1 Duties and Damages
Charlie Webb
I. Duty and Sanction
The common law is notoriously reluctant to make us do what it says we are duty bound to do. If I contract to sell you my shoes but refuse to hand them over, it’s not likely that you will get a court to compel me to give them to you. If I take yours, the likelihood once more is that a court will not order that I give them back. Instead all a court will do is order that I pay you damages, aimed at making good the losses I have thereby caused you. Some think this reveals something important about the content of our duties. Though we may say that I owe a duty to perform my contracts and to refrain from interfering with your assets, what courts in fact do tells a different story. When it matters, a court will not make me perform my contracts or refrain from interfering with your assets but will only require me to compensate you for the losses I cause if I do not. And if this, in the end, is all I will be required to do, we would do better to recognize that the law leaves me free to break my promise and free to take your things, my duty, at most, to make good your losses.
This view of what duties we owe reflects a particular understanding of what it means to owe a duty in law. On this understanding, to owe a legal duty is to face a legal sanction and, if we want to know what duties I owe, we need to know what sanctions I will face. So the conclusion that I owe no duty to perform my contract follows from the fact that no court will order me to do this. If the law will not compel it, I am free not to do it. The most notable proponent of this view, or something like it, was Holmes.1 Holmes’ motivation was to keep a clear division between law and morality. One way to maintain such a divide was to show how law mattered even to those who cared nothing for morality. When I am making decisions, moral considerations carry only such weight as I give to them. If I choose to give them no weight at all, there is nothing stopping me. Not so with law. For law cannot simply be ignored or wished away,
1 Oliver Wendell Holmes, “The Path of the Law” (1897) 10 Harv. LR 457.
Charlie Webb, Duties and Damages In: Oxford Studies in Private Law Theory: Volume 1. Edited by: Paul B. Miller and John Oberdiek, Oxford University Press (2020). © The Several Contributors. DOI: 10.1093/oso/9780198851356.003.0001
but will forcibly intrude into our lives. If the law is going to take my liberty or possessions if I act in a given way, that is a reality to which I cannot but attend when deciding what action to take. And so, when the law says that I owe a duty to φ, the one unavoidable, undeniable reality of this duty is the sanction which I will face if I do not.
This view proposes not so much a revision of the idea of a legal duty as its rejection.2 When the law recognizes a duty, it is not simply foreshadowing the imposition of some penalty or other coercive measure on those who fail so to act; it is telling us that this is something we must do, marking out the relevant conduct as non-optional. So the intended message of the “no parking” sign across the road is not “parking here will cost you,” let alone “you can park here but you’ll have to pay for it,” but “don’t park here.” If I choose to park and to pay the fine, I have not cut through to the law’s true, bottom-line message; I have ignored it. Now, given that some will not heed this message, and given that the law’s objective in imposing duties is ordinarily to prompt us to do as the duty directs, sanctions serve a useful purpose. The law wants me not to park here; indeed in making this obligatory, it is saying not just that I should choose not to park here but that the option of parking here is not open to me, so that there is, in a sense, no choice to be made. If, nonetheless, I remain open to the idea, the law’s threat to visit unwanted consequences on me is there to motivate me to choose differently. But, as this brings out, while sanctions are often necessary and often decisive, they remain a secondary means of the law securing conformity, practically significant only for those who do not accept, or are insufficiently motivated to act on, the obligation the law identifies.3
To think our duties are determined by the sanction the law imposes is to assume the law’s only language is coercion. But while this may be all some people hear, it is not the only language the law speaks. There are no doubt some who take nothing from the “no parking” sign save that they will have to pay some money if they do, who more broadly see the law’s penalties as simply prices. But this perspective offers not a clearer, sharper view of the law but one which is radically incomplete. What makes it incomplete is precisely its failure to account for the way the law purports to bind us, which is to say, its failure to accommodate, or even make sense of, the idea of a legal duty. So the proper implication of the approach Holmes proposed is not that my contractual duty is disjunctive—to perform or to pay damages—or conditional—to pay damages
2 See too H. L. A. Hart, The Concept of Law (3rd ed., OUP 2012) 38–42, 82–91, John Finnis, Natural Law and Natural Rights (2nd ed., OUP 2011) 320–30.
3 See too Leslie Green, “Introduction” in Hart (n 2) xxx: “Sanctions are the law’s Plan B.”
if I do not perform—but that I owe no duty (as the law understands it) at all. True, a court will order me to pay damages if I do not perform. But so what? If I have the option of breaking my contract, do I not also have the option of not paying?4 Of course, the law says I must pay, but then it also said I must perform. If I do not pay, the court will order my employer to deduct the sum I owe from my wages or authorize a bailiff to seize and sell such of my assets as needed to generate that sum. But as all this happens, I just stand by. In this race to the bottom, duty is redundant, the only reality to law is the brute application of state power.
II. Actionability, Enforceability
So Holmes’ account tells us nothing at all about legal duties. Now, the understanding of, for instance, contractual duties attributed to Holmes—that my duty is not to perform but only to pay damages—may yet be true. But if it is true, it is not by virtue of the sanctions attached to those who do not perform. Rather we need to look to what the law tells me I must do: does it tell me that I must do as I undertook (with damages then payable as the sanction for breach) or does it tell me that I am at liberty not to, so long as I compensate you for any losses I thereby cause? This is a distinction we just cannot see if we look only to sanctions. In answering this question, it is significant that the law, through the language and reasoning of legal officials and the broader structure of legal doctrine, presents performance of one’s contractual undertakings not as an option but as a duty, with failure to perform a breach of duty and damages its remedy.
Still, one might ask, if it is true that the law sees performance of contractual undertakings as not optional but obligatory, why is it so rarely prepared to take action to see that this happens? No doubt, there is often nothing that can now be done to secure performance. A court cannot compel timely performance when it is already late, some defects and harms cannot be undone. But even when performance remains possible and desired by the recipient, courts will typically limit a claimant to damages. Does this not undermine the suggestion that the law does indeed recognize a duty to perform? No, or at least not without more. The message “you must φ, but we won’t make you” is not equivocal. On the question of whether φ-ing is obligatory or optional, it is clear: it is obligatory. Nonetheless, while rejecting Holmes’ view that our legal duties
4 See too Finnis (n 2) 324.
are to be found in the sanctions the law attaches to us, we might yet think that these duties and sanctions should, where possible, match up. If the law’s position is that I must φ, then it should, at least all else equal, follow that through by taking what action it can to see that I φ. So, for many, the law’s failure to back up its duties with corresponding sanctions involves if not a contradiction then a sort of inconsistency: the law saying one thing and doing another. And, on this basis, we can see how the law’s failure to compel performance of its duties might be thought to call into question whether it really means what it says.
This view is not wholly wrong but it is too quick. To see why we need to recognize that our duties and the sanctions which attach to their breach answer different questions. What I should do (or what the law should tell me that I should do) is not the same question as what a court should compel me to do and we should not think that these different questions should receive the same answer. So for contracting parties the question is “must I do as I promised?” For judges charged with resolving claims for breach of contract, however, the question is “should I, as a state official and backed by its coercive machinery, now compel that defendant to perform?” There will sometimes be good reason for the state to tell us to do things which it would be unreasonable for it to compel us to do, and so it may be reasonable for the law to tell us we must perform our contracts even where it would not be reasonable to enforce that duty by compelling performance.5 To put the point the other way around: that it would, for example, be oppressive and costly to compel me to perform my employment contract are not reasons for the law to tell me that turning up for work is optional.
It is a mistake to think that our legal duties extend only so far as what the law compels. But, as we can now see, it is no less a mistake to think that the law should compel us to do whatever it says is our legal duty. Often, by the time the opportunity for the law to apply a sanction arises, circumstances have changed and performance of the original duty is not possible or, though possible, it may no longer be desirable or indeed reasonable. And where the defendant’s performance of that duty remains possible and reasonable, it may yet be unreasonable for the law to compel that performance. Indeed there are occasions where the proper response for the law to a breach of duty is to take no action whatsoever. Some breaches of duty are inconsequential. In such cases, there may be no need for the law to intervene, nothing for it to do if it did. The common law
5 See too Stephen A. Smith, “The Law of Damages: Rules for Citizens or Rules for Courts?” in Ralph Cunnington and Djakhongir Saidov (eds.), Contract Damages: Domestic and International Perspectives (Hart Publishing 2008).
sometimes resists this conclusion, preferring to allow a claim but to award only nominal damages, thereby giving some token expression to the wrong that has been done. But it is not the award of nominal damages which makes the conduct wrongful and where the law holds wrongs to be actionable only on proof of loss or some other consequence, it is not thereby saying that there was no wrong, no breach of duty unless loss results. Just as there is nothing contradictory or equivocal in the law saying “you must φ, but we won’t make you,” so there need be nothing contradictory, equivocal, or even inconsistent in it saying “you were wrong (i.e. in breach of duty) not to φ, but no action now lies against you.”
A good example of courts saying just this comes in cases of harmless negligence. The tort of negligence imposes duties of care, duties breached where reasonable care is not taken. Unless, however, the breach of duty leads to recognized injury, no claim arises. This rule of actionability is taken by some to reflect back on the content of the duty: the fact that liability arises only if and when injury occurs shows my duty is not to take care; rather it is a duty not to injure carelessly—a duty of non-injury rather than non-injuriousness— breached only if and when my carelessness results in your injury.6 But there is, to repeat, no reason at all to think that the fact the law provides no claim suggests, let alone entails, that there was in truth no breach of duty. Indeed the law’s position here—that there has been a breach of duty, but no claim lies—is not only clear but readily explicable. There is good reason for the law to tell me to take care when, say, driving: it is bad if people get hurt, driving is dangerous, I am more likely to hurt someone if I do not take care. Now, sometimes my careless driving will do harm, sometimes it will not. If it does, besides any penalty I may face, it is right that the law should provide my victim with a means of obtaining compensation from me. But what if no harm results? While it may still be reasonable to penalize me, there is no harm to undo or loss to make good. The law reasonably concludes, therefore, that no claim lies: my careless driving cannot itself be undone, there are no consequences to that carelessness which the law might demand I correct or offset.
6 See, e.g., Donal Nolan, “Deconstructing the Duty of Care” (2013) 129 LQR 559, 561–62; James Plunkett, The Duty of Care in Negligence (Hart Publishing 2018) 94–95; cf. John Gardner, “Obligations and Outcomes in the Law of Torts” in Peter Cane and John Gardner (eds.), Relating to Responsibility: Essays in Honour of Tony Honoré on his 80th Birthday (Hart Publishing 2001) 122; Arthur Ripstein and Benjamin C. Zipursky, “Corrective Justice in an Age of Mass Torts” in Gerald J. Postema (ed.), Philosophy and the Law of Torts (CUP 2001) 218–21; John C. P. Goldberg and Benjamin C. Zipursky, “Unrealized Torts” (2002) 88 Virginia LR 1625, 1652; Robert Stevens, “Rights and Other Things” in Donal Nolan and Andrew Robertson (eds.), Rights and Private Law (Hart Publishing 2011) 118.
The argument that duties of care are duties of non-injury sometimes channels a deeper misunderstanding. Thus far I have talked about breaches of duty without further qualification. But, as some are keen to stress, the duties private law deals in are specifically relational duties. My duties not to stamp on your toes or to trespass on your land are duties I owe to you. Accordingly, if I do these things, I do not just commit a wrong; I wrong you. These duties I owe to you correspond to rights you have against me and my breach of such a duty is simultaneously a violation of your rights. This may appear to tell against the view that duties of care are duties of non-injuriousness. For, so the argument goes, while there is no doubt a violation of your rights when my carelessness injures you, what right of yours is violated when I merely put you at risk? Do we have, in addition to our rights to our person and our property, a right not to be subjected to unreasonable risks? And, if we do, must we also then say that, when my carelessness does cause you injury, I commit not one wrong but two: one when I expose you to the risk of injury, another when that risk materializes? This looks a lot like double counting.
The trouble here comes from an indiscriminate use of the language of rights.7 Sometimes rights describe a sort of practical conclusion, identifying how others must act in their dealings with us. It is when understood in this way that rights correlate with and are equivalent to duties.8 Our contract puts me under a duty to you to hand over the shoes and this duty corresponds to a right you have that I hand them over. “I owe you a duty to φ” and “you have a right that I φ” are just different ways of saying the same thing. Accordingly, if I owe you a duty of care when driving, you necessarily have a right that I drive carefully. The right does not help us make the case for or against me owing that duty; it just gives us an alternative way of asking the question and expressing our conclusion. At other times, however, we see rights used to capture a different idea. Here rights do not provide an alternative expression of the duties we owe each other, rather they identify a reason we might owe these duties, telling us not, or not just, what duties we owe but why we owe them. So, on one popular view, for you to have a right is for some aspect of your well-being, some interest of yours, to be sufficiently important that its protection or advancement justifies imposing duties on others.9 Call these respectively rights1 and rights2.
We can see why some doubt that we have rights2 not to be subjected to unreasonable risks. Someone who simply endangers me does not appear to set
7 See further Charlie Webb, “Three Concepts of Rights, Two of Property” (2018) 38 OJLS 246.
8 See, e.g., Wesley Newcomb Hohfeld, Fundamental Legal Conceptions as Applied in Judicial Reasoning (Walter Wheeler Cook (ed.), YUP 1919) 39.
9 See, e.g., Joseph Raz, The Morality of Freedom (OUP 1986) 180.
back any interest of mine, does not compromise my well-being. Risk matters only because harm matters, an unrealized risk is a harm to no-one. If this is correct, we cannot justify duties of care on the basis that we have rights2 not to be subjected to unreasonable risks.10 But the truth is that they do not need this justification. For your interest in, and so your right2 to, your bodily safety is more than enough to justify the conclusion that I must avoid acting in ways which unjustifiably threaten to injure you. Moreover, this duty to take care is a duty I owe to you, whatever view we take on what this means: the duty serves and is grounded in your interests; it is one you are empowered to waive or, where possible, enforce. Accordingly, the duty I owe you to take care, say when driving, corresponds to a right1 of yours that I drive carefully. In short, you do not need to have a right2 not to be exposed to unreasonable risks to have a right1 that I not expose you to unreasonable risks (e.g., when driving); your right2 to your bodily safety is itself sufficient to ground this right1 and my corresponding duty of care.
As this brings out, we need to take care when thinking about the law of torts, and private law more generally, in terms of rights. Take the following three propositions:
1. The duties recognized in the law of torts are enforced by claimants bringing private actions, seeking private remedies, against defaulting defendants.
2. These private duties and rights of action advance or protect important interests of these claimants.
3. There can be no breach of such a duty, no such wrong, unless and until some interest of the claimant’s is affected or set back.
These three propositions are not equivalent, nor do they follow from one another. 1 can be true without 2 or 3 also being true. 1 and 2 can, as we have seen, be true without 3. But framing them each in terms of “rights” invites their elision or at least suggests, wrongly, that they come as a package deal: torts are breaches of duties owed to others, as is evident from the basic form of tort litigation, therefore torts are rights-violations (since a breach of duty just is the violation of a right), therefore there can be no tort unless and until some right of the claimant’s is violated and injury is caused. The consequence is that “rights-based” accounts of private law often move far too quickly from accurate
10 Is it correct? For an argument that it is not, see John Oberdiek, Imposing Risk: A Normative Framework (OUP 2017) 93–130.
observation of the form and structure of private law actions to claims about private law’s proper content. The truth, I think, is that the basic form of private law litigation places few, if any, constraints on the sorts of reasons (principles, policies) which might bear on the range of practical questions private law addresses and so on the content of the duties private law recognizes.11
III. Performance Damages
We have seen that for me to owe a legal duty to φ entails neither than the law will take further action to see that I φ nor that it should. The conclusion that I was in breach of duty is a determination of how I should have conducted myself at some earlier point in time. But the central practical question for a court charged with resolving a claim founded on that breach of duty is not what I should have done then but what it should do now. These questions may receive different answers either because circumstances have now changed, so that what was reasonable to expect of me then is not reasonable now, or because there are considerations which bear on the reasonableness of the court compelling me to φ which do not bear, or bear differently, on the reasonableness of me φ-ing.
Sometimes it is too late to see that I φ as φ-ing is no longer possible. Here, necessarily, any action the law takes must be directed toward achieving some other result. Even if it remains possible for me to φ, it may be that the passage of time and the changed circumstances in which we now find ourselves mean that the balance of reasons no longer supports this and so the law should not take action to see that I φ, even if it could do so cost-free. Say we contract for me to give you a lift to the airport tomorrow. You have a flight to catch, taking you to a friend’s wedding overseas. (I may know this, I may not.) I don’t show up. Should a court order me to take you to the airport at the next available opportunity? You have missed your flight and missed the wedding, you have no interest in going to the airport for any other reason. Whether this is because the reasons that supported my initial obligation to take you to the airport no longer apply or because other reasons—you no longer want that performance, indeed you would now be an unwilling party to its performance—override
11 For a fuller argument to a similar conclusion, see John Gardner, “Backwards and Forwards with Tort Law” in Joseph Keim-Campbell, Michael O’Rourke, and David Shier (eds.), Law and Social Justice (MIT Press 2005).
those reasons, the upshot is that it would be unreasonable now for a court to order me to do this.
Even where it remains possible for me to φ and unreasonable for me not to, the law may still reasonably decline to take action to see that I do on account of the costs of taking such action. Here we see the standard reasons the courts give for refusing specific relief: compelling performance would be too great an infringement of the defendant’s liberty, would be too costly to police, would invite further acrimony and litigation, and so on. In such instances, the courts are not saying that it is reasonable for me not to perform, only that it would be unreasonable for them to take further action to see that I provide that performance. On occasion, the denial of specific relief may be based on a blend of both sorts of consideration. I covenant with you not to build on the vacant plot of land I own adjoining yours. Nonetheless I do. By the time the case comes to court, the work is already complete; an apartment block sits on the once-vacant plot. It is not too late to return the land to its prior undeveloped state, the block could be demolished, performance of the covenant could, to this extent, still be secured. But doing this would mean the destruction of good housing at a time when housing is in short supply. The fact that I cannot now perform my covenant without significant social cost is a consideration which rightly bears on whether such performance should now be considered obligatory, even reasonable. (Some influential accounts of private law disagree. Their wilful blindness to the relevance of such considerations testifies to their inadequacy.) And, even if these costs are not sufficient to defeat any obligation to return my land to its prior state, such that this remains, all considered, what I really ought to do, we might still conclude that it would be unreasonable for a court now to take action to secure this, intervening to order that these homes be destroyed.12 Nonetheless, when it comes to deciding how the law should respond to breaches of duty, the law’s determination that, on the balance of reasons at the time, I was under a duty to φ counts for something. Though what I should do will sometimes change by the time the law has the chance to respond, often it will not. The reasons which make it unreasonable for me to play my music loud late into the night do not diminish if I have now been doing this for weeks. It is unreasonable to drive my car onto your foot and it remains just as unreasonable to keep it there once I have. And while there are always costs involved in the law intervening, there are also benefits to be had, not just in securing my conformity with my duties but also in encouraging others’ conformity with theirs. So, it is not surprising that, where the costs of compelling performance
12 Cf. Wrotham Park Estate Co. Ltd. v. Parkside Homes Ltd. [1974] 1 WLR 798 (Ch.).
are low, and certainly no higher than any other meaningful remedy, the courts will tend to do so. We see this in their routine enforcement of duties to pay money, which do not raise the same concerns about supervision and forced labor which attend many other forms of specific relief.
It may be that the law is sometimes too cautious here, too ready to cede to worries about the costs of compelling performance, too slow to see its benefits. The common law’s default approach outside duties to pay money is to consider specific relief only where damages would be inadequate, rather than to look beyond specific relief only where it would be too costly. Much depends on how good a substitute for specific relief damages provide. As typically formulated, the standard aim of damages is to put the claimant in as good a position as she would have occupied but for the breach. But this formulation is ambiguous.13 One way to put a claimant in a position as good as had there been no breach is through compensation. Even if the specific injuries and disadvantages the claimant has incurred cannot or will not be reversed, they may yet be offset or counterbalanced by providing the claimant with alternative benefits—benefits distinct from and additional to those which the defendant’s performance of her duty would have secured—which advance her interests, considered in the round, to the equivalent extent as the defendant’s breach set them back. (This assumes the possibility of a sort of netting of our diverse interests.) So the pain and suffering my breach caused you cannot be undone, the assets I damaged cannot be repaired or replaced, but by paying you damages I provide you with funds which then enable you to purchase other items you value, satisfying other interests of yours. The upshot is that by adding such advantages to the disadvantages caused by the breach, you can be placed in a position which, though different, is no worse than the position you would have been in had I performed my duty.
But there is another, and in some ways more straightforward, way of putting the claimant in as good a position as she would have occupied but for the breach: namely to ensure that she obtains the very advantages which the defendant’s performance of her duty would have given her. This is, of course, the effect of granting specific relief: my duty is to hand over the shoes and an order of specific performance sees that you get those shoes from me. But much the
13 See further Charlie Webb, “Justifying Damages” in Jason W. Neyers, Richard Bronaugh, and Stephen G. A. Pitel (eds.), Exploring Contract Law (Hart Publishing 2009). The distinction, drawn there and here, between different ways a claimant can be put in as good a position as had the defendant performed her duty largely mirrors the distinction identified in Adam Slavny, “Negating and Counterbalancing: A Fundamental Distinction in the Concept of a Corrective Duty” (2014) 33 Law and Philosophy 143, though see n 32 and surrounding text for discussion of examples where negation of the claimant’s harms is possible though performance of the defendant’s primary duty is not.
same result can also, on occasion, be achieved through an award of damages. You contract to buy 1,000 widgets from me. The widgets I supply are defective. You sue. The remedy you will be awarded is damages assessed at the difference in market value between the widgets I supplied and those you contracted for. These damages may appear straightforwardly compensatory. I have not supplied you with the promised widgets, so the law orders me to pay you a sum of money which can apply to acquire other benefits which will offset the losses this has caused you. But the award can also be viewed in a different light. You still want the correct widgets. You can still get them by going back into the market and buying some from another supplier. The damages award is premised on you doing just this.14 If you sell the defective widgets I supplied, you get some of the money you need to buy replacements; the damages ensure that the additional money you have to provide to make up the difference is recovered from me. In so doing the award covers the expenses you incurred as a result of my breach. But, more than that, it sees that, at the end of the day, you obtain—or would have obtained had you taken the steps the law expected you to take—the (or such) widgets as I undertook to supply and that it is me who funds this. The end result is not simply that I compensate you for the losses you suffer as a result of my breach, but that I supply you with some part of the performance you were, and remain, due.
Here these two functions of damages awards run together: an award aimed at compensating you for your losses, quantified by reference to the difference in value between the widgets you contracted for and those I in fact provided, also enables you to obtain the widgets you wanted from another source. But we get a clearer sense of the true distinctiveness and significance of damages aimed at securing partial performance where the two come apart. I contract to install a new kitchen in your home. You go away while I do the work. When you return you see that the units I have installed are not the color you had asked for. The cost of removing these units and putting in the right ones is substantial. The difference in market value between the two sets of units may well be nothing, however, and, while you have a preference for the color you chose, the difference in value to you—as reflected in how much less you would have paid for units I in fact installed—falls well short of the sum it would take to replace them.
My breach here, though it leaves you no poorer in balance-sheet terms, does indeed put you in a worse position than had I performed, your interest in
14 This aspect of these awards is brought out in Andrew Dyson and Adam Kramer, “There is No ‘Breach Date Rule’: Mitigation, Difference in Value and Date of Assessment” (2014) 130 LQR 259.
having a kitchen which meets your esthetic preferences set back. But to compensate you for this loss, we need only to provide you with a sum of money which reflects this difference in the value you place on the kitchen as promised compared to the kitchen as delivered, a sum which then allows you to purchase other benefits, advancing other interests of yours, sufficient to offset the setback I have caused. Such an award would see that you are left in no worse/ as good a position as had I performed. What it would not do is to provide you with the kitchen you contracted for. This end result remains possible, however, and, if the court is not prepared to compel me to do this work myself, it could still see that I put you in the position to get the kitchen you wanted by funding someone else to come in and complete the job. This is what a cost of cure damages award would achieve.
It is, therefore, a mistake to see cost of cure as, or as simply, one possible measure of or formula for calculating a claimant’s losses and hence the compensation she is due. True, the award of cost of cure damages is premised not only on the claimant’s continuing right to the defendant’s performance but also on this money then being used to secure that performance. The effect of the award is therefore to cover the out-of-pocket expenses which the claimant has incurred or is expected to incur. But to see such awards as compensation for these expenses sets us off on the wrong foot. We can see this from the English cases.15 Cost of cure damages are awarded only where the court concludes that this is the true measure of the losses suffered by the claimant.16 This will be the case where the cost of cure is equal to or lower than the difference in value between the performance promised and the defective performance in fact provided. Where, however, the cost of cure exceeds the difference in value, it will be seen as the proper measure of the claimant’s losses only where the court considers that it would be reasonable for the claimant to incur such costs. This is effectively an application of the mitigation principle: claimants are expected to take reasonable steps to mitigate their losses; if they do not, their damages will not extend to the losses they would have avoided if they had. By the same token, when determining whether a claimant who has yet to get the defective performance put right should recover as damages the sum it would take for her to do so, the court must consider whether incurring such costs would be reasonable. Accordingly, cost of cure damages will be refused where the court
15 For a fuller version of the argument made in this section and further analysis of these cases, see Charlie Webb, “Performance Damages” in Graham Virgo and Sarah Worthington (eds.), Commercial Remedies: Resolving Controversies (CUP 2017).
16 Ruxley Electronics and Construction Ltd. v. Forsyth [1996] AC 344 (HL).
considers that these costs are disproportionate to the benefits that would be derived from having this work done.
But the mitigation principle only kicks in if what you are seeking is compensation for your losses. When we are dealing with a claim for compensation, we are in the business of determining what losses you have suffered, or are likely to suffer, and who should bear them. Where there are losses which you could reasonably have avoided or could yet avoid, the law might reasonably conclude that they should be borne by you and not shifted onto me. But the case for awarding you cost of cure damages set out here is not premised on this being the proper measure of the losses you have suffered as a result of not receiving the performance you were promised. On the contrary, the award is designed to see that you obtain (the better part of) that performance. Here the question for the court is not who should bear what losses but whether I should be ordered to provide you with means to secure the performance you were, and are still, due. Your right to performance, and my duty to perform, is not conditional on this being the most cost-effective way I might advance your interests. That the cost to me of performing the contract exceeds the benefits you take from its performance does not make it any less my duty to do as I undertook. And since, or so long as, this performance remains my duty and given the enforcement costs, in terms of administration and supervision, are no higher here than those which attend other damages awards, the courts should not channel you toward compensation simply because this is the cheaper option for me.
It does not follow that the fact that compensation is the cheaper option is irrelevant. The burden placed on defendants required to meet the costs of funding repairs and replacements is a consideration which bears on the reasonableness of compelling them to do so and there will be times where not only is the extent of this burden disproportionate to any benefit that comes from having this work done but it is also unfair to the defendant to make her bear it.17 By framing the question as one of mitigation, the courts misidentify the relevant question here, which is not whether the claimant is acting reasonably so as to limit her own losses but rather whether it would be unreasonable to require the defendant to meet the costs of performing her duty. The consequence is that the courts move too quickly from the position that damages are a preferable form of remedy to specific relief to the conclusion that any remedy is properly directed to, and so should be limited by reference to, compensation of the claimant’s losses. As a result, the law is doing not only less than it could but less than it should to secure performance of these duties.
17 Jacob & Youngs, Inc. v. Kent 230 NY 239 (1921) may be such a case, though this is not the only way we might support the decision.
IV. Continuity
Of course, however creative the courts are in finding ways to see that I perform my duties, there will be cases where performance is now impossible. In such instances, any action the law takes against me must have some other aim. Where I can no longer perform, or where a court is now unwilling to take action to secure that performance, I might yet be able to undo or offset some of the consequences of my breach, most clearly by compensating you for the losses I have thereby caused. But, just as my contractual duty is not the disjunctive duty often associated with Holmes—to perform or to pay damages—so the court’s order that I compensate you for the losses caused by my failure to perform is not an alternative means of securing that performance. Rather, the effect, if successful, is that my duty remains unfulfilled but you are left no worse off for it.
Of course, even on this view, these duties are not independent. It is my breach of my initial (primary) duty which triggers and grounds the (secondary) duty to compensate you for the losses caused by that breach.18 On occasion, the relationship between the primary duty I breached and my secondary duty to pay compensation is closer still. Imagine that our contract for the installation of the kitchen provides expressly not only for the work I must do but also for the compensation I must pay if I do not. My duty to install the kitchen and my duty to compensate you for your losses if I do not install it remain distinct duties, detailing different actions and bringing about different results. And, as before, my failure to do the work to the specifications we have agreed is, without more, a breach of contract, and it is this breach which then triggers my duty to compensate you for the losses this causes. Here, however, the payment of compensation is provided for in the contract, my duty to pay that compensation is what performance of our contract now requires, your right to performance now a right to that compensation.
Some think that this is the rule rather than the exception: that secondary duties are not simply triggered by the breach of a primary duty but are always versions of that primary duty, remodeled in light of the effects of the breach; or, at least, that the primary duty always continues in some form, always prescribing some sort of secondary, corrective duty. To say this is not simply to say that these secondary duties arise only where and to the extent that the primary duty is unperformed, such that they, in this sense, stand in for the primary duty. This much is true on any view. What is distinctive about the claim that these are remodeled versions or continuations of the primary duty is not what it reveals about the incidence or
18 Which is not to say that all duties to compensate are triggered by breaches of duty. The fact that I acted wrongfully in causing you loss is not the only reason why it may be just to require me to bear it.
function of secondary duties but what it suggests about their justification. Where our contract expressly provides for me paying you compensation in the event of breach, both my primary duty to do the work and my secondary duty to pay compensation are grounded in the undertakings I made to you; both applications or expressions of my broader duty to do as I undertook.19 The reasons I came under a duty to do the work are the very same reasons I now owe you a duty to compensate you for your losses.
This gives us what Gardner calls the continuity thesis:20
Once the time for performance of a primary obligation is past, so that it can no longer be performed, one can often nevertheless still contribute to satisfaction of some or all of the reasons that added up to make the action obligatory. [Once we know the rationale of the primary obligation] we also have the rationale, all else being equal, for a secondary obligation, which is an obligation to do the next-best thing. . . . [T]he secondary obligation is a rational echo of the primary obligation, for it exists to serve, so far as may still be done, the reasons for the primary obligation that was not performed when its performance was due.
And so:21
The normal reason why one has an obligation to pay for the losses that one wrongfully occasioned (ie that one occasioned in breach of obligation) is that this constitutes the best still-available conformity with, or satisfaction of, the reasons why one had that [primary] obligation.
Is the continuity thesis sound? No doubt reasons do not disappear simply because they are not acted on and are not conformed to. Again, the reasons not to play music loud late into the night are not extinguished by my doing so. If
19 This reveals the limitations, or perhaps simply the error, of Gardner’s claim that duties are individuated by the actions they make obligatory (John Gardner, “What is Tort Law For? Part 1. The Place of Corrective Justice” (2011) 30 Law and Philosophy 1, 29–30), while supporting his conclusion that, if secondary duties are a continuation of primary duties, the essential continuity is of the reasons which ground these duties (ibid. 30–34).
20 Ibid. 33. Others who have put forward similar views express the relevant continuity in terms of rights rather than reasons: Ernest J. Weinrib, Corrective Justice (OUP 2012) 87–98; Arthur Ripstein, Private Wrongs (HUP 2016) 233–54. As noted above (text to nn 7–9), there are different concepts of rights. While these accounts reveal some inconsistency in their use of the term, the claim that my secondary duty to compensate you for the losses caused by my breach is a continuation of your pre-breach right can work only if the “right” here describes a sort of reason. As such rights-continuity is best seen as an example of reasons-continuity.
21 Gardner (n 19) 33–34.
I have started, I should stop. It is also clear that changes in my circumstances can make a difference to what I have reason to do. If it is raining I have reason to put up my umbrella, if it is dry I (usually) do not. Moreover, what I have reason to do may change without my reasons changing. The same reason can support different actions depending on my circumstances. The fact that it is raining is a reason not only for putting up my umbrella when I am walking to work, but also for choosing to take the bus rather than walking or for delaying my journey. Indeed I think we should say that, while changes in the weather make a difference to whether I should put up my umbrella, this is not because the reasons which count in favor of putting up an umbrella only bear on me on rainy days. What changes is not the applicability of these reasons but rather their application: what, if anything, they actually count in favor of here and now. How, one might object, can the fact that it is raining be a reason (for putting up an umbrella, for anything) when it is not, in fact, raining? But the fact that it is raining does not give a reason (for putting up an umbrella, for anything) in isolation, but only in conjunction with some value which is served by responding to this fact in this, or some such, way. So the reason for putting up an umbrella in the rain is not the bare fact of the rain falling but, say, that it is good to be comfortable and healthy and getting cold and wet is uncomfortable and risks illness.22 While the rain comes and goes, the value of being comfortable and healthy does not, and the reasons it provides are ever-present even if, right now, there is nothing I need do to conform to them.
So the continuity thesis is, to this extent, true to the way reasons work: a failure to conform to the reasons which apply to us does not make those reasons disappear and changes in our circumstances, such as those brought about by our breaches of duty, may make a difference to what these reasons give us reason to do. Some reason grounds a duty on me to φ at T1. In breach of that duty, I do not. I cannot now act in accordance with that reason at T1; history cannot be rewritten. But that does not mean I cannot act in accordance with that reason now, at T2. We know that reason continues to apply to me. We also know that I do have reason to compensate you for any losses my breach caused. Still, what we need to know is whether the two marry up: whether it is the reason which grounds my duty to φ at T1, and which still applies to me, which gives me reason to compensate you at T2
Some respond: how could it not? It would be “absurd,” Weinrib suggests, were my duty to be discharged, your right destroyed, by its breach.23 For
22 Making this point and with this example: John Finnis, “On Hart’s Ways: Law as Reason and as Fact” (2007) 52 Am. J. Juris. 25, 44–45.
23 Weinrib (n 20) 90.
Ripstein, there would be “no sense” in which you have a right, if it could be unilaterally dissolved by its violation.24 And so, for both of them, it must be that the payment of damages is the performance that duty now requires, the demand your right now supports, for otherwise that right and duty would come to nought. But none of this is true. I kick you on the shin at T1. I cannot now, at T2, undo the kick at T1. The fact that it cannot be undone does not make it any less a breach of duty, any less a violation of your right. That would be absurd. Moreover, the reasons which supported my duty not to kick you at T1 continue to apply to me at T2, and I remain under a duty not to kick you. Accordingly, before we get to any question of what I might owe you for the harm I have done you and why I owe it, there is no sense in which my duty, your right, or the reasons which ground them have been discharged, destroyed, or dissolved. This way of thinking might seem to have greater purchase where the effect of my breach is that, now at T2, I can no longer do what I was duty-bound to do at T1. I snatch your coffee and pour it away. Again, I cannot now undo what I did. But while I could at least avoid kicking you again, here the coffee is gone and there is nothing left of my duty not to interfere with it. Accordingly, we can say that, as a result of my breach, I do not owe you a duty I would otherwise have owed. But this much is true on any view: Weinrib and Ripstein do not think I still owe you a duty not to spill what is already spilled. The relevant duty I owe you now is to compensate you for your losses. Again, however, nobody doubts this. What distinguishes Weinrib and Ripstein’s position is their claim that this duty does not just stand in for my duty not to spill the coffee but is the very same duty in revised form. But it is hardly absurd or senseless to question whether this is true.
So what arguments can be made in favor of the view that these are the same duty, grounded in the same reasons? Weinrib and Ripstein proceed by showing how my duty to compensate you can be viewed as an alternative expression or effectuation of the right of yours which grounded the duty I breached. So, for Weinrib, your right can be “restored” not only “qualitatively”—as when the remedy sees that I provide the very performance I was initially duty-bound to provide—but also in “quantitative form,” whereby you obtain not that performance but its monetary equivalent as damages.25 For Ripstein, the rights recognized by private law mark out and protect our “means”: things (property) and attributes (our bodily and mental powers) which are ours to employ in the pursuit of our chosen ends.26 When I wrong you, injuring your person, taking
24 Ripstein (n 20) 248.
25 Weinrib (n 20) 94.
26 Ripstein (n 20) 29–33.
or damaging your assets, I deprive you of some of these means. The payment of damages sees that I restore equivalent means to you.27
I doubt this takes us much further. Describing your right to damages as just the quantitative form of your right that I not kick your shin or spill your coffee only begs the question. Similarly, the idea that damages provide you with means equivalent to those I deprived you of is an effective demonstration less of how those damages protect your initial right than of how they are able to offset the losses caused by my violation of that right. Indeed, it is worth repeating the point made at the outset: my primary duty is not that I either not kick you or pay you damages. Equivalently, your right is that I do not kick you, not that I ensure that you receive full value for any kicks I give or that I preserve the sum of means at your disposal. Of course, Weinrib and Ripstein are not closet Holmesians. The quantitative form of the right and duty comes out only after breach. Till then, there is no qualitative dimension to my duty or your right, no way in which they view your means as fungible. But, once we reject Holmes’ reductionism, the stretch to see paying damages as a sort of next-best performance of my primary duty looks unmotivated, unnecessary.
Gardner makes a different case. He starts with an example.28 I promise my children that I will take them to the beach today. However I forget or some distraction intervenes. What should I now do? Take them tomorrow or, if not tomorrow, then at the next possible opportunity. Why? Not, so Gardner sees it, because I thereby keep some part of my promise. What I promised to do—to take them to the beach today—was one thing, what I am to do instead—taking them to the beach tomorrow—something else. Nonetheless what is required of me now closely resembles what was required of me before. Can it be, Gardner asks, that:29
I [am] bound to take them to the beach tomorrow for reasons that are entirely different from the reasons that I had to take them to the beach today? Surely not. Why me? Why the children? Why the beach? Why tomorrow?
Accordingly my taking them tomorrow amounts not to partial performance of my primary duty but rather to partial conformity to the reasons which ground that duty. This, Gardner believes, then provides a general model for secondary duties: wherever I breach a duty, the reasons which grounded that primary
27 Ibid. 245–46.
28 Gardner (n 19) 28–29.
29 Ibid.
