University of Notre Dame

Specific Performance: On Freedom and Commitment in Contract Law

March 25, 2023

View PDF



Hanoch Dagan* & Michael Heller**

When should specific performance be available for breach of contract?  This question—at the core of contract—divides common-law and civil-law jurisdictions and it has bedeviled generations of comparativists, along with legal economists, historians, and philosophers.  Yet none of these disciplines has provided a persuasive answer.  This Article provides a normatively attractive and conceptually coherent account, one grounded in respect for the autonomy of the promisor’s future self.  Properly understood, autonomy explains why expectation damages should be the ordinary remedy for contract breach.  This same normative commitment justifies the “uniqueness exception,” where specific performance is typically awarded, and the personal services exclusion, where it is not.  For the most part, the boundaries of specific performance in the common law track our underlying commitment to autonomy.  But not entirely.  There’s still work to be done on both sides of the common‌/civil-law divide, and this Article points the way with doctrinal reforms that can better align specific performance with its animating principles.


When should specific performance be available for breach of contract?  Civil-law systems make it the primary remedy for breach, while the common-law jurisdictions treat it as a humble exception with limited application.  The same sharp split exists in legal theory on this foundational question of contract, with philosophers tending to endorse the civil-law tradition and economists praising the common-law.1  The net result: the law and theory of specific performance oscillates between incompatible values and reforms.

There is a better approach.  This Article shows how autonomy, rightly understood, explains when specific performance should be available, and it offers well-grounded reforms that can bring doctrine closer to its animating principles.  By anchoring this contract remedy in a conceptually coherent and normatively attractive framework, we can break the decades-long logjam in one of the field’s foundational debates.

We understand our task in this Article as an exercise in charitable interpretation and reconstruction of private law.  Like other reconstructive efforts, our reading builds on existing practices, reaffirms much of existing law, and offers targeted proposals for justified reforms.  We do not pretend to divine the intention of the lawmakers and judges who developed the doctrine we analyze.  We do not focus on the accidental historical origins of common-law specific performance in equity courts.  Nor do we claim that the common law is, in any systemic sense, superior to its civil-law counterpart—indeed, in another private-law context, we’ve criticized the common law and shown how the civilian tradition is truer to the law’s liberal commitments.2  In this Article, what we do offer is an understanding of how freedom and commitment drive contract law.  We do this by decoding specific performance and grounding it firmly in the most fundamental normative commitments of contract law in a liberal polity.

At root, contract is an empowering practice that is, and should be, guided by an autonomy-enhancing mission.3  Contract’s operative doctrines—including the choice of remedy—allow people legitimately to recruit others to their future plans by committing their own future selves in return.  This commitment necessarily curtails the self-determination of the promisor’s future self—and it’s the key to understanding specific performance.

Contract-keeping is justified because and only to the extent that the claimed dominion of the present self over the future self can itself be justified.  This seemingly simple proposition encapsulates both the moral premise of the common law and its challenges.  The common-law baseline of compensation—and not specific performance—serves as a stronghold for the autonomy of promisors’ future selves.  Covering a promisee’s expectation interest is qualitatively less imposing on the future self’s self-determination.  Therefore, other things being equal, contract’s autonomy-enhancing mission requires that disappointed promisees should be entitled to damages, rather than specific performance.

The requirement that “other things be equal” must be unpacked.  This is the terrain on which we do the analytic work to delineate specific performance’s proper boundaries.  Most contracts can achieve their mission of facilitating promisees’ plans by liquidating breach into money.  Where this is true, allowing the promisor’s current self to encumber her future self with the obligation to specifically perform, rather than to cover the promise’s value, cannot be justified by reference to her self-determination.  This means, at least for liberal contract law, that it cannot be justified, period.

Other things, however, are not always equal.  The first challenge therefore is to identify categories of cases in which liquidating the promisor’s performance does significantly frustrate contract’s function as a planning tool.  Those categories, at least a priori, do justify specific performance.  For the most part, the common law correctly identifies these categories.  A second, related challenge is to help parties signal cases in which they consider the contract’s actual performance significant for their own particular plans, even though their contract does not fall within the usual categories.  Here, the common law falls substantially short.  Parties have a hard time ensuring they will get specific performance when that’s what they want from their contracts.  The third challenge pushes in the opposite direction.  Here, contract law faces categories of cases, notably involving employment, where specific performance is bound to threaten the self-determination of the future self to such a degree that it cannot be justified—even if excluding such a remedy diminishes contract’s empowering potential.

These three challenges in turn yield three practical takeaways.  First, we show how the so-called “uniqueness” exception in the common law—covering cases in which specific performance is regularly provided—can and should be refined so it more carefully tracks its (reconstructed) normative foundation.  As a practical matter, this means, for example, distinguishing in real estate transactions between sales of residential and commercial property, and between breaches by sellers and buyers.  Second, we criticize the resistance of the common law to parties’ attempts to opt into specific performance (and to penalty clauses, as we’ll show) to remedy breach of their contract.  One instance where specific performance should be more readily available is when an employer breaches a promise to continue to employ an employee.  Finally, we highlight what may well be the most challenging normative question for liberal contract law: how to address cases in which empowering the current self’s constitutive exercise of self-authorship threatens self-determination of the future self.  Noncompete agreements vividly illustrate the problem that arises when law overly facilitates the current self’s pursuit of welfarist interests.  We show the path forward.

Let us state our bottom line plainly.  From the standpoint of autonomy, (1) specific performance must not be the default remedy; (2) specific performance should nonetheless be available where monetary recovery cannot substantially avoid the disruption breach causes to a promisee’s plan; (3) translating #2 into a workable rule implies that specific performance should be the default if the promisee is a buyer of a unique good for personal use, paradigmatically, a personal residence; (4) because #3 is only a proxy for #2, parties should be able to opt into specific performance, so long as they do not violate #5; (5) specific performance should not be awarded against providers of personal services.

The dominant economic analysis of specific performance also arrives at these five principles, raising the question whether our account is just economic analysis in disguise.  It is not.  Our five principles derive from contract’s autonomy-based telos and lead to a distinct reform agenda.  In our account, #2 and #3 are normative defaults.  Contra the economists, they do not arise from or depend upon the current majoritarian preferences of contracting parties (as we discuss in subsection II.B.3 below).4  On #4, relational justice constrains party opt in, an autonomy-regarding limit missing from the economic account.  Finally, #5 yields a mandatory rule.  Contra the economists, it is not contingent on people’s imperfect foresight, which technology may ameliorate.

Taken together, these normative commitments constitute an autonomy-enhancing law of specific performance—one that supports contracting parties as they trade off terms and remedies in the service of their self-determination.  We show that the common-law baseline5 deserves moral praise—contra the economists’ pragmatic apology and the philosophers’ moral condemnation.6

Our account suggests a new perspective on specific performance, one that offers useful legal reforms and highlights new questions that must be addressed if contract law is to be fully loyal to its liberal commitments.  Part I briefly outlines the existing terrain of specific performance; Part II shows how autonomy, particularly, the challenge of respecting the future self, unlocks the key to understanding specific performance; Part III lays out the doctrinal and reform takeaways.

I.     The State of the Art in Specific Performance

We briefly outline the existing terrain of specific performance, first considering its evolving place in the common law, and then critically reviewing the sharply conflicting legal economic and philosophical responses.  Although the common-law doctrine has been quite stable, its economic advocates have been on the defense of late.  They argue against expanding specific performance primarily on pragmatic grounds and to respect contractors’ likely preferences ex ante.  Philosophers have the law against them, but more momentum analytically.  Neither side though gets specific performance right because both miss the normative underpinnings that justify its legitimacy in the first instance and that actually drive its direction in a liberal polity.

A.   The Common Law in Comparative Perspective

1.   The Civil Law Comparison

We start with the civil-law tradition, which offers a stark contrast to the common law.  Both systems agree contractors are expected to perform their contractual promises, all else equal.  In civil-law systems, this truism is understood to suggest that specific performance will be granted as a matter of course such that courts “literally” order the breaching party to “complete the contract” thus “replicat[ing] the entirety of the claimant’s threatened right.”7

This approach is indeed the basic principle of modern civil law systems.  As one commentator argues, because “the promisor is obligated to perform his duty under the contractual obligation,” the promisee has, in the case of breach, a “right to enforce this duty, while it is possible and conscionable.”8  Indeed, the claim to specific performance is regarded in these systems as “the normal remedy,” which is an “inherent and normal right flowing from a contract.”9

While there are differences, at times significant, among different civil law jurisdictions,10 what matters for our purposes is their common denominator: “in civil-law systems, the right to performance is asserted to be a fundamental right of a creditor, emanating from the adagium ‘pacta sunt servanda’ itself.”11

Common-law systems begin from nearly the opposite starting point.  There are again subtleties that somewhat qualify the drama, but Ernst Rabel’s old description of the gap between these two great traditions as an “abyss” still holds, by and large.12  As Stephen Smith recently noted, courts and commentators in common-law jurisdictions “often describe specific performance as a ‘secondary’ or ‘exceptional’ remedy for breach of contract”; and indeed “specific performance is rarely available for the breach of a contractual duty to provide goods or services, even where that duty remains alive.”13

This stark doctrinal difference between common- and civil-law systems appears to be eroding, however, at least in certain practice areas.14  Convergence in practice is not surprising given the shared liberal commitments of both legal families and the overlapping commercial ties among contracting parties across the legal divide.

2.   The Common-Law Baseline

The Restatement (Second) of Contracts declares that after a contract has been breached, the injured party is entitled to damages as a matter of right.  Even if the breach creates no loss, the injured party will still be awarded nominal damages.15  By contrast, specific performance, as the Restatement notes, is “granted in the discretion of the court,”16 and it “will not be ordered if damages would be adequate to protect the expectation interest of the injured party.”17  The Restatement’s list of “Factors Affecting Adequacy of Damages” may lead to the impression that the availability of specific performance is, or at least may be, relatively wide.18  But as commentators have repeatedly noted, the main exception to the baseline rule of no specific performance involves sales of unique goods or land.19

The Uniform Commercial Code (UCC) sought to expand this exception, allowing that specific performance may be decreed both “where the goods are unique,” and “in other proper circumstances.”20  Importantly, the UCC has added output and requirement contracts as categories for which specific performance is available.21  Beyond that, though, the UCC has had little practical effect.  The Code vaguely notes that the uniqueness test is to be determined by “the total situation which characterizes the contract”22 and adds that the category of unique goods includes “property which is not readily obtainable due to scarcity.”23  While most courts do read the UCC’s language as indicating the Code’s relatively “liberal attitude”24 toward specific performance, this liberalization by and large does not depart from the common law’s parameters.25

The final component of the common-law doctrine—one that is shared by its civil-law counterpart26—involves employment contracts.  As the Restatement categorically states, “[a] promise to render personal service will not be specifically enforced.”27  Furthermore, to ensure that contracts of this type are not being de facto specifically enforced, the Restatement adds that 

[a] promise to render personal service exclusively for one employer will not be enforced by an injunction against serving another if its probable result will be to compel a performance involving personal relations the enforced continuance of which is undesirable or will be to leave the employee without other reasonable means of making a living.28

B.   The Best Economic Account and Its Limits

One of the hallmarks of contract theory is its practitioners’ long-standing fascination regarding the proper scope of specific performance.  The story of the moves and counter-moves of judges and legal academics offering arguments for and against the common-law baseline is rich and at times complex.  Fortunately, its elaboration is unnecessary for our current task.  What is needed, rather, is to consider critically the state of the art.

This Section and the next present the most powerful critiques and defenses of specific performance.  On one side, we have Oliver Wendell Holmes’s famous dictum that “[t]he duty to keep a contract at common law means a prediction that you must pay damages if you do not keep it,—and nothing else.”29  To illustrate the opposing view, we can look to Seana Shiffrin’s proposition that “a commitment to perform, morally, entails a commitment to perform rather than a commitment to perform or pay.”30  Criticizing each position on its own terms is important in setting the scene for our autonomy-enhancing approach.

1.   The Winding Path to Joint Maximization

We begin with the modern Holmesian position, which has been carried forward by economic analysts of the law.  Fifty years ago, Richard Posner started propagating the idea that the common law’s baseline is justified by reference to its salutary overall efficiency effects.  Limiting a promisee’s remedy to expectation damages (the monetary loss of expected profit), he wrote, generates “an incentive to commit a breach” when profit from breach exceeds “the expected profit to the other party from completion of the contract”—and this, he added, is exactly what a welfare-maximizing law should do.31

More recently, however, economic theorists have rejected this first-generation law-and-economics theory of efficient breach, labeling it “vacuous” to aim at the common law’s overall efficiency.32  Instead, they emphasize the incompleteness of contracts, and then base the choice of the default remedy on an assessment of what the parties would likely have wanted.  The relevant question under this approach is whether specific performance would be the remedy of choice for the majority of contracting parties, or—in a slightly different formulation33—whether it properly reflects (and exhausts) the respective obligations the parties can reasonably be deemed to have undertaken.34  The most persuasive economic defense of the common-law doctrine of specific performance, in our view, follows exactly this path.  It combines two insights—one from Anthony Kronman, the other from Robert Scott and Jody Kraus.35

The New York Court of Appeals embraced Kronman’s argument in one of the most famous specific performance cases.36  Kronman begins with the difficulty of making sense of the uniqueness test given that “every good has substitutes, even if only very poor ones.”37  The key for understanding why uniqueness is nonetheless a proper legal test, Kronman argues, lies in how it testifies to “the volume, refinement, and reliability of the available information about substitutes for the subject matter of the breached contract.”38  As he explains, when a court asserts “that the subject matter of a particular contract is unique and has no established market value,” it is “really saying that it cannot obtain, at reasonable cost, enough information about substitutes to permit it to calculate an award of money damages without imposing an unacceptably high risk of undercompensation on the injured promisee.”39  Uniqueness, in other words, turns out to be a test that can help courts distinguish between cases in which there is a well-developed, thick market for substitutes and ones in which the relevant market is imperfect or thin.40

The next step of the argument is to show that most contracting parties—“were they free to make their own rules concerning remedies for breach and had they deliberated about the matter at the time of contracting”—would indeed draw this line between specific performance and money damages.41  Scott and Kraus provide this link.

“Once a regret contingency has occurred,” they claim, the promisor faces a choice between two principal options—“perform and lose” and “breach and pay”—and is thus “motivated to choose the least costly option.”42  If both parties are able to acquire substitute goods on the same terms, then promisors are likely to perform, all else equal.  Why?  Because breach would entail not only liability in the promisee’s expectation, but also “bad feelings, loss of business reputation, good will, etc.”43  Other things are not always equal, however.  Thus, where the promisor has a reason to believe that the promisee “is better able to cover on the market and thus reduce [her] anticipated losses on the contract,” she is likely to choose the option of breach.44  This “‘benign’ vision of breaching behavior,” Scott and Kraus conclude, conceptualizes breach as a “cry for help” by the promisor: a request for the promisee “to salvage the broken contract at least cost and to send [her] the damage bill.”45

A regime in which specific performance is granted as a matter of course—even in a thick market where substitute goods are readily available—“invites” an opportunistic response by the promisee.  By limiting the availability of specific performance to cases of a thin market, the common law preempts such a response and thereby minimizes the parties’ costs of regretted contracts.  This means that the common law reduces the ex ante cost of the parties’ contracts and increases the size of the pie for both—exactly what most contracting parties would likely want contract law to require.46

2.   The Built-In Limits of Welfarist Analysis

This is a subtle and (in our view) convincing argument.  Yet, it is important to clarify the limits of its power and indeed of its ambition.  First, this account does not purport to offer a normative case for the common-law position.  Rather, it operates within the canonical economic view in which contract law’s gap-filling apparatus in its entirety is understood in majoritarian terms.  The raison d’être of this understanding is familiar: writing contracts is costly, so setting up rules that mimic the preferences of most contracting parties minimizes transaction costs and thus maximizes the contractual pie.

The second limitation emerges from the empirical premise of the argument: the distinction between thick and thin markets.  This distinction is indeed critical for the dynamics of contract performance, and economic analysts are therefore correct to study its implications and the proper response of an attentive contract law.  But it is worthwhile recalling that this account relies on Kronman’s observation regarding the dramatic differences between “the volume, refinement, and reliability of the available information about substitutes” for the subject matter of different types of contract.47  This means that changes in technology may affect these differences.  Some changes may even overturn the analysis outright.

When Kronman first put the spotlight (more than four decades ago) on how the distinction between thick and thin markets affects the law of specific performance, the contingency of the distinction was not that important.  But now things are different.  Big data and algorithmic processing increasingly make individual preferences computable.48  In turn, this means that the economic argument for the common law of specific performance is by its nature contingent rather than principled—unlike our approach set out in Part II.49  

For economists, contract remedies—nominal, restitution, reliance, expectation, super-compensatory damages and specific performance (and everything in between)—all lie on a single spectrum, differing only in their distinctive ex ante and ex post cost and benefits.50  In this view, which we reject, there is no difference between specific performance and liquidated damages large enough to be considered penalties.  That courts draw a line between money damages and specific performance has no particular moral valence.  It simply reflects contingent differences in the predicted cost structure of each remedy.51

But this contingent quality of the economists’ position has practical consequences:  it limits their ability to respond to the philosophers’ challenge.  By linking restrictions on specific performance to majoritarian preferences, the economist friends of the common law can only deflect the civil-law-cum-moralist challenge.  If majoritarian preferences shift, then the common-law approach loses its justification.  The economists do not, and cannot, contest the normative power of the moralist critique.52

C.   The Most Adamant Critique and Its Pitfalls

To illustrate the moralist critique of the common law, we focus on Shiffrin’s influential account—an approach that shares the same deontological normative structure and crucial pitfalls with other dominant consent and transfer contract theories.53

1.   Contract and Promise

Shiffrin argues that existing specific performance doctrine epitomizes the common law’s moral bankruptcy.54  To get there, she starts from what she calls the troubling “divergence of contract and promise.”55  The morality of promises, Shiffrin argues, implies that “a promisor is morally expected to keep her promise through performance,” and that only “if, for good reason, what was promised became impossible, or very difficult, to perform,” might financial substitutes be appropriate.56  “Otherwise, intentional, and often even negligent, failure to perform appropriately elicits moral disapprobation.”57  Promissory morality can be satisfied only by “the consent of the promisee,” and not by supplying “the financial equivalent of what was promised.”58

Under this account, when breach occurs, it generally requires not expectation damages, but specific performance (and at times punitive damages).59  Shiffrin concedes that in some cases there may be “distinctively law-regarding grounds,”60 such as “the difficulties of judicial supervision, risks of arbitrary enforcement, and in some cases, the hazards of involuntary servitude,”61 which could justify “reluctance to order specific performance.”62  But these discrete reasons are local and they do not, and indeed cannot, “question the general proposition that specific performance is the appropriate moral response to breach.”63  Nor do they justify the mitigation doctrine, which “places the burden on the promisee to make positive efforts to find alternative providers instead of presumptively locating that burden fully on the breaching promisor.”64  Because existing doctrine “fails to use its distinctive powers and modes of expression to mark the judgment that breach is impermissible as opposed to merely subject to a price,” the common law’s treatment of transgressions, Shiffrin concludes, is indefensible.65

Shiffrin is not impressed by the economists’ account in which majoritarian logic drives what may constitute a breach and what should be the proper remedial response.  Even if it were justified to allow rational actors, who care only about profit-maximizing, to opt into a regime that compromises “the moral significance of a broken promise,” contract law must not embrace this position in its doctrine.66  By re- (or rather mis-) presenting efficient breach as part of the parties’ contract, the economists’ account (like the common law that it seeks to justify) robs contract of its moral foundation.  “Part of the underlying . . . value of promises is that promises transfer, rather than hoard, discretionary power.  The ‘perform or pay’ promise, however, retains a good portion of that discretion.”67  This makes the common-law vision of a promise “shabby and second-rate.”68  Encoding this vision in the law—“encourag[ing] promisors to make only this thinnest of commitments”—cannot possibly be conducive to a proper “moral culture.”69

2.   The Transfer Theory Core, and Why It’s Wrong

The key to Shiffrin’s moral condemnation of the common law, and the core of our critique, lies in understanding her heavy reliance on the “transfer theory of contract”—spelled out in the italicized text in the previous paragraph.  Following a tradition that goes back to Kant and Hegel, Shiffrin analyzes “the moral structure of promise[]” as a “transfer[] of decision-making power.”70  The binary transfer of authority explains, in this view, the “bindingness” of promises and the wrongfulness of breaking a promise:

By promising to φ, the promisor transfers his or her right to act otherwise to the promisee.  To not φ, then, is to act in a way the promisor has no right to do, and to φ is to act in a way the promisee has a right that she (the promisor) do.71

Thus, “[i]n light of this transfer, the promisee has a right to expect (and often to demand) performance and has the concomitant power to use her transferred power or decision to waive or excuse the promisor’s obligation of performance.”72

In other words, as Peter Benson recently explained, for transfer theory, contract is “a form of transactional acquisition—a transfer of ownership between the parties—that is contractually specified and complete at . . . formation.”73  Formation of contract in this vision is a “representational medium of mutual promises,” through which each party moves “a substantive content” from her “rightful exclusive control” to the other’s control.74  This means that the performance of a contract is of no normative significance: it merely delivers to the promisee’s factual possession what was rightfully hers already.  This is why for transfer theory a failure to perform must be understood “as an interference with the plaintiff’s exclusive right,”75 which implies, as Shiffrin’s analysis assumes, a “striking parallel” between breach of contract and conversion.76

Transfer theory, however, is deeply flawed, because it misses the nature and the point of contract, which is at core a power-conferring body of law (as we’ve argued in detail elsewhere).77  Contract attaches legal consequences to certain acts thus enabling people to affect their entitlements, if they so desire.  This power-conferring aspect is what makes contract law dramatically different from tort law.  Tort law doctrines, at least those dealing with our bodily integrity, are duty-imposing.78  Assuming people have certain pre-legal and pre-conventional rights, tort law affirms the correlative duties against their violation.79  But contract law is different: rather than vindicating existing rights, contract law first and foremost confers the power to create new rights, rights that are crucial to people’s autonomy, as we will soon show.

Duties not to interfere with people’s rights are surely relevant to contract law.  But they are secondary.  Duty-imposing rules that safeguard contracts’ voluntariness (dealing for example with duress, fraud and the like) would be meaningless in the absence of (power-conferring) contracts: the duties’ role is to protect our ability to apply the powers enabled by contract, and they would be pointless in a world that did not recognize the power to contract.80  This means that—in sharp contrast to torts—the relevant question for a liberal theory of contract is not “what constraints to people’s autonomy are legitimate,” but rather “how should contract law enhance people’s autonomy.”81

By dogmatically asserting that a contract necessarily needs to take the form of a complete assignment to the promisee of the right to the promisor’s future actions—by perceiving any qualification as at least prima facie unwarranted hoarding of discretionary powers, in Shiffrin’s terms82—transfer theory obscures this crucial question.83  It thus fails to consider the ways in which law can facilitate forms of bilateral voluntary obligations that are conducive to contract’s autonomy-enhancing telos.84

This inquiry, to which we now turn, is not quantitative—it is not about maximizing the amount of autonomy in the world.  But it is teleological nonetheless.85  Contract is not worth keeping in and of itself.  Rather, its value derives from its contribution to our autonomy which is valued for its own sake.86  So, we are looking for specific performance rules that are as conducive as possible to people’s autonomy—rules that generate the most autonomy-friendly implications.

II.     Specific Performance and the Future Self

The familiar debate on specific performance points to two troubling conclusions.  First, the problem with the moralist critics of the common law—and their implied normative defense of the civil law87—is not their moralism.  Their mistake lies instead in misstating the underlying morality of contract and indeed of promise.88  But we cannot overcome this failure, dramatic though it is, by resorting to the economists’ defense of the common law.  Hence, our second conclusion: even the best economic account does not provide a principled normative justification for the limits of specific performance.  Nor does it purport to provide such a justification.  The economists’ account is contingent on mere computational difficulties that technology may be able to overcome.  And their view is founded on the presumed preferences of (certain type of) contractors, rather than on the moral virtue of the position they seek to vindicate.89

It is time to start afresh.  This Part shows how a thoroughgoing commitment to autonomy explains and justifies much of the current law of specific performance, charitably understood.

A.   Autonomy in Contract90

1.   The Moral Virtue of Expectation Damages

Contract is not a transfer—at least not in the sense the philosophers require for their approach to cohere.91  Rather, it is a plan co-authored by the parties in the service of their respective goals; a joint undertaking of a cooperative arrangement.92  Contract’s significance lies in its service to planning since the capacity to plan is what makes a person an agent, rather than a mere object of powers, effects, and circumstances.93  To be sure, much of what happens to us is beyond our control—it derives from our natural and social endowments as well as life’s vicissitudes.  But contract is exactly the instrument that expands the scope of what we can and do have control over.94  Law’s justification for enforcing the parties’ agreement is accordingly grounded in its commitment to enhance their self-determination, and both its animating principles and its operative doctrines are guided by this autonomy-enhancing mission.95

Thinking about contract in these terms implies that formation is not the only normatively relevant moment in the life of a contract—as transfer theorists posit.96  Rather, viewing contract as a coauthored plan reinstates the full significance of contract’s intertemporal dimension.  Focusing on the times of contract brings to light the core achievement of contract—the prevalence of expectation damages, which makes it so vital to planning—and highlights its central challenge—that is, showing proper respect for the autonomy of the contractors’ future selves.97

Contract is the means through which we can legitimately enlist others to our own goals, purposes, and projects—both material and social.98  By ensuring the reliability of contractual promises for future performance, contract law enables people to join forces in their respective plans into the future.  An enforceable agreement is the parties’ script for this co-operative endeavor, and contract law provides them with the indispensable infrastructure that both facilitates this risky venture and ensures its integrity.99

This account explains why vindicating the promisee’s expectation interest is contract’s signature commitment.  The ability to develop protected expectations is what makes contract a crucial tool for planning and thus for self-determination.  Therefore, a contractual right is the right to expect.100  Contract’s autonomy-enhancing services admittedly impose on promisors an extra burden: an exposure to liability that goes beyond other people’s actual harm.101  Promisees’ claims are not capped by their actual reliance.  But this burden is quite modest—it only requires putative promisors to apply some additional caution while making promises—and it falls squarely within the obligation of reciprocal respect for self-determination that underlies private law as a whole.102

Our core claim is that contract is both justified and best interpreted by reference to this autonomy-enhancing telos.103  Contract is, as Charles Fried argues, “a kind of moral invention” exactly because it extends people’s reach in this way.104  By expanding the available repertoire of secure interpersonal commitments beyond the realm of close-knit interactions, contract law dramatically augments people’s ability to plan.  In offering that, contract makes a crucial contribution to our autonomy because self-determination involves planning.  People, to be sure, may change their plans, and autonomous persons must be entitled to do so.  But having a set of plans arranged in a temporal sequence is typically key to the ability to carry out higher-order projects, that is, to self-determine.

2.   Why Choice Matters

This conception of contract entails three principles—addressing rangelimit, and floor—that must guide contract law in a liberal society:  (1) Law should proactively facilitate the availability and viability of multiple contract types in each sphere of human endeavor.  (2) Contract law must respect the autonomy of a party’s future self, that is, it must take seriously the ability to re-write the story of one’s life.  And, (3) to justify coercive enforcement by the state, all contracts must comply with the demands of relational justice.105

We unpack these three principles over the course of a book and many articles that together set out “choice theory”106—and explain the jurisprudential structure of contract law as a whole.  The core rules governing the life of contracts, from inception to breakdown, all follow the three principles of choice theory because all partake in the same autonomy-enhancing mission.107  Here, we limit our discussion to just two pages summarizing the first and third principles for readers unfamiliar with the major findings of choice theory.  Then we turn our focus to the second principle, concerning the future self.  This is the principle that most directly grounds our account of the common-law doctrine of specific performance. 

The first principle, of proactive facilitation, manifests in the numerous contract doctrines that go beyond safeguarding the parties’ independence.  These doctrines seek to empower people by expanding the scope of cooperative engagements available to advance the parties’ future plans.108  A foundational example here is the canonical status of the “objective” approach to party intention that guides the rules on contract formation.  This approach is best explained by the qualitative difference between the limited autonomy-enhancing potential of a subjective theory of contract and the far more impressive potential of its objective counterpart.109  But the objective theory is only the tip of the iceberg: many other doctrinal features of modern contract law follow suit.

Perhaps the most powerful example of the significant facilitative role of modern contract law is its extensive gap-filling apparatus.  This apparatus sharply departs from the traditional common-law reluctance to enforce incomplete agreements           .110  Contemporary contract law is not satisfied with providing enforcement services only to parties who fully specify the terms of their engagement.  Rather, it goes out of its way to facilitate transactions by offering defaults that can fill gaps—even regarding crucial aspects of a transaction, such as price.111  A significant subset of the current contract law canon belongs to this category—including most rules dealing with performance and breach.112  The same concern for proactive facilitation underlies contract law’s characteristic supply of a variety of contract types.  When properly implemented, these types generate an inventory of diverse frameworks for interpersonal interaction for people to choose from in each major sphere of contracting.113

The third animating principle of choice theory shifts gears to the interpersonal dimension of contracting.  Contract law requires attention to relational justice—that is, to reciprocal respect for self-determination.114  This obligation arises from people’s foundational right of self-determination, the same right that underlies the legitimacy of contract in the first instance.115 Therefore, when someone relies on contract law, that party is also necessarily—inescapably—undertaking the obligation to respect the other party’s self-determination.116

This principle has important consequences for the structure of contract law.  In particular, it means that when people use contract law’s empowering potential, their uses should be limited to interactions that show reciprocal respect for self-determination.  This obligation of respect cannot be too onerous, but neither is it limited to a negative duty of noninterference117—as some philosophical accounts have mistakenly suggested.118  Consider contract law’s careful, but important, deviations from the laissez-faire mode of regulating the parties’ bargaining process.119  For example, note the expansion of the law of fraud beyond the traditional categories of misrepresentation and concealment to include affirmative duties of disclosure.120  Also, note modern rules dealing with unilateral mistake, duress, anti–price gouging, and unconscionability.121  Concern for relational justice also best explains key rules during the life of a contract, as epitomized by the duty of good faith and fair dealing.  This duty, now read into every contract, protects the parties against the heightened interpersonal vulnerability that contract performance engenders and solidifies a conception of contract as a cooperative venture.122

Now, we shift from autonomy’s interpersonal demands to intrapersonal ones.  And this circles us back to choice theory’s second animating principle, the one that matters most for our purposes, concerning the autonomy of a party’s future self.  This principle of regard for the future self focuses on what autonomy demands within each person across time, that is, intra-personally and inter-temporally.  In simple terms, what are the limits of the legitimate dominion of my current self over my future self?  How much and how far can one’s current self legitimately commit her future self?

B.   The Current and Future Self123

1.   The Freedom to Change Your Mind

Self-determination requires that people have the right to write the story of their lives.  As Michael Bratman explains, people are planning agents and planning agency implies that people’s “prior intentions provide a rational default for present deliberation.”124  A liberal law necessarily follows suit.125  It offers people the normative power to make contractual commitments, and it properly assumes that insofar as these commitments are indeed part of the current self’s plan, the future self is presumed to adhere to them.126  Thus, contract law “takes seriously the voluntary commitments individuals undertake”: it requires them to “make good on their promises” and is not “moved by sheer regret” following bad choices.127

This means that self-determination necessarily entails some authority of a person’s current self over her future self.  But this authority must not be boundless.  To see why, recall that self-determination also requires that people have the right to rewrite the story of their lives.128  The intertemporal constancy that planning agency requires needs to be, in other words, sensitive to the fact that “sometimes an agent supposes there are conclusive reasons for change.”129  While new “ordinary desires and preferences” may not suffice, the constancy that planning agency implies should nonetheless be “defeasible constancy: constancy in the absence of supposed conclusive reason for an alternative.”130

This is why a liberal legal regime—one that offers people the normative power to make contractual commitments so as to enhance their autonomy—cannot fully ignore the impact of such contracts on the parties’ future selves.  It is true that enhancing people’s autonomy in their capacity as promisees requires, as noted, vindicating their expectations (and not only reliance).131  But respecting their autonomy in their capacity as promisors also implies that contract law must be careful in defining the scope of the obligations it enforces and in circumscribing their implications.132  Why?  Law must allow some space for the defeasibilityof intertemporal constancy.133  In other words, people sometimes must be free to change their minds.

A liberal contract law, beyond enabling us to make credible commitments, should always be alert to its potentially detrimental implications for the autonomy of the parties’ future selves.134  Accordingly, choice theory’s second principle requires that the same law that empowers people with the ability credibly to commit themselves through contracts, cannot ignore the impact of these contracts on their future selves.  In a genuinely liberal legal regime, contract’s invaluable empowerment service must not end up as a carte blanche for allowing people’s current self to fully dominate their future selves.135  

This prescription of respecting autonomy of the promisor’s future self does not make a sham of the canonical doctrinal obligation to perform a contractual promise; neither does it imply a Holmesian disjunctive duty to perform or pay.136  As we noted early on, promisees are entitled to nominal damages even if breach creates no loss—a rule reflecting this interpersonal obligation.137  But law also recognizes that the obligation is not absolute.  It must accommodate the defeasibility of agents’ intertemporal constancy because the obligation to perform is itself premised on contract’s autonomy-enhancing telos.  This is why specific performance is not readily available in cases where expectation damages do not significantly disrupt the contractual script.  In these circumstances, as we explain below, the promisor’s change of mind justifiably excuses the duty to specifically perform.138

This concern for the autonomy of promisors’ future selves,139 as we frame it, should be carefully distinguished from two competing ways of investigating the implications of the time dimension of contract.

First, our account does not rely on people’s imperfect foresight.140  While we do not deny the relevance of systemic behavioral limitations to contract law, we think that reliance on such imperfections can neither explain nor justify contract doctrine.  As a matter of positive law, the claim that imperfect foresight limits the power to bind is over inclusive, because it also covers many cases of mistaken judgment (such as a bad gamble) that contract law does not hesitate to enforce.  Further, and more fundamentally, our normative claim is that even if behavioral limitations were to be completely eliminated—say, through new technology or legal techniques—liberal contract law would not, or at least should not, authorize the current self’s complete domination of the future self.

Second, choice theory’s concern for the future self does not imply an endorsement of the idea of “multiple selves,”141 that is, the idea of the disintegration of the self.142  Quite the contrary.  Choice theory rejects this position.  Indeed, its core claim regarding the significance of planning to self-determination implies that the current self and the future self are the same self.143The integrity of the self, rather than its separation into different selves, is what drives choice theory’s justification for contract enforcement, and is thus a necessary feature of its account of the telos of contract.144  The discussion of the future self is a discussion of the self in the future and the liberal requirement that it be able to rewrite its course.145

2.   The Line-Drawing Challenge

 Because any act of self-authorship constrains the future self, the obligation of the liberal state to enhance autonomy implies that contract law must both bolster and limit people’s ability to commit.146  This is a subtle task, and there is no easy formula for resolving this difficulty.  But this does not necessarily lead us to an impasse, nor does it imply that its resolution needs to be done on an ad hoc basis.  Instead, liberal contract law can and should apply qualitative judgments and identify categories of limitations on promisors’ freedom to change their minds—consider, for example, indentured servitudes—that should not be enforceable (in general or under certain conditions) because they overly undermine the autonomy of their future selves.147

Even more fundamentally, law should be particularly vigilant in ensuring that contractual liability does not attach in categories of cases where contractual commitments actually do not significantly serve the parties’ current selves.  Categories of commitment that are not autonomy enhancing to people’s current selves should not be used to constrain their future selves.148  This prescription is the normative foundation of doctrines governing failures of both parties’ basic assumptions, namely, mutual mistake, impossibility, impracticability, and frustration.149

While contract theory has long grappled with transaction facilitation and interpersonal justice, it has not shown similar concern for autonomy of the future self.  We think this is a mistake.  If contract is—as indeed it should be in a liberal setting—first and foremost about enhancing people’s self-determination, then the mission of empowering our current selves while safeguarding the right of our future selves to re-write the story of our lives must be the (or at least a) major challenge of contract law.150

3.   What About Contractors’ Preferences?

Legal economists may be impatient at this point.  Contract law, they often insist, should be designed—absent systemic externalities or behavioral concerns—so it reflects what most contractors are likely to want.151  But our account does not seem to apply to commercial contracts between sophisticated legally informed firms, which this objection anticipates.  Because it appears to focus instead only on the limited domain of individual contracting, whatever its normative purchase may be, it is irrelevant to general contract law.

To see why the objection misfires, we need to return to basics, and appreciate contract’s justificatory challenge, which focuses on what most lawyers may perceive as obvious, but isn’t: the enforceability of wholly executory contracts.  Contract law authorizes a promisee to constrain the freedom of a reneging promisor even absent detrimental reliance, applying the law’s coercive power in the promisee’s service.  Additionally, contract law heightens this threat to people’s negative liberty by adhering to the “objective” approach and by applying a robust apparatus of default rules—elements that operate to the potential detriment of idiosyncratic and legally uninformed parties, respectively.

Choice theory appreciates the difficulty of justifying these seeming deviations from liberal principles.  It responds to this challenge by relying, as we’ve emphasized throughout, on people’s fundamental right to autonomy and on a liberal state’s obligation to enhance individual self-determination.  This individual right also entails modest interpersonal duties—including those arising from the enforceability of wholly executory contracts, the objective approach, and the default rule apparatus—to the extent these duties are crucial to contract law’s autonomy-enhancing function.

In short, we premise contract’s justification as a practice, along with each of its particular rules, on its service to people’s autonomy, defined as the ability to write and rewrite one’s life plans (and the many mini plans they include).152  This premise means the sheer reference to most parties’ presumed intentions, which underlies the legal economist objection, cannot provide a sufficient justification.  Rules are justified if and only if—and to the extent that—they enhance people’s autonomy both in making commitments crucial to self-determination and in being able to start afresh.153

This conclusion does not mean that all contract rules should be mandatory.  When legally informed, sophisticated firms set up commercial contracts, they should be able to adjust their contract terms according to the type of cost and benefit calculus that economic analyses highlight.  But contract rules apply to idiosyncratic and legally uninformed parties as well, even in the context of wholly executory contracts.  Therefore, general contract law—as opposed to statutes or doctrines strictly addressed to commercial contracts—must respond first and foremost to the real people whom it serves and upon whom it exercises coercive power and authority.  

To close the loop on this argument, many contract law rules, including the rules on specific performance, are best regarded as normative defaults chosen because of their freestanding value,154 and not as majoritarian defaults adopted because of their correspondence to party preferences.155  This account requires that contract law should set up its rules with individuals in mind, and then allow sophisticated commercial firms to adjust as they see fit in the ordinary course of legally informed contract drafting.

Concerns of institutional comparative advantage dovetail with this autonomy-enhancing approach: judges are well-positioned to set rules—and they should set rules—based on what seems normatively justified for transactions between individuals.  At the same time, judges should allow legally sophisticated contracting firms to opt in and out of (most of) these rules.  Why?  Because such firms are relatively more expert than judges in pricing and negotiating terms and remedies that will likely serve their wealth-maximizing aims.

With the economists’ seemingly powerful objection behind us, we come to our core question: what should be the reach and limit of specific performance?  As we noted in the Introduction, our answers echo many of the economists’ prescriptions, but we ground them in a more robust normative foundation and we derive different reform implications.

C.   Justifying the Common-Law Baseline

To start, we agree with the economists that contract law should resist the moralists’ expansive approach to specific performance.156  Indeed, it should take nearly the opposite approach: ensuring that doctrinal limits to the self-determination of the promisor’s future self are necessary to enhancing her current self’s ability to plan.157  Excessive remedial responses—remedies that go beyond what is required to empower the parties’ current selves—are autonomy reducing.  Entrenching such remedies into contract law would impinge upon, rather than bolster, its morality.

The mission here is to find the rules that are as conducive as practicable to enhancing people’s autonomy.  And, as usual, this is a complex task, requiring the sort of qualitative judgment that typifies any credible autonomy talk.158  Still, addressing this challenge head-on proves immensely fruitful: it helps vindicate the moral underpinnings of the common-law baseline of no specific performance along with the uniqueness exception and the treatment of service contracts.  In other words—and this is the core finding of this Article—both the common law’s resilient reluctance toward specific performance and the pockets where such a remedy is readily available are best explained and justified by reference to contract’s innermost normative commitments,159specifically, to choice theory’s principle of concern for the future self.

We recognize that both expectation damages and specific performance constrain the future self: remedy law’s imposition on the future self is the inevitable implication of its function of making people’s contractual commitments credible so that they can properly serve as planning devises.  This is why contract law does not allow the future self to rewrite the past whenever she regrets her commitments:160 such a rule would deter others from dealing with her, thus undercutting her ability to plan using the empowering device of contract.161  Nevertheless, there is a qualitative distinction between the constraint of expectation damages and that of specific performance.

Requiring the promisor to cover the promisee’s expectation limits her course of action: she now needs either to divert some of her resources for that purpose rather than using them for her own, or to undertake some further obligations in order to cover the cost of substituting her performance with another’s, which the expectation measure of recovery represents.  

Specific performance, in turn, goes much further than that because it dictates a course of action through an affirmative duty.  To be sure, in welfare terms the difference between these remedies is still quantitative.  Moreover, there are cases in which the cost of specific performance for the promisor may even be lower than that of covering expectations (this may explain some contractors’ preference for “repair and replace”).  But while these remedies may be continuous in terms of welfare, they are nonetheless different in kind from the viewpoint of contract’s autonomy-enhancing telos—because specific performance qualitatively imposes more on promisors’ self-determination.162

Specific performance compels a promisor to act in accordance with the contractual script.  The fact this is a script she previously coauthored certainly mitigates this compulsion.  But because self-determination requires both the right to write one’s future plans and the right to re-write them and start afresh, the additional constraint which specific performance entails nonetheless needs to be carefully scrutinized and properly justified.  Moralists and other friends of the civil law tradition do not even attempt to offer such a justification.

This may not be surprising because it is hard to see how they could justify this excess.  Consider the core case in which the civil law and the common law diverge: a contractual duty to provide a characterless good (the famous widget), for which the former regime allows specific performance whereas the latter limits the promisee to her expectation interest.  If receiving a widget from a specific source was part of a person’s plan, then a check in the mail that allows her effortlessly to receive a perfect substitute elsewhere is surely a change in the plan.  But it is a minimal change, one that does not substantively affect contract’s function as a planning device.  It is not a change that amounts to the kind of autonomy-based reason—a real disruption—that can qualify as a justification for constraining the future self.163

Admittedly, in real life the difference between performance and expectation damages may not be minimal even in this type of contract.  But as long as the difference does not affect the essence of the contractual plan, the autonomy-based analysis need not substantially change.  After all, while some degree of stability is essential for people’s plans to be meaningful, plans need not be fully immune from changes.  Machines are supposed to follow operating protocols to the letter, but human beings who make plans anticipate some changes.  Our plans, big and small, tend to evolve and can often adapt without undermining our self-determination.  Further, insofar as the expected change is purely financial—which liquidating the promisor’s performance entails—the parties can arrange for recovery by resorting to liquidated damages, say, for incidental costs an injured party incurs that may be hard to monetize.164

Thus, other things being equal or close to equal for the promisee, contract in the common law tradition rightly opts not to compel the promisor to act in accordance with the contractual script, allowing her to choose between doing so and covering the promisee’s expectation.165  Alas, other things are not always equal.

D.   The Uniqueness Exception and Personal Services Exclusion

The same autonomy-based commitments that justify the common-law baseline also explain much of the remaining terrain of specific performance, in particular the uniqueness exception and the personal services exclusion.  We start with uniqueness. 

1.   What is “Uniqueness”?

If the widget case is the canonical example in which common law affords only damages, a purchase of a residential dwelling is the paradigmatic case in which specific performance is granted as a matter of course.166  Our analysis explains why.  In the widget case, there is no autonomy-based reason for encumbering the autonomy of the promisor’s future self with the constraint specific performance adds over and above damages.  By contrast, in the residential dwelling case, such a reason is readily available.

Like other major transactions, purchasing a home has significant financial implications.  But it involves more than that.  A person’s residence is understood in contemporary society as the paradigmatic safe haven; as a bastion of individual independence and a symbol of the self.  It provides an almost sacrosanct private sphere that serves as a prerequisite to people’s personal development and autonomy.167  Many legal doctrines protect individuals in their homes—from rent controls and homestead exceptions in bankruptcy law to restrictions on search in the Fourth Amendment.

The constitutive role of a home in people’s ordinary experience implies that a purchaser’s expectation in a residential transaction typically transcends the financial stakes.  Buying a unique good—most notably a residential dwelling—involves extensive planning for how that good will be integrated into the next chapters of one’s life.  As the Nebraska Supreme Court noted in a frequently cited case, 

a purchaser of a particular piece of land may reasonably be supposed to have considered the locality, soil, easements, or accommodations of the land generally, which may give a peculiar or special value to the land to him, that could not be replaced by other land of the same value, but not having the same local conveniences or accommodations.168  

Thus, when a seller reneges on the promise to deliver a dwelling, she is relatively more likely to disrupt the purchaser’s life plan than if she breaches in other ordinary contracting contexts.

In autonomy terms, the “uniqueness” exception captures the core category of cases in which breach amounts to such a disruption—in contrast to the economic approach in which a good’s uniqueness is a function of information costs.  Unlike the minor changes in plan that breach of a widget contract prompts, disruption in the residential context is qualitatively different and typically cannot be sufficiently ameliorated by expectation damages.  Even where no financial setback is involved, disruptions around residential sales threaten to frustrate contract’s ability to function as a planning tool.

Happily, law can preempt this autonomy-undermining result by granting specific performance.169  Because there is an autonomy-based reason for the additional constraint of this remedy, liberal contract law properly follows suit for residential sales transactions and other categories of goods in which breach imposes the same plan-disrupting consequences—that’s what makes these goods “unique.”  (Section III.A sets out our proposed reforms regarding real estate sales.)170  To be sure, even in these cases, specific performance may be refused if it “would cause unreasonable hardship or loss to the party in breach.”171  But unlike the widget paradigm, this limited exception does not arise from the intra-personal limit on the dominion of promisors’ current selves over their future selves, but rather from choice theory’s third principle of relational justice.

Conceptualizing uniqueness as a proxy for “disruptions to the promisee’s plan” may explain, and indeed justify, the way courts use the liberalized language of the UCC to expand the exception to the no-specific-performance rule so that it includes cases of significant scarcity.172  This proxy can likewise—even more importantly—inform the Restatement’s vague standard in which specific performance will be granted only if damages cannot “adequate[ly]” protect the injured party’s expectancy.173  When are damages not adequate?  In categories of cases in which breach is likely to be too disruptive to promisee’s important life plans.

This definition ensures that when parties invoke the contract convention for their coauthored plan, they do not encumber their future selves’ self-determination more than this type of plan conventionally requires.  A liberal contract law entrenches such a rule because of its freestanding autonomy-securing value, not because it is what most parties prefer or can be assumed to prefer.  In other words, the common law’s scheme does not depend on its accordance with majority preferences—like in the economic account.  Rather, the limits of specific performance derive directly from contract’s normative foundations.

2.   Uniqueness as a Default

That said, liberal law should recognize that the borderline contract sets for uniqueness is indeed conventional.  This means that there may be contractual scripts that fall outside the uniqueness category (even in its expanded configuration) and yet their actual performance is of the essence for the specific parties who coauthored them.

A liberal contract law should respect those idiosyncratic plans and thus refrain from thwarting them by mandatingboundaries for specific performance.  And yet, parties’ attempts to opt into specific performance in cases that go beyond the (expanded) uniqueness category are not to be treated casually, as if they were garden-variety overrides of default rules that merely mimic majoritarian preferences.  Recognizing the normative weight that justifies liberal contract’s expectation damages baseline justifies some caution when departing from it.

Treating this baseline as a normative default implies that contracting parties should be able to override it and choose to have their script specifically enforced—if and only if they meet two conditions: first, they signal the distinctiveness of their plan by using “apt and certain words,”174 and second, that there are no conflicting obstacles such as difficulties of judicial supervision.  (We detail how this proposed reform would work in Section III.B.)175

3.   Personal Service Contracts

Thus far, we’ve stayed close to the role uniqueness plays in the existing law of specific performance.  Uniqueness, we’ve argued, can and should stand for conventional categories of breaches that are tantamount to substantive disruptions in the promisee’s plan, as enshrined in the contractual script.  But there is another doctrinal pocket in which the idea of uniqueness affects the common law (as well as civil law), even though this concept is not explicitly used.  Contract law accords special treatment to agreements whose performance uniquely involves the person of the promisor—that is, to personal service contracts.176

This personal services category may be understood as a mirror image of the unique goods category.  In the latter case, the uniqueness exception implies that breach disrupts the promisee’s plan.  In the former, it suggests that performance might undermine the ability of the promisor to abandon a plan.  Personal service contracts are the paradigmatic case in which specific performance might trigger autonomy-inhibiting effects.  Ordering a worker specifically to perform her employment contract compels her to take a specific course of action.  Additionally, that particular course of action requires her to do (and not only to deliver) specific things and thus involves her personal cooperation with another person’s project.177  It is therefore no surprise that contract in both common law and civil law—loyal to the innermost liberal commitment to self-determination178—resists granting specific performance for personal service contracts.179  (Sections III.B and III.C nonetheless offer some refinements.)

III.     Reform Implications and Hard Cases

American contract law is, at root, committed to enhancing peoples’ autonomy, that is, their ability to write and rewrite the story of their lives.  Seen through this prism, much of the existing terrain of specific performance in the common law snaps into focus—the preference for expectation damages, the uniqueness exception, and the personal services exclusion.  This area of law has been so resistant to change in part because it already adheres closely to an appealing normative framework—not the contingent, ever-shifting one of the economists, nor the wrongly conceived approach of the philosophers.

That said, there is still room for improvement.  Careful attention to the normative foundation of the law points to three areas where specific performance should be reformed so the doctrine better complies with its deepest animating principles.

A.   Circumscribing the Land Sale Exception

1.   The Traditional View

Most American courts adhere to the “traditional view” that “accord[s] a special place” to contracts for the sale of land “in the law of specific performance.”180  Although this traditional approach comes in different formulations and varying degrees of rigidity, it stands for the crystallization of a more-or-less bright line rule in which specific performance is granted as a matter of course to the injured party in all agreements for the sale of land.181  The question whether the breaching party is the buyer or the seller is irrelevant in this view.  Additionally, in cases where the seller breached, courts pay no attention to whether the buyer intended the land for private use or was in the business of buying and selling land.182

Thus, for some courts, “an agreement of sale of real estate . . . vests in the grantee . . . an equitable title to the real estate,” so that from the moment of its execution, “the vendor is considered as a trustee of the real estate for the purchaser and the latter becomes a trustee of the balance of the purchase money for the seller.”183  Other courts simply state, for example, that “[w]here land is the subject-matter of the contract, the damage is held to be irreparable as a matter of law,”184 or that in such cases “the inadequacy of the legal remedy is well settled,185 so that no further discussion is required.186  Either way, most courts conventionally refuse to investigate the purchaser’s intended use187 or to consider withholding specific performance where the injured party is the seller, rather than the buyer.188

2.   The Path to Reform

But not all courts continue to adhere strictly to the traditional doctrine—and these deviations point to some fruitful areas for reform.  Some courts have held that the seller “is not automatically entitled to specific performance as a matter of right or law”189—in sharp distinction from the buyer of real estate.  The reason seems straightforward: the “rationale underlying the grant of specific performance in real estate transactions” does not apply to sellers,190 where the seller’s recovery can simply follow the conventional formulas for damages.191

Other courts have challenged the traditional view even as it applies to cases in which the injured party is the buyer.  As these courts explain, when “plaintiffs are faced with the loss of commercial, and not residential, property[,] [t]hey are . . . threatened with an economic loss which is compensable in large part if not entirely, in damages.”192  In other words (as a recent case put it),

if land is purchased merely to be resold and‌/or rented, it is being treated by the purchaser as a fungible commodity, and such a party can be made whole via money damages regardless of whether he might have been entitled to equitable relief had he intended to keep and use the land.193

Canadian courts have taken a bolder turn.  Following the lead of a few lower courts’ cases, the Canadian Supreme Court discarded (in a long dictum) the traditional view in which “every piece of real estate was generally considered to be unique.”194  The “progress of modern real estate development,” said the court, implies that “this is no longer the case” and there are cases in which damages are an adequate remedy.195  Therefore, it concluded, specific performance should no longer “be granted as a matter of course absent evidence that the property is unique to the extent that its substitute would not be readily available.”196

We hesitate in recommending the Canadian path insofar as it suggests open-ended equitable discretion, which the traditional approach successfully curbs.197  Overly vague standards offend the rule of law because they effectively authorize courts to exercise unconstrained power, which in turn inhibits law’s ability to guide the behavior of its addressees.198  This general concern matters even more specifically to contract because law’s effective guidance is intimately related to people’s ability to form reasonable expectations and plan for the future.199  The rule of law quest for predictability is grounded in contract’s own autonomy-enhancing telos.

That said, the Canadian view correctly warns against the over-inclusiveness of the broad category of land sale contracts.  Is it possible to narrow the category appropriately without embracing ad hoc discretion?  Yes.  The emerging minority position in American courts deviates from the traditional view’s rigid strictures and does so in a way that helpfully suggests how the shortcoming of the Canadian approach can be remedied.  Bringing these points together, we suggest reading the minority cases not as a way of returning to specific performance’s equitable-cum-discretionary origins,200 but rather as a salutary step in a gradual process, typical to the common-law tradition, of more carefully circumscribing the category of cases in which specific performance is awarded.201

The underlying, autonomy-based reason that justifies specific performance in contracts for the sale of land also suggests limiting its reach to the category of cases in which (1) the injured party is the buyer and (2) the purchased land is intended as a residence.202  Where both of these conditions obtain, breach is likely to disrupt the promisee’s life plans in ways that damages alone will not address.  By contrast, there is no good autonomy-based reason to continue specifically enforcing real estate contracts on behalf of sellers or in fungible commercial cases even where their subject-matter is land.

B.   The Proper Limits for Opt-Ins

1.   Expanding the Scope of Opt-In

 The uniqueness exception can be overinclusive—as we’ve just seen.  Alan Schwartz points to a mirror-image problem.  He would expand specific performance from the other side of the autonomy equation.  A promisor’s autonomy, he argues, “is not seriously compromised by a specific performance decree if the promisor sells roughly fungible goods or is in the business of selling unique goods.”203  The reason is simple: in such cases “the goods are assets to the promisor much like cash; requiring their delivery is not relevantly different from requiring the delivery of cash.”204  By the same token, Schwartz contends, “requiring a sizable corporation that renders services to perform for a given promisee does not violate the corporation’s associational interests or the associational interests of its employees.”205

Schwartz’s conclusion is that, excepting in cases of personal service contracts, “specific performance should be available as a matter of course to those promisees who request it.”206  We think that this conclusion is too quick.  Behind the “artificial persons” of corporate bodies stand real people, and at times the law ultimately serves the choices and plans those real people are seeking to make.207  Moreover, as Schwartz’s careful language suggests (the part we italicized above), in some cases it would be difficult to determine whether specific performance seriously jeopardizes the promisor’s autonomy or not.  But when liberal contract law sets its background rules—guided, as it should be, first and foremost by its fundamental commitment to enhancing people’s autonomy—it should avoid, absent any countervailing autonomy-based consideration, imposing rules that might be autonomy-reducing.  A normative default, like the common-law baseline, should not be too easily discarded.

That said, Schwartz’s claims provide a powerful case for allowing contracting parties affirmatively to opt into specific performance when their agreement indeed fits Schwartz’s description of the transaction.  That is, specific performance should be an option when it does not implicate—and thus might not endanger—the self-determination of the promisor’s future self.  The economic analysis discussed above may suggest that rational maximizers would, more often than not, refrain from making this choice.208  But at times, even for these parties, actual performance may be of the essence.  Where the parties affirmatively indicate that their specific plan requires the backstop of specific performance, a liberal contract law should—absent conflicting considerations—facilitate their choice.209

2.   Equity’s Troubling Legacy

Unfortunately, the common law does not comply with this position.  As the Idaho Supreme Court noted in a recent case, “[a] contract clause which gives a non-breaching party the right to elect the remedy of specific performance does not require a court to award specific performance.”210  Instead, its only effect is to provide “some additional support to finding that specific performance is equitable.”211

We agree there are important reasons (addressed below) that may justify overriding the parties’ choice to include a specific performance clause in their contract.  And we agree these reasons validate the conclusion that such a clause should guide, and not bind, the court.  But courts can and should go further than they have been willing to go in accommodating party choice.  For now, they just indicate that the parties’ resort to a specific performance clause “may guide a trial court’s equitable determinations.”212  That’s not good enough.  Courts need to explain the circumstances that would justify their refusal to be guided by the parties’ choice—and there is only a limited list of acceptable explanations.

Courts’ plenary discretionary power regarding the enforceability of specific performance clauses is a product of the equitable origin of specific performance.  “Parties cannot by contract compel a court of equity to exercise its powers,”213 says one court.  The Restatement adds, “[b]ecause the availability of equitable relief was historically viewed as a matter of jurisdiction,” the parties are deemed incapable to “vary by agreement” the preconditions of that jurisdiction.214  But as Ian Macneil noted, the doctrine’s arbitrary historical origins obscure the present-day effect of this excessive discretionary power: it unjustifiably curtails contract’s autonomy-enhancing function.215

Disabling the parties from affecting contract law’s remedial scheme is one of the few areas in which the common law does get close to transfer theory—but this is a vice, not a virtue in our view.  For transfer theorists, as Benson explains, breach is tantamount to conversion, so contract remedies must not be understood as “the contingent product of the parties’ individual or joint decisions.”216  In other words, there is—there must be—a “legally categorical difference between terms and remedies.”217  Remedies represent “the law’s coercive response to the civil wrong of breach.”218  Therefore, the parties can inform the court of facts that may be relevant for that response, but they do not have the power to determine the parameters of that response.219

Transfer theory is, however, wrong on this fundamental point.  As we’ve shown above (and in more depth elsewhere), breach is not tantamount to conversion.220  Therefore, there is no reason categorically to deprive the parties of the power to determine remedies in the ordinary course of contracting.  Quite the contrary.  The parties’ plans should, as they often do, cover the eventuality of breach as well.  If they ex ante decide together that specific performance is the best response for this contingency,221 an autonomy-enhancing contract law should generally not hesitate to provide this remedy.

Though it goes beyond the scope of this Article, our approach equally supports (absent concerns of relational injustice) enforcement of liquidated damages clauses that amount to penalties.  Penalty clauses can have the effect of coercing specific performance, and to the extent they do so, both should be evaluated through the same normative framework.

3.   The Proper Limits on Party Choice

We can and should discard the excessive discretion that arises from the doctrine’s equitable origin.  Fixing the law in this way does not imply, however, that courts must simply rubber-stamp parties’ specific performance clauses.222  As always, the hard work is to find the most autonomy-enhancing line between categorical extremes.  

In brief, judicial scrutiny should respond to party choice, but not in an ad-hoc manner.  Refusal to enforce specific performance clauses should comply with the rule of law, which is, as noted, particularly important for contract’s planning function.223  This means courts should be guided by fairly precise rules or by guidance-friendly standards.  Such rules and standards enable their addressees (or their lawyers) to figure out their intended content in advance and thus to predict future effects and possible applications.224

Our discussion thus far implies four autonomy-based reasons for a court to refuse to enforce a specific performance clause—but if none of these applies, then the parties’ remedial choice should be respected:

  1. A specific performance clause should not be enforced where this is simply impossible, a proviso that applies even in civil law systems.225
  2. Parties must not be able to opt into this remedy where it necessarily threatens self-determination, such as with service contracts whose performance—as noted above and elaborated below226—intimately implicates the person of the promisor.
  3. Parties should not be entitled to burden courts with excessive costs of supervision.227  This guideline reflects a broader category of limits on freedom of contract in cases where the parties’ agreement imposes substantial negative external (third-party) effects.228
  4. Finally, as with any other contractual term, a specific performance clause should be unenforceable when it goes below the floor of relational justice.229  An important example for this category—and one in which, unfortunately, American courts do grant specific performance—is mandatory arbitration and no-class-action clauses which upset consumers and employees’ background expectations of access to courts or to reasonably equivalent procedures for dispute resolution.230

Enumerating these reasons here is sufficient to demonstrate the viability of a predictable and justifiable legal framework for judicial scrutiny of specific performance (and penalty) clauses, one that ensures an autonomy-enhancing residual rule of enforcement.  Some of these reasons may and probably should be further refined by courts, legislatures, and code drafters.231

Be that as it may, remedies can be, as we’ve claimed, material to a contract’s substantive terms; terms can depend on remedies.  Parties may care about both.  It’s unprincipled to refuse to recognize party choice simply because of a remedy’s accidental historical origin in equity jurisdiction.232  Respecting autonomy means parties should be able to elect specific performance ex ante, and courts should grant the remedy ex post, subject to the four caveats just noted.

C.   Employment Contracts

1.   The Autonomy Basis for the Personal Services Exclusion

We turn now to the last major component of specific performance doctrine, dealing with exclusion of employment contracts.  The recent Restatement of Employment Law states a bright-line rule: “An employer may not obtain specific performance of the employee’s promise to work.”233  We assume that, like us, many readers would find this rule nearly a truism.  But this wasn’t the law until the nineteenth century.  Quite the contrary.  Employment contracts were enforced through both specific performance and criminal penalties—it took a lengthy and concerted effort to create the personal services contract exclusion.234

Banning specific enforcement of employment contracts has not been cost-free—even to employees.  As Christopher Wonnell argues, “rational reasons exist for employees to bind themselves to particular employers for a specific period of time.”235  Such arrangements, he explains, “can avoid the high costs of delayed productivity, prevent the risk of situational monopolies due to detrimental reliance, and shift some of the risks of one’s employment productivity to the employer.”236  Therefore, he concludes, current law disempowers employees by preventing them from extracting “more favorable terms from employers in exchange for enforceable promises to fulfill their parts of the bargain.”237

In some contexts, the proenforcement view may be readily dismissed because of employers’ market power or other concerns of relational injustice in the formation of employment contracts.  But the rule is not limited to such cases or even to labor markets typified by these characteristics.  Thus, if this rule’s plenary scope is to be justified, we must look elsewhere.

Autonomy is the right starting point.  As Mindy Chen-Wishart argues, “the bar to specific performance of contracts of personal services (where damages are most likely to be inadequate)” is best explained by reference to individuals’ right “to reassess and to break from past commitments, especially long-term or personal commitments.”238  Indeed, as we’ve noted above, the right to re-write the story of one’s life is most impinged where specific performance means that one’s future self is compelled to do specific things—by requiring someone personally to cooperate with another’s project.239  This autonomy-based reason is why modern (liberal) contract law is justified in limiting the ability of employees’ current selves to commit—even where this limitation means they pay a cost for their future selves’ freedom.

Wonnell counters that this limit is a form of paternalism.240  In turn, paternalism is unjustified because it distrusts people’s agency and thus offends their autonomy.241  This line of reasoning might be a devastating critique of our approach—but only if we relied on people’s imperfect insight to ground the no-enforcement rule.

Fortunately, our account does not assume imperfect insight, as we discussed earlier,242 and thus it is not vulnerable to this paternalism charge.  In choice theory, contract enforcement is justified—and therefore circumscribed—by reference to its autonomy-enhancing function.  This means attempts to use this instrument that are likely to be autonomy-reducing must be treated as ultra vires (at least prima facie).  In other words, contract cannot legitimately contravene the autonomy of the future self, properly understood.

Employment is not the only context in which the current self, attempting to use contract as a planning device for long-term interactions, nevertheless faces limits on committing the future self’s autonomy.  The law governing co-ownership of land strictly limits people’s ability to use contract to lock themselves together: the rights to sell one’s share of co-owned land and to initiate a partition action are semi-inalienable, and can thus be suspended contractually only for limited periods.243  Likewise, the law of spousal contracts refuses to enforce arrangements that jeopardize a spouse’s decision to exit, prohibiting any “provision that by its terms disfavors a party because that party initiates the divorce action.”244

Such limits on the power to commit make successful cooperation more challenging and thus costlier.  A strong right to exit tends to undermine interpersonal sharing and trust by exacerbating the difficulty of collective action, inviting opportunism, and threatening cooperation even in long-term relationships.245  But this burden the parties’ current selves incur—these limits on their ability to contract—is the inevitable price of liberal contract law’s commitment to autonomy.  An autonomy-enhancing contract law—committed to people’s right both to write and re-write their life story—must ensure some ability to withdraw or to refuse to further engage; to dissociate, to cut oneself out of a relationship with other persons.246

There is much to say beyond the scope of this Article on how liberal law can (and to some extent does) accommodate its loyalty to the self-determination of people’s current and future selves in the contexts of co-ownership and spousal contracts.247  Here, we aim simply to highlight the common denominator among the core examples.  All three cases—marriage, co-ownership of land, and employment—typically involve “ground projects,” that is, the projects that make people who they are and give meaning to their lives.248  This constitutive quality is why liberal law treats the future self’s change of mind as a “conclusive reason,” one that justifiably overrides the current self’s choices.249

Indeed, at least since the decline of feudalism, work has figured prominently in people’s adult lives not only as a means to an end.  Although it is surely a means as well, work is also a ground project; for many, it is the quintessential one.250  This is why excessively limiting the promisor’s control over her work is autonomy-defying and thus, by definition, beyond the justifiable limits of contract.

2.   The Asymmetry of Employer and Employee

Understanding the liberal foundation to the bar on specific performance of personal service contracts leads to an important reformist payoff: the autonomy criterion does not bind symmetrically in the employment context—unlike the co-ownership and marriage cases in which it applies equally to all contracting parties.251  As a consequence, the law should not necessarily apply the same rules to employers as it does to employees.  In many cases, the labor market is typified by a corporate employer with many employees (at times, thousands) with whom it has no personal connection.  In these cases, a bright-line immutable bar to specific performance when the employee is the injured party cannot be justified, and indeed should be rejected.

We do not state this reform as a blanket rule.  There are some contexts—think of a small business with a few employees or a small family corporation—in which the employment relationship is indeed intimately connected to the employer’s life project.  In such cases, the “traditional unwillingness of courts of equity to enforce contracts for personal service either at the behest of the employer or of the employee”252 remains fully justified.  As always, the challenge is to draw sufficiently precise boundaries such that people can reliably plan ex ante, while remaining true to contract’s animating principles.

The crucial difference between employee and employer implies that, at the very least, absent any other conflicting reason (such as excessive judicial supervision costs), specific performance should be available to employees where the parties agree to such a remedy.253  Unfortunately, here again the common law fails—as with the other legitimate attempts consensually to expand the scope of specific performance we discussed earlier.254  As the Restatement notes, in nonstatutory cases, “a court will almost never grant reinstatement of an individual providing services in a personal-services contract.”255  And courts consistently refuse to specifically enforce “an employer’s agreement, promise, or statement” that it will continue to employ an employee.256  This common-law position is wholly unwarranted—and it stands in contrast with statutory contexts, particularly in collective bargaining agreements, in which unlawfully dismissed employees are routinely reinstated.257

The significance of work to employees’ self-determination implies that where the employment relationship is purely instrumental to their employers, the alternative of opting into specific performance should be readily available to employees and not to employers.

3.   Noncompete Agreements and Autonomy

This critique leads us to our last point: the great challenge that employee noncompete agreements—which have become endemic in recent years258—present to an autonomy-enhancing contract law.  Some (perhaps many) of these agreements are abusive, but not all of them.  The abusive ones are easy cases:  they should not be enforceable.  The most pointed normative difficulty arises instead from nonabusive agreements.  Many come about where the current self makes a plan that is genuinely empowering—in exchange for agreeing to a noncompete, the employee not only earns more, but also gains upgraded skills that may open up new professional horizons.259

The existing doctrine governing noncompetes is complex, varies widely across jurisdictions, and is currently in substantial flux.260  The unifying point, however, is that where a noncompete imposes a significant encumbrance on the future self, specific performance is not granted even where such quid pro quos are generally available.  Liberal contract law cannot remain agnostic toward severe limitations on the ability of the employee’s future self to rewrite the story of her life.

Setting the proper limits for specific performance here, however, is genuinely difficult.  Safeguarding the future-self’s right to rewrite her life story may not only impose costs on the current self, but also, as just noted, limit the autonomy-enhancing potential that an employment contract could have generated in the first instance.261  The significant challenge that liberal contract law faces in this context is to develop rule-of-law-friendly informative standards and categories that properly accommodate the conflicting autonomy claims of employees’ current and future selves.

The basic thrust of the informative standards now applied in many jurisdictions—examining the reasonableness of noncompetes in terms of occupational, geographic, and temporal scope262—seems, well, reasonable, at least as a first approximation.  Additionally, the rapid pace of reform among American states on noncompetes is pointing toward more autonomy-friendly boundaries for the doctrine, such as, for example, categorically refusing enforcement against low-wage, seasonal, and unskilled employees.263  For the moment—and this is quite a recent phenomenon—both legislative and judicial reforms are better aligning noncompete law with its underlying autonomy imperatives.

 Concluding Remarks

The preference for damages over specific performance in Anglo-American law goes way back, at least to 1616, with Lord Coke’s opinion in Bromage v. Genning.264  For over 400 years, common-law courts have hemmed in specific performance, with a handful of exceptions and exclusions—in sharp contrast with the civil law tradition.  Today, comparativists catalog small oscillations—convergences and divergences—among these regimes, while historians trace the contingent path of specific performance through courts of equity and law.  Legal economists offer contingent reasons to endorse the current state of affairs, while legal philosophers bemoan the resilience of the expectation principle.  But none of these disciplines persuasively answers the question we started with: when should specific performance be available for breach of contract?

This Article provides the answer: respect for autonomy of the future self explains why damages rather than specific performance are the ordinary remedy for contract breach.  The same normative commitment to the contracting parties’ autonomy explains the “uniqueness exception” and the personal services exclusion.  For the most part, the boundaries of specific performance track the common law’s fundamental normative structure.  But not entirely.  There’s still work to be done, and this Article points the way.

©2023 Hanoch Dagan & Michael Heller.  Individuals and nonprofit institutions may reproduce and distribute copies of this Article in any format at or below cost, for educational purposes, so long as each copy identifies the authors, provides a citation to the Notre Dame Law Review, and includes this provision in the copyright notice.

*Stewart and Judy Colton Professor of Legal Theory and Innovation & Director of the Edmond J. Safra Center for Ethics, Tel-Aviv University.

**Lawrence A. Wien Professor of Real Estate law, Columbia Law School. Thanks to Ronen Avraham, Oren Bar-Gill, Samuel Becher, Itzik Benbaji, Leora Bilsky, Brian Bix, Yishai Blank, Avihay Dorfman, Vic Goldberg, Jim Gordley, Sharon Hannes, Matthew Harding, Ron Harris, Robin Kar, Roy Kreitner, Tamar Kricheli-Katz, Kaiponanea Matsumura, Menny Mautner, Yifat Naftali, David Schorr, Irit Samet, Alan Schwartz, Stephen Smith, and workshop participants at Tel-Aviv and Villanova for their helpful comments.

1In common law, specific performance is a remedy in equity.  Our reference here and throughout to common law is a shorthand to “common-law jurisdictions.”

2See Hanoch Dagan & Michael A. Heller, The Liberal Commons, 110 Yale L.J. 549 (2001) (celebrating the civil law approach to co-ownership).

3For the canonical statement of this “choice theory” approach, see Hanoch Dagan & Michael Heller, The Choice Theory of Contracts (2017).  For refinements to the concept of autonomy that bear on specific performance, see Hanoch Dagan & Michael Heller, Freedom, Choice, and Contracts, 20 Theoretical Inquiries L. 595 (2019); Hanoch Dagan & Michael Heller, Why Autonomy Must Be Contract’s Ultimate Value, 20 Jerusalem Rev. Legal Stud. 148 (2019) [hereinafter Dagan & Heller, Why Autonomy]; Hanoch Dagan & Michael Heller, Autonomy for Contract, Refined, 40 Law & Phil. 213 (2021) [hereinafter Dagan & Heller, Autonomy Refined]; Hanoch Dagan & Michael Heller, Choice Theory: A Restatement, in Research Handbook on Private Law Theory 112 (Hanoch Dagan & Benjamin C. Zipursky eds., 2020) [hereinafter Dagan & Heller, Restatement].

4On the distinction between normative and majoritarian defaults, see infra subsection II.B.3; see also Hanoch Dagan & Avihay Dorfman, Justice in Contracts, 67 Am. J. Juris. 1 (2022).

5As the text clarifies, by “common-law baseline” we refer to the doctrinal baseline of the common law, rather than to its historical baseline.

6Not all economists advocate for the common-law baseline.  For example, in an influential article, Steven Shavell supports the distinction in French contract law between conveying existing property and producing new goods.  See Steven Shavell, Specific Performance Versus Damages for Breach of Contract: An Economic Analysis, 84 Tex. L. Rev. 831 (2006).  Conversely, not all philosophers condemn the common law.  For examples premised on Hegelian philosophy and Scanlonian contractualism, see Jennifer Nadler, Freedom from Things: A Defense of the Disjunctive Obligation in Contract Law, 27 Legal Theory 177 (2021); Robin Kar, Contract as Empowerment, 83 U. Chi. L. Rev. 759 (2016).  We cannot address these specific philosophical accounts here, but have criticized them in other contexts elsewhere.  See Hanoch Dagan, Liberalism and the Private Law of Property, 1 Critical Analysis L. 268 (2014); Hanoch Dagan, The Value of Choice and the Justice of Contract, 10 Juris. 422 (2019).

7For this framing of specific performance, see Stephen A. Smith, Rights, Wrongs, and Injustices: The Structure of Remedial Law 163 (2019).  As the Restatement notes, “[a]n order of specific performance . . . will be so drawn as best to effectuate the purposes for which the contract was made.”  Restatement (Second) of Conts. § 358(1) (Am. L. Inst. 1981).

8Charles Szladits, The Concept of Specific Performance in Civil Law, 4 Am. J. Compar. L. 208, 212 (1955).

9Id. at 217, 221.

10See, e.g., Shael Herman, Specific Performance: A Comparative Analysis (1), 7 Edinburgh L. Rev. 5 (2003); see also Shavell, supra note 6.

11Gerard De Vries, Right to Specific Performance: Is There a Divergence Between Civil- and Common-Law Systems and, If So, How Has It Been Bridged in the DCFR?, 17 Eur. Rev. Priv. L. 581, 581 (2009).

12Ernst Rabel, A Draft of an International Law of Sales, 5 U. Chi. L. Rev. 543, 559 (1938); see also, e.g., John P. Dawson, Specific Performance in France and Germany, 57 Mich. L. Rev. 495, 537–38 (1959).

13Smith, supra note 7, at 164–65; see also, e.g.Edward Yorio & Steve Thel, Contract Enforcement: Specific Performance and Injunctions § 2.4, at 2–12 (2d ed. Supp. 2020); Mindy Chen-Wishart, Specific Performance and Change of Mindin Commercial Remedies: Resolving Controversies 98, 108 (Graham Virgo & Sarah Worthington eds., 2017).  Smith claims, however, that “it is misleading to describe specific performance as secondary or exceptional,” since “one of the most common private law remedies—an order for a sum due—is in substance, even if not in name, specific performance.” Smith, supra note 7, at 164.  Our account of specific performance better explains the difference between these two types of orders.  The difference is not in name only; quite the contrary.  The gap between the rarity of specific performance orders and the prevalence of orders for the sum due vindicates our thesis:  what accounts for the former phenomenon is its unique effect on the autonomy of our future selves, an effect that is substantially absent from orders for the sum due.  See also Chen-Wishart, supra, at 125.

14See Theodore Eisenberg & Geoffrey P. Miller, Damages Versus Specific Performance: Lessons from Commercial Contracts, 12 J. Empirical Legal Stud. 29, 38–41 (2015). 

15Restatement (Second) of Conts. § 346 (Am. L. Inst. 1981).

16Id. § 357(1).

17Id. § 359(1).  Further, while specific performance “will not be refused merely because there is a remedy for breach other than damages,” such a remedy may be considered as a factor in the court’s discretion.  Id. § 359(3).

18See id. § 360 (referring to “(a) the difficulty of proving damages with reasonable certainty, (b) the difficulty of procuring a suitable substitute performance by means of money awarded as damages, and (c) the likelihood that an award of damages could not be collected”).  The Restatement further adds that specific performance will be refused if such relief “would cause unreasonable hardship or loss to the party in breach or to third persons,” id. § 364(1)(b), as well as “if the character and magnitude of the performance would impose on the court burdens in enforcement or supervision that are disproportionate to the advantages to be gained from enforcement and to the harm to be suffered from its denial.”  Id. § 366.

19See, e.g., 5A Arthur Linton Corbin, Corbin on Contracts § 1142 (1964); Smith, supra note 7, at 168. 

20U.C.C. § 2-716 (Am. L. Inst. & Unif. L. Comm’n 2021); see also id. § 2-716 cmt. 2; see also, e.g., Coop. Ins. Soc’y v. Argyll Stores (Holdings) Ltd.1998 AC 1 (HL) (appeal taken from Eng.).

21See also, e.g., E. Air Lines, Inc. v. Gulf Oil Corp., 415 F. Supp. 429, 442–43 (S.D. Fla. 1975).

22U.C.C. § 2-716 cmt. 2 (Am. L. Inst. & Unif. L. Comm’n 2021).

23Cumbest v. Harris, 363 So. 2d 294, 297 (Miss. 1978); U.C.C. § 2-713 cmt. 3 (Am. L. Inst. & Unif. L. Comm’n 2021); see also, e.g., Sedmak v. Charlie’s Chevrolet, Inc., 622 S.W.2d 694, 700 (Mo. Ct. App. 1981); Chadwell v. English, 652 P.2d 310, 314 (Okla. Civ. App. 1982); King Aircraft Sales, Inc. v. Lane, 846 P.2d 550, 555–57 (Wash. Ct. App. 1993). 

24Ruddock v. First Nat’l Bank, 559 N.E.2d 483, 487, 488–89 (Ill. App. Ct. 1990); see also, e.g., Kaiser Trading Co. v. Associated Metals & Mins. Corp., 321 F. Supp. 923, 932–33 (N.D. Cal. 1970); Tower City Grain Co. v. Richman, 232 N.W.2d 61, 66 (N.D. 1975).  Other courts, as the text implies, do not take to heart the Code’s additional language.  See, e.g., Klein v. PepsiCo, Inc., 845 F.2d 76, 80 (4th Cir. 1988); Bander v. Grossman, 161 Misc. 2d 119, 123–125 (N.Y. Sup. Ct. 1994); Scholl v. Hartzell, 20 Pa. D. & C.3d 304, 308–09 (Pa. Ct. Com. Pl. 1981).

25See U.C.C. § 2-716 cmt. 1 (Am. L. Inst. & Unif. L. Comm’n 2021).

26See, e.g., Szladits, supra note 9, at 226.

27Restatement (Second) of Conts. § 367(1) (Am. L. Inst. 1981).

28Id. § 367(2).

29O.W. Holmes, The Path of the Law, 10 Harv. L. Rev. 457, 462 (1897).

30Seana Shiffrin, Could Breach of Contract Be Immoral?, 107 Mich. L. Rev. 1551, 1551 (2009); see also, e.g., Daniel Friedmann, The Efficient Breach Fallacy, 18 J. Legal Stud. 1, 2 (1989).

31Richard A. Posner, Economic Analysis of Law 57 (1972).  Posner was not the first to introduce this analysis.  See, e.g., Robert L. Birmingham, Breach of Contract, Damage Measures, and Economic Efficiency, 24 Rutgers L. Rev. 273, 284 (1970).

32See, e.g., Daniel Markovits & Alan Schwartz, The Myth of Efficient Breach: New Defenses of the Expectation Interest, 97 Va. L. Rev. 1939, 1944 (2011).

33The text implicitly assumes a (conceptual and normative) continuity between contractual terms and contract remedies.  For a defense of this assumption, see infra text accompanying notes 220–21.

34See, e.g., Markovits & Schwartz, supra note 32, at 1948.

35Because our aim here is to outline the most powerful economic defense of the common law, we leave aside influential economic analysts such as Steven Shavell, who advocates for an approach like in French law distinguishing between producing new goods and conveying already-existing ones.  See Shavell, supra note 6.

36See Van Wagner Advert. Corp. v. S & M Enters., 492 N.E.2d 756, 759–60 (N.Y. 1986).

37Anthony T. Kronman, Specific Performance, 45 U. Chi. L. Rev. 351, 359 (1978).

38Id. at 362.


40Id. at 362–63 .

41Id. at 365.  Kronman’s own account of this step, which is based on a calculus of renegotiation costs, has been convincingly criticized.  See Robert E. Scott & Jody S. Kraus, Contract Law and Theory 114 (5th ed. 2013).

42Scott & Kraus, supra note 41, at 114. 

43Id. at 115. 



46Id. at 114–15.

47Kronman, supra note 37, at 362.

48For the implications of this phenomenon on the economic understanding of market mechanisms, and more specifically contract, see Hanoch Dagan, Why Markets?  Welfare, Autonomy, and the Just Society, 117 Mich. L. Rev. 1289, 1297–1300 (2019); Dagan & Heller, Why Autonomysupra note 3, at 158–59; see also Przemysław Pałka, Algorithmic Central Planning: Between Efficiency and Freedom, 83 Law & Contemp. Probs. 125 (2020).

49See infra Part II. 

50See Robert Cooter & Thomas Ulen, Law & Economics 287–93, 307–08 (6th ed. 2012).

51The contingency of economic accounting is well demonstrated by the detailed analysis of the (mirroring) costs and benefits of damages and injunctions in Judge Posner’s opinion in Walgreen Co. v. Sara Creek Property Co., 966 F.2d 273 (7th Cir. 1992).

52For perhaps the most sustained efforts by economists to contest the philosophers’ critique, see Richard Craswell, Contract Law, Default Rules, and the Philosophy of Promising, 88 Mich. L. Rev. 489 (1989) and Steven Shavell, Why Breach of Contract May Not Be Immoral Given the Incompleteness of Contracts, 107 Mich. L. Rev. 1569 (2009) (replying to Shiffrin, supra note 30).  Craswell’s article persuasively shows the irrelevance of then-existing autonomy-based theories, but does not speak to more powerful, still-flawed, later accounts such as Shiffrin’s that aimed to improve on Charles Fried’s version and to rebut Craswell’s.  Shavell’s exchange with Shiffrin is also off point: his criticisms there do not reach to the account we develop in Part II, below.  We explore this debate in more detail in Dagan & Heller, supra note 3, at 25–47.

53Because we have demonstrated these points in detail elsewhere for contract law as a whole, here we focus on the aspects tied to specific performance.  Note also, there are dissenting views that seek to accommodate the common-law approach to specific performance with transfer theory, but we find them unconvincing.  See Hanoch Dagan, Two Visions of Contract, 119 Mich. L. Rev. 1247, 1265–67 (2021).  And conversely, there are efforts to justify the morality of the common-law view, premised on Hegelian philosophy and Scanlonian contractualism.  Our responses to these approaches lie outside the scope of this Article.  See supra note 6. 

54Seana Valentine Shiffrin, The Divergence of Contract and Promise, 120 Harv. L. Rev. 708 (2007).


56Id. at 722.

57Id. Lionel Smith, Understanding Specific Performancein Comparative Remedies for Breach of Contract 221, 233 (Nili Cohen & Ewan McKendrick eds., 2005) (stating that specific performance “prevents a defendant from expropriating a plaintiff’s patrimonial entitlement to performance”).

59Shiffrin, supra note 54, at 722–24.

60Id. at 733.

61Shiffrin, supra note 30, at 1568.

62Shiffrin, supra note 54, at 733. 


64Id. at 710.

65Id. at 724.

66Id. at 729.

67Seana Valentine Shiffrin, Must I Mean What You Think I Should Have Said?, 98 Va. L. Rev. 159, 176 (2012).



70Seana Valentine Shiffrin, Is a Contract a Promise?, in The Routledge Companion to Philosophy of Law 241, 242–44 (Andrei Marmor ed., 2012). 

71Seana Valentine Shiffrin, Promising, Intimate Relationships, and Conventionalism, 117 Phil. Rev. 481, 517 (2008).

72Shiffrin, supra note 70, at 244. 

73Peter Benson, Justice in Transactions: A Theory of Contract Law 41 (2019).

74Id. at 321.

75Id. at 251.

76Id. at 247; see also, e.g., Arthur Ripstein, Kantian Perspectives on Private Law, in The Oxford Handbook of the New Private Law 69, 74 (Andrew S. Gold, John C.P. Goldberg, Daniel B. Kelly, Emily Sherwin & Henry E. Smith eds., 2021) (explaining that a contract right “gives one person the entitlement to have another person’s action available for the first person’s purposes. . . .  [I]t is one person’s entitlement to determine (that is, demand or compel) a particular action from another . . . .”).

77See Dagan & Hellersupra note 3, at 36–40; Dagan & Heller, Autonomy Refinedsupra note 3, at 215–38.  The remainder of this Section heavily draws on these sources, reproducing and citing content where appropriate. 

78See Dagan & Hellersupra note 3, at 37.

79Id. at 36.

80Id. at 38.

81Id. at 39.

82See Shiffrin, supra note 54.

83See Dagan & Hellersupra note 3, at 37. 

84Id. at 39.  Thus, in presupposing that contract must necessarily fit the pattern of a one-shot reassignment of future entitlements and risks, transfer theory is unable to account for many contract types that go beyond this limiting barter-like pattern.  In these types, a party’s current self obligates her future self to cooperate with the other party, which inevitably creates the intrapersonal dilemmas that doctrines like specific performance address.

85Dagan & Heller, Autonomy Refinedsupra note 3, at 230.

86Id. at 215.

87See supra text accompanying note 11.

88A recent study shows that, as an empirical matter, “most people do not perceive breach of contract followed by compensation for the promisee as immoral.”  See Sergio Mittlaender, Morality, Compensation, and the Contractual Obligation, 16 J. Empirical Legal Stud. 119, 119 (2019).

89See supra notes 47–49 and accompanying text.

90This section heavily draws on Dagan & Hellersupra note 3; Dagan & Heller, Restatementsupra note 3; Dagan, supra note 53, reproducing and citing content where appropriate.

91For those interested in the jurisprudential heavy lifting that nails down our refutation of consent and transfer theories, unpacks its implications, and responds to critics, see Dagan & Hellersupra note 3, at 19–47; Dagan & Heller, Autonomy Refinedsupra note 3.

92Dagan, supra note 53, at 1253. 


94Dagan & Heller, Restatementsupra note 3, at 120.

95Dagan, supra note 53, at 1248. 

96Id. at 1253. 


98Id. (citing Charles Fried, Contract as Promise: A Theory of Contractual Obligation 8 (2d ed. 2015)).


100The law in action is admittedly often different: at times, parties waive this right and in other cases they resort to nonlegal sanctions for its vindication.  But this observation does not negate the significance of the legal right to expect.  Rather, it points to the way the right often functions in the background, rather at the foreground, of social life.  SeeScott & Kraus, supra note 41, at 594; cf. Dagan & Heller, supra note 2, at 577–81.

101Dagan & Heller, Restatementsupra note 3, at 116. 

102This last proposition is not self-evident.  One of us has defended it at some detail in work coauthored with Avihay Dorfman.  See, e.g., Hanoch Dagan & Avihay Dorfman, Just Relationships, 116 Colum. L. Rev. 1395 (2016) [hereinafter Dagan & Dorfman, Just Relationships]; Hanoch Dagan & Avihay Dorfman, The Domain of Private Law, 71 U. Toronto L.J. 207 (2021); Dagan & Dorfman, supra note 4; Hanoch Dagan & Avihay Dorfman, Precontractual Justice, 28 Legal Theory 89 (2022) [hereinafter Dagan & Dorfman, Precontractual Justice]; see also Hanoch Dagan, A Liberal Theory of Property 114–47 (2021); Hanoch Dagan, Autonomy, Relational Justice and the Law of Restitution, in Research Handbook on Unjust Enrichment and Restitution 219 (Elise Bant, Kit Barker & Simone Degeling eds., 2020); Avihay Dorfman, Relational Justice and Torts, in Research Handbook on Private Law Theorysupra note 3, at 321.

103Dagan & Heller, Restatementsupra note 3, at 112. 

104Charles Fried, The Ambitions of Contract as Promise, in Philosophical Foundations of Contract Law 17, 20 (Gregory Klass, George Letsas & Prince Saprai eds., 2014).

105Dagan & Heller, Restatementsupra note 3, at 112.

106See Dagan & Hellersupra note 3. 

107Dagan & Heller, Restatementsupra note 3, at 118. 

108Id. at 119.


110Dagan, supra note 53, at 1255. 

111Id. at 1256 (first citing U.C.C. §§ 2-204(3), 2-305 (Am. L. Inst. & Unif. L. Comm’n 2017); and then citing Restatement (Second) of Conts. § 33 (Am. L. Inst. 1981)).

112See Hanoch Dagan, Types of Contracts and Law’s Autonomy-Enhancing Role, in European Contract Law and The Creation of Norms 109 (Stefan Grundmann & Mateusz Grochowski eds., 2021). 

113See Dagan & Hellersupra note 3, at 65–137.  Indeed, our initial work in developing choice theory focused squarely on the essential, and previously overlooked, role of types in contract law.

114Dagan, supra note 53, at 1260–61.

115Dagan & Heller, Restatementsupra note 3, at 126.



118See, e.g., Bensonsupra note 73, at 364, 367, 377.

119See Dagan & Heller, Restatementsupra note 3, at 128.



122For a detailed defense, both normative and positive, of the role of relational justice in contract law, see Dagan & Dorfman, Precontractual Justice, supra note 102; Dagan & Dorfman, supra note 4.

123This section heavily draws on Dagan & Hellersupra note 3; Dagan & Heller, Restatementsupra note 3; Dagan, supra note 53; Hanoch Dagan & Ohad Somech, When Contract’s Basic Assumptions Fail, 34 Canadian J.L. & Juris. 297 (2021), reproducing and citing content where appropriate.

124See Michael E. Bratman, Time, Rationality, and Self-Governance, 22 Phil. Issues 73, 74 (2012) (explaining that planning agency implies “diachronic rationality constraints”).

125Dagan & Heller, Restatementsupra note 3, at 123.


127Dagan & Somech, supra note 123, at 310 & n.59, 315.  Such cases of regret should be carefully distinguished from cases in which the parties were mistaken regarding the basic assumptions on which their contracts were based.  See id. at 315.

128Dagan & Heller, Restatementsupra note 3, at 123.

129Bratman, supra note 124, at 82.


131Dagan & Somech, supra note 123, at 311.



134Cf. Aditi Bagchi, Contract and the Problem of Fickle People, 53 Wake Forest L. Rev. 1, 3 (2018); Dori Kimel, Personal Autonomy and Change of Mind in Promise and in Contractin Philosophical Foundations of Contract Law, supra note 104, at 96, 99–101.

135Dagan & Heller, Restatementsupra note 3, at 112.

136See supra subsection II.B.1.

137See supra note 15 and accompanying text. 

138Exiting a relationship admittedly affects values other than autonomy, notably community and efficiency.  But for a genuinely liberal contract law, these values are either constitutive of or instrumental to autonomy, which is contract’s ultimate value.  This means that, but for the most extreme cases, autonomy enjoys a lexical priority.  See Dagan & Heller, supra note 3, at 84–85.

139Being a limit on people’s legitimate jurisdiction to undertake future commitments, this principle focuses on promisors, rather than promisees, as the text emphasizes.  Oftentimes, of course, each contractor is both a promisor and a promisee, which means that the principle does apply to both, but then it applies to each in her capacity as a promisor.

140See, e.g., Susan Rose-Ackerman, Inalienability and the Theory of Property Rights, 85 Colum. L. Rev. 931, 937–41 (1985).

141Contra Kaiponanea T. Matsumura, Binding Future Selves, 75 La. L. Rev. 71, 77, 98, 113 (2014).

142Dagan & Somech, supra note 123, at 311 n.62.




146Dagan & Heller, Restatementsupra note 3, at 124.


148See Dagan & Somech, supra note 123, at 311.

149Id. at 298.

150Dagan, supra note 53, at 1267.

151See, e.g., Alan Schwartz & Robert E. Scott, Contract Theory and the Limits of Contract Law, 113 Yale L.J. 541, 544 (2003).

152Dagan, supra note 53, at 1265.

153See id. at 1266.

154For more on the notion of normative defaults and its doctrinal manifestations, see Dagan & Dorfman, supra note 4.

155Some readers may object, arguing that contractors are likely to prefer the rule we advance, which limits the authority of the present self over the future self to cases in which it is in fact needed for the former’s self-determination.  But the meaning of taking this proposition as a given—rather than as an empirical hypothesis that may be disproved if, for example, it turns out that people’s actual preferences are dominated by their welfarist interest—is to subscribe to, rather than refute, our position.  For more on the relationship between autonomy and preferences, see Hanoch Dagan & Roy Kreitner, Economic Analysis in Law, 38 Yale J. on Regul. 566 (2021).

156See supra text accompanying notes 59–70.

157Cf. Curtis Bridgeman, Contracts as Plans, 2009 U. Ill. L. Rev. 341, 384–85.

158Cf. H.L.A. Hart, Between Utility and Rights, 79 Colum. L. Rev. 828, 834–35 (1979) (Liberals should recognize the significance of the “unexciting but indispensable chore” of distinguishing “between the gravity of the different restrictions on different specific liberties and their importance for the conduct of a meaningful life.”); Hanoch Dagan, The Jurisprudence of Liberal Property: A Reply, 13 Jurisprudence 668, 675–76 (2022) (discussing the notion of “autonomy-based audit”).

159As the text implies, we insist that the concern for the autonomy of the future self is not external to contract.  But cf. Dori Kimel, From Promise to Contract: Towards a Liberal Theory of Contract 103–04, 108 (2003); Ewan McKendrick, The Common Law at Work: The Saga of Alfred McAlpine Construction Ltd v Panatown Ltd, 3 Oxford U. Commonwealth L.J. 145, 172 (2003).

160Such termination power may nonetheless be appropriate where the promisor is a government, whose authority stands for the self-determination of the nation at large, which means, in a democracy, a rather robust right to re-write the story of our collective life (as represented by the practice of periodic elections).  See Dagan & Heller, supra note 3, at 100.

161Notice that this concern is inapplicable in cases in which the past that the agent wishes to rewrite is purely self-regarding, which may explain the basic restitutionary rule in mistaken payments law.  See Hanoch Dagan, Mistakes, 79 Tex. L. Rev. 1795, 1800–03 (2001).

162Cf. Dagan & Dorfman, Just Relationshipssupra note 102, at 1455; Kimel, supra note 134, at 114.

163The burden here is comparable to the restitutionary burden that the basic, uncontroversial rule of the law governing mistaken payments imposes on their recipients.  See Dagan, supra note 102, at 222–25.

164See Alan Schwartz, The Case for Specific Performance, 89 Yale L.J. 271, 276 (1979).

165Cf. Chen-Wishart, supra note 13, at 119; Stephen A. Smith, Future Freedom and Freedom of Contract, 59 Mod. L. Rev. 167, 179–80 (1996).

166See Restatement (Second) of Conts. § 360 cmt. e (Am. L. Inst. 1981).

167See, e.g., Margaret Jane Radin, Property and Personhood, 34 Stan. L. Rev. 957, 991–92, 1013 (1982); see also, e.g., Lorna Fox, The Meaning of Home: A Chimerical Concept or a Legal Challenge?, 29 J.L. & Soc’y 580 (2002).

168Gartrell v. Stafford, 11 N.W. 732, 734 (Neb. 1882).

169Cf. Kar, supra note 6, at 795; Nadler, supra note 6, at 193–96.

170See intra Section III.A. 

171Restatement (Second) of Conts. § 364(1)(b) (Am. L. Inst. 1981).

172See supra note 25 and accompanying text.

173See Restatement (Second) of Conts. § 359(1) (Am. L. Inst. 1981); supra note 17 and accompanying text.

174Jacob & Youngs, Inc. v. Kent, 129 N.E. 889, 891 (N.Y. 1921).

175See intra Part III.B. 

176See supra text accompanying notes 26–28. 

177Cf. Anthony T. Kronman, Paternalism and the Law of Contracts, 92 Yale L.J. 763, 780–85 (1983).  Kronman’s point is different from ours in that he focuses on the ex post threat that compulsory performance poses to the promisor’s integrity or self-respect in case her values have changed dramatically since she entered the contract so that she can no longer identify it as hers.  Id.

178But see Smith, supra note 7, at 310; Arthur Ripstein, The Contracting Theory of Choices, 40 Law & Phil. 185 (2021).

179As Douglas Laycock notes, courts deny specific relief of personal service contracts even if damages would be inadequate.  See Douglas Laycock, The Death of the Irreparable Injury Rule, 103 Harv. L. Rev. 687, 746 (1990).

180Restatement (Second) of Conts. § 360 cmt. e (Am. L. Inst. 1981).  The traditional view seems to also prevail elsewhere, notably in Australia.  See Pianta v Nat’l Fin & Trs (1964) 180 CLR 146, 151 (Austl.) (Barwick, C.J.); Zhu v Snell 2014 NSWSC 468, paras. 209–11 (Austl.); Fairborne Pty Ltd v Strata Store Noosa Pty Ltd 2009 QSC 250, paras. 14–15 (Austl.); Turner v Bladin (1951) 82 CLR 463, 473–74 (Austl.).  But see Rofiza Pty Ltd v Gangley Pty Ltd 2002 NSWSC 986, para. 36 (Austl.).

181See, e.g., Md. Clay Co. of Baltimore City v. Simpers, 53 A. 424, 426 (Md. 1902); Mohrlang v. Draper, 365 N.W.2d 443, 446–47 (Neb. 1985).

182See Restatement (Second) of Conts. § 360 cmt. e (Am. L. Inst. 1981).

183Payne v. Clark, 187 A.2d 769, 770 (Pa. 1963); see also, e.g., Cox v. RKA Corp., 753 A.2d 1112, 1128–29 (N.J. 2000).

184Kann v. Wausau Abrasives Co., 129 A. 374, 378 (N.H. 1925); see also, e.g., Beaver v. Brumlow, 231 P.3d 628, 637 (N.M. Ct. App. 2010).

185SMS Fin., LLC v. CBC Fin. Corp., 417 P.3d 70, 75 (Utah 2017) (quoting Cummings v. Nielson, 129 P. 619, 624 (Utah 1912)).

186See Keystone Sheep Co. v. Grear, 263 P.2d 138, 142 (Wyo. 1953).

187See, e.g., Justus v. Clelland, 651 P.2d 1206, 1207–08 (Ariz. Ct. App. 1982); Phillips v. Homer (In re Smith Tr.), 745 N.W.2d 754, 759 (Mich. 2008); Texaco, Inc. v. Creel, 314 S.E.2d 506, 512 (N.C. 1984).

188See, e.g., Lonas v. Metro. Mortg. & Sec. Co., 432 P.2d 603, 606 (Alaska 1967); Vincent v. Vits, 566 N.E.2d 818, 819–20 (Ill. App. Ct. 1991); Metro Holdings One, LLC v. Flynn Creek Partner, LLC, 25 N.E.3d 141, 163–64 (Ind. Ct. App. 2014); Thompson v. Kromhout, 413 N.W.2d 884, 885 (Minn. Ct. App. 1987); Osborne v. Bullins, 549 So. 2d 1337, 1340 (Miss. 1989); Morgan & Bro. Manhattan Storage Co. v. Balin, 364 N.Y.S.2d 904, 907 (N.Y. App. Div. 1975); Shuptrine v. Quinn, 597 S.W.2d 728, 730 (Tenn. 1979); Ash Park, LLC v. Alexander & Bishop, Ltd., 783 N.W.2d 294, 304–05 (Wis. 2010); see also Dan B. Dobbs, Handbook on the Law of Remedies: Damages—Equity—Restitution 796, 861 (1973); Larissa Katz, Equitable Remedies: Protecting “What We Have Coming to Us, 96 Notre Dame L. Rev. 1115, 1117 (2021).

189Wolf v. Anderson, 334 N.W.2d 212, 215 (N.D. 1983); see also, e.g., Bailey v. Musumeci, 591 A.2d 1316, 1318 (N.H. 1991).

190Kesler v. Marshall, 792 N.E.2d 893, 897 (Ind. Ct. App. 2003); see also, e.g., Centex Homes Corp. v. Boag, 320 A.2d 194, 196 (N.J. Super. Ct. Ch. Div. 1974).

191See, e.g., Trachtenburg v. Sibarco Stations, Inc., 384 A.2d 1209, 1211–12 (Pa. 1978); Manning v. Bleifus, 272 S.E.2d 821, 824 (W. Va. 1980).

192Geneva Ltd. Partners v. Kemp, 779 F. Supp. 1237, 1241 (N.D. Cal. 1990); see also Goldie’s Bookstore, Inc. v. Superior Ct., 739 F.2d 466, 471 (9th Cir. 1984); M&T Bank v. SFR Invs. Pool 1, LLC, No. 2:17-CV-1867, 2019 WL 3577645, at *4 (D. Nev. Aug. 6, 2019); Medgar Evers Houses Assocs., L.P. v. Carro, No. 01-CV-6107, 2001 WL 1456190, at *8 (E.D.N.Y. Nov. 6, 2001); Simone v. N.V. Floresta, Inc., Nos. 98 Civ. 0268, 98 Civ. 1970, 1999 WL 429504, at *10–11 (S.D.N.Y June 18, 1999); Watkins v. Paul, 511 P.2d 781, 783 (Idaho 1973); Hilton v. Nelsen, 283 N.W.2d 877, 881 (Minn. 1979); H.K. & Shanghai Banking Corp., Ltd. v. Kallingal, 2005 Guam 13 ¶ 22 (quoting Geneva Ltd. Partners, 779 F. Supp. at 1241).

193Bayview Loan Servicing, LLC v. Alessi & Koenig, LLC, 5 F. Supp. 3d 1218, 1220 (D. Nev. 2014) (citing Watkins, 511 P.2d at 783); see also Miller v. LeSea Broadcasting, Inc., 87 F.3d 224, 230 (7th Cir. 1996).  New Zealand’s Court of Appeals has taken a similar position.  See Landco Albany Ltd v. Fu Hao Constr. Ltd. 2005 NZCA 293, 2006 2 NZLR 174 at 43 (N.Z.) (“[T]he respondent’s interest in the land is plainly commercial rather than private or sentimental.  It must have entered into the transaction in order to make a profit and in those circumstances damages would be an adequate remedy.”).

194Paramadevan v. Semelhago 1996 2 S.C.R. 415, 425 (Can.).

195Id. at 428.

196Id. at 429; see also Domowicz v. Orsa Invts. Ltd. (1993), 15 O.R. 3d 661 (Can. Ont. Gen. Div.); Alberta Ltd. v. Tiberio 2008 ABQB 328, paras. 2, 10–11 (Can.), aff’d, 2008 ABCA 341 (Can.); Alberta Ltd. v. Trail S. Devs. Inc. 2001 ABQB 442, para. 50 (Can.); Chaulk v. Fairview Constr. Ltd. (1977), 14 Nfld. & P.E.I.R. 13 (Can. Nfld. C.A.); McNabb v. Smith (1981), 124 D.L.R. 3d 547 (Can. B.C.S.C.), aff’d, (1982), 132 D.L.R. 3d 523 (B.C.C.A.).

197See Chen-Wishart, supra note 13, at 124.

198See Hanoch Dagan, Doctrinal Categories, Legal Realism, and the Rule of Law, 163 U. Pa. L. Rev. 1889, 1898–1905 (2015); see also Malcolm Lavoie, Canada’s ‘Unique’ Approach to Specific Performance in Contracts for the Sale of Land: Some Theoretical and Practical Insights, 12 Oxford U. Commonwealth L.J. 207, 207–212 (2012) (noting that courts’ new discretionary authority leads real estate buyers to seek damages rather than risk seeking specific performance).

199See Joseph Raz, The Rule of Law and Its Virtuein The Authority of Law: Essays on Law and Morality 210, 213, 220, 222 (1979).

200Cf. Jody P. Kraus & Robert E. Scott, The Case Against Equity in American Contract Law, 93 S. Cal. L. Rev. 1323 (2020).

201Cf. Dagan, supra note 198, at 1905–08, 1910–11.  Joseph Raz’s analysis of the phenomenon of distinguishing cases brings home a similar point. See Raz, supra note 199, at 183–97.

202This admittedly still leaves courts with some borderline cases, notably of units in condominiums and cooperative apartments, which often are identical to hundreds with the same layout in the same (or similar) building, but at times are adapted to buyers’ specifications.  See, e.g., Schwinder v. Austin Bank of Chi., 809 N.E.2d 180, 196 (Ill. App. Ct. 2004); Lezell v. Forde, 891 N.Y.S.2d 606, 616 (N.Y. Sup. Ct. 2009).

203See Schwartz, supra note 164, at 297 (emphases added).


205Id.see also Szymon Osmola, Reflective Choices: A Liberal Theory of Consumer Law 123–128 (Oct. 26, 2022) (Ph.D. thesis, European University Institute, Florence) (on file with author).

206Schwartz, supra note 164, at 306.  Most of Schwartz’s reasons for this conclusion are economic.  With Scott and Kraus, however, we think that the economic analysis leads to the opposite conclusion.  See supra note 46 and accompanying text.

207As the text implies, this is surely not always the case.  We do not offer here an autonomy-based account of incorporated persons that transcends their economic function, an account that is urgently needed in private law theory. 

208See supra Section I.B; see also James J. White & Robert S. Summers, Uniform Commercial Code 321 (6th ed. 2010) (doubting whether “the authorization for specific performance where the contract so provides would have real impact.”); Yonathan A. Arbel, Contract Remedies in Action: Specific Performance, 118 W. Va. L. Rev. 369 (2015) (indicating that plaintiffs tend to opt out of specific performance due to difficulties of execution, even in a jurisdiction (Israel) where it is readily available); cf. Eisenberg & Miller, supra note 14 (studying 2347 contracts of public corporations, and observing that a majority of contracts do not include specific performance clauses, and that, among those which do, there is substantial variation among different contract types).  But see Theresa Arnold, Amanda Dixon, Hadar Tanne, Madison Whalen Sherrill & Mitu Gulati, “Lipstick on a Pig”: Specific Performance Clauses in Action, 2021 Wis. L. Rev. 359, 359 (analyzing “a dataset of more than 1,000 M&A contracts,” and concluding that basically, everyone in this market uses specific performance clauses and that “the most consistent explanation . . . for why parties want specific performance as a remedy [i]s that they [don’]t think that judges would give them the appropriate amount of money damages their bargain demanded”).

209See supra text accompanying note 181; see also, e.g., Kronman, supra note 37, at 371, 376; Jonathan Morgan, On the Nature and Function of Remedies for Breach of Contractin Commercial Remedies: Resolving Controversies, supra note 13, at 23, 43–44.

210Fazzio v. Mason, 249 P.3d 390, 397 (Idaho 2011).


212Reeder v. Carter, 740 S.E.2d 913, 919 (N.C. Ct. App. 2013) (emphasis added); see also, e.g., Ritchie Bros. Auctioneers (Am.), Inc. v. Best Rental Corp. SE, No. 14cv36, 2014 WL 896992, at *2 (M.D. Ala. Mar. 6, 2014).

213Black v. American Vending Co., 238 S.E.2d 420, 421 (Ga. 1977).

214See Restatement (Second) of Conts. § 359 cmt. a (Am. L. Inst. 1981).

215See Ian R. Macneil, Power of Contract and Agreed Remedies, 47 Cornell L.Q. 495, 521–23 (1962).

216Benson, supra note 73, at 261–62.

217Id. at 313.

218Id. at 255.

219Id. at 207–09, 212–13 (discussing liquidated damages).  But see Friedmann, supra note 30, at 23 (perceiving remedies as default rules notwithstanding an implicit endorsement of transfer theory).

220See supra notes 78–80 and accompanying text; see also Dagan & Heller, supra note 3, at 36–40; Dagan & Heller, Autonomy Refinedsupra note 3, at 222–24, 228–29.

221See, e.g., Fazzio v. Mason, 249 P.3d 390, 397 (Idaho 2011) (“[T]he inclusion of the clause shows that specific performance was within contemplation of the parties and may have been part of reason the Fazzios entered into the settlement agreements and allowed Mason to extend the closing date.”).

222In other words, courts indeed “ought to consider and reflect other interests in devising a system of contract remedies.”  Kakaes v. George Washington Univ., 790 A.2d 581, 584–85 (D.C. 2002) (emphasis omitted) (quoting Edward Yorio, Contract Enforcement: Specific Performance and Injunctions § 19.2.3, at 444 (1989)).

223See supra note 198 and accompanying text.

224Cf. Jeremy Waldron, Vagueness and the Guidance of Action, in Philosophical Foundations of Language in the Law 58, 65–66, 69 (Andrei Marmor & Scott Soames eds.,2011).

225See supra notes 8–9 and accompanying text.

226See supra notes 176–79 and accompanying text; infra notes 246–58.

227See supra note 18.

228See generally Dagan & Heller, Autonomy Refinedsupra note 3, at 243–44.  Cf. Omri Ben-Shahar, David A. Hoffman & Cathy Hwang, Nonparty Interests in Contract Law, 171 U. Pa. L. Rev. (forthcoming 2023).  A typical category, which falls somewhere between the first and second guidelines, involves cases where a promise has been conveyed to an innocent third party.  As Kronman correctly argues, the proper response in these cases is to “impose a constructive trust for the promisee’s benefit on the profit realized by the resale (that is, the difference between the resale price and the original contract price), even though this may exceed the damages the promisee has suffered.”  Kronman, supra note 37, at 377.

229See supra note 122 and accompanying text.

230See Dagan & Heller, supra note 3 at 112.

231For example, as the Restatement notes, courts refuse specific performance if the provision was the result of oppression or imposition, or if the agreement was, in general, one-sided or otherwise unfair.  See Restatement (Second) of Conts. § 364(1)(b) (Am. L. Inst. 1981).

232To clarify: there is nothing objectionable in celebrating equity’s capacity to refine the rules of contract law so that they are properly fine-tuned to their autonomy-enhancing task.  But celebrating this capacity is very different from endorsing an authority to apply ad-hoc judgment prospectively.  In a liberal system, it is justified to require people to incur some of the costs of having public officials normatively refine the law; but it is not justified to require people to be subject to ad-hoc discretion by these officials.

233Restatement of Emp. L. § 9.08(a) (Am. L. Inst. 2015).

234See Robert J. Steinfeld, The Invention of Free Labor: The Employment Relation in English and American Law and Culture, 1350–1870, at 4 (1991).

235Christopher T. Wonnell, The Contractual Disempowerment of Employees, 46 Stan. L. Rev. 87, 115 (1993).


237Id. at 145.

238Chen-Wishart, supra note 13, at 121, 117.

239See supra note 177 and accompanying text.

240See Wonnell, supra note 235, at 88.

241See Seana Valentine Shiffrin, Paternalism, Unconscionability Doctrine, and Accommodation, 29 Phil. & Pub. Affs. 205, 207–220, 231 (2000).

242See supra note 140 and accompanying text.

243See Dagan & Heller, supra note 2, at 568–69, 597–600. 

244Principles of the L. of Fam. Dissolution: Analysis and Recommendations § 7.08(3) & cmt. c (Am. L. Inst. 2002) [hereinafter ALI Principles].  In some states, “covenant” marriage sanctions some cooling-off by allowing spouses to commit to a time-limited waiting period before divorce.  See, e.g., La. Stat. Ann. §§ 9:272–275.1, 9:307–309 (2022).  This option is unobjectionable if (but only if) it allows immediate exit from psychologically or physically abusive relationships.

245See Dagan & Heller, supra note 2, at 574–77.  Exit is particularly threatening to egalitarian communities because it exacerbates the challenges of brain-drain, adverse selection, and free-riding.  See Ran Abramitzky, The Mystery of the Kibbutz: Egalitarian Principles in a Capitalist World 250–51, 263 (2018).

246See Dagan & Heller, supra note 2, at 567–69 (citing Laurence H. Tribe, American Constitutional Law 1400–09 (2d ed. 1988)).  Interestingly, people seem to share this position.  See Hanoch Dagan & Tamar Kricheli-Katz, Long-Term Contractual Commitments and Our Future Selves, 47 Law & Soc. Inquiry 1 (2022).

247See Dagan & Heller, supra note 2, at 581–602; Hanoch Dagan, Intimate Contracts and Choice Theory, 18 Eur. Rev. Cont. L. 104 (2022).

248See Dagan & Dorfman, Just Relationshipssupra note 102, at 1419.

249Cf. supra text accompanying notes 124, 127, 129.

250See Dagan & Dorfman, Just Relationshipssupra note 102, at 1419, 1442.  The term “ground project” in this context should not be equated with career but with (at least minimally) meaningful work.  Without some such measure, selling one’s labor—promising to comply with the employer’s directives—is tantamount to a consensual subordination to another’s authority.

251Contra Chen-Wishart, supra note 13, at 117.

252Sampson v. Murray, 415 U.S. 61, 83 (1974).

253Cf. Dobbs, supra note 188, at 929–30; Yorio & Thel, supra note 13, §, at 14-28.  As the text suggests, it is unclear whether this asymmetry suffices in cases of structural inequality of power between employers and employees.  We think that the answer to this important concern—the acceptability of the prevailing at-will default regime—depends on the presence of mandatory countermeasures strong enough to secure relational justice.  Cf. Aditi Bagchi, The Employment Relationship as an Object of Employment Law, in The Oxford Handbook of New Private Law, supra note 76, at 361.

254See supra Section III.B.

255Restatement of Emp. L. § 9.04(a) cmt. b (Am. L. Inst. 2015); cf. id. § 2.02 (listing contractual variations from at-will employment).

256Id. § 9.04(a).

257See id. § 9.04(a) cmt. c (mentioning that “[e]mployment statutes often provide as a presumptive remedy reinstatement of employees to positions the employees held before their employment was unlawfully terminated” and that “[c]ollective bargaining agreements . . . typically empower arbitrators to award similar relief,” but adding that in all other cases, which are “governed by common law” courts, “as a general matter have not awarded reinstatement as a remedy”).

258See The White House, Non-Compete Agreements: Analysis of the Usage, Potential Issues, and State Responses 3 (2016).

259See, e.g., Jonathan M. Barnett & Ted Sichelman, The Case for Noncompetes, 87 U. Chi. L. Rev. 953, 969–71, 1029–30 (2020) (positing that noncompetes encourage firms to invest in cultivating intellectual and human capital).  But see Orly Lobel, Talent Wants to Be Free: Why We Should Learn to Love Leaks, Raids, and Free Riding (2013).

260See Viva R. Moffat, Making Non-Competes Unenforceable, 54 Ariz. L. Rev. 939 (2012); J.J. Prescott, Norman D. Bishara & Evan Starr, Understanding Noncompetition Agreements: The 2014 Noncompete Survey Project, 2016 Mich. St. L. Rev. 369.

261See Stewart E. Sterk, Restraints on Alienation of Human Capital, 79 Va. L. Rev. 383 (1993).

262See Restatement (Second) of Conts. § 188 cmt. d (Am. L. Inst. 1981); see also, e.g., Suresh Naidu, Eric A. Posner & Glen Weyl, Antitrust Remedies for Labor Market Power, 132 Harv. L. Rev. 536, 595 (2018).

263For a recent analysis, collecting citations, see Karla Walter, The Freedom to Leave: Curbing Noncompete Agreements to Protect Workers and Support Entrepreneurship, Ctr. for Am. Progress (Jan. 9, 2019), https://‌‌/article‌/the-freedom-to-leave‌/ [https://‌‌/L3J9-NWE5].

264(1616) 81 Eng. Rep. 540; 1 Rolle 368.