This essay is a revised version of a paper prepared for the “Obligations VI” conference at the University of Western Ontario in July 2012. The conference theme was “Challenging Orthodoxy”. I would like to thank the conference participants for their comments and questions. Special thanks also go to Nicola Langille and Zain Naqi for research assistance.

1. “Almost everything of mine now needs calling back for burning”: Birks, Peter, Unjust Enrichment, 2d ed (Oxford: Oxford University Press, 2005)CrossRefGoogle Scholar [Birks, Unjust Enrichment] at xii. Birks was referring primarily to his abandonment of the “unjust factors” approach to establishing unjust enrichments (which he had defended in An Introduction to the Law of Restitution ((Oxford: Oxford University Press, 1985) [Birks, An Introduction to the Law of Restitution]) in favour of a “no legal basis” approach).

2. To mention just one example, the law governing “restitution for wrongs” is discussed in Birks, An Introduction to the Law of Restitution, ibid, but not in Birks, Unjust Enrichment, supra note 1.

3. Compare Birks, Unjust Enrichment, supra note 1 to Birks, An Introduction to the Law of Restitution, supra note 1. The American Law Institute’s recent restatement of the subject adopts both labels: Restatement of the Law Third: Restitution and Unjust Enrichment (St Paul, MN: American Law Institute Publishers, 2011) [Restatement].

4. An excellent overview of the major debates can be found in Burrows, Andrew, The Law of Restitution, 3d ed (Oxford: Oxford University Press, 2011) at 28–44 Google Scholar, 86-116. Many authors doubt that the subject even exists, that is to say, that it constitutes a distinct legal field: see, e.g., Hedley, Steve, Restitution: Its Division and Ordering (London: Sweet & Maxwell, 2001)Google Scholar; Jaffey, Peter, The Nature and Scope of Restitution (Oxford: Hart, 2000)Google Scholar; Dietrich, Joachim, Restitution: A New Perspective (Sydney, Australia: The Federation Press, 1998).Google Scholar

5. A consequence of the above-mentioned debates is that even writers who agree that unjust enrichment law is a distinctive subject disagree, sometimes radically, about the subject’s scope. In this essay, “unjust enrichment law” refers to the rules that Peter Birks discussed in Birks, Unjust Enrichment, supra note 1.

6. The qualification “for the most part” is necessary because Birks believed that some unjust enrichments may give rise to new proprietary rights: Birks, Unjust Enrichment, supra note 1 at 38. Others disagree: see, e.g., Virgo, Graham, The Principles of the Law of Restitution, 2d ed (Oxford: Oxford University Press, 2006) at 11–17 CrossRefGoogle Scholar.

7. Thus, I say nothing about the rules governing things like restitution for services, restitution for wrongs, or restitution for a failure of consideration. Birks described this essay’s subject matter as cases involving an “initial failure of basis” (as opposed to a “subsequent failure”): Unjust Enrichment, supra note 1 at 185. In addition to mistake and duress, Burrows includes as examples of (what I label) impaired transfers, transfers that arise from undue influence, exploitation of weakness, incapacity, ignorance, necessity, illegality, and public authority ultra vires demands: Burrows, supra note 4 at 201-317, 403-522.

8. There is no accepted label for this body of law. I will refer to it as the “law governing impaired transfers”, although strictly speaking I am only concerned with transfers that are impaired in some way notwithstanding that title to the relevant property (or other right) is successfully conveyed.

9. Under current English law, all monetary awards and other awards that contemplate the transfer of property from the defendant to the plaintiff are framed as orders: “It is therefore ordered that [the defendant] must pay the Plaintiff [the sum of X]” (The Civil Procedure Rules 1998 (UK), para C17-001, Form N30(1)). Previously, monetary awards were described as “judgments” and were framed as abstract pronouncements that “the plaintiff shall have and recover” a certain sum from the defendant (a practice that continues in other common law jurisdictions: see, e.g., Dobbs, Dan B, Law of Remedies: Damages, Equity, Restitution, 2d ed (St Paul, MN: West, 1993))Google Scholar. The distinction that I draw in Part 1 between the different meanings attached to rules and orders exists equally if the comparison is between rules and judgments.

10. I first developed this argument in an attempt to understand the terminology of the Restatement Third: Restitution and Unjust Enrichment, supra note 3, in particular the Restatement’s practice of describing individuals who have been unjustly enriched as subject to liabilities (not duties) to make restitution. The product of that attempt, on which this essay builds, can be found in Smith, Stephen, “The Restatement of Liabilities in Restitution” in Mitchell, C & Swadling, W, eds, The Restatement Third, Restitution and Unjust Enrichment: Comparative and Critical Essays (Oxford: Hart) [forthcoming in 2013]Google Scholar [Smith, “The Restatement of Liabilities in Restitution”].

11. See Part 3.

12. See Part 4.

13. See, e.g., Morgan v Ashcroft, [1938] 1 KB 49 (CA).

14. “The previous paragraphs have shown that a striking feature of the liability to make restitution of mistaken payments is that the liability is strict.” (Birks, Unjust Enrichment, supra note 1 at 8). Similar language is found in Virgo, supra note 6; Webb, Charlie, “Property, Unjust Enrichment, and Defective Transfers” in Chambers, Robert, Mitchell, Charles & Penner, James, eds, Philosophical Foundations of the Law of Unjust Enrichment (Oxford: Oxford University Press, 2009) 335 CrossRefGoogle Scholar at 354; McFarlane, Ben, “Unjust Enrichment, Rights, and Value” in Nolan, Donal & Robertson, Andrew, eds, Rights and Private Law (Oxford: Hart, 2012) 581.Google Scholar Rare examples of authors who consistently describe unjust enrichments as giving rise to duties to make restitution include Dennis Klimchuk (“The Normative Foundations of Unjust Enrichment” in Chambers, Mitchell & Penner, supra note 14 at 81) and Saprai, Prince (“Restitution without Corrective Justice” (2006) 14 Restitution L Rev 41).Google Scholar

15. Restatement, supra note 3. It is clear that the liability mentioned in this section is a liability to a court order because whenever the Restatement does not use liability terminology it describes unjust enrichments as giving rise to “a claim in restitution”: see, e.g., § 6, 7, 8, 9, 10, 11, 12, 18, & 19. Similar language is used consistently throughout the Restatement: see Smith, “The Restatement of Liabilities in Restitution”, supra note 10 at n 27.

16. Though theoretically possible, this interpretation is vulnerable to objections that I explain below at note 68.

17. The assumption that court orders, including restitutionary orders, replicate existing duties underlies the entirety of Birks’ essay, “Rights, Wrongs, and Remedies” (2000) 20 Oxford J Legal Stud 1 [Birks, “Rights, Wrongs, and Remedies”].

18. I say “at a minimum” because the duty model is also consistent with believing that recipients of impaired transfers do not fall immediately under a duty to make restitution but instead are only “liable” to fall under such a duty when they are informed of the impairment: see below at note 68 and accompanying text.

19. Birks, Unjust Enrichment, supra note 1 at 169.

20. Ibid at 208-19.

21. Scholars who adopt this view must also presumably suppose that no duty arises at all in cases where, prior to knowledge, the recipient changes position so as to dissipate the entire benefit.

22. Hohfeld, Wesley, “Fundamental Legal Conceptions” (1913) 23 Yale LJ 29 at 44–54.Google Scholar

23. Section 15.2(1) of the federal Divorce Act, RSC 1985, c 3 (2d Supp) (which states that a court may “on application by either or both spouses, make an order requiring a spouse to secure or pay, or to secure and pay, such lump sum or periodic sums, or such lump sum and periodic sums, as the court thinks reasonable for the support of the other spouse”). See also s 16(1) of the same Act, and the Family Relations Act, RSBC 1996, c 128. The English equivalents are ss 22 & 23 of Matrimonial Causes Act, 1973.

24. See, e.g., Baker v Courage & Co, [1910] 1 KB 56 at 65 [Baker].

25. I explain in Part 2A why payment in advance, even when not part of a settlement, is sufficient to negative a prima facie liability.

26. I discuss these requirements in more detail in Smith, Stephen, Contract Theory (Oxford: Oxford University Press, 2005) at 13–15 Google Scholar, 24-32.

27. Hart, HLA, The Concept of Law (Oxford: Clarendon Press, 1961) at 82–91 Google Scholar; Raz, Joseph, The Authority of Law (Oxford: Oxford University Press, 1979) at 3–232 Google Scholar; Raz, Joseph, Ethics in the Public Domain (Oxford: Oxford University Press, 1994) at 210–37Google Scholar.

28. My point is not that the reason the law presents legal duties as morally obligatory is that it presumes there is an obligation to obey the law. The point is rather that legal duties are presented as requiring actions that are morally obligatory independent of whether there is a moral duty to obey the law.

29. As mentioned earlier, in England today all awards that require the performance of an action are framed as “orders”: see supra note 9 and accompanying text.

30. For present purposes, I include within the category of replicative orders those that command the performance of a close substitute—for example, the payment of a sum equal to the value of an object that the defendant had a legal duty to deliver or return.

31. As any parent knows, commands often motivate when rules fail: see Smith, Stephen, “Why Courts Make Orders (And What This Tells Us About Damages)” (2011) 64 Current Legal Probs 51 at 60–63.Google Scholar

32. I explore these explanations in greater detail in Smith, Stephen A, “Duties, Liabilities, and Damages” (2012) 125 Harv L Rev 1727 Google Scholar at 1741-49.

33. I am not suggesting that theorists attempting to explain creative orders must invoke instrumental (or “utilitarian”) theories. The explanation can be “moral” in a broad sense—it is simply that the moral reasons it invokes cannot be reasons that apply to the defendant. In other words, the theory cannot be about the defendant’s moral obligations. Retributivist theories of punishment are non-instrumental.

34. See, e.g., Colonial Bank v Exchange Bank of Yarmouth, Nova Scotia (1885) 11 App Cas 84 at 90-92; Baker, supra note 24 at 65; Banque Financiere de la Cité v Parc (Battersea) Ltd, [1999] AC 221 (HL) at 221.

35. Birks, Unjust Enrichment, supra note 1 at 169, citing Baker, supra note 24 at 65.

36. On the Romanist framework, see Dedek, Helge, “The Relationship Between Rights and Remedies in Private Law: A Comparison Between the Common and the Civil Law Tradition” in Sharpe, Robert J & Roach, Kent, eds, Taking Remedies Seriously (Montreal: Canadian Institute for the Administration of Justice, 2010) 63.Google Scholar Birks’ adoption of this framework is clearly set out in his Introduction to English Private Law ( Birks, Peter, ed (Oxford: Oxford University Press, 2000)).Google Scholar

37. This assumption underlies almost the entirety of Birks’ essay “Rights, Wrongs, and Remedies”, supra note 17.

38. Even if legislatures were to mandate that punitive damages be paid immediately following a wrong, how could wrongdoers know how much to pay? It also seems clear that, as a matter of positive law, there is no duty in the common law to pay damages of any kind prior to a court ordering such payment: see Smith, supra note 32 at 1741-49.

39. Sempra Metals Ltd v Inland Revenue, [2007] UKHL 34 [Sempra Metals]; Senior Courts Act 1981 (UK), c 54, s 35A.

40. Wadsworth v Lydall, [1981] 2 All ER 401.

41. Burrows, supra note 4 at 23-24.

42. Westdeutsche Landesbank Girozentrale v Islington London Borough Council, [1996] AC 669 at 696, 736; Sempra Metals, supra note 39 at 34-50, 116, 132, 178-79. See also Birks, Unjust Enrichment, supra note 1 at 53; Burrows, supra note 4 at 55-56.

43. See Burrows, supra note 4 at 525, 537-40.

44. Strictly speaking, it is knowledge of the possibility of a liability or duty that is relevant because, as I explain below, recipients cannot be certain of the existence of either prior to a court judgment (which, as I explain further, is also an argument against the duty model).

45. See Burrows, supra note 4 at 526-27; Weinrib, Ernest, Corrective Justice (Oxford: Oxford University Press, 2012)CrossRefGoogle Scholar, chapter titled “Unjust Enrichment”.

46. It is difficult to find clear authority for this proposition, but the counter-proposition appears never to have been contemplated by courts or commentators.

47. Edmunds v Lloyds Italico & l’Ancora Compagnia di Assicurazione e Riassicurazione SpA, [1986] 1 WLR 492 at 496 (confirmed in Fast Ferries One SA v Ferries Australia Pty Ltd, [2000] 1 Lloyd’s Rep 534). See also Tettenborn, A, The Law of Damages (London: LexisNexis UK, 2003) at 211.Google Scholar

48. Again, while there is no clear authority for this proposition, the counter-proposition appears never to have been contemplated.

49. Burrows, supra note 4 at 97-99. Even those who believe England has adopted the unjust factors approach accept that this happened only recently. For Birks, the change occurred with the “swaps” cases: Birks, Unjust Enrichment, supra note 1, 108-17. Canada appears to have shifted to the no legal basis approach: Garland v Consumers’ Gas Co, 2004 SCC 25, [2004] 1 SCR 629. It could be argued that the unjust factor requirement is satisfied in cases where a repayment is made because the recipient mistakenly assumed he had a legal duty to repay. But if this argument is accepted, it follows (contrary to the duty model) that a recipient who pays in the mistaken belief that the limitation period has not expired also satisfies this requirement.

50. I write “appears” because the Limitation Act 1980 ((UK), c 58) does not expressly apply to claims for restitution (“The overall picture of the law is de Pressingly uncertain”: Burrows, supra note 4 at 605), and because there is no direct authority for this particular rule.

51. (UK), c 60. Canada lacks a legislative equivalent, but Canadian courts appear to interpret ordinary thefit as including the situations contemplated in section 5(4): see R v Smith, [1993] 3 SCR 635; confirming the reasons of Finlayson JA in R v Smith, (1992) 77 CCC (3d) 182, 59 OAC 390.

52. I am grateful to Nicholas McBride for bringing section 5(4) to my attention.

53. This interpretation is supported by comparison to the apparently similar case of a debtor who intentionally refuses to pay a debt. If section 5(4) is aimed at defendants who have intentionally breached a duty to pay money, then its principle should apply to the defaulting debtor; indeed, the principle of 5(4) should apply more strongly to the defaulting debtor because his obligation was undertaken voluntarily. The only apparent difference between the two situations appears to be that in the case of a mistaken payee it is possible to argue that the payor retains a kind of proprietary interest.

54. See, e.g., Att-Gen’s Reference (No. 1 of 1983), [1985] 1 QB 182.

55. For example, section 66(3) of the Sex Discrimination Act 1975 ((UK), c 65) provides that “[a]s respects an unlawful act of discrimination falling within section 1(1)(b) [indirect discrimination]… no award of damages shall be made if the respondent proves that the requirement or condition in question was not applied with the intention of treating the plaintiff unfavourably on the ground of his sex.”

56. The unavailability of damages for failing to pay damages is not another example because there is no legal duty to pay damages (merely a liability to be ordered to pay damages): see supra note 38.

57. The philosophical underpinnings of this view and the difficulty of reconciling it with restitutionary duties are described in Klimchuk, supra note 14 at 81. Klimchuk focuses in particular on Ernest Weinrib’s attempted reconciliation (which invokes a notion of hypothetical consent), as set out in Weinrib, Ernest, “The Normative Structure of Unjust Enrichment” in Rickett, Charles & Grantham, Ross, eds, Structure and Justification in Private Law (Oxford: Hart, 2008) 1 Google Scholar at 37.

58. The idea that English law recognizes duties that arise from committing wrongs is more controversial because, as noted earlier, there is no duty to pay damages in English law: see supra note 38.

59. With the exception of duties to vote, these duties also differ from duties to make restitution in that they can, in theory anyway, be avoided by not engaging in the duty-creating activity (earning taxable income, buying a house).

60. See, e.g., Charter of Human Rights and Freedoms, RSQ c C-12, s 2 (Quebec).

61. For an overview, see the Introduction in Reinhard Zimmerman & Whittaker, Simon, Good Faith in European Contract Law (Cambridge: Cambridge University Press, 2000).Google Scholar

62. Ibid.

63. See, e.g., Kant, Immanuel, The Metaphysics of Morals, translated by Gregor, Mary (Cambridge: Cambridge University Press, 1991) at 196 Google Scholar; Stuart Mill, John, On Liberty (London: John W Parker and Son, 1859).Google Scholar

64. Birks, Unjust Enrichment, supra note 1 at 208.

65. Ibid at 183.

66. Indeed, as John Gardner notes, the law does not care one way or the other, when determining if a strict duty has been breached, how much care the defendant took; all that matters is whether the defendant succeed in fulfilling the duty. As part of his defence of strict duties, Gardner further argues that in some activities success is more likely if you take less than reasonable care (Gardner’s example is walking barefoot over hot coals) and that in all activities there comes a point at which taking more care becomes counter-productive (Gardner’s example is driving): “Obligations and Outcomes in the Law of Torts” in Cane, P and Gardner, J, eds, Relating to Responsibility: Essays for Tony Honoré (Oxford: Hart, 2001) 111 Google Scholar at 116. It is not clear to me that walking quickly when traversing hot coals or not being unduly cautious when driving actually amounts to taking less care, but even accepting Gardner’s general point it remains the case that anyone who wants to ensure compliance with a strict duty should take whatever steps will promote such compliance. In the case of restitutionary duties, these steps would clearly involve taking extreme care by, for example, constantly checking one’s bank account.

67. The law does, of course, recognize strict liabilities—for example, I am strictly liable if I trespass on your property or convert your goods. It might be argued that the existence of such liabilities demonstrates that, whatever theoretical objections might be thought to arise, in practice the law regularly recognizes duties that are every bit as strict as duties to make restitution. This argument is fawed, however, because it cannot be assumed that strict liabilities correlate to strict duties. Although this issue cannot be explored here, a more natural way to understand strict liabilities is that they are exactly what their name suggests—liabilities. The fact that a defendant who has done X is strictly liable, as a consequence, to pay damages does not prove that the defendant was under a strict duty not to do X. Doing X may simply trigger a liability to pay damages. I explore this issue in more detail in “Strict Duties and the Rule of Law” in Austin, Lisa & Klimchuk, Denis, eds, Private Law and the Rule of Law (Oxford: Oxford University Press)Google Scholar [forthcoming in 2013].

68. One possible way to avoid this objection (though not the other conceptual objections) is to suppose that the liability that arises on transfer is not a liability to a court order but instead a liability to fall under a duty in the case that the recipient learns of the impairment, or the transferor makes a demand, etc. This interpretation can explain why courts and commentators regularly describe recipients as falling under liabilities to make restitution, but it cannot explain why courts also say that the cause of action arises on transfer, since a cause of action is a liability to a court order. Another difficulty is that it cannot strictly be the case that the recipient is under a “liability” to fall under a duty since liabilities are activated by the exercise of a legal power, and it is normally assumed that, exceptional cases aside, legal powers can only be exercised by actions that reflect an intention to exercise the power. Mere acquisition of knowledge does not satisfy this test, nor does a demand for restitution or the initiation of a lawsuit.

69. Another odd consequence of assuming that a duty to make restitution arises if (and only if) credible evidence of an impairment is presented to the recipient is that transferors could then bring actions to enforce duties that do not yet exist. Transferors are under no obligation to disclose the facts upon which they intend to rely in court. Yet if the plaintiff has not disclosed the facts, it follows that the plaintiff is not suing to enforce a duty since the duty does not arise until the facts are revealed. In such a case, the transferor appears to be suing on the basis of a pure liability.

70. See supra note 23.

71. Writers who defend this view include Stoljar, Samuel, The Law of Quasi-Contract, 2d ed (Sydney, AU: The Law Book Company of Australasia, 1989) at 5–9 Google Scholar; Watts, Peter, “Restitution— A Property Principle and a Services Principle” [1995] Restitution L Rev 49 Google Scholar at 49-70; Dietrich, supra note 4 at 208-13; Jaffey, supra note 4 at 275-79; Webb, supra note 14; McFarlane, supra note 14.

72. Webb, supra note 14 at 335.

73. “On the model proposed here, the liability principle [which requires that defective transfers be reversed] is justified at one remove. The function of the principle … is to mitigate the effects on C [the plaintiff] of other legal rules”: McFarlane, supra note 14 at 607. McFarlane’s explanation is more properly described as a “legal-rights-based” explanation rather than a property-based explanation because it applies—correctly, in my view—to transfers of legal rights generally, not just to transfers of property rights. Interpreting the law of defective transfers as a corrective to other legal rules helps explain, inter alia, why the law is so complex and why it is often described as “subsidiary” to other bodies of law. It also helps explain why defective transfers are thought to lead to “unjust” enrichments. In its core and most common usage, to say that something is unjust means that it is thought to be unfair and, moreover, that it is unfair because of something that an authority, such as the state, has done or not done. I discuss these observations in more detail in Smith, “The Restatement of Liabilities in Restitution”, supra note 10.

74. “It hardly needs stating that no person is obliged to litigate and that, therefore, a defendant is perfectly free to refrain from responding to the service of the claim”: Zuckerman, Adrian, Zuckerman on Civil Procedure, 2d ed (London: Sweet & Maxwell, 2006) at 269.Google Scholar