The Holmesian bad man flubs his entrance.

• Author: Fried, Barbara H.
• Position: 'Contract as Promise' at 30: The Future of Contract Theory

1. Introduction

   Thirty years after its publication, Contract as Promise remains the canonical presentation of a liberal, autonomy-based conception of contractual obligation. In Charles Fried's words, "The moral force behind contract as promise is autonomy: the parties are bound to their contract because they have chosen to be," and their "rights and duties [are] as far as possible a function of their own will and not of standards of justice external to that will." (1) While other strains of liberal contract theory (consent-based, obligation-based) may differ from Prof. Fried's "will" theory of contracts in other respects, they all share his foundational commitment to the view that promissory obligations, unlike most other forms of obligation, are voluntarily assumed. The same is true of most liberal, autonomy-based conceptions of promissory obligation in the moral realm. (2)

   The question I wish to pursue here is this: Having established the voluntary nature of promissory obligation, has liberal contract theory (LCT) put itself out of a job? What further role, if any, does it have to play in elaborating the nature and content of promissory obligation?

   It clearly has something to say about why promising to do X imposes obligations of some form on the promisor, where a mere expression of a future intention to do X would not. Indeed, of the enormous philosophical literature on promising, the overwhelming majority is addressed to just this question (or so my casual survey suggests). It presumably also has something to say about the preconditions for concluding that a promise was freely, rationally and deliberately--that is, voluntarily--made. This is no small subject, and implicates a host of issues concerning procedural fairness--duty to disclose, misrepresentation, economic duress, etc.--as well as mental capacity.

   Finally, LCT does not itself require that the expressed wills of the parties always take precedence over other considerations. In extreme cases, liberal contract theorists could well conclude that those expressed desires should be trumped by paternalistic, public-policy, or fairness considerations. But all of these considerations are, in Prof. Fried's terms, "standards of justice external to that will," and hence not matters on which will theory itself can shed any light. (3)

   The question is, does LCT have anything to say about the permissible content of X, beyond saying that parties may give X any content that they wish and courts should, as far as they are able, resolve any dispute between the parties in accordance with the parties' own intentions?

   I think the answer is no. Rereading Contract as Promise in full thirty years after its publication, I was surprised to discover that for the most part (the other) Prof. Fried agrees. Much of the book is devoted to two lines of argument, both consistent with that conclusion.

   The first line of argument repudiates, as inconsistent with the will theory of contracts, common-law doctrines that override the parties' clear intent. Prof. Fried's treatment of consideration is illustrative here. Acknowledging the irreducible contradiction between "the liberal principle that the free arrangements of rational persons should be respected" and the requirement, embedded in the doctrine of consideration, that in order to be respected those arrangements must include what the law recognizes as a "bargain," Prof. Fried concludes, in effect, so much the worse for the consideration doctrine: "Freedom of contract is freedom of promise, and...the intrusions of the standard doctrines of consideration can impose substantial if random restrictions on perfectly rational projects." (4)

   Second, Fried argues that once we are unable to discern the parties' intent, will theory runs out, and any remaining gaps in the contract must be filled by resort to other (external) notions of justice. Prof. Fried's thoughtful treatment of the doctrines governing changed circumstances--mistake, frustration, impossibility, impracticability--is illustrative here. As Prof. Fried notes, the defining problem presented by changed-circumstances cases is that the situation the parties now find themselves in was one they failed to foresee when they entered into the agreement. As a consequence, "[t]he one basis on which these cases cannot be resolved is on the basis of the agreement--that is, of contract as promise. The court cannot enforce the will of the parties because there are no concordant wills. Judgment must therefore be based on...nonpromissory standards of justice" like restitution, reliance or loss-splitting. (5) I concur.

   There is, however, one substantive issue about which most, if not all, proponents have taken LCT to have something important to say: the consequences that should follow in the event the promisor fails to perform on the main subject of the promise. In Contract as Promise, Prof. Fried famously argues that the promisee's right to an expectation measure of damages is logically entailed in the will theory of contract. Other liberal contract theorists have argued, with equal vehemence, that specific performance is the appropriate legal response, and that promisors have a corollary moral duty to perform voluntarily, if they reasonably can. (6)

   Both views, I believe, are mistaken, and derive from the same error: thinking that the consequences that should flow from nonperformance of the main subject of the contract are external to the parties' agreement, rather than an integral part of that agreement. Once one views each party's obligations in the event it does not go forward with the proposed exchange as just another term in the agreement, from the perspective of LCT what one party owes the other should it choose not to go forward is just whatever the parties themselves agreed it would owe under those circumstances.

   In the event the contract is silent on a party's options in lieu of performance, LCT implies that the court should choose the term it thinks the parties themselves would have said was implicit in their deal, had they spoken to the issue ex ante. Most of the cases in which liberal contract theorists have strong convictions that X--specific performance, expectation damages, etc.--is the appropriate term for the court to impose can best be explained, I believe, not by a first-order belief that X is what a (generic) promisor owes a (generic) promisee in the event of nonperformance, but instead by an intuition that X is what this particular promisor and promisee had in mind, or would have said they had in mind, if they had spoken to the issue ex ante. Those intuitions may well be right much of the time. But right or wrong, they are an exercise in contract interpretation, not first-order morality.

   Finally, if the court has no evidence from which to infer the parties' intent, it will have to choose a gap-filling term on some other basis. At that point, LCT is out of a job, for the reasons eloquently stated by Prof. Fried with respect to changed circumstances and other unprovided-for cases: Once we are outside the scope of the parties' voluntarily assumed obligations, we are outside the domain of LCT.

   The same analysis applies in the moral as opposed to the legal realm, leading to the same conclusion. The massive literature on the moral bindingness of promises establishes only that if you promised to do X or led others reasonably to believe you promised to do X, you should do it if it is reasonably within your power to do it. It says nothing about what the content of X is. The strict reading that liberal theorists have given to promissory language in the moral realm, like the strict reading they have given in the legal realm, I believe, is best understood as a judgment not about the sorts of commitments people ought to make to each other, but rather about the sort that they did make. That is to say, it too is an act of interpretation. In particular, it is based on a surmise that when parties make promissory noises about doing X, they usually mean to commit themselves to do X (or, on an objective interpretation of meaning, can reasonably be understood by the promisee to have done so) and, unless they expressly condition their promise, they intend (or can reasonably be understood by the promisee to intend) that commitment to hold unless it becomes impossible or extremely burdensome to do so.

   While that strict interpretation of promissory language may get the parties' intentions right much of the time in the legal realm, in the moral realm it is very likely to get things wrong. The reason for this difference has to do with the different contexts in which promissory-type noises are made. Much of the literature on the morality of promising focuses on gratuitous promises between intimates (friends, family). In the context of those relationships, what parties mean when they make promissory noises is idiosyncratic, often opaque to third parties, and (I believe) typically more forgiving than a strict interpretation presumes.

   In short, then, LCT, like its analogue in the moral realm, has put itself out of a job when it comes to determining the appropriate consequences of nonperformance. I am hardly the first to make this argument. Richard Craswell has urged it with respect to all default contract rules, noting that the "fidelity principle," which asserts that parties should be held to their commitments, "is consistent with any set of background rules because those rules merely fill out the details of what it is a person has to remain faithful to, or what a person's prior commitment is deemed to be." (7) More recently, in an important article that has not received the attention it deserves in the "philosophy of promising" literature, Robert Scott and George Triantis have elaborated the argument with respect to remedies in particular. The traditional default remedies--expectation damages, reliance damages, specific performance--they argue, should be regarded as part of a much larger...