119-124.

11 Birks (n 2) xii dramatically writes: "[A]lmost everything of mine now needs calling back for burning".

12 Birks (n 2) 116.

13 Birks (n 2) 129. These are primarily concerned with theft or where the claimant is ignorant of the defendant being enriched at the claimant's expense.

14 Birks (n2) 129. This category covers cases of payments made under liability mistakes, money paid pursuant to void contracts, and of taxes demanded by public authorities acting ultra vires. Other circumstances invalidating the obligation under this head include undue influence, illegitimate pressure, and compulsion: see p.136.

15 Birks (n 2) 129. This category covers payments made in anticipation of contracts, and for the purpose of creating a trust or making a gift.

16 Birks (n 2) 116.

17 Birks (n 2) 116.

18 McInnes characterizes this choice as "the most profound shift imaginable within the law of unjust enrichment": 'Taxonomic Lesson for the Supreme Court of Canada', in Structure and Justification in Private Law, C. Rickett and R. Grantham, (eds), (Oxford: Hart Publishing, 2008), 92.

19 Birks (n 2) 114.

20 Birks (n 2) 44.

21 L. Smith, 'The Mystery of 'Juristic Reason'' (2000) 12 SCLR (2d), 214. See also T. Krebs, 'In Defence of Unjust Factors', in Unjustified Enrichment: Key Issues in Comparative Perspective, D. Johnston and R. Zimmermann, (eds.), (Cambridge: Cambridge University Press, 2002), 97.

22 M. McInnes, 'Making Sense', (n 6) 402. See Krebs, 'Unjust Factors', (n 21) 99 for a colouful explication of the same point.

23 J. Edelman, 'Meaning of 'Unjust', (n 3) 326. For other expressions of the incompatibility of the two approaches, see Burrows, 'Basis', (n 3); G. Dannemann ,'Unjust Enrichment as Absence of Basis: Can English Law Cope?'; and S. Meier, 'No Basis: a Comparative View', in Mapping the Law: Essays in Memory of Peter Birks, Andrew Burrows and Lord Rodger of Earlsferry (eds.), (Oxford: Oxford University Press, 2006).

24 Burrows, 'Basis' (n 3) 33.

25 Smith observes that much of the confusion in Canadian unjust enrichment law stems from the fact that it blended a civilian and an unjust factors approach: Smith, 'Mystery' (n 21) 243. I discuss this in detail in Section III.

26 Burrows, 'Basis' (n 3) 36-44. See also: Birks (n 2) 44; Stevens, 'Birksian Approach' (n 3) 272.
Virgo takes issues with the fifth example, above, arguing that in such legal compulsion cases the "restitutionary claim against the defendant cannot be analysed in terms of absence of basis, because the claimant has discharged a real legal liability owed to the defendant's creditor": G. Virgo, 'Pyramid' (n 5) 8. Birks justified restitution under absence of basis by characterising the enrichment to the defendant in such circumstances as a by-benefit: "The discharged debtor can offer no explanation at all as to why he should retain his enrichment... [T]here was no reason at all why [the defendant] should reap the by-benefit": Birks (n 2) 159. The difference here between Birks and Virgo lies in whether the explanatory basis must exist between the claimant and defendant (Birks) or if it is sufficient if it exists as a basis for payment between a claimant and a third party (paid on behalf of the defendant). In my view, absence of basis can justify the defendant's retention of the benefit on the simple grounds that the claimant's purpose in transferring the funds was fulfilled. Regardless of whether the Birks or the Virgo position is to be preferred on this question, the debate itself reveals a fundamental flaw in the absence of basis model, namely the lack of clarity as to what constitutes a basis. This is discussed fully in Section II.

27 This is not an exhaustive list. For instance, Dannemann asserts that absence of basis has difficulty preventing restitution for a variety of non-contractual enrichments (such as lending one's car): Dannemann, 'English Law' (n 23) 364-67; and Virgo argues that Birks' analysis suggests it would allow recovery for partial failure of consideration: Virgo, 'Pyramid' (n 5) 4-8.

28 Indeed, Edelman observes that no legal system applying either a civilian or the traditional approach awards recovery in these circumstances: Edelman, 'Meaning of Unjust' (n 3) 320.

29 Edelman, 'Meaning of Unjust' (n 3) 321. See also the dicta of Lord Halsbury in Ruabon Steamship Co v The London Assurance [1900] AC 6, 10, rejecting in principle that a man who cut down trees on his own land could seek contribution because this act improved the latter's view.

30 Burrows, 'Basis' (n 3) 44; Edelman, 'Meaning of Unjust' (n 3) 319-21; L. Smith, 'Demystifying Juristic Reasons', (2007) 45 Can Bus LJ 291. It is noteworthy that even Baloch, one of the few commentators supportive of absence of basis, acknowledges the obstacle by-benefits posed to this theory: T. Baloch, 'The Unjust Enrichment Pyramid', (2007) 123 LQR 653.

31 Birks opted for the artificial reasoning approach, identifying a basis for retention by characterizing by-benefits as "grudging gift(s)": Birks (n 2) 159. Meier rejects Birks' explanation here as "not entirely convincing", but then offers an equally exotic explanation for the heating example: "The best explanation seems to be that D's enrichment is not at C's expense. Causation alone is not enough...Heat which escapes a flat is not attributed to the owner of the flat or to the person generating it; it belongs to no one", Meier, 'No Basis' (n 23) 356. However, even if Meier's reasoning is accepted, it would not account for the incidental benefit conferred on D in the land improvement example, for surely C's labour belongs to him. Sheehan has argued that in neither the heating nor the improvement to the neighbour's land examples would absence of basis necessarily result in prima facie unjust enrichment. According to Sheehan, in neither case is there a "putative purpose" to the transfer which then fails, and therefore no prima facie right to restitution arises: D. Sheehan, 'Unjust Factors or Restitution of Transfers Sine Causa', (2008) Oxford U Comparative L Forum 1 at ouclf.iuscomp.org, text after note 128. In my view, Sheehan's argument fails to take into account that, unless Birks' notion of grudging gift is invoked, unjust enrichment may still arise under the non-participatory category of enrichments. The lack of agreement here as to whether these examples lead to prima facie recovery under a civilian model highlights the uncertainty of its application.

32 Meier notes that "[o]ne of the problems which German authors faced after the Code introduced a general enrichment action (similar to absence of basis) was to explain why not every incidental benefit can be recovered as being without legal ground": Meier, 'No Basis' (n 23) 355-6. Krebs notes that it took German jurists "almost a half century" to work out a delicate balance of defences and distinctions to limit the inherent expansionary nature of the German 'without legal ground' provision: Krebs, 'Unjust Factors' (n 8) 80. In Canada, the expansionist tendencies of 'no juristic reason' have been controlled through the ambiguous defenses of 'policy' and 'reasonable expectations', Garland v Consumers' Gas Co. [2004] 1 SCR 629, discussed in Section III.

33 M. McInnes, 'Unjust Enrichment, Juristic Reasons and Palm Tree Justice: Garland v. Consumers' Gas Co.', (2004) 41 Can Bus L J 120.

34 See Virgo, Principles (n 3) 144.

35 See Virgo, Principles (n 3) 85-6, citing, among other cases, Rowe v Vale of White Horse DC [2003] EWHC 388, in which Lightman J explicitly recognised the previously controversial doctrine of free acceptance.

36 Virgo, Principles (n 3) 85 and 70-2; Meier, 'No Basis' (n 23) 358.

37 The phrase 'restitutionary response' is borrowed from Chen-Wishart, who uses it to argue that the protective policies underlying certain unjust factors help explain whether a change of position defence should be available to certain defendants, and that this sensitivity to fact is lost under an absence of basis approach: see M. Chen-Wishart, 'In Defense of Unjust Factors: A Study of Rescission for Duress, Fraud and Exploitation', in Unjustified Enrichment: Key Issues in Comparative Perspective, D. Johnston and R. Zimmermann, (eds.), (Cambridge: Cambridge University Press, 2002), p.160; published also as (2000) Oxford U Comparative L Forum 2 at ouclf.iuscomp.org.

38 CTN Cash and Carry Ltd v Gallaher Ltd [1994] 4 ALL ER 714 (CA). If the payment is a tax liability, however, not a purported contractual obligation, then the 'policy' unjust factor is engaged and restitution will be available: Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL).

39 L. Smith, 'Public Justice and Private Justice: Restitution After Kingstreet', (2008) 46 Can Bus L J 14. See also Lord Hope in Kleinwort Benson v Lincoln City Council [1999] 2 AC 389 (HL) 410: 'A state of doubt is different from that of mistake. A person who pays when in doubt takes the risk that he may be wrong... whether the issue is one of fact or of law".

40 Dannemann ,'English Law' (n 23) 374-5; Burrows, 'Basis' (n 3) at n 56.

41 The claimant may still be able to recover if the basis of the transfer was communicated to the recipient and falls within the principles of total failure of consideration: Meier, 'No Basis', (n 19), 352; and Virgo, Principles (n 3) 144.

42 Virgo, 'Pyramid' (n 5) 7.

43 M. McInnes, 'Bases for Restitution: A Call for Clarity with Unjust Factors', (1996) 10 J Contract L,77-8; and Virgo, 'Pyramid' (n 5) 7, citing Birks' conclusion to this effect inBirks (n 2) 145.

44 See, for instance, R. Zimmermann, 'Unjustified Enrichment: The Modern Civilian Approach', (1995) 15 Oxford J Legal Stud 403, 416; S. Meier, 'Unjust factors and Legal grounds' in Unjustified Enrichment: Key Issues in Comparative Perspective, D. Johnston and R. Zimmerman, (eds.), (Cambridge: Cambridge University press, 2002), 37; S. Meier and R. Zimmerman, 'Judicial Developments of the Law, error iuris and the Law of Unjustified Enrichment', (1995) 115 LQR. 556; and Birks (n 2) 160.

45 Birks (n 2) 160.

46 Zimmermann, 'Unjustified Enrichment' (n 44) 416.

47 Chen-Wishart, 'Defense of Unjust Factors' (n 37) 163. See also Virgo, Principles (n 3) 182-3: "[I]f the claimant's mistake was sufficient to vitiate the contract it must follow that the claimant's mistaken belief in the validity of the contract was the operating cause of his or her payment to the defendant".

48 Edelman, 'Meaning of Unjust' (n 3) 322, citing Birks (n 2) 126. Likewise, Stevens observes that English law, "...could not leave such questions as when is pressure illegitimate' or 'does a misrepresentation as to the future give rise to an absence of basis' to be dealt with within contract law. First, such questions are still conceptually part of the law of unjust(-ified) enrichment, although they may arise in the context of contract. Second, books are not written...on the law of non-contractual transfer. The law concerning when it is possible to set gifts and other non-contractual transfers aside must be dealt with in books on unjust enrichment. There is nowhere else for them to go": Stevens, 'Birksian Approach' (n 3) 272.

49 J. Beatson, 'Restitution in Canada: A Commentary', in Restitution: Past, Present & Future, W. Cornish, et al. (eds.), (Oxford: Hart Publishing, 1998), 298.

50 Dannemann, 'English Law' (n 23) 376.

51 Beatson, 'Restitution in Canada' (n 49) 298; Dannemann ,'English Law' (n 23) 376; Smith, 'Mystery' (n 21) 243; McInnes, 'Palm Tree' (n 33) 116.

52 See Section I.

53 Sheehan, 'Transfers Sine Causa' (n 31) 11, citing s. 814 of the German Civil Code.

54 Sheehan, 'Transfers Sine Causa' (n 31) 11.

55 "[A] scheme can be made to appear elegant if it no longer has to answer many difficult questions": Burrows, 'Basis' (n 3) 47.

56 Virgo, 'Pyramid' (n 5) 9.

57 Virgo, 'Pyramid' (n 5) 9.

58 Burrows, 'Basis' (n 3) 46-7.

59 Virgo, 'Pyramid' (n 5) 9.

60 Virgo, 'Pyramid' (n 5) 9. This problem has emerged in Canada, as I discuss in Section III.

61 "Even as regards the swaps cases Birks acknowledged that it would be necessary to consider whether the claimant was mistaken when paying the money...since this would indicate that the claimant had not willingly taken the risk that the money was not due. But this emphasis on risk-taking, which appears to incorporate aspects of the law concerning the established grounds of restitution, undermines the elegant pyramidal structure of Birks' approach to absence of basis. For, as regards his analysis of the swaps cases, he asserted that the unjust enrichment claim could be grounded simply on the invalidity of the underlying transaction", Virgo, 'Pyramid (n 5) 9.

62 Which requires an intention to give and applies to transfers of property not services.

63 Burrows, 'Basis' (n 3) 47.

64 K. Barker, 'Responsibility for Gain: Unjust Factors or Absence of Legal Grounds? Starting Points in Unjust Enrichment Law', in Structure and Justification in Private Law, C. Rickett and R. Grantham, (eds.), (Oxford: Hart Publishing, 2008), 65.

65 Birks (n 2) 148 (gift appearing in its own right as a participatory basis of enrichment) and 158 (gift serving to deny restitution in context of non-participatory transfers).

66 Burrows, 'Basis' (n 3) 47; Meier, 'No Basis' (n 23) 349; Virgo, 'Pyramid' (n 5) 8; Stevens, 'Birksian Approach' (n 3) 272; Sheehan, 'Transfers Sine Causa' (n 31) 10.

67 Birks (n 2) 125-6.

68 Birks implicitly acknowledged that such logic was a necessary fudge to fit absence of basis into the exiting legal framework: "If we were to say that terminability and voidability were insufficient in themselves and that the absence of valid basis only began from the termination or the rescission, we would not be able to explain the origin of the powers to terminate or avoid, which are not granted by contract and cannot be said to arise from their own effects. These powers arise from the enrichment rendered unjust by the invalidity of the contract under which it was transferred", per Birks (n 2) 126.

69 Virgo, 'Pyramid' (n 5) 8. Similarly, Burrows writes: "[I]t is odd to think of there being an absence of basis before termination or rescission": ' Basis' (n 3) 47.

70 Birks (n 2) 126.

71 Meier, 'No Basis' (n 23) 349, is critical of making special exceptions to the notion of basis for one party's right to recover.

72 Birks (n 2) 116.

73 Birks (n 2) 129.

74 So, while it is easy to say the claimant's purpose was to make a gift, and likewise easy to say the gift was ineffective and therefore prima facie unjust enrichment is established, in order to actually explain why the gift was ineffective, and enable this legal determination, we must have regard to mistake, undue influence or other unjust factors. See W. Tang, 'Natural Obligations and the Common Law of Unjust Enrchment', (2006) 6 Oxford U Commonwealth L J no. 2, 138.

75 Garland v Consumers' Gas Co. (2001), 208 DLR (4th) 494 (Ont. CA).

76 Pacific National Investments Ltd v Victoria (City), [2004] 3 SCR 575, at para. 31.

77 This was the observation of Borins J.A., dissenting, in Garland v Consumers' Gas Co. (2001), 208 DLR (4th) 494 (Ont. CA). As Smith observes, such an approach is, of course, "...not really... a different approach at all. It ...[is] the traditional approach, with 'absence of juristic reason' serving only a kind of rhetorical function": Smith, 'Juristic Reasons' (n 30) 285.

78 Tang, 'Obligations' (n 74) 138.

79 Virgo, 'Pyramid' (n 5) 10.

80 Sheehan, 'Unjust Factors or Restitution of Transfers Sine Causa' (n 31) 10.

81 Barker, 'Responsibility' (n 64) 63.

82 Burrows, 'Basis' (n 3) 47. Birks was aware of this criticism, recognizing that, because setting out all the possible bases "verges on the impossible", the concept of basis risks expression as "jejune abstraction", which "has the unfortunate consequence that it leaves the law of unjust enrichment abstract and further out of touch with the Clapham omnibus", Birks (n 2) 127.

83 "Natural obligations" refers to the following bars to recovery listed by Lord Mansfield in Moses v Macferlan (1760) 2 Burr 1005: "...[M]oney paid by the plaintiff, which is claimed of him as payable in point of honor and honesty, although it could not have been recovered from him by any course of law; as in payment of a debt barred by the Statute of limitations, or contracted during his infancy, or to the extent of principal and legal interest upon an usurious contract, or, for money fairly lost at play: because in all of these cases, the defendant may retain it with safe conscience...".

84 Birks (n 2) 258.

85 Dannemann ,'English Law' (n 23) 368. Note that Tang takes the opposite view, preferring to regard natural obligations as providing defences in cases of gambling losses and payments of time-barred debts: Tang, 'Obligations' (n 74) 156.

86 Dannemann ,'English Law' (n 23) 368.

87 In Moses v Macferlan, Lord Mansfield identified four categories of natural obligation (time-barred debt; contract made during age of minority; usurious contract; gambling loss). Dannemann accepts the first of these, rejects the second, and identifies a possible third natural obligation (contracts not complying with statutory formalities) that may exist depending on scrutiny of the applicable legislation: Dannemann ,'English Law' (n 23) 369-72. Tang recognizes only the first and fourth of Lord Mansfield's list of natural obligations: Tang, 'Obligations' (n 74) 156. Birks seemingly endorses the entire list proposed by Lord Mansfield: Birks (n 2) 258.

88 Birks, while acknowledging that natural obligations can become an "unruly horse", does little to clarify the boundaries of the concept when he writes, without elaboration, "the law of unjust enrichment would be stultified if...[restitution were allowed in circumstances where] behind the technicalities of the law, there was a moral obligation to pay": Birks (n 2) 258. Tang proposes analyzing "...whether the policies behind the doctrine or law which avoided the obligation will be furthered by allowing or denying restitution" in order to ascertain which of Lord Mansfield's list should be given effect as a defence: Tang, 'Obligations' (n 74) 150. Dannemann appears to take a similar approach: Dannemann ,'English Law' (n 23) 369-72. Sheehan, in contrast, seems to focus on whether the underlying unenforceable obligation is itself worthy of recognition (as in the case of promises performed): D. Sheehan, 'Natural Obligations in English Law' [2004] L.M.C.L.Q. 172, 185-90.

89 Logically, Birks must be taken to implicitly reject treating natural obligations as bases because he treats them as defences. This would not be necessary if they were bases in their own right.

90 Garland v Consumers' Gas Co. (2001), 208 DLR (4th) 494 (Ont. CA), para. 46.

91 Krebs, 'Unjust Factors' (n 21) 79, quoting B.S. Markesinis, W. Lorenz and G. Dannemann, The German Law of Contract and Restitution (Oxford: Clarendon Press 1997), 713.

92 Barker, 'Responsibility' (n 64) 63.

93 Birks (n 2) 113, 263-4.

94 Indeed, Birks acknowledged the shifting importance under absence of basis from the third question (unjust) to the fifth (defences), Birks (n 2) 113, 263. Krebs confirms that this has been the experience in Germany: Krebs, 'Unjust Factors' (n 21) 79-80.

95 Virgo, Principles (n 3) 129.

96 Krebs, 'Unjust Factors' (n 21) 79-80. As McInnes observes, "[b]ecause of the need for careful integration, wholesale adoptions [of defences] from civilian jurisdictions...are simply out of the question": McInnes, 'Palm Tree' (n 33) 116.

97 Birks (n 2) 240.

98 Points '(i)' and '(ii)' are Virgo's: 'Pyramid' (n 5) 10.

99 Birks acknowledged that his examples of stultification are not exhaustive (p. 240), "leave much to be desired" and "need tidying up" (p. 263): Birks (n 2). It is beyond the scope of this paper to evaluate the merits of individual examples of stultification identified by Birks. Rather, it is the dangers of stultification as an open -ended, policy- based super-category, which parallels the residual open-ended 'policy' category of defense created by the Supreme Court of Canada in Garland v Consumers' Gas Co. [2004] 1 SCR 629, that will be examined here and in Section III, below.

100 Birks (n 2) 258.

101 These policies aim to "protect...the minor from being taken advantage of by unscrupulous adults and saving the minor from her own imprudence": Tang, 'Obligations' (n 74) 143.

102 Birks (n 2) 258.

103 Birks (n 2) 261-2.

104 Rathwell v Rathwell [1978] 2 SCR 436, 455. Dickson J. gave two examples of juristic reasons, suggesting a civilian orientation: "contract or disposition of law". Smith speculates that, ironically, Dickson J. chose 'absence of juristic reason' rather than 'unjust' in order to emphasize the reasoned basis of his decision and avoid complaints of palm tree justice. See Smith, 'Mystery' (n 21) 219. This speculation met with some approval by the Supreme Court in Garland v Consumers' Gas Co. [2004] 1 SCR 629, para. 40.

105 P. Maddaugh and J. McCamus, The Law of Restitution, (looseleaf ed.), (Toronto: Canada Law Book Inc., 2008), 3-24

106 P. Maddaugh and J. McCamus, Restitution (n 105) 3-26.

107 Pettkus v. Becker, [1980] 2 SCR 834, 847-8.

108 McInnes, 'Making Sense' (n 6) 403.

109 Smith, 'Mystery' (n 21) 217.

110 See Smith, 'Mystery' (n 21) ; McInnes, 'Making Sense' (n 6) ; and P. Perell, 'A Survey of the Case Law About the Cause of Action for Unjust Enrichment', (2004-5) 29 Advocates' Q. 458, for a discussion of these two lines of authority pre Garland.

111 See, for instance, the reasoning of McLachlin J. (premising her judgment on an absence of juristic reason) and Cory J. (basing his on the unjust factor free acceptance) in Peter v Beblow [1993] 1 SCR 980.

112 McInnes, 'Making Sense' (n 6) 405.

113 Smith has detailed the various juristic reasons that developed in this pre Garland period, and shown why many of them, such as 'legitimate expectations' and 'the rules of property law' have no place in unjust enrichment law: Smith, 'Mystery' (n 21) 220-8. See also Perell, 'Unjust Enrichment' (n 110) for a list of juristic reasons.

114 Smith, 'Mystery' (n 21) 214.

115 McInnes, 'Making Sense' (n 6) 405.

116 [2004] 1 SCR 629.

117 Garland, paras. 44-6. It is beyond the scope of this paper to consider all of the problems associated with this reformulation, such as the insistence that unjust enrichment is an equitable doctrine and that certain defences will operate in some cases but not in others (without any guidance provided). For a criticism of these aspects, see Smith, 'Juristic Reasons' (n 30) ; McInnes, 'Palm Tree' (n 33); and J. Neyers, 'One Step Forward, Two Steps Back: Unjust Enrichment in the Supreme Court of Canada', (2005) Lloyd's Maritime and Comm. L Q 435.

118 Garland, para 42.

119 As evidenced by the use of the word "includes" and the category of "other obligations" on the list, Garland, para. 44. Recall that Birks' list includes the residual category "other purposes", Birks (n 2) 152.

120 Donative intent sits uncomfortably on the list of juristic reasons, because while every other example is a legal figure outside the law of restitution, it refers instead to a question of fact: Smith, 'Juristic Reasons' (n 30) 290.

121 Garland cites Peter v Beblow [1993] 1 SCR 980 for recognizing 'donative intent' as a juristic category, but the phrase does not appear in that case. Peter v Beblow instead uses the term gift. Does donative intent mean legal gift? This is unclear. If the terms are synonymous, then why invoke the vague concept of donative intent? Furthermore, while the Court in Peter v Beblow assumed that the provision of domestic services leading to enrichment could be a gift if there was proof of intent to offer them gratuitously, this does not accord with the current law of gifts which is limited to gratuitous transfers of property: see Halsbury's Laws of England, (4th ed.), vol. 20, 16; and Dannemann ,'English Law' (n 23) 366. Recall that Birks' use of 'grudging gift' is also inconsistent with the law of gifts, as it applies to services and cases where the transferor is ignorant of the transfer.

122 Garland, para. 46.

123 P. Birks, 'Misnomer', in Restitution: Past, Present & Future, W. Cornish, et al. (eds.) (Oxford: Hart Publishing, 1998), 12; and M. McInnes, 'Reflections on the Canadian Law of Unjust Enrichment: Lessons From Abroad', (1999) 78 Can Bar Rev 430.

124 Birks (n 2) 116-7.

125 I will consider the unworkability of 'policy' when I discuss defences.

126 Indeed, under Garland once a policy defence is recognized under the second branch of the test, it may become a juristic reason in its own right, presumably falling in future cases under the first branch, see Garland, para. 46.

127 Recall that Birks treats both a busker's services and escaping heat as 'grudging gifts' despite neither concerning the transfer of property: Birks (n 2) 160 and 158, respectively. In Peter v Beblow the Court contemplated the possibility that domestic services leading to enrichment could be retained if there was proof of intention to gift them.

128 Virgo, 'Pyramid' (n 5) 9.

129 While Birks continued to view risk-taking as relevant to absence of basis, his 'grudging gift' concept now seems to do much of that work. The distinction between the concepts is problematically subtle. Confusingly, while Birks treats escaping heat as a grudging gift, he conflates gift and risk taking when discussing the officious busker: "The busker wants to be paid but takes the risk of getting nothing. The basis of his offering is not reciprocal payment but gift", Birks (n 2) 158-60. It is a fiction to regard the busker as gifting his services. As Virgo observes: "The real problem with the busker is that he is a risk- taker and the notion of risk has become unacceptably wrapped up with Birks' absence of basis analysis": Virgo, 'Pyramid' (n 5) 22.

130 Peter v Beblow [1993] 1 SCR 980.

131 Her failure to insist on adding her name to title in exchange for providing services suggests her willingness to risk receiving nothing. In other words, so long as she perceived, subjectively, that she may not receive any proprietary interest in consideration for performing household services, but continued to perform them in any event in the mere hope of compensation, then risk-taking should operate to bar recovery. In this respect, see Virgo, Principles (n 3) 40: "Where the claimant can be considered to have taken the risk...[that] the defendant may pay for the benefit without being obliged to do so, it is not for the law of restitution to reallocate the risk so that the defendant is required to make restitution... Where the claimant has taken the risk he or she should be considered to have acted officiously".

132 Virgo, 'Pyramid' (n 5) 22.

133 (1999), 173 DLR (4th) 270, para. 43.

134 Tang reminds us that Birks was hostile to using the notion of morality in the law of unjust enrichment, Tang, 'Obligations' (n 74) 146, citing P. Birks, 'Annual Miegunyah Lecture: Equity, Conscience and Unjust Enrichment' (1999) 23 MULR 1, 20-3.

135 Birks (n 2) 257-8. See Tang, 'Obligations' (n 74) 146.

136 Dannemann, 'English Law' (n 23) 368.

137 2004 SCC 75.

138 Pacific National Investments v. Victoria (City), 2003 BCCA 162, para 26.

139 PNI, para. 51.

140 For example, intent to contract was vitiated though mistake, or donative intent was ineffective due to duress. See, McInnes, 'Making Sense' (n 6) 409: 'Garland's absence of juristic reason necessarily builds upon the types of consideration that historically served as unjust factors". See also R. Grantham, 'Absence of Juristic Reason in the Supreme Court of Canada', (2005) 13 RLR, 107.

141 R. Chambers, 'Canada', (2006) 14 RLR, 147.

142 McInnes, 'Palm Tree' (n 33) 107.

143 PNI was argued less than four months after Garland.

144 PNI, para. 13.

145 PNI, para. 31.

146 PNI, para. 31, emphasis mine.

147 Garland, paras. 49-53.

148 McInnes, 'Making Sense' (n 6) 405.

149 Grantham speculates that the reference to mistake in PNI is a manifestation of the 'limited reconciliation' between unjust factors and absence of basis embedded in Birks' pyramid: Grantham, 'Absence of Juristic Reason' (n 140) 107.

150 [2007] 1 SCR 3.

151 Kingstreet, para. 4.

152 Smith, 'Kingstreet', n.35, 23.

153 Kingstreet, para. 31. There are many problems associated with the Court's reasoning here, but, as they are unrelated to the interpretation and application of the Garland framework, it is beyond the scope of this paper to canvass them. For a fulsome discussion, see Smith, 'Kingstreet' (n 39).

154 Smith, 'Kingstreet' (n 39) 22, citing, among others, Air Canada v British Columbia, [1989] 1 SCR 1161.

155 Kingstreet, para. 38.

156 Kingstreet, para. 39.

157 Kingstreet, para. 58.

158 Smith, 'Juristic Reasons' (n 30) 301.

159 Grantham, 'Absence of Juristic Reason' (n 140) 106; Smith, 'Juristic Reasons' (n 30) 301.

160 This is supported by the Court's reference to Garland as 'very complex', despite the fact that applying absence of basis to ultra vires taxes is, as Smith notes, "extremely straightforward": Smith, 'Kingstreet' (n 39) 23. It is also noteworthy that Bastarache J. states (para. 36) that Garland was simply a 'restatement and refinement" of previous Canadian jurisprudence, with reference to Peel (Regional Municipality) v Canada, [1992] 3 SCR 762. This belies what Garland really achieved and reveals a deep misunderstanding either of Garland itself or of what came before (or both). Indeed, as McInnes notes: "Garland was momentous precisely because it replaced the common law model of unjust enrichment with a civilian counterpart", M. McInnes, 'Restitution for Ultra Vires Taxes', (2007) 123 LQR 368. See also Neyers, 'Two Steps Back' (n 117) for a similar view that Garland was a "final and forceful" break from the traditional approach.

161 Whereas the House of Lords in Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 7 employed constitutional principles to establish a new policy based unjust factor, Kingstreet is different because it removes the question from the law of unjust enrichment alltogether. The implications of this are considerable, including uncertainty as to what extent other principles from the law of restitution would continue to apply, and, more seriously, the possibility that in future situations the presence of a constitutional remedy will stop a plaintiff from pursuing an otherwise available claim in unjust enrichment: Smith, 'Kingstreet' (n 39).

162 (2005) 285 NBR (2d) 201.

163 Chambers, 'Canada' (n 141) 147.

164 2009 SCC 15

165 BMP, para. 23

166 BMP, para. 22

167 [1979] 3 All ER 522 Q.B., 541: "If a person pays money to another under a mistake of fact which causes him to make the payment, he is prima facie entitled to recover it as money paid under a mistake of fact".

168 BMP, para. 52

169 M. McInnes, 'Forged Cheques, Tracing and Restitution in the Supreme Court of Canada', (2009) 125 LQR 556.

170 Barker reminds us to consider the jurisprudence following the creation of the tort of negligence, which also had an expansionary momentum and non-comprehensive defences: Barker, 'Responsibility' (n 64) 65.

171 Recall that stage two of Garland creates a residual category of defence based on public policy and the reasonable expectations of the parties. The latter factor is doctrinally problematic. See McInnes, 'Palm Tree' (n 33). However, as it does not have an equivalent place in Birks's scheme, it is beyond the scope of this paper to discuss it further.

172 See discussion in Section II, above.

173 Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (HL).

174 Garland, para. 45; Birks (n 2) 116.

175 Garland, para. 45. Birks was equally clear that policies favouring restitution are irrelevant to absence of basis and must be supplanted, at the defence stage, by policies favouring retention: Birks (n 2) 116

176 Garland, para. 57.

177 PNI, para. 57.

178 Kingstreet, para. 38.

179 [1989] 1 SCR 1161.

180 Kingstreet, para 38.

181 McInnes, 'Making Sense' (n 6) 413.

182 Kingstreet, para. 38.

183 Lipkin Gorman (a firm) v Karpnale Ltd [1991] 2 AC 548, 578.

184 McInnes, 'Palm Tree' (n 33) 133.

185 McInnes, 'Making Sense' (n 6) 400, citing Attorney General v Blake, [2001] AC 269 291 (HL).

186 S. Hedley, 'Unjust at Common Law: So Many Concepts, So Little Clarity', (2006) 3 ERPL 407.

187 Beatson, 'Restitution in Canada' (n 49) 298-9.

© 2009 Chris D. L. Hunt. This HTML edition © 2009 University of Oxford.
The contents of this page may be downloaded and printed out in single copies for individual use only. Making multiple copies without permission is prohibited.

Your must accept our disclaimer before you continue.
The Oxford University Comparative Law Forum is sponsored by CMS Hasche Sigle.