Emory Law Journal

Contracts and Friendships
Ethan J. Leib *Visiting Associate Professor of Law, University of California—Berkeley Law School (Spring 2010); Associate Professor of Law, University of California—Hastings College of the Law; Scholar-in-Residence, Columbia University School of Law (Spring 2009). Thanks to the 1066 Foundation and the Roger Traynor Scholarly Publication Award for financial support; to Verity Winship, Steve Vladeck, Bob Scott, Alice Ristroph, Eric Posner, Michael Pratt, Michael O’Hear, Stewart Macaulay, Dan Markel, Julian Ku, Katy Kuh, Adam Kolber, Paul Ingram, Adil Haque, Jay Feinman, Ira Ellman, Bill Dodge, Reza Dibadj, I. Bennett Capers, Dave Campbell, Mike Cahill, and Curtis Bridgeman for comments on earlier drafts; to the participants of a workshop at Hofstra Law School (“Prawfsfest!”) for feedback; to the participants of the Bay Area Forum for Law & Ethics (BAFFLE) at Berkeley; and to Doug Ecks for wonderful research assistance.

Abstract

This Article aims to give the relational theory of contract new life, sharpening some of its claims against its competitors by refracting its theory of relational contracts through an analogy to friendship. In drawing the analogy between friendships and relational contracts and revealing their morphological similarities, this Article offers a provocative window into friendship’s contractual structure—and into relational contracts’ approximation of friendships. The analogy developed here is poised to replace the “relational contract as marriage” model prevalent among relationalists. This new model is more honest to relational contract theory and to marriage—and helps relational contract theory produce some new insights, support old ones, and revise some of its normative agenda.

Introduction

It is likely true that marriages are types of contracts. 1For a maximalist version of this thesis, see, for example, Neil G. Williams, What To Do When There’s No ‘I Do’: A Model for Awarding Damages Under Promissory Estoppel, 70 Wash. L. Rev. 1019, 1047 (1995) (“Modern marriages are . . . tantamount to contractual relationships.”). But contracts are not generally like marriages. Friendships are also, in part, types of contracts. Yet many contracts are also like friendships. At the least, it is more illuminating to think of contracts as friendships than as marriages. Let me explain.

For some time, so-called “relational” contract theorists 2This group is usually taken to include Ian Macneil, Stewart Macaulay, Richard Speidel, and Jay Feinman. This Article largely addresses the work of those relational contract theorists who are essentially legal scholars, not theorists of relational contracting in economics, organizational theory, history, and sociology, though some aspects of these approaches do inform several of my arguments here. For some examples of the work in this latter group, see Naomi R. Lamoreaux et al., Beyond Markets and Hierarchies: Toward a New Synthesis of American Business History, 108 Am. Hist. Rev. 404 (2003) (explaining how relational contracting can be a useful—and causal—response to changing economic organization); Charles F. Sabel & Jonathan Zeitlin, Neither Modularity nor Relational Contracting: Inter-firm Collaboration in the New Economy, 5 Enterprise & Soc’y 388 (2004) (criticizing some ways of writing business history and some ways of thinking about informality within relational contracting); George Baker et al., Contracting for Control (Mar. 21, 2006) (unpublished manuscript), available at http://www.law.columbia.edu/null/Contracts+Conf+-+April+7-8,+2006+-+Paper+-+Gibbons?exclusive=filemgr.download&file_id=961193&showthumb=0 (arguing that although many firms contract with informality in relational contracting environments, firms are not indifferent to enforcement mechanisms and will try to contract to control temptations to renege); Oliver Hart & John Moore, Contracts as Reference Points (Harvard Law Sch., John M. Olin Ctr. for Law, Econ. & Bus., Discussion Paper No. 572, 2006), available at http://www.law.harvard.edu/programs/olin_center/papers/pdf/Hart_Moore_572.pdf (exploring some of the challenges of developing efficient long-term contracts in light of elements of such contracts that simply cannot be enforced by courts). Thanks to Bob Scott for some guidance in this literature. —theorists who urge us to focus on “relational” elements in contracts in order to devise a theory and law of contracts 3For a recent effort to specify the commitments of relational contract theory clearly, see Dori Kimel, The Choice of Paradigm for Theory of Contract: Reflections on the Relational Model, 27 Oxford J. Legal Stud. 233 (2007). —have sought to convince us that the majority of contracts, particularly the most relational ones, are types of marriages. 4 See, e.g., Ian R. Macneil, The New Social Contract 15, 22, 29, 68, 91 (1980); Robert W. Gordon, Macaulay, Macneil, and the Discovery of Solidarity and Power in Contract Law, 1985 Wis. L. Rev. 565, 569 (“[P]arties [to relational contracts] treat their contracts . . . like marriages . . . .”); Kimel, supra note 3, at 245; John Wightman, Intimate Relationships, Relational Contract Theory, and the Reach of Contract, 8 Feminist Legal Stud. 93, 100 n.23, 102 (2000); see also Gillian K. Hadfield, Problematic Relations: Franchising and the Law of Incomplete Contracts, 42 Stan. L. Rev. 927, 964 (1990) (exploring how franchising relationships—quintessentially treated as relational contracts—are like marriages); Stewart Macaulay, Non-contractual Relations in Business: A Preliminary Study, 28 Am. Soc. Rev. 55, 65 (1963) (“A breach of contract law suit may settle a particular dispute, but such an action often results in a ‘divorce’ ending the ‘marriage’ between the two businesses.”); Marjorie Maguire Shultz, Contractual Ordering of Marriage: A New Model for State Policy, 70 Cal. L. Rev. 204 (1982) (exploring how marriages are like relational contracts). More modestly, perhaps, they suggest that marriage is a good analogy for relational contracting. Even those sympathetic to the law and economics movement have suggested the equivalence between relational contracts and marriages. 5 Eric A. Posner, Law and Social Norms 160 (2000) (“A contract is a kind of marriage.”).

The core idea of “relational contract as marriage” is that most contracts, like marriages, share the following properties: (1) they ensnare parties into—and control party compliance through—a thick set of social norms; (2) they generally give parties autonomy in resolving internal disputes through those social norms; (3) they are very costly to exit; (4) they leave many terms open; (5) they tend to be incomplete in specifying obligations, especially on the issue of how the parties will cooperate over time; (6) parties tend to presume good faith and best efforts in performance and tend to allow for easy modifications for changed circumstances; and (7) they establish interdependence, partiality, and solidarity. Indeed, some have even examined the equivalence between marriages and relational contracts to determine what the content of marital law ought to be. 6Elizabeth S. Scott & Robert E. Scott, Marriage as Relational Contract, 84 Va. L. Rev. 1225 (1998). The purchase of the analogy for relationalists, however, is tied up with their most general point—that discrete, one-shot transactions between strangers have occupied the center of “classical” contract law and contract theory to its detriment. In reality, as relationalists have demonstrated since the 1960s, contracts involve thicker relationships than those among strangers. Indeed, we cannot understand the social or legal practice of contracting as a practice among strangers; this insight alone has dramatic implications for our entire market economy and our effort to regulate it through contract law and otherwise.

This Article makes an effort to replace the marriage analogy in relational contract thinking with a story that ties contracts and friendships closer together. Part I is a brief overview of relational contract theory (as it is conceptualized by contracts scholars in law schools). Part II highlights how friendships are very much like relational contracts and how friendship can be illuminated by exposing its morphological similarity to relational contracts. Part III argues that friendship rather than marriage should be viewed as the paradigmatic relational contract. And Part IV explores some of the legal and theoretical ramifications for contract law and theory in light of the view that friendships are types of relational contracts and relational contracts are types of friendships. Ultimately, this window into relational contracts and friendships furnishes a few important lessons. First, it supports the legal enforceability of some contracts, both explicit and implicit, between friends. Second, it supports and revises certain prescriptions of relational contract theory. Third, it helps us better understand the relationship between social norms and legal norms. Finally, understanding relational contracts as friendships helps us make better sense of the under-compensatory nature of contractual remedies.

Let me not overstate the thesis here: I am not building a general theory of friendship or contract. This Article only suggests that putting these social practices side-by-side and exposing their morphological similarities can help illuminate each practice. Certainly, if pressed, some relationalists might concede that the place of marriage in relational contract theory is only as a suggestive analogy. This Article proposes that friendship offers a better and more illuminating analogy that can do some modest work in advancing our understanding of contract doctrine and theory.

I. Relational Contract Theory: A Sympathetic Reconstruction

As many have observed, it is a difficult business to summarize relational contract theory. 7 See, e.g., Dori Kimel, From Promise to Contract: Towards a Liberal Theory of Contract 80–81 (2003) (noting that relational contract theory literature “does not speak in one voice”). There are many “relationalists” and even some who view themselves as critics of relational contract theory concede that “we are all relationalists now.” 8Although often credited to Robert E. Scott, The Case for Formalism in Relational Contract, 94 Nw. U. L. Rev. 847, 852 (2000), the phrase appeared earlier in a 1992 essay by Randy E. Barnett. Randy E. Barnett, Conflicting Visions: A Critique of Ian Macneil’s Relational Theory of Contract, 78 Va. L. Rev. 1175, 1200 (1992); see also Richard E. Speidel, The Characteristics and Challenges of Relational Contracts, 94 Nw. U. L. Rev. 823, 845 n.86 (2000) (“Scott now states that ‘[w]e are all relationalists.’ Nonsense. Those advocating the virulent strain of the new formalists are . . . [not] relationalists and the sooner we say so the better.” (citation omitted)). For my purposes, a quick sketch of relational contract theory should be sufficient.

At the center of relational contract theory is an empirical observation about the real world of contracting, a related effort to reorient the paradigm under which contract theory is organized and analyzed, and a set of normative and doctrinal prescriptions for how to apply contract law to disputing parties. It is probably fair to say that it is only the empirical observation (and a moderate form of it, to boot) that the critics of relational contract theory are willing to concede. The other dimensions of the theory are embraced by those I describe as “relationalists”—adherents of relational contract theory.

A. The Empirical Claim

The empirical observation that commands consensus is that many contracts in the real world involve long-term, complex relationships and are not merely “one-shot,” discrete transactions. In these “relational contracts,” social norms often seem to play a larger role in controlling the parties’ conduct than the threat of legal sanctions. 9This insight is often traced to the pioneering work of Stewart Macaulay. See, e.g., Stewart Macaulay, Changing a Continuing Relationship Between a Large Corporation and Those Who Deal with It: Automobile Manufacturers, Their Dealers, and the Legal System, 1965 Wis. L. Rev. 483; Stewart Macaulay, Elegant Models, Empirical Pictures, and the Complexities of Contract, 11 Law & Soc’y Rev. 507 (1977); Macaulay, supra note 4. People also like to cite Bob Ellickson’s important book here, along with an early piece by Lisa Bernstein. See Robert C. Ellickson, Order Without Law: How Neighbors Settle Disputes (1991); Lisa Bernstein, Opting Out of the Legal System: Extralegal Contractual Relations in the Diamond Industry, 21 J. Legal Stud. 115 (1992); see also David Charny, Nonlegal Sanctions in Commercial Relationships, 104 Harv. L. Rev. 375 (1990).It should be noted, however, that some have found that even in relational contracts, legal and formal contracting serve a core function of forging trust. See Simon Deakin et al., Contract Law, Trust Relations, and Incentives for Cooperation: A Comparative Study, in Contracts, Co-operation, and Competition 110 (Simon Deakin & Jonathan Michie eds., 1997). Trust and social solidarity tend to underwrite and support (and, in turn, tend to be underwritten and supported by) these types of contracts, and informal cooperation, coordination, and collaboration are typical in contractual relations. In these types of relationships, parties expect some form of loyalty.

Although contracts between strangers are, of course, possible, relational contract theory emphasizes that many contracts do not show this pattern of “strangership.” In contrast, relationalists highlight just how often contracts are formed between parties who are otherwise connected in pre-existing and ongoing relationships. 10For some recent evidence of the affective nature of many business relationships, see Paul Ingram & Xi Zou, Business Friendships, 28 Res. Org. Behav. 167 (2008); Paul Ingram & Arik Lifschitz, Kinship in the Shadow of the Corporation: The Interbuilder Network in Clyde River Shipbuilding, 1711-1990, 71 Am. Soc. Rev. 334 (2006); and Paul Ingram & Peter W. Roberts, Friendships Among Competitors in the Sydney Hotel Industry, 106 Am. J. Soc. 387 (2000). Ingram & Zou observe that “[b]usiness friendships represent substantial benefits to careers and organizational performance, and they are becoming more common.” Ingram & Zou, supra at 182. They are, however, clear that this fact is not without costs: “[B]usiness friendships also represent notable psychic, social and cultural tensions . . . .” Id. Still, they are also clear that “the option of segregating economic and social worlds is not available to contemporary business people.” Id.It is worth noting that much of Ingram’s work is devoted to horizontal ties in business—affective relationships among competitors—rather than vertical ties among buyers and sellers, who are likely to be engaged in contractual relations as well. For some useful work on vertical ties in business, see Paul DiMaggio & Hugh Louch, Socially Embedded Consumer Transactions: For What Kinds of Purchases Do People Most Often Use Networks?, 63 Am. Soc. Rev. 619 (1998); Vincenzo Perrone et al., Free to Be Trusted? Organizational Constraints on Trust in Boundary Spanners, 14 Org. Sci. 422 (2003); Brian Uzzi, Embeddedness in the Making of Financial Capital: How Social Relations and Networks Benefit Firms Seeking Financing, 64 Am. Soc. Rev. 481 (1999) [hereinafter Uzzi, Financial Capital]; and Brian Uzzi, The Sources and Consequences of Embeddedness for the Economic Performance of Organizations: The Network Effect, 61 Am. Soc. Rev. 674 (1996) [hereinafter Uzzi, Sources and Consequences]. Moreover, relationalists observe that when parties find themselves in such ongoing relationships, contractual behavior tends to be driven by informal and implicit dimensions of their mutual understandings 11The best source on the implicit dimensions of contract broadly understood is Implicit Dimensions of Contract: Discrete, Relational, and Network Contracts (David Campbell et al. eds., 2003) [hereinafter Implicit Dimensions of Contract]. within the relationship, rather than “paper” deals and strict adherence to formalities. 12Stewart Macaulay, The Real and the Paper Deal: Empirical Pictures of Relationships, Complexity and the Urge for Transparent Simple Rules, in Implicit Dimensions of Contract, supra note 11, at 51.

This is not terribly controversial stuff, though it took a while for these insights to be fully understood and incorporated into mainstream contracts thinking. 13Even now, some contract theorists cannot resist challenging the empirical claim. See Kimel, supra note 7, at 81 n.50. In a recent article on relational contract theory, however, Kimel backs off the challenge to the modest form of the empirical claim. See Kimel, supra note 3, at 234. To the extent that one hears that “we are all relationalists now,” all this statement means is that most people have embraced the relationalists’ empirical observation. Two significant cleavages between relationalists and non-relationalists have nevertheless emerged from this now widely accepted empirical observation. The disagreements are about what follows from the observation as a matter of general contract theory and as a normative matter within contract law.

B. The Analytic Claim for a Paradigm Shift in Contract Theory

Relationalists contend that relational contracts are so prevalent in the real world that the general theory of contract must acknowledge this practical reality. That is, contract theory must be oriented in such a way as to put relational contracts at its center. This can mean several things.

First, owing to the recognition that many contracts (some relationalists would say most, if not all, contracts) are relational, 14Some relationalists are committed to the view that exchange itself is deeply relational and that all contractual intercourse is shaped in meaningful ways by relational features because contracts could not get off the ground without sufficient social solidarity. Ironically, this view may be too moderate to be at the core of relational contract theory. In short, once all contracts get folded into this form of relationalism, the claim is no longer strong enough to warrant the disparate treatment of different forms of contracts, which seems central to the normative thrust of relational contract theory. Indeed, non-relationalists do not bother opposing this form of relationalism. Non-relationalists are only too happy to concede that all contracts should be treated under the same rules: if all contracts are relational, all contracts are subject to the same law. Nevertheless, as the text immediately following this note reveals, there is still room for differential treatment because of the possibility of a continuum of relational factors. relationalists argue that it would be useful to analyze contracts on some continuum or gradation from the most discrete to the most relational. By utilizing this method of analysis, one can differentiate more easily among types of contracts and apply the proper policy approach based on the contractual context. Second, because relational contracts are so common, relationalists argue that the general theory of contracts needs to respect the nature of contracts in the real world and not operate with a presumption that the law of contracts should be designed as a law for strangers, where there is little background trust. Because relationalists think that relational contracts are pervasive, they argue that orienting contract law as a method to provide trust for discrete and formal transactions is inappropriate. Rather, a new paradigm is needed to account for the degree to which relational contracts and trust are central to the enterprise of contracting.

One can easily understand why relational contract theory’s analytic move is more controversial than its underlying empirical claim. Some relationalists might think this analytical claim about reorienting contract theory follows from the empirical claim, but that is not necessarily true. There is little controversy surrounding the rather modest empirical claim that some contracts are deeply relational. All that follows from this claim, perhaps, is that general contract theory must make room for analyzing and respecting a class of relational contracts. 15For an analysis of how the empirical claim was absorbed into mainstream “neo-classical” contract law, see generally Jay M. Feinman, Relational Contract Theory in Context, 94 Nw. U. L. Rev. 737 (2000). Some of the ways the UCC can be read to absorb relationalism in laws governing performance, modification, interpretation, and formation are discussed infra note 34. But that hardly requires the sort of paradigm shift in theory and law that one tends to associate with the urgings of hard-core relationalists. At least part of the resistance to the analytical claim comes from the sense that the empirical claim can easily be accommodated by slight modifications to reigning contract theories. However, this resistance generally works, perhaps, because non-relationalists simply do not accept the more aggressive empirical claim that most contracts are meaningfully characterized by relational properties and can be usefully analyzed based upon where they fall on the discrete to relational continuum. 16In a recent clarification, Macneil changed terminology, hoping to get a “relational/as-if-discrete” continuum off the ground to replace the relational-discrete continuum. See Ian R. Macneil, Relational Contract Theory: Challenges and Queries, 94 Nw. U. L. Rev. 877, 894–900 (2000) [hereinafter Macneil, RCT: C & Q]. Macneil has always been fairly clear that “discreteness” is a “relative” property from the perspective of a relational approach. See, e.g., Ian R. Macneil, Relational Contract: What We Do and Do Not Know, 1985 Wis. L. Rev. 483, 485; Ian R. Macneil, Values in Contract: Internal and External, 78 Nw. U. L. Rev. 340 (1983). However, Macneil’s recent clarification demonstrates his ultimate commitment to the “strongest” form of the empirical claim that all contracts are relational. The problem with this form of the empirical claim, perhaps, is that it seems to dilute what is uniquely special about “relational contracts” since many relationships forged by and underwriting long-term relational contracts are rather thin. See Kimel, supra note 3, at 239 (“[T]he further we move towards [the empirical claim’s] most extreme form—all contracts are relational—the more it depends for its plausibility on an understanding of ‘relationship’ so watered-down as to make the debate entirely uninteresting.”).Macneil has also quite recently tried to change his most basic terminology, calling his theory “essential contract theory,” rather than “relational contract theory.” Macneil, RCT: C & Q, supra at 877, 881. That re-branding has not been successful, and I prefer to stick with the old terminology and its thirty years of history and development.

The analytic claim becomes more pressing if the empirical claim is acknowledged to be more thoroughgoing. If relational contracts are very pervasive (though the degree of pervasiveness to make the analytic claim plausible is very hard specify), it becomes harder to embrace traditional contract theory, which presumes contracts between strangers as its paradigm. At least if one’s meta-theoretical commitments require contract theories to “fit” the underlying practice of contracting, it becomes hard not to orient one’s contract theory to match the reality of the social practice. Indeed, most relationalists explicitly or implicitly embrace a stronger form of the empirical claim than do non-relationalists.

The underlying empirical dispute is difficult to resolve, however. It is virtually impossible to get a sense of what is happening in most contracts, and it is very hard to say how pervasive a contract type needs to be to require reorienting the theory of contract. These difficulties are at least part of the reason we are unlikely to see much conciliation between the camps anytime soon. 17There is a useful cut at the empirical question in DiMaggio & Louch, supra note 10, at 620, 622. In their close look at the 1996 General Social Survey economic sociology module, they find that “people buy and sell goods and services from friends and kin” and that “substantial percentages of major transactions take place between friends, relatives, or compound ties.” Id. More specifically, “[a]lmost one-half of all used-car purchases from individuals (46.0 percent) and home purchases where no agent is used (46.8 percent) are transactions between relatives, friends, or acquaintances.” Id. at 622–23. Approximately one-fourth of purchases of legal services and home maintenance services are similarly within a social network. Id. at 623.Does this evidence resolve the central empirical claim of relational contract theory? Of course not. But it does at least help focus attention on contractual contexts where consumer transactions tend to involve the use of strong social networks.

Still, it should be noted that one can contest a strong form of the empirical claim and nevertheless subscribe to a form of the analytic claim. One could concede that true relational contracts are actually a rather small class of contracts, all things considered, but that they are so normatively attractive in their foregrounding of cooperation, collaboration, and solidarity that we should want to model other contracts on that particularized, even if somewhat atypical, form of contracting. 18An example of this type of argument was once explained by Robert Gordon:Some theorists have even advanced a bold hypothesis of a specific historical causal relationship between the fantasy world of political-legal ideological discourse about contracts and the social world of contracting: they contend that encouraging people to deal with one another as strangers progressively erodes the underlying relations of solidarity, reciprocity and trust upon which capitalist economies essentially depend.Gordon, supra note 4, at 578–79; see also Mark Vernon, The Philosophy of Friendship (2005) (arguing that friendship and capitalism can sit in tension). Brian Uzzi asks: “What modern institutions and . . . arrangements need exist if embedded [i.e., relational] exchange systems are to arise and prosper in a society?” Uzzi, Sources and Consequences, supra note 10, at 695. It may be that a more thoroughly relational contract law could be the answer. Indeed, DiMaggio & Louch generally find a high degree of satisfaction in embedded transactions, which is perhaps a reason for society to encourage them. DiMaggio & Louch, supra note 10, at 633–34. One could argue that modeling contracts on deep relationships of collaboration and solidarity would better orient parties in their transactions and could even maximize their surplus. 19Daniel Markovits’s contract theory of “contract as collaboration,” for example, argues for orienting contract theory along the dimension of collaboration. Daniel Markovits, Contracts and Collaboration, 113 Yale L.J. 1417 (2004). However, he is not really a relationalist because he seems to think that parties generally come to contracts as strangers, id. at 1435, and that it is the contract itself that establishes the cooperative relation between the parties, one that has no affective dimension, id. at 1432, 1451. Add to that the fact that he explicitly claims distance from Ian Macneil (though the position he ascribes to Macneil and rejects—that contracts are not “freestanding” sources of obligation—is not obviously one Macneil endorses). Id. at 1450 n.68. This is a way to defend the core relationalist analytical claim without needing to prove the empirical claim in any strong form.

The opposite is also true, however. One could concede the strong form of the empirical claim and still resist the analytical claim on normative grounds. One could say that preserving forms of “strangership” is essential to our economy and society, 20I have explained the role of “strangership” in Ethan J. Leib, Friendship & the Law, 54 UCLA L. Rev. 631, 663−64 (2007), drawing upon two articles by Allan Silver. See Allan Silver, “Two Different Sorts of Commerce”—Friendship and Strangership in Civil Society, in Public and Private in Thought and Practice: Perspectives on a Grand Dichotomy 43 (Jeff Weintraub & Krishan Kumar eds., 1997); Allan Silver, Friendship in Commercial Society: Eighteenth-Century Social Theory and Modern Sociology, 95 Am. J. Soc. 1474 (1990). and, accordingly, contracts should be modeled on personal distance. If we design contracts and orient our contract theory as something categorically different from our personal relationships which are governed largely by moral and social norms, we are most likely to avoid the juridification or contractualization of intimate personal relations. 21Forms of this argument against the analytic claim are actually quite common among non-relationalists. See generally Kimel, supra note 7 (arguing that contract facilitates detachment from personal relations); Melvin A. Eisenberg, Why There Is No Law of Relational Contracts, 94 Nw. U. L. Rev. 805, 820 (2000) (“This massive contractualization of human relationships [which is effected by relational contract theory] undesirably obscures critical differences between economic and affective relationships, between explicit and tacit reciprocity, between relationships that should be enforceable by both law and social norms and relationships that should be enforceable only by social norms.”); Kimel, supra note 3; Aditi Bagchi, Contract v. Promise (Univ. Pa. Law Sch., Public Law and Legal Theory Research Paper Series, Paper No. 07-35, 2007), available at http://ssrn.com/abstract=1012150; see also Gunther Teubner, Expertise as Social Institution: Internalising Third Parties into the Contract, in Implicit Dimensions of Contract, supra note 11, at 333, 343–46 (recognizing that relational contract theory risks contractualizing social life and arguing that would be bad).It is worth noting that this type of argument, if correct, goes a long way to address Seana Shiffrin’s recent concern about the “divergence” of contract norms from moral norms of promising. See Seana Valentine Shiffrin, The Divergence of Contract and Promise, 120 Harv. L. Rev. 708 (2007). If contracts are best designed as something different than promises in “private” life, there is less reason to worry that there is norm divergence. This argument is spelled out at length by Aditi Bagchi. See Bagchi, supra. Although we lose something in the fit of the theory (if the strong empirical claim is, after all, true), we gain some protection of the private sphere from the law, and that is, arguably, normatively desirable. 22This argument is similar to the “crowding” thesis that because the law will tend to crowd out authentic trust relations, we should be wary of interposing the law into trust relations. In its extreme form, the argument counsels for no legal intervention where there is real trust between the parties. The merits of this argument are addressed further infra Part IV. I discussed (and rejected) the “crowding” thesis in Ethan J. Leib, Friends as Fiduciaries, 86 Wash. U. L. Rev. 665 (2009). See also Anthony J. Bellia, Jr., Promises, Trust, and Contract Law, 47 Am. J. Juris. 25 (2002) (arguing against the thesis that legal enforceability degrades the social norm of trust); Frank B. Cross, Law and Trust, 93 Geo. L.J. 1457, 1545 (2005) (“Whatever the intuitive appeal of the claims that legalization undermines trust, they cannot be sustained once they are subjected to scrutiny and empirical testing.”); Claire A. Hill & Erin Ann O’Hara, A Cognitive Theory of Trust, 84 Wash. U. L. Rev. 1717 (2006). Most generally, there is a real problem trying to separate neatly the spheres of contract and law on the one hand and intimacy on the other. See generally Viviana A. Zelizer, The Purchase of Intimacy (2005). This Article seeks to undermine the “separate spheres” story too.

One could also concede the empirical claim and still think that the apparatus of relational contract theory is asking all the wrong questions since it does not concern itself directly with wealth maximization or incentivizing parties’ behavior through default rules. 23This “asking the wrong questions” objection appears in Richard Craswell, The Relational Move: Some Questions from Law and Economics, 3 S. Cal. Interdisc. L.J. 91 (1993). Craswell wants to orient contract theory toward economic analysis to focus on consequences (of price and product safety) and sees relational analysis as sociological and indifferent to consequences. Id. at 92, 98−99. It is not at all clear that this is a fair description, as many relationalists are, after all, sensitive to such consequences. See, e.g., 1 Stewart Macaulay et al., Contracts: Law in Action 643–44 (2d ed. 2003); Jay M. Feinman, Relational Contract and Default Rules, 3 S. Cal. Interdisc. L.J. 43, 55 (1993); Stewart Macaulay, Private Legislation and the Duty to Read—Business Run by IBM Machine, the Law of Contracts and Credit Cards, 19 Vand. L. Rev. 1051 (1966); Ian R. Macneil, Bureaucracy and Contracts of Adhesion, 22 Osgoode Hall L.J. 5 (1984). Nevertheless, it is likely true as a general matter that the law and economics crowd prefers ex ante reasoning in resolving contracts disputes, and the relationalist crowd tends to prefer ex post reasoning. See Speidel, supra note 8, at 844.Others in the law and economics tradition have acknowledged the pervasiveness of relational contracts but have urged an orientation to contract theory guided by value maximization of the relationship between the parties so that they can produce the most surplus. See Charles J. Goetz & Robert E. Scott, Principles of Relational Contracts, 67 Va. L. Rev. 1089 (1981); Robert E. Scott, A Relational Theory of Default Rules for Commercial Contracts, 19 J. Legal Stud. 597 (1990). For that matter, neither does it concern itself primarily with wealth redistribution or welfare. 24 See Hugh Collins, The Research Agenda of Implicit Dimensions of Contract, in Implicit Dimensions of Contract, supra note 11, at 1, 22–23 (highlighting how difficult it is to see relationalists as having a singular welfarist philosophy); James W. Fox, Jr., Relational Contract Theory and Democratic Citizenship, 54 Case W. Res. L. Rev. 1, 4–5 (2003) (listing variants of the theory and their underlying political theories). If those are a contract theorist’s desiderata, relational contract theory may not be especially useful and those with other normative aspirations could reasonably reject this as a paradigm.

But there is also another way of contesting the strong form of the empirical claim while still arguing for a version of the analytical claim. It is plausible that one could find orienting contract theory on the discrete-relational continuum so useful as a way of distinguishing among types of contracts that one could be relatively indifferent to how strong the empirical claim is and still want to use the classificatory apparatus from relational contract theory. 25 See generally Ian R. Macneil, Economic Analysis of Contractual Relations: Its Shortfalls and the Need for a “Rich Classificatory Apparatus, 75 Nw. U. L. Rev. 1018 (1981). Some relationalists—Ian Macneil is exemplary on this score, delineating a laundry list of dimensions of relationality (or “common contract norms”) 26The list changes from time to time and is not intended to be exhaustive. The standard elements in Macneil’s most recent articulation are: “(1) role integrity (requiring consistency, involving internal conflict, and being inherently complex); (2) reciprocity (the principle of getting something back for something given); (3) implementation of planning; (4) effectuation of consent; (5) flexibility; (6) contractual solidarity; (7) the restitution, reliance, and expectation interests (the ‘linking norms’); (8) creation and restraint of power (the ‘power norm’); (9) propriety of means, and (10) harmonization with the social matrix, that is, with supracontract norms.” Macneil, RCT: C & Q, supra note 16, at 879–80. —have developed sophisticated ways of describing the relational features of contracts and have helped create categories of analysis that divide the real world of contract practice in a way that might prove analytically useful.

The trappings of relational contract theory may be useful for categorical neatness, for theoretical integrity and exhaustiveness, or for more general philosophical reasons (apart from “fit”). Since the apparatus of the theory can accommodate promise-based liability, consent-based liability, reliance-based liability, and relationship- or status-based liability, its very pluralism may be attractive, especially to those who tire of seeking a fully coherent promissory or reliance-based theory of contractual obligation. 27 But see Kimel, supra note 3, at 252–53 (arguing that relational theory disables us from distinguishing between different sources of obligation and helps us organize our priorities when they conflict in interesting ways). Kimel’s critique misses the mark. To say, as relationalists do, that promissory sources of obligation and reliance sources of obligation all come within contract is not to say that we are always disabled from producing hierarchies of source of obligations to privilege in any given context. It is only to say that we cannot—as Kimel and Bagchi do, for example—build a contract model with promise at its center, allowing occasional “exceptions” for reliance-based liability. Relationalists think the practice is too capacious to be organized around any singular principle of liability. Many other contract theorists use only a singular organizing principle to account for a social practice that is rather heterogeneous, and the appeal of relational theory may be its willingness to welcome multifarious sources of obligation and deep fragmentation of the field of contract. 28Relational contract theory’s commitment to contextualization and fragmentation is explained in Feinman, supra note 15. Although, to be sure, Macneil’s complicated matrices probably scare away more contract theorists than they invite into the fold, the approach of seeing contracts through their relational dimensions has proven to be a rich method of dividing the often too-unified world of contract theory. 29For my assessment of how contract theory is too unified, see Ethan J. Leib, On Collaboration, Organizations, and Conciliation in the General Theory of Contract, 24 Quinnipiac L. Rev. 1 (2005). Relational contract theorists can win over adherents if contracts can be usefully mapped and ordered based on their relational elements, even if only a small fraction of contracts are actually of a “purely” relational type.

C. The Normative Claim About Contract Law

Whereas the first relationalist claim—the empirical one—is really sociological and the second—the analytic one—is really meta-theoretical, the third is a claim about what the law should be. There is plenty of controversy about relational contract law, however. Consider Mel Eisenberg’s summary of the doctrinal suggestions that relationalists embrace:

(1) Rules that, in the case of relational contracts, would soften or reverse the bite of the rigid offer-and-acceptance format of classical contract law, and the corresponding intolerance of classical contract law for indefiniteness, agreements to agree, and agreements to negotiate in good faith. (2) Rules that would impose upon parties to a relational contract a broad obligation to perform in good faith. (3) Rules that would broaden the kinds of changed circumstances (impossibility, impracticability, and frustration) that constitute an excuse for nonperformance of a relational contract. (4) Rules that would give content to particular kinds of contractual provisions that may be found in relational contracts, such as best-efforts clauses or unilateral rights to terminate at will. (5) Rules that would treat relational contracts like partnerships, in the sense that such contracts involve a mutual enterprise and should be construed in that light. (6) Rules that would keep a relational contract together. (7) Rules that would impose upon parties to a relational contract a duty to bargain in good faith to make equitable price adjustments when changed circumstances occur, and would perhaps even impose upon the advantaged party a duty to accept an equitable adjustment proposed in good faith by the disadvantaged party. (8) Rules that would permit the courts to adapt or revise the terms of ongoing relational contracts in such a way that an unexpected loss that would otherwise fall on one party will be shared by reducing the other party’s profits. 30Eisenberg, supra note 21, at 817–18 (citing David Campbell & Donald Harris, Flexibility in Long-Term Contractual Relationships: The Role of Co-Operation, 20 J.L. & Soc’y 166 (1993)); Robert A. Hillman, Contract Excuse and Bankruptcy Discharge, 43 Stan. L. Rev. 99, 132–33 (1990); Robert A. Hillman, Court Adjustment of Long-Term Contracts: An Analysis Under Modern Contract Law, 1987 Duke L.J. 1; Ian. R. Macneil, Contracts: Adjustment of Long-Term Relations Under Classical, Neoclassical, and Relational Contract Law, 72 Nw. U. L. Rev. 854 (1978); Richard E. Speidel, Court-Imposed Price Adjustments Under Long-Term Supply Contracts, 76 Nw. U. L. Rev. 369 (1981)); see also Richard E. Speidel, Article 2 and Relational Contracts, 26 Loy. L.A. L. Rev. 789 (1993) [hereinafter Speidel, Article 2 and Relational Contracts]; Speidel, supra note 8.

We do not need to take each suggestion apart to get the gist: relationalists respond to the reality and centrality of relational contracts by applying a set of special doctrines that better resemble loose standards than formalistic rules. Most importantly (and this is not reflected in Eisenberg’s list of “doctrines”), as Macneil has recently emphasized, the key relationalist move on the doctrinal or legal front is to suggest that courts, when faced with disputing parties in relational contracts, should seek to apply the intrinsic norms of the relationship to settle the dispute, even if an application of classical contract law to the parties’ paper deal might call for a different outcome. 31 See Macneil, RCT: C & Q, supra note 16, at 900; see also Speidel, Article 2 and Relational Contracts, supra note 30, at 793 (arguing that relational contract theory’s normative dimension involves internalizing parties’ social norms). Again, relationalists emphasize the implicit dimension of contractual undertakings and the unfolding nature of the contractual relationship, which develops and is modified over time even if the paper deal does not reflect such developments. 32 See, e.g., Macneil, supra note 25, at 1041. Relationalists instruct judges to bring social norms to bear upon disputing parties since the social norms are the real scaffolding for the parties’ relationship and must be given respect and effect to support relational contracting properly. This does not, however, mean that relationalists reject the application of all external norms; applying legal norms is always, in part, the application of external norms. Relationalists merely require heightened sensitivity to calibration with internal norms as well. 33I thank Jay Feinman for reminding me about this important dimension of relational contract theory that is often under-emphasized.

In relational theory’s most modest form, its adherents urge incorporating trade usages, courses of dealing, courses of performance, and a general good faith obligation. But this urging has largely already been adopted by modern contract law through the Uniform Commercial Code—another way we are all relationalists now. 34 See, e.g., Macaulay, supra note 12, at 59; Speidel, Article 2 and Relational Contracts, supra note 30. Courses of performance, courses of dealing, and usages of trade are incorporated by, inter alia, U.C.C. § 1-102(2)(b) (1977) (establishing an “underlying purpose[] and polic[y]” of the UCC to be “the continued expansion of commercial practices through custom, usage and agreement of the parties”); id. § 1-201(3) (defining “agreement” to include these); id. § 1-205 (defining “course of dealing” and “usage of trade”); id. § 2-202 (allowing terms of agreements to be supplemented and explained by these); id. § 2-208 (allowing courts to consider these in determining the meaning of an agreement); and id. § 2-302(2) (allowing courts to consider evidence of “commercial setting” in assessing unconscionability). Commercial reasonability more generally pervades the code as a benchmark for proper behavior and is one of the code’s core gap-fillers. E.g., id. §§ 2-305, 2-309, 2-311, 2-503, 2-504, 2-602, 2-610, 2-704, 2-706, 2-709, 2-714, 2-715, 2-716, 2-718. Moreover, the UCC’s recognition of largely indefinite and vague agreements is a move in the relationalist direction. See id. §§ 2-204, 2-206 (relaxing the offer and acceptance and definiteness requirements of the common law); id. §§ 2-305, 2-306, 2-716 cmt. 2 (recognizing requirements, outputs, and open-price contracts, which might have been too vague to be enforceable under classical contract law or might have lacked consideration under classical rules). Perhaps the single largest relationalist incorporation into the code occurs in U.C.C. § 1-201(3): an agreement may be found solely “by implication,” a method of contract formation from which most formalists would recoil. Moreover, Macneil neatly demonstrates how § 2-207’s abrogation of the strict mirror-image rule for acceptance and its embrace of formation through conduct in the case of the “battle-of-the-forms” is a partial incorporation of elements of relational contract theory. See Macneil, supra note 4, at 72–75. The common law also already embraces many of these minimalist relationalist prescriptions. True devotees of relational contract theory, however, believe that these modest incorporation strategies do not go far enough. The relational contract theory camp recommends a much more substantial effort to mine parties’ relationships for implicit understandings and social norms and to analyze their relational properties to help resolve disputes. These implicit understandings can play a role in adjudicating questions of formation, performance, modification, interpretation, or remedies.

The set of normative claims that relationalists offer about their preferred approach to contract law has met substantial opposition. It can be very difficult to envision how to operationalize a special set of judicial principles to govern relational contracts without, as a threshold matter, having an operationalized definition of the relational contract. Non-relationalists argue that it is difficult to pin down with any degree of clarity which contracts should count as relational ones. According to Eisenberg, “it is impossible to locate . . . a definition that adequately distinguishes relational and nonrelational contracts in a legally operational way—that is, in a way that carves out a set of special and well-specified relational contracts for treatment under a body of special and well-specified rules.” 35Eisenberg, supra note 21, at 813. Although duration is often the test used as a shortcut by relationalists, it is clearly an insufficient one. Some short-term transactions trigger substantial interdependence, and some long-term transactions still remain “discrete” and controlled primarily by legal documents and formalities. Once the test for what counts as a relational contract becomes so diluted as to encompass every contract (because every contract establishes some relationship), there is not, as Eisenberg has argued, much room for a special relational contract law. 36 Id. at 818 (“Because there is no significant distinction between contracts as a class and relational contracts, these rules, and others like them, can be separated into two broad classes: those that are good for all contracts and therefore should be general principles of contract law, and those that are not good for any contracts.”); see also id. at 817 (“There can be no special law of relational contracts, because relational contracts and contracts are virtually one and the same.”).

But Eisenberg’s critique of relational contract theory, 37Eisenberg’s critique is generally embraced by opponents of relational contract theory. See, e.g., Ira Mark Ellman, “Contract Thinking” Was Marvin’s Fatal Flaw, 76 Notre Dame L. Rev. 1365, 1369 (2001); Kimel, supra note 3, at 235–36. a sorites argument of sorts, 38The sorites paradox was a rhetorical strategy used by the ancient Skeptics against the Stoics to prove that their confidence in the possibility of knowledge was unfounded. The argument proceeds by trying to prove that although one thinks that one knows what a “heap” is, it turns out that there are no heaps. (Sorites means heap in Greek.) One might think 10,000 grains of sand make a heap. Yet, since (1) one grain of sand is no heap and (2) for every n grains of sand that are not heaps, n+1 grains of sand will also not be a heap, there are no heaps. For more on the sorites, see Ethan J. Leib, On the Sorites: Toward a Better Understanding of Chrysippus, 21 Ancient Phil. 147 (2001). The sorites argument probably fails because proposition (2) is false. Just because a person may have some difficulty in figuring out where to draw the line between a few grains of sand and a heap does not mean we cannot distinguish the concepts at the extremes. Fuzziness of categories does not mean that they cannot be used, even when it comes to the law. misses its mark in several ways. First, it is only one brand of the normative claim that seeks special treatment for a small class of relational contracts; more thoroughgoing relationalists might very well concede that a special set of standards that would only apply to “relational contracts” is not needed. These relationalists would have contract law develop a generalized law to apply to all or most contracts, in light of their relational nature. Thus, Eisenberg’s concern about the need for an operationalized legal definition of the relational contract may not be necessary after all, if all contracts should be treated as relational. 39And even if a relational contract law would apply only to a smaller class of contracts, a multi-factor test proposed by Speidel (drawing from Lewis Kornauser’s summary of Macneil’s work) is as coherent as any and raises questions about Eisenberg’s general belief in the uselessness of trying to define relational contracts:The degree to which a contract is relational depends upon the presence of some or all of the characteristics discussed below. Most notable are whether “(1) the transaction extends over time, (2) parts of the exchange cannot be measured or specified precisely [at the time of contracting], and (3) the interdependence of the parties to the exchange extends at any given moment beyond any single discrete transaction to a range of social interrelationships.”Speidel, Article 2 and Relational Contracts, supra note 30, at 792 (quoting Lewis A. Kornhauser, The Resurrection of Contract, 82 Colum. L. Rev. 184, 190 (1982)). Undoubtedly, there are difficulties in classification at the margin, but fuzziness of a concept’s boundaries does not mean that no meaningful distinctions can be drawn in clear cases. See supra note 38.

Alternatively, if we take Macneil’s most recent suggestion for the content of relational contract law—to apply internal relational norms to contractual disputes, whatever the relationship and whatever the contract—no complicated work needs to be done to separate the world into relational and non-relational contracts. Rather, to the extent there are any relational norms, the law should remain sensitive to them. That may not be an easy task, but it does not implicate Eisenberg’s concern about differentiating relational and non-relational contracts.

More importantly, perhaps, Eisenberg fails to prove that there is no way to operationalize a law of relational contracts. Although he seems to think a “spectrum approach” of seeing contracts along a spectrum from the discrete to the relational is “acceptable . . . from a sociological and economic perspective,” 40Eisenberg, supra note 21, at 814. Eisenberg seems to contest the centrality of the spectrum approach to Macneil’s relational contract theory. Id. at 813 (“Macneil sometimes treats discreteness as an end of a spectrum rather than as a definition of a body of contracts.”). However, it really is difficult to read Macneil as ever suggesting anything other than a spectrum approach. To be fair, others (like Speidel and Hillman, for example) have tried to treat “relational contracts” as a reasonably distinct category the law could identify and treat specially and differently. In what follows, when I discuss “relational contracts,” I mean “relatively relational contracts” and embrace Macneil’s spectrum approach, which I do not think fails Eisenberg’s challenge. he concludes that a spectrum cannot be used within contract law because “many or most contracts will have both relational and discrete elements” and because “[r]ules whose applicability depended on where a contract is located in a spectrum—that is, on how many relational indicia the contract has and of what kind—would be rules in name only.” 41Eisenberg, supra note 21, at 814. Yet this entire argument hinges on a substantive view about the desirability of rules over standards, something relationalists tend to contest as part of their normative claim. Indeed, Eisenberg offers no proof that standards are not capable of being legally operationalized; he simply states that it would be hard to develop clear rules using a spectrum approach, betraying his preference for rules over standards.

Macneil has always challenged us to see that in dispute resolution certain types of contract mixtures (between discrete and relational properties) will counsel for giving greater effect to the formal planning of the parties and their manifestations of consent, while other types of mixtures will counsel for more flexibility and solidarity enhancement. 42 See, e.g., Macneil, supra note 4, at 17. Although it is not easy to figure out which norms to foreground and when, relationalists try to imbue decisionmakers with this sensitivity. That a multi-factor set of inquiries is involved does not render the inquiry less law-like than clear formal rules, though it is also clear why “rule of law” arguments are routinely invoked against relational contract theory’s doctrinal and normative prescriptions, which sound in standards over rules. 43 See, e.g., Barnett, supra note 8, at 1191.

To be sure, one may want clear legal rules for all sorts of reasons and decide to reject relational contract theory because its approach relies too heavily on standards. Indeed, Macneil has already admitted that good practitioners of relational contract analysis would have to become “anthropologists, sociologists, economists, political theorists, and philosophers.” 44 Macneil, supra note 4, at 70. This is a tall order, indeed. Yet Eisenberg’s dispute is ultimately about the merits, not an explanation for the impossibility of a relational contract law. It is debatable whether it is wise to instruct judges to apply the intrinsic norms between parties to a contract dispute and whether a judge could ever carry out these instructions well in light of evidentiary difficulties. But we could not really say that such an instruction is not capable of being legally operationalized. 45 But see Oliver E. Williamson, Transaction-Cost Economics: The Governance of Contractual Relations, 22 J.L. & Econ. 233, 236 (1979) (“Serious problems of recognition and application are posed by such a rich classificatory apparatus.”). Macneil replies: “A ‘rich classificatory apparatus’ of some kind is essential if contractual relations are to be both understood and subjected to successful, realistic, and reasonably consistent analysis.” Macneil, supra note 25, at 1025 n.26 (quoting Williamson, supra). One can choose to be a non-relationalist because one likes rules over standards. Nevertheless, preferring rules over standards does not tell us, in the words of Eisenberg’s title, “why there is no law of relational contracts.” 46Eisenberg, supra note 21.

There are two related substantive objections to directing judges, in accordance with relational contract theory, to apply relational norms to contractual disputes. One is an argument deriving from the path-breaking work of Lisa Bernstein, which tends to show that the law cannot easily incorporate trade usages and customs because few actually exist and few are generally observed. 47 See generally Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms, and Institutions, 99 Mich. L. Rev. 1724 (2001); Lisa Bernstein, The Questionable Empirical Basis of Article 2’s Incorporation Strategy: A Preliminary Study, 66 U. Chi. L. Rev. 710 (1999); Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search for Immanent Business Norms, 144 U. Pa. L. Rev. 1765 (1996). For a related critique about courts’ inability to incorporate norms with any degree of uniformity, see Robert E. Scott, The Uniformity Norm in Commercial Law: A Comparative Analysis of Common Law and Code Methodologies, in The Jurisprudential Foundations of Corporate and Commercial Law 149 (Jody S. Kraus & Steven D. Walt eds., 2000). This skepticism that the law can discover internal norms within relational contracts is quite relevant for relational contract theory. If relationalists are committed to the normative claim that the law should incorporate and seek internal norms between parties, evidence that tends to show that these norms are rarely mutually understood between parties and rarely embraced by parties surely complicates the viability of the relationalists’ fundamental normative prescription.

Still, there is reason to believe that Bernstein’s work can be read more as a cautionary note about evidence than as decisive proof against relational contract theory as such. In short, Bernstein’s empirical studies are obviously limited to a few areas of inquiry. And although they are fascinating windows into certain trade communities, it hardly follows that there are no immanent business norms that exist, that can be discovered, that are generally followed, and that can be enforced by courts. 48This line of defense is on display in Stewart Macaulay, Relational Contracts Floating on a Sea of Custom? Thoughts About the Ideas of Ian Macneil and Lisa Bernstein, 94 Nw. U. L. Rev. 775, 784–804 (2000). Macaulay also analyzes Scott’s related critique. Macaulay, supra note 12, at 59–68. Some have taken Bernstein’s empirical findings to support the rejection of any incorporation strategy, whether of the UCC or of the relationalists more generally. Yet that seems an unnecessary jump to a conclusion about the viability and desirability of incorporation, and it certainly is not required by the available evidence. In any case, another plausible response to Bernstein’s findings, offered by Macneil himself, is to find ways to get courts to send relational disputes to mediation, arbitration, or agency management since these institutions are likely more capable than courts of discovering such norms, if and when they exist. 49Macneil, RCT: C & Q, supra note 16, at 906.

A related concern that underwrites the argument against the normative claim stems from what Eric Posner calls the possibility and likelihood of “radical judicial error.” 50 See generally Eric A. Posner, A Theory of Contract Law Under Conditions of Radical Judicial Error, 94 Nw. U. L. Rev. 749 (2000). The worry here is slightly different from Bernstein’s. Whereas she emphasizes that there may be no immanent norms between parties, Posner emphasizes that even if there were such norms, judges would not be competent to discern or apply them. Posner is quite clear that this assumption of radical judicial incompetence is just an assumption, 51 Id. at 754. Although Scott abjures from signing on to the radical judicial error hypothesis, concerns about judicial competence certainly animate his argument for formalism as well. See, e.g., Scott, supra note 8, at 858–59, 861. though he offers a bit of empirical evidence to prove the incompetence of judges in at least one area, consumer credit contracts. 52 See Jeffrey E. Allen & Robert J. Staaf, The Nexus Between Usury, “Time Price” and Unconscionability in Installment Sales, 14 UCC L.J. 219 (1982). The gist of his argument is that judges are going to resolve disputes essentially at random if they are stuck trying to figure out parties’ internal norms. However, judges are capable of engaging in formalistic decision-making, and they should stick to that task within their competence. 53There are many nuances to Posner’s view, but to discuss them all is beyond the scope of this Article. Posner’s argument—and arguments regarding judicial incompetence more generally—underwrites a formalistic approach to contract law, which sits in substantial tension with relationalists’ normative claim about how contract law should be administered. To be sure, formalism may be attractive for other reasons. Perhaps non-relational norms in contract law support the underlying relationships best, and judicializing relational norms may “destroy the very informality that makes them so effective in the first instance.” 54Scott, supra note 8, at 852. However, judicial incompetence seems to be central to the non-relationalist view that implicit social norms ought not govern contractual disputes in court.

This is clearly not the place to attempt to settle the age-old debates between formalists and non-formalists. But a few observations seem appropriate in light of the recent effort to rejuvenate formalism in response to the relationalist challenge. First, the assumption of radical judicial error is quite difficult to swallow. 55Even Posner seems to acknowledge that if “courts can determine the parties’ intentions from context and common sense . . . . then courts should ignore form.” Posner, supra note 5, at 162. That concession also raises the possibility of a perfectly acceptable third way that might appeal to some formalists and some relationalists alike: when courts can determine parties’ intentions from context and common sense, they ought to gap-fill accordingly; when they cannot, they should endeavor to be formalistic. For his part, Macaulay would accept a different third way: he would allow courts to be formalistic only when both parties had competent counsel and clearly were able to bargain meaningfully with one another (rather than through battling of standardized forms). See Macaulay, supra note 12, at 62 (citing Binks Mfg. Co. v. Nat’l Presto Indus., Inc., 709 F.2d 1009 (7th Cir. 1983)). Campbell and Collins also develop a nuanced third way of deciding when the implicit dimensions of a contract should control in court. David Campbell & Hugh Collins, Discovering the Implicit Dimensions of Contract, in Implicit Dimensions of Contract, supra note 11, at 25, 42. Admittedly, the empirical evidence available to settle the question of the error-prone nature of judicial decision-making is remarkably thin. Yet it would seem like the claim that judges would do no better than a random coin toss when it comes to using their common sense to understand the underlying deals in contractual contexts should bear the burden of proof. 56To be sure, others in law and economics have not made such radical assumptions about judicial incompetence. See Gillian Hadfield, Judicial Competence and the Interpretation of Incomplete Contracts, 23 J. Legal Stud. 159 (1994). Posner has not met that burden yet, so presuming that properly instructed judges can get it right much of the time (or at least on average better than random) is probably not outlandish. Indeed, even if radical judicial error is possible in a particular class of cases, those are the cases in which formalism might be appropriate. Relational norms can still be applied in other cases where the parties’ intentions can be made relatively clear through context and common sense. 57 See supra note 55. Of course, if courts or parties don’t know when judges are likely to err, it is hard to create a two-track model along the lines suggested. There is also a real cost associated with attempting to learn about party norms and expectations—that cannot be discounted fully, though the cost itself (in judicial time and energy) is not particularly easy to measure.

Second, if judges are, in fact, so incompetent at divining business deals, there would be reason to suspect they would be similarly incompetent at applying the rigors of formalism. After all, formalism’s contributions to contract law—consideration, the statute of frauds, the parol evidence rule, the plain meaning rule, offer-and-acceptance principles, and indefiniteness rules—are hardly uniformly applied and can be unpredictable for parties and lawyers alike. Maybe it is somewhat easier to enforce a paper deal than it is to enforce a real deal, which is hard to discover, especially when parties have incentives to lie once the relationship breaks down. 58Posner has emphasized that the fact-finding processes in judicial adjudications do not inspire confidence and can lead to many judicial errors. Posner, supra note 5, at 153. However, it is still altogether plausible that judges would not be very good at formalism either. They prize their discretion and are not any better at applying potentially undesirable acontextual rules than they are at discerning the understanding of the parties based on the context. 59 See Macaulay, supra note 12, at 62, 77 (drawing upon “realism” to reject the possibility of formalism). If they are not very good at formalism, formalism is not a great alternative to the complexity of discovering relational norms. 60Posner insists that although judges are incompetent at discovering and applying internal relational norms, they are perfectly capable of deciding whether parties have an intention to have legally available remedies in the first place. Posner, supra note 5, at 156; Posner, supra note 50, at 762–63, 767. This central distinction is necessary to make the argument work. But it preys on the ability of courts to stick with formalism, since Posner thinks judges can competently make these determinations through the use of formal reasoning. I am somewhat less optimistic since like most relationalists, I am a realist too—I think judges use play in the joints to manipulate formal categories. If that is correct, there will be—and must be to make his argument work—a fair bit of judicial “error” on the very threshold question about which Posner thinks judges are competent. Consider Judge Richard Posner on the matter:

There has never been a time when the courts of the United States, state or federal, behaved consistently in accordance with [legal formalism]. Nor could they, for reasons rooted in the nature of law and legal institutions, in the limitations of human knowledge, and in the character of a political system. 61Richard A. Posner, What Am I? A Potted Plant?, New Republic, Sept. 28, 1987, at 23, 23.

Third, formalism’s premise that parties will be properly chastened and put all they want in their paper deals anticipating the response of formalistic courts is not obviously viable. It is ultimately an empirical question whether a judicial strategy of formalism can actually succeed in getting parties to put all their important matters into formal arrangements, since the choice to use formalities is a complex one. Parties choose not to use formalities for many reasons. 62 See generally Charny, supra note 9 (discussing the role of non-legal sanctions in the formation of non-formalized agreements). For example, contracting with forms is very costly, and since few contracts actually result in legal disputes, it is a cost that is often not worth incurring because of the low risk of contractual failure. Moreover, in many cases, a long-term relationship will be presumed to require accommodation over time that cannot easily be anticipated at some moment of formal execution. Finally, in the case of many contracts, the relationship is premised on social and interpersonal ties (even though the parties would want legal sanctions if the relationship ends). Being formalistic about the paper deal up front will lead parties not to cooperate at all, costing both parties the surplus they otherwise could have achieved from the relationship as long as it lasts. 63 See Ingram & Zou, supra note 10, at 170. Given that these factors contribute to the sort of incomplete and gap-ridden contracts we see in relational contracting, formalism might actually serve not to chasten parties at all but only to prevent these forms of cooperation altogether. 64There is, it should be noted, some empirical evidence that some relational contracts actually become more detailed over time. See Laura Poppo & Todd Zenger, Do Formal Contracts and Relational Governance Function as Substitutes or Complements?, 23 Strategic Mgmt. J. 707 (2002); Michael D. Ryall & Rachelle C. Sampson, Do Prior Alliances Influence Alliance Contract Structure?, in Strategic Alliances: Governance and Contracts 206 (Africa Arino & Jeffrey J. Reuer eds., 2006); Michael D. Ryall & Rachelle C. Sampson, Formal Contracts in the Presence of Relational Enforcement Mechanisms: Evidence from Technology Development Projects, 55 Mgmt. Sci. 906 (2009). Although this finding seems to stand in some tension with the idea that relational governance relies on loose and open-ended contract terms, it makes some sense too. As the relationship deepens, one can more easily get more specific about the nature of the relationship, which needed to remain more incomplete in its infancy. Macneil suggests as much in his work on relational contract theory. See Macneil, supra note 25, at 1041 (explaining that contract choice “is thus an incremental process in which parties gather increasing information and gradually agree to more and more as they proceed”). The same is probably true in friendships. As time goes on, we certainly get clearer about our expectations and responsibilities within the relationship. Relational contractors are already the ones most immune to formal planning, so it is an odd fit to insist on formalism in these contractual contexts. 65Relationalists may very well embrace formalism for relatively discrete contracts, since the norms of planning and consent are considered especially important by relational contract theory in relatively discrete contracts. See Macneil, RCT: C & Q, supra note 16, at 879–80.

Alternatively, formalism might succeed in part, but then the parties risk getting stuck in their formalities and thus will be unable to rely on the flexibility and good faith that make relational contracts so successful in the first place. Legal norms, when they are formalistically applied, risk eroding the underlying relationship, which does so much of the work keeping parties in line.

Let me elaborate on the last point. It is common to see arguments along the following lines: “[I]n ongoing intimate relationships, legal mechanisms are imperialistic and do not function effectively in concert with extralegal forces” or social norms internal to a relationship. 66Scott & Scott, supra note 6, at 1295. I revisit this debate infra Part IV. However, this is not a necessary conclusion. Indeed, the very same authors who endorse this thesis (one of whom is now clearly a formalist) 67 See Scott, supra note 8. also concede that “legal enforcement of long-term commitments is a particularly valuable method of bolstering extralegal norms.” 68Scott & Scott, supra note 6, at 1330. See also Posner, supra note 5, at 8 (“[M]any legal rules are best understood as efforts to harness the independent regulatory power of social norms.”). But see id. at 64 (“[J]udicial enforcement would interfere with trust relationships.”). Presumably, they view legal enforcement as particularly useful for bolstering social norms when they are weak and as imperialistic when social norms are particularly strong. Although these scholars are to be commended for providing nuance to a generally oversimplified analysis of the interaction between social and legal norms, I suggest that the manner of enforcing legal norms also will likely effect how legal and social norms will interact. My suggestion is that formalism with respect to legal norms is likely to prove more imperialistic and will have a greater likelihood of displacing and interfering with social norms. This is a play on Scott’s argument for formalism—that judicializing social norms will destroy what makes relational contracts so effective. To the contrary, forcing parties to get formalistic and enforcing their commitments in formalistic ways risks accomplishing that very same result.

There is a related argument against formalism from within relational contract theory. Relationalists’ normative claims are underwritten in some measure by an orientation toward using contract law to support “organic solidarity.” This means that the law must be made resonant with our affinities and symbolize to us our general sense of obligations or risk illegitimacy. 69Collins, supra note 24, at 1, 16; Macaulay, supra note 48, at 777–96. See generally Macaulay, supra note 12. The “resonance” idea might be said to find some support in scholarship that traces the relationship between compliance and the perceived substantive fairness of legal institutions. See generally Janice Nadler, Flouting the Law, 83 Tex. L. Rev. 1399 (2005) (arguing that harmonizing law and social norms can help to breed compliance, and the opposite can breed noncompliance); Paul H. Robinson & John M. Darley, The Utility of Desert, 91 Nw. U. L. Rev. 453 (1997). Organic solidarity is also described (with a somewhat different focus) in Gordon, supra note 4, at 569–70. It is an empirical question (though, again, one quite difficult to test) whether formalism interferes with the ultimate legitimacy of contract law and whether members of society will feel that a formalistic approach to contract does not resonate with their aspirations as relational contractors. But relationalists argue that we are more likely to achieve contractual justice, however defined, and contractual legitimacy by tailoring contract law’s application in light of real relationships between parties, not fictitious and incomplete ones on paper. Of course, formalists understand that there are real costs associated with formalism, 70 E.g., Scott, supra note 8, at 861. but relationalists see those costs as so deeply undermining contract law’s purpose and legitimacy that they cannot stomach it. 71 See Collins, supra note 24, at 1, 8 (“[A] process of legal recognition of implicit [and relational] dimensions [of contract] is necessary and inevitable in any system of law.”); Macneil, supra note 16, at 508 (“[P]romise-centered theories [are] inadequate to deal with complex contractual relations without distortion and omission.”). More practically, although formalists emphasize all the gains in predictability and “rule of law” that might result from a commitment to enforcing only clear paper deals that are bargained-for promises, relationalists (that is, relationalists who concede this premise) focus on a different kind of “calculability”: the predictability for the parties’ themselves rather than for their lawyers. 72Some relationalists would just parry this argument with realism and say there is good reason to be skeptical about judges’ capacities to adhere to truly formalistic reasoning. The capitalist system—if that is one’s point of departure for designing a contract—can work better, they might argue, if parties can bank on their real deals, not idiosyncratic paper deals drawn up by their adversarial lawyers with little sensitivity to the underlying relationship between the parties. 73 See Collins, supra note 24, at 10.

* * * * *

I do not expect to resolve all of these central debates of contract theory here. I only hope to give a flavor of the status of the current debate. The argument that follows supports the relational theory of contract by highlighting how relational contracts are like friendships (Parts II and III) and employs that analogy to illuminate some ongoing debates within contract law and theory (Part IV).

II. Friendship as Relational Contract

To argue that friendships can be illuminated by relational contract theory is not to argue that the entirety of the social institution of friendship can be understood in contractualized terms. That is, it is so obvious that friendship is something more than a “mere” contract, even a relational one, that the argumentative endeavor here might be viewed as a category mistake. 74For my “rich classificatory apparatus” about how to define the friend, see Leib, supra note 20, at 638–53; and Ethan J. Leib, Friends and the Law: Can Public Policy Support the Institution of Friendship?, 145 Pol’y Rev. 55 (2007). I also use a particular emphasis in defining friendship in Leib, supra note 22, but limit the analysis there to very close friends. It is fair to say that the working concept of friendship in what follows is a broader category than the one I have utilized in previous work (though not as broad as the type of friendship I explore in Ethan J. Leib, The Politics of Family and Friends in Aristotle and Montaigne, 31 Interpretation: J. Pol. Phil. 165 (2004)). I am not the first, of course, to see the fiduciary relationship as a special case of the relational contract. See, e.g., Goetz & Scott, supra note 23, at 1126–28; Scott & Scott, supra note 6, at 1265–66. Still, with the appropriate caveats in place (and I shall put some more in place after the affirmative case is made for thinking of friendship as a relational contract), this can be an illuminating window into the interpersonal relationship of friendship.

A. Friendship’s Similarity to Relational Contracts

Friendships and relational contracts have several basic structural features in common. In both types of relationships, there is an understanding that parties contemplate a “long-term commitment to pursue shared goals, the fulfillment of which will enhance the joint welfare of the parties.” 75Scott & Scott, supra note 6, at 1229, 1255. It is, of course, true that the “term” and intensity can vary greatly. Short-term friends frequently pursue friendship during, say, summer camp or military training. However, so long as there is some duration and a predicted future, such relationships can qualify as real friendships for the purposes of the comparison. Although parties are never fully expected to engage in complete selflessness in fulfilling their mutual commitment to one another in either friendships or relational contracts, they are generally supposed to be taking their partners’ interests as independent reasons for action. 76For a description of friendship that coheres with this conception, see Jeanette Kennett & Steve Matthews, What’s the Buzz? Undercover Marketing and the Corruption of Friendship, 25 J. Applied Phil. 2, 8 (2008) (“My reasons for action where [my friend] is concerned do not depend on any contingent similarity of interests, and neither are they derivative of other personal or professional commitments. In the case where my close friend’s interests diverge from mine her interests continue to have active guiding force for me, since in close friendship it is her interests as such that are important, not her interests as filtered though my assessment of my own profit or amusement.”).

In pursuing their relationships in the cases of both friendships and relational contracts, parties will often furnish gifts and favors to one another, 77 See Jack L. Carr & Janet T. Landa, The Economics of Symbols, Clan Names, and Religion, 12 J. Legal Stud. 135, 156 (1983); Janet T. Landa, The Enigma of the Kula Ring: Gift-Exchanges and Primitive Law and Order, 3 Int’l Rev. L. & Econ. 137, 152–56 (1983); Ian R. Macneil, Exchange Revisited: Individual Utility and Social Solidarity, 96 Ethics 567, 568 (1986). Those in relational contracts display just this need to distance themselves from “mere” commercial interaction. See, e.g., Uzzi, Sources and Consequences, supra note 10, at 680 (providing evidence of how gifting works within relational exchange). Indeed, there is evidence that gifts can be valued differentially, depending on whether the recipient thinks the gift is based on affect, role, or cost-benefit analysis. See, e.g., Daniel R. Ames et al., It’s the Thought That Counts: On Perceiving How Helpers Decide to Lend a Hand, 30 Personality & Soc. Psychol. Bull. 461 (2004). and the nature of the gift-giving is rarely a matter of pure altruism. 78 See Wightman, supra note 4, at 120 (arguing that although gift-giving can be associated with altruism, it is more generally reciprocal within interpersonal relationships and relational contracts, and its reciprocity complicates the story of altruism); see also Marcel Mauss, The Gift (W.D. Halls trans., Routledge 2005) (1954) (exposing the gift as a form of exchange); Posner, supra note 5, at 50, 55–56 (arguing that gift-giving is rarely deeply altruistic); Eric A. Posner, Altruism, Status, and Trust in the Law of Gifts and Gratuitous Promises, 1997 Wis. L. Rev. 567. To be fair, in some contractual settings, gifts are refused, precisely to avoid muddying the waters in a commercial exchange. See, e.g., Jordan I. Siegel et al., Egalitarianism, Cultural Distance, and FDI: A New Approach 8 (Am. Law & Econ. Ass’n Annual Meetings, Working Paper No. 133, 2008), available at http://law.bepress.com/cgi/viewcontent.cgi?article=2268&context=alea. Siegel and co-authors quote from a form letter that a “prospective customer or partner” could expect to receive if he or she gave a gift to a Motorola employee in Korea:I hereby wish to thank you for your kindness in sending a gift. While I know that your intentions were positive, in order to avoid even the appearance of anything but a business relationship built on the business value offered by our respective organizations, Motorola’s Code of Business Conduct requires that we not accept gifts from our business partners, other than low cost promotional items. Therefore, I am respectfully returning your gift. I would like to take this opportunity to emphasize that it is really unnecessary to send gifts and thank you once again for your gesture of friendship. These reciprocated gifts and favors signal commitment and consideration of one another’s interests over and above the commitment and consideration that result from “mere” commercial exchange in one-shot discrete transactions. There is a deliberate effort in both types of relationships to show especial partiality. 79Id. This is a fascinating but quite unusual business practice. As Siegel and co-authors concede, this unusual strategy forces Motorola Korea to lose business, since it is “de rigueur” within Korea to accept and give such gifts. Id. at 8.There is a large philosophical literature discussing the partiality that is constitutive of friendship. See generally Lawrence A. Blum, Friendship, Altruism, and Morality (1980); Marilyn Friedman, What Are Friends For? 192 (1993); Troy A. Jollimore, Friendship and Agent-Relative Morality (2001); David B. Annis, The Meaning, Value, and Duties of Friendship, 24 Am. Phil. Q. 349 (1987); Neera Badhwar Kapur, Why It Is Wrong to Be Always Guided by the Best: Consequentialism and Friendship, 101 Ethics 483 (1991); Marcia Baron, Impartiality and Friendship, 101 Ethics 836 (1991); Dean Cocking & Justin Oakley, Indirect Consequentialism, Friendship, and the Problem of Alienation, 106 Ethics 86 (1995); John Cottingham, Partiality, Favouritism and Morality, 36 Phil. Q. 357 (1986); Neera Kapur Badhwar, Friends as Ends in Themselves, 48 Phil. & Phenomological Res. 1 (1987); Peter Railton, Alienation, Consequentialism, and the Demands of Morality, in Consequentialism and Its Critics 93, 98–99 (Samuel Scheffler ed., 1988); William H. Wilcox, Egoists, Consequentialists, and Their Friends, 16 Phil. & Pub. Aff. 73 (1987); Susan Wolf, Morality and Partiality, 6 Phil. Perspectives 243 (1992). But see Shelly Kagan, The Limits of Morality 367 (1989) (denying that love and friendship require favoritism).

The requirements for behavior in either relationship are rarely well-specified; we enter these relationships without fully knowing what we shall be called upon to do. 80Macaulay, supra note 12, at 78 n.65. There is also a high degree of interdependence between the parties, both relationships are complex, and both require a set of rather varied duties. Uncertainty about our basic responsibilities is constitutive of the relationships, and they rely heavily on implicit and tacit understandings. 81One other legal scholar has pursued the morphological similarity between friendship and at least one type of relational contract: the employer-employee relationship. See Gary Chartier, Friendship, Identity, and Solidarity. An Approach to Rights in Plant Closing Cases, 16 Ratio Juris 324 (2003) (focusing upon the implicit understandings in each type of relationship).

When one does have to specify particular conduct, parties tend to do so in very general, broad, or vague terms. 82Goetz & Scott, supra note 23, at 1092–93. Most often, parties simply cannot allocate risks of their mutual endeavor at the start of a relationship because so much is uncertain, and so much trust-building is necessary to get the relationship off the ground. 83Scott & Scott, supra note 6, at 1248. Indeed, the very incompleteness of the deal between the parties is central in defining what counts as a relational contract, 84 See, e.g., Goetz & Scott, supra note 23, at 1091 (arguing that contracts are relational to the extent that “parties are incapable of reducing important terms of the arrangement to well-defined obligations”). Again, it is important to draw attention to some of the new findings in the business literature, cited supra at note 64, that prior relationships can lead to more customized and specific agreements than are traditionally associated with relational contracting. The full implications of that finding must await a future date, but I do not think it undermines the morphological similarity between relational contracts and friendships; both lead to more information over time that will allow parties to fill in details of the nature of their expectations. and friendship shares this incompleteness in the specification of parties’ rights and duties. 85Wightman has argued that the reason for or source of incompleteness in relational contracts and intimate relationships, like friendships, is different. Wightman, supra note 4, at 108−09. In relational contracts, he argues that incompleteness results from trade usages, incomplete information, trust, transaction costs, and/or information asymmetry. Id. By contrast, in intimate relationships, “leaving it open” is constitutive of the relationship. Id. I am not sure I agree that the difference is so marked since virtually all of the considerations in play in commercial contexts are likewise relevant in intimate relationships. As Wightman also notes, power inequality within both types of relationships can cause incompleteness. Id.

Yet there are some general principles of conduct that can be presumed in both types of relationships. First, there is an assumption that parties will behave in “good faith” throughout the often uncertain term of the relationship (or risk destroying the relationship). Second, both parties in both relationships will expect “best efforts” in producing joint value from the relationship. And finally, both parties in both relationships will expect parties to adjust their assumed responsibilities reasonably if underlying circumstances change. 86These features of relational contracts are explored and explained in Goetz & Scott, supra note 23, at 1114 (describing the requirement of parties to relational contracts to maximize joint value from the relationship); id. at 1149–50 (describing the “best efforts” requirement); Scott & Scott, supra note 6, at 1265–67 (describing the “best efforts” requirement); Speidel, supra note 8, at 840–42 (exploring the heightened duty of “good faith” in connection with relational contracts); id. at 838 (arguing that courts should facilitate adjustments in relational contracts); and Wightman, supra note 4, at 103 (noting that within relational contracts, new evolutions will lead to flexible adjustment of the underlying deal). See also Hillman, Court Adjustment of Long-Term Contracts: An Analysis Under Modern Contract Law, supra note 30, at 8. In sum, both relational contracts and friendships display high degrees of trust, interdependence, flexibility, reciprocity, and solidarity.

In light of the nature of the obligations that arise from within these relationships, parties generally do not seek legal enforcement of entitlements that flow from them. As Bob Gordon has written of relational contracts, “In bad times the parties are expected to lend one another mutual support, not stand on their rights; each will treat the other’s insistence on literal performance as willful obstructionism . . . .” 87Gordon, supra note 4, at 569. So long as the relationship is still in place, meddling by the law is generally disfavored, 88Scott & Scott, supra note 6, at 1269−70. and most parties to these relationships would think it bizarre and “countercultural” to seek legal enforcement. 89Of course, this is generally credited as Macaulay’s insight about relational contracts in the sources cited supra note 9. Some theorists of friendship have emphasized its “voluntary” nature, suggesting that friends would find the intervention of the law into their rights and duties inappropriate. See generally Scott Feld & William C. Carter, Foci of Activity as Changing Contexts for Friendship, in Placing Friendship in Context 136 (Rebecca G. Adams & Graham Allan eds., 1998); Allan Silver, Friendship and Trust as Moral Ideals: An Historical Approach, 30 Eur. J. Soc. 274 (1989) (arguing that friendships are quintessentially voluntary). See generally Aristotle, Nicomachean Ethics bk. VII, ch. 13, ll. 1162b22–b32; bk. IX, ch. 1, ll. 1164b12–b15 (Terence Irwin trans., 1999) (arguing that friends must live by trust, not by law). I develop and critique Aristotle’s views on the subject at some length in Leib, supra note 20, at 647–53. That said, upon dissolution of the relationship or upon a substantial breach of an agreement evidencing betrayal or opportunism, seeking remedies at law is not altogether uncommon, and the law will routinely find a way to resolve such disputes. 90Disputes between friends and ex-friends are common. See generally Leib, supra note 22, at 700−20; Leib, supra note 20, at 684−97. Even though friendships and relational contracts will generally not seem to the parties like legal relationships when the parties are in them 91 See Scott & Scott, supra note 6, at 1298 (“Thus, the parties will agree to live under two sets of rules: a more flexible set of rules for social and relational enforcement during the marriage, and a stricter set of rules for legal enforcement upon divorce.”). —whether because they are successful or because resorting to the law seems unpleasant and counter to the nature of the relationship itself—legal enforcement of duties is reserved as a rare remedy for certain types of defections from the parties’ duties of good faith, best efforts, and reasonable adjustment. Sometimes parties will seek to formalize some portion of their deal, and sometimes they won’t, but the threat of legal sanctions is rarely the motivating factor for compliance. Even when some formality is sought for a symbolic reason, the formality itself takes on its own import, and legal enforceability is not obviously or necessarily contemplated on account of the formality. 92 See Terence Daintith, The Design and Performance of Long-Term Contracts, in Contract and Organization: Legal Analysis in the Light of Economic and Social Theory 164, 187–88 (Terence Daintith & Gunther Teubner eds., 1986); Campbell & Collins, supra note 55, at 39.

The fact that resort to the law is rare does not mean that friends and those in relational contracts lack mechanisms for enforcing the obligations that flow from the relationship or within the relationship. Social norms internal to the relationship play a strong role in keeping parties in line. They facilitate the parties’ cooperation over time, preventing defection and opportunism. Again, although selflessness is rarely demanded of parties to these relationships, social norms against overly selfish conduct will help parties avoid violating their obligations to their counterparts. These social norms can be enforced through practical concerns about keeping the relationship alive or about reputational costs a defecting party may incur in the future as well as general stigmatic concerns about being a good friend or a good business partner. 93Scott & Scott, supra note 6, at 1288, 1292 (“Social sanctions include reputational harms from gossip, reduced social acceptance and ostracism.”). Friendship certainly “constrain[s] the parties’ freedom and influence[s] them toward trustworthiness, fidelity, [and] honesty. . . . The threat of social censure discourages conduct inconsistent with the announced norms.” 94 Id. at 1289; see also Joseph Epstein, Friendship: An Exposé 69 (2006) (“Whatever else it has to do with, friendship entails obligation—sometimes ample and demanding, sometimes miniscule and subtle, but always, I believe, present.”). Moreover, the social “norms encourage interdependence and trust, which makes extrication from the relationship costly, and they promote mutual oversight by each [party] of the other’s activities.” 95Scott & Scott, supra note 6, at 1290.

How does that oversight occur? The very internalization of norms by the parties serves to prevent defection in some measure. However, parties’ intimate disclosures and sincerity also serve as bonding and monitoring mechanisms to reinforce trust that emerges from the relationship: Truth telling and open communication help parties create the intimacy and faith that sustain the relationship, in part because the risk of disclosure to others as a punishment for defection is salient in such relationships. 96 Id. For some work on the role of truth telling and disclosure within friendship, see Epstein, supra note 94, at 40 (“[O]ne of the things one looks for in a friend . . . is the possibility of easy candor in conversation.”); Michel de Montaigne, Of Friendship, in The Complete Essays of Montaigne 135, 136 (Donald M. Frame trans., 1958) (“Friendship feeds on communication . . . .”); Michel de Montaigne, Of the Art of Discussion, in The Complete Works of Montaigne 705 (Donald M. Frame trans., Stanford Univ. Press 1967) (arguing that friendship “delights in the sharpness and vigor [of verbal] intercourse . . . . It is not vigorous and generous enough if it is not quarrelsome, if it is civilized and artful, if it fears knocks and moves with constraint.”); Andrew Sullivan, Love Undetectable: Notes on Friendship, Sex, and Survival 204 (1998) (“[T]he more you know a friend, the more a friend he is.”); id. at 216 (“What do we tell our friends? We tell them everything.”); Dean Cocking & Jeanette Kennett, Friendship and the Self, 108 Ethics 502 (1998) (explaining, though rejecting, the “secrets” view of the friend); and Allan Silver, Friendship and Sincerity, 4 Sozialersinn 123 (2003); Silver, supra note 20, at 1477 (“Friendship . . . turns on intimacy—the confident revelation of one’s inner self to a trusted other—and on essentially expressive and consummatory behavior.”).

This set of dynamics—very complex and under-specified relationships, social norms being responsible for policing friendships and relational contracts, and legal norms playing only a bit part—also results in another important commonality between the two relationships: resistance to formalizing parties’ agreements and understandings. Although agreements and understandings between parties to relational contracts and friendships are not at all rare, the parties frequently wish to avoid formalities. It is not in the spirit of the relationship to get too formal, and the very request for a formality from one of the parties signals to the other that perhaps the relationship is taking on a rather different character. 97We may ask people to join our list of “friends” on Friendster or Facebook with a formal invitation, but these are not generally thought of as core examples of real friendship since nothing about these lists requires an actual relationship with any intimacy. Nevertheless, these invitations do sometimes bring the invitee into a circle of confidence, in which personal details are disclosed to an “in-group.” I discuss these issues more extensively in the first chapter of my forthcoming book, Friend v. Friend: The Transformation of Friendship and What the Law Has To Do with It.

In sum, many aspects of friendship are clarified when viewed through the lens of relational contract theory. When we think of friendship as a relational contract, we can get a better feel for why obligations within friendships are so hard to specify, why we tend to specify them at a degree of generality and vagueness, what they amount to, why we do not seek to use the law to control conduct within friendships, and how we are still successful in maintaining compliance though social norms. Although this window into describing friendship cannot be exhaustive, it nevertheless proves to be a useful perspective for understanding an implicit dimension of our friendships, revealing their underlying structure, both practical and moral. We have numerous portraits of friendship from sociology, 98 See, e.g., Graham A. Allan, A Sociology of Friendship and Kinship (1979); Graham Allan, Friendship: Developing a Sociological Perspective (1989) [hereinafter Allan, Sociological Perspective]; Graham Allan, Kinship and Friendship in Modern Britain (1996) [hereinafter Allan, Modern Britain]; Claude S. Fischer, To Dwell Among Friends: Personal Networks in Town and City (1982); Anthony Giddens, The Transformation of Intimacy: Sexuality, Love & Eroticism in Modern Societies (1992); Ray Pahl, On Friendship (2000); Placing Friendship in Context (Rebecca G. Adams & Graham Allan eds., 1998) [hereinafter Context]; Liz Spencer & Ray Pahl, Rethinking Friendship: Hidden Solidarities Today (2006). psychology, 99 See, e.g., The Company They Keep: Friendship in Childhood and Adolescence (William M. Bukowski et al. eds., 1996); Conversations of Friends: Speculations on Affective Development (John M. Gottman & Jeffrey G. Parker eds., 1986); The Development of Children’s Friendships (Steven R. Asher & John M. Gottman eds., 1981); Beverley Fehr, Friendship Processes (1996); William K. Rawlins, Friendship Matters: Communication, Dialectics, and the Life Course (1992); Lillian B. Rubin, Just Friends: The Role of Friendship in Our Lives (1985). literary studies, 100 See, e.g., Allan Bloom, Love and Friendship (1993); Caleb Crain, American Sympathy: Men, Friendship, and Literature in the New Nation (2001); Ronald A. Sharp, Friendship and Literature: Spirit and Form (1986); Wayne C. Booth, “The Way I Loved George Eliot”: Friendship with Books as a Neglected Critical Metaphor, Kenyon Rev., Spring 1980, at 4. See generally The Norton Book of Friendship (Eudora Welty & Ronald A. Sharp eds., 1991). religious studies, 101 See, e.g., Alan Bray, The Friend (2003); C.S. Lewis, The Four Loves (1988); Gilbert Meilaender, Friendship: A Study in Theological Ethics (1981); Gene Outka, Agape: An Ethical Analysis (1972). political theory, 102 See, e.g., Hauke Brunkhorst, Solidarity: From Civic Friendship to a Global Legal Community (Jeffrey Flynn trans., Mass. Inst. Tech. Press 2005) (2002); Jacques Derrida, The Politics of Friendship (George Collins trans., Verso 2005) (1994); Love and Friendship: Rethinking Politics and Affection in Modern Times (Eduardo A. Velásquez ed., 2003); James R. Martel, Love Is a Sweet Chain: Desire, Autonomy, and Friendship in Liberal Political Theory (2001); Carl Schmitt, The Concept of the Political (George Schwab trans., expanded ed., Univ. Chi. Press 2007) (1932); Jacques Derrida, The Politics of Friendship, 85 J. Phil. 632 (1988); James V. Schall, Friendship and Political Philosophy, 50 Rev. Metaphysics 121 (1996); Jason A. Scorza, Liberal Citizenship and Civic Friendship, 32 Pol. Theory 85 (2004). anthropology, 103 See, e.g., Robert Brain, Friends and Lovers (1976). and philosophy, 104 See, e.g., David Bolotin, Plato’s Dialogue on Friendship: An Interpretation of the Lysis, with a New Translation (1979); John M. Cooper, Aristotle on Friendship, in Essays on Aristotle’s Ethics 301 (Amélie Oksenberg Rorty ed., 1980); Horst Hutter, Politics as Friendship (1978); Immanuel Kant, The Metaphysics of Morals 261–64 (Mary Gregor trans., Cambridge Univ. Press 1991) (1797); Paul Schollmeier, Other Selves: Aristotle on Personal and Political Friendship (1994); Lorraine Smith Pangle, Aristotle and the Philosophy of Friendship (2003); Suzanne Stern-Gillet, Aristotle’s Philosophy of Friendship (1995); Simon Keller, Friendship and Belief, 33 Phil. Papers 329 (2004); Anthony Kronman, Aristotle’s Idea of Political Fraternity, 24 Am. J. Juris. 114 (1979); Sibyl A. Schwartzenbach, On Civic Friendship, 107 Ethics 97 (1996); Sarah Stroud, Epistemic Partiality in Friendship, 116 Ethics 498 (2006); Jennifer E. Whiting, Impersonal Friends, 74 Monist 3 (1991); see also sources cited supra note 79. but the structure of the relationship and its attendant obligations can be further illuminated by understanding its morphological similarity to relational contracts.

B. Some Cognitive Dissonance Managed

That it is useful to think of friendships as relational contracts does not, of course, mean that one can make a clean one-to-one correlation between these two types of relationships. There are many different sorts of friendships and many kinds of contracts. Here I pursue a few discontinuities between the relationships, attempting to downplay their significance for this project.

1. Value

From the perspective of human valuing, contractual relationships and friendships surely occupy different places in our lives. Although most people would say that both relational contracting and friendships are “valuable,” they would also likely concede that when valuing them they are operating with two different conceptions of value. In the case of relational contracting, many people would say that contracts are valued in a financial or instrumental sense (for wealth generation or capital formation); friendships, by contrast, are valued in themselves.

But what does it mean for something to be valued in itself? Philosophers are helping us answer this question. In his recent account of valuing, Sam Scheffler offers the following sketch (though he self-consciously refrains from explaining whether his account can be linked with instrumental valuing):

[V]aluing any X involves at least the following elements: (1) A belief that X is good or valuable or worthy, (2) A susceptibility to experience a range of context-dependent emotions regarding X, (3) A disposition to experience these emotions as being merited or appropriate, and (4) A disposition to treat certain kinds of X-related considerations as reasons for action in relevant deliberative contexts. 105 Samuel Scheffler, Valuing, in Equality and Tradition (forthcoming May 2010).

Central to Scheffler’s picture of valuing is a form of emotional vulnerability and a meta-requirement that the subject doing the valuing tends to regard that vulnerability as merited. 106Connecting valuing to emotional vulnerability is a feature of many philosophers’ accounts of value and valuing. See, e.g., Elizabeth Anderson, Value in Ethics and Economics 1–16 (1993); Niko Kolodny, Love as Valuing a Relationship, 112 Phil. Rev. 135, 152 (2003) (“[V]aluing a relationship involves being emotionally vulnerable to it . . . .”). Still, there are also alternative accounts. See Harry G. Frankfurt, The Importance of What We Care About, in The Importance of What We Care About 80 (1998) (focusing on caring and the idea of something being important to someone); T.M. Scanlon, What We Owe to Each Other 87–107 (1998) (defining friendship as a core example of value but not linking value with emotional vulnerability). Under this definition, it is relatively easy to see how friendship routinely meets the requirements of human valuing, and it is likewise plain that relational contracts are not obviously of the same character. Although parties to relational contracts will tend to believe that their partners are valuable or worthy and will be disposed to treat relationship-based considerations as reasons for action, their emotional susceptibility in the context of the relationship—its formation, maintenance, and success—should be much more attenuated in the standard case than the emotional susceptibility in the standard case of friendship.

The degree of attenuation of emotional vulnerability within relational contracting highlights how different it is from friendship. Although some relational contract partners might display emotional susceptibility, all friendships must display this feature in substantial measure to qualify as an instance of friendship and an instance of participation in the value of friendship. Moreover, without the disposition to experience the emotional vulnerability as being merited or appropriate, parties would have a hard time understanding themselves to be in a friendship. As a general matter this is not the case with relational contracting partners. Although some such partners may have a similar attitude toward their emotional vulnerability, we would generally find such vulnerability inappropriate in a commercial context. At the very least, we would expect those who exhibit such susceptibility (and those who endorse their susceptibility) to have a pre-existing interpersonal relationship upon which a relational contract was built.

Thus, from the perspective of valuation, there are clear differences between friendships and parties to relational contracts: friendships are essentially connected with our emotional lives (and only sometimes connected to our financial lives), and relational contracts are essentially connected with our financial lives (and only sometimes connected affectively to our emotional lives). 107 But see Jeanne L. Schroeder, Pandora’s Amphora: The Ambiguity of Gifts, 46 UCLA L. Rev. 815, 827−28 (1999) (“I argue that far from being characterized solely by the cold calculation of self-interest, markets are erotic in the Hegelian–Lacanian sense that they are driven by the desire for recognition. Contract, being mutual, reflects the true love relation in which recognition is freely granted and received by equals.”). Nevertheless, this does not prevent us from seeing friendships as relational contracts, so long as we keep the analogy in perspective. 108Even this differential form of valuing is, however, probably contingent. See generally Katharine W. Swett, “The Account Between Us”: Honor, Reciprocity and Companionship in Male Friendship in the Later Seventeenth Century, 31 Albion 1, 4 (1999) (arguing that early-modern era male friendships were characterized by “potentially conflicting values” and “contradictory elements”). That each type of relationship is valued differently does not render it useless to think of the relationships side-by-side. Although the type of value that each creates differs from the other in a fundamental way, they are both relationships that are valued by the parties within them, however attenuated emotional vulnerability may be in relational contracting.

There is, however, another point worth considering. For some, it would seem that the particular valuing process within friendship exists because it is not thought of as contractual in any way. 109 See, e.g., Ira Ellman, Why Making Family Law Is Hard, 35 Ariz. St. L.J. 699, 711 (2003) (“The problem with [treating intimate friendships as contracts] is that it does not acknowledge that duties can arise from relationships themselves, apart from an exchange of promises that constitutes a contract.”). The problem with Ellman’s view is that his contract theory puts explicit “exchanges of promises” at its center; relational contract theory does not, so it is more amenable to acknowledging relational obligations as contractual ones. Generally, friends do not self-consciously think in terms of contract, and doing so might be thought to debase the relationship itself. 110For a related argument about marriage and contract, see Ellman, supra note 37, at 1367. If the moral value of friendship consists in complete voluntariness, 111 See Peter Goodrich, The Immense Rumor, 16 Yale J.L. & Human. 199, 223 (2004) (“Banishing friendship to the rites of passage or ceremonies of death keeps it hygienically private and squarely in the safety of ‘another context.’”); Silver, supra note 20, at 1476 (“[F]riendship in modern society is a quintessentially private relationship, not normatively constituted by public roles and obligations—[and] indeed, often in distinction from them. . . .”); see also Brain, supra note 103, at 75 (“Friendship should be ‘free,’ ‘pure,’ and based on moral obligations alone.”); Pahl, supra note 98, at 63 (“There are no rules and contracts to bind us to our closest friends: we simply have to trust them.”). I have tried to complicate this story about friendship’s voluntariness in earlier work. See Leib, supra note 20, at 663–67. exploring its morphological similarity to the domain of contract—which seems to be about compliance backed by legal sanction—can risk undermining what is so special about friendship. In some measure, this is a variation of a similar argument we confronted earlier that is often launched against relational contract theory more generally. 112 See supra Part I.B, text accompanying notes 20–22. In particular, Bagchi argues that contracts risk “suffocating personal relationships that depend on their separateness from the civil order.” Bagchi, supra note 21, at 35. I take on the “debasement” thesis more directly infra Part IV.C.

But as we can more clearly see by now, relationalists take the position that it is only a crude caricature to see contracts as relationships solely backed by legal sanction and interpersonal relationships as backed only by something else. Both contracts and interpersonal relationships mix legal and non-legal forms of sanctions to maintain compliance in the real world. 113A version of this argument is usefully spelled out in Bellia, supra note 22. Real obligations flow from each type of relationship, and real freedoms exist within each relationship to defy the norms that control them. Neither of these relationships is enforced specifically by the law at all times. The law rarely requires specific performance of any contract (relational or otherwise) and would likely turn its cheek if any friend tried to enforce a “mere” social duty of friendship (like a dinner date). 114See E. Allan Farnsworth et al., Contracts: Cases and Materials 126 (6th ed. 2001) (citing Horsley v. Chesselet, a 1978 case in which the Municipal Court of San Francisco refused to enforce an agreement to go on a date, notwithstanding substantial reliance expenses). All the same, in both, the relationships can be legally enforced to some extent, under certain conditions. 115The details of this argument are deferred until Part IV. It is hard to see how we debase personal relationships by revealing that there is an implicit contractual structure in them, especially when that contractual structure is thoroughly relational. Special internal norms of heightened duties of care pervade and control both contracts and interpersonal relationships. In both instances the participants also share a complex and uncertain range of responsibilities that emerge from the relationship itself.

2. Exchange

A related concern is that contract is principally about exchange but friendship is not. Contracts are surely premised on mutual benefit and a thin form of reciprocity, but treating friendships as forms of transactional exchange, which is implied by thinking of it in terms of contract, is crass and inappropriate. Indeed, the visceral emotional reaction of taboo that some are likely to feel at the very thought of friendships being “reduced” to contractual or market relationships commands some response. 116For a discussion of the taboo, see Alan Page Fiske & Philip E. Tetlock, Taboo Trade-Offs: Reactions to Transactions that Transgress the Spheres of Justice, 18 Pol. Psych. 255 (1997). However, that something is taboo and may provoke anger or disorientation for destabilizing a presumed social order doesn’t mean it is wrong. Nor does it mean that it is avoidable. Fiske and Tetlock’s argument actually suggests that it is almost impossible to avoid making comparisons between markets and social spheres, and that we do so implicitly all the time. See id. at 282, 292–94. In any case, part of my argument here assumes that the sphere of the “market” is not as stranger-based as is traditionally assumed, so we should be less concerned about thinking of the social sphere as having some apparent market traits. And the taboo does not seem to get triggered when one suggests, as I do and will further, infra Part III, that the market sphere of relational contracting partakes in many of the value structures of friendship. See id. at 257 (“[P]eople are less disturbed by applications of Communal norms to relations that they assume should be governed by prices (such an act may even seem ‘nice’).”).To the extent we reject thinking about friendship in this way because of the taboo, the taboo that political psychologists have charted is about “pricing” the value of friendship, which is not something my approach here requires. I am dealing in analogies, not full-scale commensurability. To say that friendships can be illuminated by thinking of them as relational contracts is not to sum up the value of friendship through a contract prism.Indeed, Joseph Raz probably goes too far when he claims that “[i]t diminishes one’s potentiality as a human being to put a value on one’s friendship in terms of improved living conditions.” Joseph Raz, The Morality of Freedom 353 (1986). If one could only value friendship in instrumental terms, it may be true that one could not quite participate in friendships. But merely to notice that friendship has instrumental components that contribute in part to its value even when practiced at its best hardly limits one from having friends in the deepest sense. To concede that elements of friendship are commensurable in some ways with certain “market” relations does not mean one cannot hold on to some weaker form of the incommensurability thesis.For a useful critique of the incommensurability thesis, see Eric A. Posner, The Strategic Basis of Principled Behavior: A Critique of the Incommensurability Thesis, 146 U. Pa. L. Rev. 1185 (1998). The symposium in which Posner’s manuscript appears is a helpful introduction to the subject. See generally Matthew Adler, Symposium: Law and Incommensurability, 146 U. Pa. L. Rev. 1169 (1998).

Although, again, there may be something to the idea that different sorts of exchanges are at work in relational contracts and friendships, there is also by now a respectable window into interpersonal relations that shows them to be meaningful forms of exchange. As Richard Lempert writes: “[A] number of quite prominent sociologists and social psychologists are prepared to argue that almost all social interaction [including friendship] may be profitably viewed as exchange transactions.” 117Richard Lempert, Norm-Making in Social Exchange: A Contract Law Model, 7 Law & Soc’y Rev. 1, 2 (1972) (citing three “exchange theory” classics: Peter M. Blau, Exchange and Power in Social Life (1964); George Caspar Homans, Social Behaviour: Its Elementary Forms (1961); and George C. Homans, Social Behavior as Exchange, 63 Am. J. Soc. 597 (1958)).

To be sure, as Lempert recognizes, and as I concede, it would defy common sense to see friendship as only exchange. 118Lempert, supra note 117, at 2–3. According to Graham Allan, friendship “is not really supposed to be a relationship in which each side carefully weighs up the costs and rewards of their interaction before proceeding, in the way that happens with, say, business contracts.” 119 Allan, Sociological Perspective, supra note 98, at 19; see also Fiske & Tetlock, supra note 116, at 267: “[W]hen either party merely starts to keep track of how much they give and how much they get, a Communal Sharing relationship is in trouble. The very act of keeping accounts . . . seriously undermines the relationship. . . . Even to remind the other of asymmetries is unkind and distancing. . . .” I find this a bit too precious. Of course we keep accounts, even if just approximately. We are generally generous with one another in these relationships and hope for reciprocation over a long time horizon, since we are confident in a future. But when things get out of hand and one friend is doing a lot more than the other to keep the relationship going and intimate, no one should be surprised if one friend confronts the other for an accounting—or terminates the relationship. Consider, in this regard, a useful example drawn by Bob Ellickson:

Dinner guests, for example, commonly bring their host a gift such as a bottle of wine. But no dinner guest would, instead of bringing wine, arrive and say, “Here’s twenty dollars. I’ve learned in an Economics course that you’d undoubtedly prefer this to the usual bottle of wine.” The tender of cash would signal that the guest thought of the dinner not as an occasion among friends but as an occasion at a restaurant, where diners have a merely commercial relationship with those who serve them. 120 Ellickson, supra note 9, at 235.

But this example—and Graham Allan’s account above—relies upon a too simplified picture of business and commercial relationships. That is, very often business relationships are forged on the backs of friendships, 121 See Posner, supra note 5, at 151 (“Contracting parties are often friends.”). See generally DiMaggio & Louch, supra note 10 (examining the impact of friendships on consumer transactions); Ingram & Zou, supra note 10 (exploring “business friendships”); Ingram & Roberts, supra note 10 (arguing that friendships between business competitors are beneficial); Perrone et al., supra note 10; Uzzi, Financial Capital, supra note 10 (arguing that social ties promote trust that can facilitate lending relationships); Uzzi, Sources and Consequences, supra note 10 (exploring how social ties affect economic action in the apparel industry). where it is not easy to disentangle what is motivated by the commercial part of the relationship and what is motivated by the friendship part of the relationship—the confusion may exist not only to outside observers but to the parties as well. 122A particularly useful study of this dynamic can be found in Sally Falk Moore, Law and Social Change: The Semi-Autonomous Social Field as an Appropriate Subject of Study, 7 Law & Soc’y Rev. 719 (1973); see also Nick Rumens, Working at Intimacy: Gay Men’s Workplace Friendships, 15 Gender, Work & Org. 9, 12 (2008) (“[I]ntimacy and instrumentality are sometimes distinct but often overlapping and temporal modes of relating. There may be many pathways to intimacy, some of which may originate within or are generated from the pursuit of instrumental endeavors.”).

More significantly for my purpose here, friends’ purported desires to prove that their relationships are different from “mere” commercial or business enterprises do not vitiate the claim that there is an undeniable element of exchange embedded in every friendship. 123 Allan, Sociological Perspective, supra note 98, at 22. As I suggested earlier, much of what appears as “altruism” can be modeled as exchange. 124 See supra notes 77– 78. Friendships generate a sense of debt since our friends routinely help us out. Although we may say, when we are feeling romantic, that acts of friendships are never undertaken out of a sense of obligation, friendship’s very foundation in reciprocity, equality, and warmth virtually requires some evenness in exchange. If we take more than we give, if we fail to reciprocate with our friend’s effort over time, if gifts and favors stop going back and forth, the friendship cannot continue. 125For more on these dimensions of friendship, see Allan, Modern Britain, supra note 98, at 89 (“Friendship, in whatever form it takes, is defined as a relationship between equals.”); Allan, Sociological Perspective, supra note 98, at 108 (“The essence of friendship from a sociological standpoint is that it is a tie of equality.”); Brain, supra note 103, at 20 (“Equality . . . is part and parcel of friendship.”); Friedman, supra note 79, at 211 (“[V]oluntariness in friendship seems to require, overall, a measure of roughly equal and mutual adaptation, a synergism achieved through the combined and mutually interested adjustments of those who are becoming, or are already, friends.”); Pahl, supra note 98, at 162 (arguing that friendship is “egalitarian”); Graham Allan, Friendship and the Private Sphere, in Context, supra note 98, at 71, 77 (recognizing the importance of “reciprocity” in friendship); and Pat O’Connor, Women’s Friendships in a Post-Modern World, in Context, supra note 98, at 117, 127 (“[F]riendship is a relationship between equals.”). Friendship is not exhausted by peering at it through the lens of exchange, but it is a central feature of the relationship—and seeing friendships as relational contracts helps reveal that dimension of them.

Ellickson observes:

Fellow-feeling seems more likely to arise when [parties] are seen to act out of friendship, not out of a need to scratch each other’s backs. Close friends have such a long future ahead of them that they need not worry about minor imbalances in the reciprocated favors between them. Therefore, a person who mentions that accounts have fallen a bit out of balance indicates either a lack of intimacy or some skepticism about future solidarity. 126 Ellickson, supra note 9, at 236.

Be that as it may, this dynamic is no less true of parties to relational contracts, who are also eager to make displays of friendship within a larger network of exchange.

3. Unity of Interests

Some might suggest that the interpersonal relationship of friendship rests upon a unity of interests that could never be mirrored in relational contracts. Ultimately, friends engage in reciprocity because they wish their counterparts well for their own sake—and they share so many interests that they can be treated as second selves. 127On wishing friends well for their own sake, see Aristotle, supra note 89, at bk. VIII, ch. 3, ll. 1156b10–b12; id. at bk. VIII, ch. 2, ll. 1155b33–b34. For the literature on friends as second selves, see, for example, id. at bk. IX, ch. 9, ll. 1170b5–b8 (“[A] friend is another []self.”) and Montaigne, supra note 96, at 141, 143 (arguing that friends are of “one soul in two bodies” and maintain a “fusion of . . . wills” and are a “second self”). Sometimes, this latter dimension of friendship is described as “agreement,” “concord,” or “consensus.” See Aristotle, supra note 89, at bk. IX, ch. 6, ll. 1167a22–b15 (noting that friends have similar judgments about their common interests); Epstein, supra note 96, at 60 (citing Cicero’s suggestion that friendship requires “agreement over all things divine and human”); Goodrich, supra note 111, at 211–12 (discussing concord and agreement). No commercial relationship, the presumptive domain of contract, can unify the interests of the parties in the way friendship can. 128 But see generally Markovits, supra note 19 (arguing that contracts are collaborations that unify party interests). Markovits could, however, distinguish the unification of interests in contracts and the unification of interests in friendship by highlighting the limited scope of the unification in contracts (friendship requires a broader scope of unity) and the lack of affective association in contractual collaboration.

There is a kernel of truth to this difference between friendship and relational contracts. Indeed, the closer friends are, the more unified their interests. 129This is why, perhaps, as friends get closer it makes greater sense to treat them as fiduciaries, not as “mere” relational contractors. For the argument to view close friends as fiduciaries, see Leib, supra note 22. And the more unified their interests, the less the relationship seems like a contract, if one assumes for the moment that relational contracts do not create unified interests in the standard case. Even if contractual partners are not modeled as having adverse interests (and many contract partners need to collaborate to maximize the surplus within the relationship), it would be odd to think they have unified interests in the way we might presume friends do.

Yet some things can be said to take the sting out of this reality check. First, this too is an overly romantic conception of the standard case of friendship. In the average friendship (even if not its paradigmatic form, perhaps), friends rarely have unified interests. They have many priorities and several friends, all competing for attention. Even in our most intimate relationships, we have to juggle too much to impute perfectly unified interests to friends. It just is not possible for parties to a friendship to always have unified interests, especially when a friend’s lover, parent, employer, professor, or child is also competing for attention. Indeed, some might argue that conflict and conflict resolution techniques are also constitutive of friendship. 130I defend this view in Leib, supra note 20, at 646–47 (citing Epstein, supra note 94, at 240–41; Montaigne, The Art of Discussion, supra note 96, at 705; Pahl, supra note 98, at 86; Sullivan, supra note 96, at 204; Bernard Yack, The Problems of a Political Animal: Community, Justice, and Conflict in Aristotelian Political Thought 110 (1993)). Wishing a friend well for her own sake is, perhaps, different from taking on her interests as our own priorities on a routine basis.

Moreover, we only rarely own property in common with friends—and even if we do not find ourselves literally competing for property in our friendships, 131 See Stuart Oskamp & Daniel Perlman, Effects of Friendship and Disliking on Cooperation in a Mixed-Motive Game, 10 J. Conflict Resol. 221, 221 (1966) (suggesting that friendship can sometimes discourage parties from cooperating in a cooperation game because of competitive urges); Graham M. Vaughan et al., Bias in Reward Allocation in an Intergroup and an Interpersonal Context, 44 Soc. Psychol. Q. 37, 40 (1981) (finding that young friends will allocate as much to their friends as to an arbitrary in-group). we are bound to have disparate interests on a number of fronts, both personal and professional. Although we are undoubtedly expected to be generous with our friends and treat them with good faith, few of us expect our friends to put our interests on equal footing with their own. In a rare case (such as a “best” friend or an emergency), we might expect such loyalty, but that is not the general rule.

The institution of contract is also being a bit under-romanticized here as well. Relational contract theory highlights just how much interdependence and unity of interest can be created through the relational contract itself (and the underlying relationships that are part of the contractual matrix). Accordingly, it might be fairer than it seems at first to put friendship and relational contract side-by-side on the dimension of the unity of interests. 132One can also see this morphology through the use of Markovits’s theory of contract as collaboration. Although he overstates the distinction between contracts and interpersonal relations because he thinks contracts cannot have an affective dimension, one can see through his theory (portions of which could be adapted by relationalists, mutatis mutandis) how relational contracts can establish a similar interest-structure that prevails in friendships. See Markovits, supra note 19, at 1432, 1451. In the standard cases of each, neither establishes a perfect unity of interest, but both establish some unity, limited in scope.

4. Natural Persons, Friendship, and Organizations

We might appreciate that something is amiss in thinking about friendships as relational contracts when we realize that friendships must be limited to natural persons, but relational contracts have no such limitation. Corporations and organizations can be friends only by analogy because only natural individuals can be friends. And, yet, as I have argued in other work, it is more than likely that contracts between organizations (“Type 3” contracts) and contracts between organizations and natural persons (“Type 2” contracts) are just as much at the core of contract as are contracts between natural persons (“Type 1” contracts). 133 See Leib, supra note 29, at 2–3, 9–20. So how can a practice—friendship—that reasonably comes only in the shape of a Type 1 contract be illuminated by a practice—relational contract—that has many other shapes (in which friendship may not partake by its very nature)?

It turns out this is not much of a challenge for this portion of my argument. Perhaps this challenge shows some limits to my general argument—there are real differences between contract as a social practice and friendship as a social practice. But analogical arguments do not require perfect correspondence of the concepts being compared. Of course, differences in the subjects of the analogy may counsel for differential treatment of the two things being compared when we turn to operationalizing the upshot of the analogy. But I do not think the analogy is threatened by showing a mere difference. In any case, nothing about this failure of perfect isomorphism vitiates the usefulness of thinking about friendship as, in certain respects, a prototypical Type 1 relational contract. 134Perhaps I am throwing in the towel too quickly here. As Verity Winship observed in a conversation regarding a prior draft, we do have groups of friends to whom we relate qua group—and we also have relationships with couples (whether as couples ourselves or as individuals). These sorts of friendships might, after all, serve as plausible analogies for Type 2 and Type 3 contracts. I have not worked out how that might be played out through the rest of my analysis, but I think it is a provocative suggestion. Indeed, much of the economic sociology cited supra note 10 clearly assumes that friendships among organizations can be meaningful. See Ingram & Roberts, supra note 10, at 387. And a recent article about friendships between countries similarly suggests that organizational friendship is a meaningful concept. See P.E. Digeser, Friendship Between States, 39 Brit. J. Pol. Sci. 323 (2009). Although this observation about the shapes in which relational contracts come and the shape in which a friendship must come may raise some problems for portions of the analysis to follow—where I suggest that friendship be thought of as the paradigmatic relational contract—at least for this part of my argument, this observation presents no barrier. I am not arguing here that all relational contracts are friendships, only that all friendships display central properties of relational contracts—and that this observation is useful to understanding the dynamics within and the requirements of friendship.

III. Relational Contract as Friendship

I observed at the outset that relational contract theorists tend to want to treat relational contracts as types of marriages. 135 See sources cited supra note 4. My hope here is to expose the benefits of treating relational contracts as types of friendships rather than types of marriages. Reorienting the paradigmatic relational contract to be a friendship rather than a marriage has several benefits. And this reorientation should furnish several lessons for relational contract theory and relational contract law, to be explored in Part IV. In any case, relational contracts are more like friendships than they are like marriages.

A. De-centering Marriage

First, there are very good reasons to de-center marriage quite generally from the various roles it currently plays in our legal system. Consider some of the reasons recently suggested by scholars with quite different points of view: (1) privileging marriage excludes those who cannot partake in marriage in a way privileging equal-opportunity friendship does not; 136 See generally Nancy D. Polikoff, Beyond (Straight and Gay) Marriage: Valuing All Families Under the Law (2008) (arguing that the law should not impute more value to marriage than to other family forms); Dan Markel, Jennifer M. Collins & Ethan J. Leib, Criminal Justice and the Challenge of Family Ties, 4 U. Ill. L. Rev. 1147, 1153 (2007) (“[F]amily ties preferences can disrupt norms of equality that should otherwise prevail in an attractive regime of liberal governance.”). (2) privileging marriage contributes to gender inequality in our society in a way the promotion of friendship may not; 137 See Laura A. Rosenbury, Friends with Benefits?, 106 Mich. L. Rev. 189, 191 (2007) (“The gendered prerequisites of marriage are seen as unrelated to family law’s goal of achieving gender equality, thereby permitting the boundaries of the legal family to continue to perpetuate gender inequality.”). and (3) de-centering marriage could “enable[] new forms of commitment, responsibility, love, care, and relatedness other than those of idealized ‘mother’ and ‘father,’” and “husband” and “wife.” 138Katherine M. Franke, Longing for Loving, 76 Fordham L. Rev. 2685, 2705 (2008). Franke’s very recent turn toward friendship is exciting for those of us who have been working in this area. I have qualms about some of her claims in the article about what friendship is—and about how she imagines that we ought to orient the law towards friendship. However, I am very sympathetic to her project of re-centering our lives and law toward friendship to displace the overwhelming role of marriage.

In short, to the extent that we model the ideal form of a relational contract on marriage rather than friendship, we are unnecessarily, and to our detriment, reinforcing marriage’s centrality in our private and public law. 139 See Ariela R. Dubler, In the Shadow of Marriage: Single Women and the Legal Construction of the Family and the State, 112 Yale L.J. 1641, 1712 (2003) (“[M]arriage continues to regulate the terrain outside of its formal borders, preserving its legal and ideological supremacy as a normative model for all intimate relations and as an arbiter of which relationships deserve legal recognition and protection.”); Franke, supra note 138, at 2698 (“To borrow a term Janet Halley has recently put to much use, all of us are conscripted into the cause of ‘carrying a brief for’ marriage whether or not we so wish.” (quoting Janet Halley, Split Decisions: How and Why to Take a Break from Feminism 17 (2006)). Even if one thinks—as many relationalists do—that certain contractual obligations emerge from the complex relationships to which partners are parties rather than from the partners’ discrete promises to each other, we need not model those obligations on the status-oriented obligations married spouses have. Real obligations can stem from other relationships, too. 140On relational obligations more generally, see Samuel Scheffler, Relationships and Responsibilities, 26 Phil. & Pub. Aff. 189 (1997). Friendships can serve as this model all the same, without needing to buy into the normative agenda that props up marriage’s centrality to our lives and law.

In addition, so much of the law’s orientation to and regulation of marriage has to do with regulating sex and child-rearing that it would seem odd to compare commercial contracts and their regulation with the law’s concerns in regulating marriage. Although calling contracts types of friendships will also produce some imperfect correlations, the concern with gender, sex, and child-rearing so predominates the law of marriage that it is a poor source of relational contract law. However, some relationalists still might embrace the marriage analogy precisely because adjudication in marriage and divorce (as in some commercial contexts) often needs to weigh the interests of affected third parties (children, in particular). Centering friendship does, indeed, lose this dimension of comparison. But, in any case, the comparison is a weak one. The prevailing standard of the child third-party’s “best interests” could only be useful in adjudicating the rights of third parties to relational contracts in extremely rare circumstances. Furthermore, friendship as a source of relational contract law would not cause quite as much of a distortion since it is a gender-neutral practice, 141I discuss gender and friendship in Leib, supra note 20, at 667–69. There, I make clear that it is actually controversial to see friendship as gender-neutral. Still, unlike marriage under federal law and the laws of most states, friendship is not gendered by law. in which sex and child-rearing are not considered essential components of the relationship.

B. Exit Costs

Relationalists seek to highlight the similarity between relational contracts and marriages because exit costs from both are quite high. Getting divorced is usually quite an elaborate affair, and the decision to divorce is rarely made casually. Parties grow interdependent throughout a marriage, and that interdependence is the very feature relationalists claim also exists within relational contracts. As David Friedman writes:

Once a couple has been married for a while, they have made a lot of relationship-specific investments, [bearing] costs that will produce a return only if they remain together. Each has become, at considerable cost, an expert on how to get along with the other. Both have invested, materially and emotionally, in their joint children. . . . [T]hey are now locked into a bilateral monopoly with associated bargaining costs. 142 David D. Friedman, Law’s Order: What Economics Has to Do with Law and Why It Matters 172 (2000).

But friendship is actually the better analogy here. Friends also display the relevant sort of interdependence with high exit costs (relative to the exit costs generally associated with relations between strangers). But the exit costs of marriage are too high to serve as a baseline for relational contracts. Friendship’s exit costs are also substantial, to be sure, with real prices to pay for extricating oneself, 143Indeed, unlike marriage, there is rarely a moment of divorce in friendship. In the standard case, friends don’t really break up; it is a slow process of developing new, more intimate ties elsewhere. Still, although it is easy to point out when a marriage ends officially, the actual process of breaking up and growing apart over time does seem to resemble the exiting process of friendships. Unless there is a massive betrayal in either case, it is hard to say when things have turned irrevocably against the relationship’s durability. but the costs are usually lower and much closer to what we might tend to think is true in the relational contract sphere. Parties to a friendship and parties to relational contracts are generally free to terminate the relationship in a way that marriage partners are not. Marriage partners must turn to law—and pay for it. Divorce is never free. It carries a much bigger transaction cost than terminating a friendship or a relational contract, which parties are generally free to do without the help of the law. 144To be sure, they may have to turn to lawyers and may be prevented from free exit because they have to pay damages first—in both relational contracts and friendships. But the analogy with marriage doesn’t hold; no one needs the state’s permission to terminate these other types of relationships. Additionally, many relational contracts will be like friendships in yet another way: there will be no agreement to terminate, just a walking away, a growing apart, a fizzle. And that type of ending will also rarely trigger legal involvement. This is not so in marriages, where even amicable dissolutions require elaborate legal intervention. On the dimension of exit costs, then—which is supposed to be a useful dimension of comparison between marriages and relational contracts—it turns out that friendship is the better model.

Although it is undoubtedly true that marriage “adds an overlay of legal and social sanctions that further restrict the freedom to renege and thus strengthen each partner’s commitment,” 145Scott & Scott, supra note 6, at 1255. friendships resemble relational contracts better than marriages. Relational contract partners, like friends, are bound in a web of social relations, and they do not need the law’s permission to extricate themselves therefrom, as they would in a marriage. Free exit affects the success of the social norms’ enforcement capabilities in both friendship and relational contracts, 146 Posner, supra note 5, at 77. but in both sets of relationships, social norms nevertheless do the vast majority of the work in motivating the parties to cooperate and comply with their underlying deals.

C. Exclusivity

Marriages, for the time being, can only be valid between two individuals and must be exclusive. That just isn’t the case with friendships. Even if friendships are essentially dyadic, 147I discuss the potentially dyadic nature of friendship in Leib, supra note 20, at 644 n.50. More generally, in my earlier work I conclude that friendship must be “exclusive,” but that exclusivity is of a different type than the one that must prevail in marriage. See id. After all, infidelity in marriage is actually criminalized in many states and the District of Columbia. Jennifer M. Collins, Ethan J. Leib & Dan Markel, Punishing Family Status, 88 B.U. L. Rev. 1327, 1411 (2008) (citing Elizabeth F. Emens, Monogamy’s Law: Compulsory Monogamy and Polyamorous Existence, 29 N.Y.U. Rev. L. & Soc. Change 277, 290 n.49). we can certainly have more than one friend. Since the realm of contract clearly does not limit parties to one partner, friendships are again a better analogy than marriages.

There is something more here too—and it relates to an earlier point about interdependence. Although relational contracts tend to breed and feed off of interdependence, the degree of that interdependence is not commensurate with the degree of interdependence in purely exclusive relationships and bilateral monopolies. To be sure, some relational contracts are bilateral monopolies, but they need not be at their core. Friendship is more true to the nature of the relational contract because it is not a betrayal to have other partners, so long as one deals with counterparts in good faith. This point also relates back to the discussion about the lack of the unity of interests in relational contracts and friendships. Marriages command a type of extreme internal egalitarianism 148For more on this feature of marriage, see Carolyn J. Frantz & Hanoch Dagan, Properties of Marriage, 104 Colum. L. Rev. 75, 81–98 (2004). that friendships simply do not (even though they too are predicated on a certain form of equality). 149For sources on equality in friendship, see supra note 125.

D. Common Ownership

In marriage, common ownership of property is a very common arrangement. At marriage dissolution, most states will enforce some version of common ownership of all marital property. 150 See Scott & Scott, supra note 6, at 1270–71. Yet, in neither relational contracts nor friendships would we generally assume that the parties intend to share their property in joint ownership. To be sure, some very close friends and business partnerships have such arrangements. But friendships and relational contracts do not generally display this ownership pattern, which is quite typical within marriages. This difference between marriages and relational contracts is instructive for how we might set up a law to regulate relational contracts on the one hand and marriages on the other, so it is best to model the relational contract on the type of property assumptions that prevail within friendships. To treat all relational contracts as marriage partnerships would not be true to the prevailing ethos within relational contracts, which are closer to friendships in this respect.

E. Default Rules

There is a more general point to make about default rules and marriage. For the most part, marriage law is full of mandatory rules, not default rules that parties may circumvent by agreement if they choose. 151 See, e.g., Posner, supra note 5, at 79. By contrast, relationalists have argued for relatively few mandatory rules (like robust good faith and best efforts requirements). There is an assumption that parties may still contract around whatever other default rules the law will create for them. Friendship seems to work this way too—there are precious few mandatory rules (though robust good faith and best efforts in performance may very well be two of them, for without both it is hard to call the relationship a friendship in the first instance). But other than these two mandatory duties, most friendships, like relational contracts and unlike marriages, admit a wide range of choices about how to structure and perform duties in the relationship. As friendships get closer to marriages in intimacy and interdependence, they get closer to fiduciary relationships—with their attendant set of more elaborate mandatory duties. 152 See generally Leib, supra note 22 (arguing that close friends can be thought of as fiduciaries). It should be noted that there is a danger in having most or many contracts trigger the fiduciary duties we tend to associate with spouses; we might cause a “stifling despotism of virtue” within contractual relations. See Roberto Mangabeira Unger, The Critical Legal Studies Movement, 96 Harv. L. Rev. 561, 641 (1983). This is part of why it is generally best to see relational contracts as “mere” friendships (and vice-versa)—and only very close friends as fiduciaries. But the general case of relational contract seems much more analogous to the general case of friendship than to the general case of marriage.

F. Formalities

There is an important difference between marriages and relational contracts from the standpoint of relational contract theory—and the difference is so central that it is rather surprising that relational contract theorists so often talk about contracts as marriages. Marriages generally cannot be entered or exited without clear formalities that are extremely easy for outsiders to spot. 153Common law marriage is clearly the exception to this rule. But common law marriage is rare (it exists in nine states and the District of Columbia) and getting rarer still even though five states have grandfathered couples under earlier, and now repealed, common law marriage laws. See Nat’l Conference of State Legislatures, Common Law Marriage (2009), http://www.ncsl.org/programs/cyf/commonlaw.htm. Moreover, even if a couple “marries” through a common law marriage, the parties must make formal showings to receive treatment as a married couple (by, for example, changing their last names to match one another’s, filing joint tax returns, or holding themselves out to the public as married), and may only dissolve their relationship through a formal divorce. Yet relationalists generally argue that although those who enter relational contracts sometimes use formalities, they do not rely on them exclusively. For relationalists, there need not always be clear formalities between parties to establish contractual obligation, and courts should not necessarily require formalities to find contractual liability. 154 See Eisenberg, supra note 21, at 817–18. Indeed, this element of relational contract theory is the one most opposed by formalists, who certainly do not want courts to find that contracts have been formed without first establishing that the parties have engaged in the relevant formalities—like offers, mirror-image acceptances, writings (where required by the statute of frauds), and consideration. Relationalists are much more comfortable with implied agreements found through conduct and other informal acts. Perhaps unsurprisingly, the law and economics scholars who embrace the idea of relational contract as marriage and marriage as relational contract have veered toward formalism since marriage is very formalistic at both formation and break up. 155 Compare Posner, supra note 5, at 69–70, 160 (embracing, from the point of view of a legal economist, the model of relational contract as marriage), and Scott & Scott, supra note 6 (discussing how marriage can be thought of as a relational contract), with Posner, supra note 50 (endorsing formalism from the same point of view of a legal economist who embraces the relational-contract-as-marriage model), and Scott, supra note 8 (embracing formalism from the perspective of a legal economist who embraces the relational contract-as-marriage model). I should distinguish formalism about formation from formalism about interpretation. Although these formalists embrace formalism across the board, here I am only talking about formalism about formation.

Here, friendship seems like the better analogy to relational contracts, at least from the perspective of relational contract theory. The relationship of friendship tends to be fluid (there is rarely a magic moment of formation); it does not rely on formalities for entry or exit (at least for now and at least in the United States); 156 But see Haw. Rev. Stat. §§ 572c-1 to -7 (2009) (illustrating a “reciprocal–beneficiary” statute allowing parties to contract into a form of friendship); Vt. Stat. Ann. tit. 15, §§ 1301–1306 (2002) (also a “reciprocal–beneficiary” statute); Brain, supra note 103, 106–07 (urging more ceremony surrounding friendship formation); David L. Chambers, For the Best of Friends and for Lovers of All Sorts, A Status Other than Marriage, 76 Notre Dame L. Rev. 1347, 1348 (2001) (arguing for a “designated friend” registry, enabling the state to keep track of who is entering friendships); Leib, supra note 20, at 683 n.267 (explaining the German custom of Brüderschaft trinken, which is a friendship ceremony of sorts and considering the value of establishing more ceremonies for entering friendships). Under the Hawaii statute cited above, “reciprocal beneficiary status [is] available to heterosexual couples unable to marry, such as brother-and-sister, father-and-daughter, and mother-and-son couples, as well as to same-sex couples. . . . [and provides] many benefits available to married couples including . . . the right to hold property by tenancy of the entireties, eligibility for health insurance coverage as family members, wrongful death and loss of consortium claims, [and] family and funeral employment leave. . . .” Lynn D. Wardle, Counting the Costs of Civil Unions: Some Potential Detrimental Effects on Family Law, 11 Widner J. Pub. L. 401, 415 (2002) (footnote omitted). The Vermont statute allows “couples related by blood or adoption to register in a new nonmarital status and receive nominal marital rights and benefits, mostly relating to health care and final medical decisions.” Id. at 417. See also Rosenbury, supra note 137 at 223 (“The status of designated friends could therefore be described as a ‘marriage–lite’ approach.” (quoting J. Thomas Oldham, Lessons from Jerry Hall v. Mick Jagger Regarding U.S. Regulation of Heterosexual Cohabitants, or, Can’t Get No Satisfaction, 76 Notre Dame L. Rev. 1409, 1430-33 (2001)). it is not regulated by law upon entry and exit (as marriage is); 157 See Posner, supra note 5, at 77. and it can create obligations in non-formalistic ways. 158 See generally Friedman, supra note 79, at 208–09 (“Friendship receives far less formal recognition in our rituals and conventions . . . than do familial relationships and is maintained by fewer of those formalities.”). Although I have argued, see Leib, supra note 20, at 685–94, and will argue infra Part IV, for establishing some legal obligations that emanate from friendship, the ones I have in mind here to establish the analogy are only ethical obligations. Formal entry and exit does not reflect the reality of practice in relational contracting, so friendship is the better paradigm.

Although contracts can be formalized, what is distinctive about relational contracts is that even when some part of the relationship is formalized, there is much more to the deal besides the formalities. One could say the same of friends, too: although surely friends can enter formal contracts (and unquestionably legally enforceable (relational!) contracts too), relational contract theory would emphasize the underlying social relationship that is an additional source of obligation. In any case, in relational contracts, as in friendship, formalities are the rare case; in marriage, formality is the standard case and is a prerequisite to getting the relationship off the ground in the first place.

G. Enforceability of Intra-relational Promises

Finally, if we model relational contracts on marriages, we risk reaching a potentially perverse result where promises within commercial relationships—even ones that are written down and bargained for—could be rendered unenforceable. This is at odds with the normative dimension of relational contract theory, which recommends the enforceability of internal norms and promises formed within relational contracts.

Only in extremely rare cases will courts enforce agreements between contracting spouses, even if they are accompanied by formalities like writings and consideration. 159 See, e.g., Balfour v. Balfour, (1919) 2 Eng. Rep. 571, 575 (K.B.) (finding agreement by separated husband and wife for support unenforceable); Miller v. Miller, 35 N.W. 464, 464 (Iowa 1887) (finding written agreement between husband and wife unenforceable). For a largely supportive argument in favor of Balfour, see Stephen Hedley, Keeping Contract in Its Place—Balfour v. Balfour and the Enforceability of Informal Agreements, 5 O.J.L.S. 391 (1985). For other (not quite relational) approaches to Balfour, see Michael Freeman, Contracting in the Haven: Balfour v. Balfour Revisited, in Exploring the Boundaries of Contract 68, 68–82 (Roger Halson ed., 1996); Peter Goodrich, Friends in High Places: Amity and Agreement in Alsatia, 1 Int’l J. L. Context 41, 44–46 (2005). Whether owing to the court’s failure to find that the parties intended to create legal relations or because of public policy, courts tend to refuse to adjudicate contractual disputes between spouses within a marriage. The conceit routinely offered is that the law wishes to afford a bubble of autonomy for the household—and this is one way to give effect to that autonomy. 160 See Posner, supra note 5, at 68. However, Posner readily concedes that the “autonomy” description is an odd one—and hard to reconcile with how the law actually regulates the family. Id. at 68–69. While many relationalists have tried to argue that this approach is mistaken, few contest the description of the law. 161 See, e.g., Wightman, supra note 4, at 94; Mary Keyes & Kylie Burns, Contract and the Family: Whither Intention?, 26 Melb. U. L. Rev. 577, 578 (2002). Accordingly, it is potentially bizarre to model the relational contract on a type of relationship that generally cannot take advantage of contract law.

By contrast, despite some underdeveloped enthusiasm for the view that friends should be wholly immune from the law, 162 See, e.g., Aristotle, supra note 89, at bk. VIII, ch. 13, ll. 1162b22–b32; id. at bk. IX, ch. 1, ll. 1164b12–b15 (discussing how some cities do not allow legal actions on contracts between friends); Franke, supra note 138, at 2705. few would really argue that contractual obligations between friends should be held unenforceable as a categorical matter. To be sure, we may agree that certain promises or commitments would not rise to the level of enforceable obligations, and that there is something distinctive about obligations within relationships of friendship that would leave some of them to be enforced principally by social norms rather than the law. Yet, most would agree that friends can undertake contractual obligations to one another through writings and formalities and should have no trouble as a general matter getting the law to enforce such agreements. 163An area of likely dispute between relationalists and non-relationalists would be whether friends can generate legally enforceable obligations to each other without formalities. See infra note 175. The relationalists might allow it because (some) social norms are meant to be incorporated by the law—and relationalists don’t fetishize promissory or formal obligation as the core of contract. The non-relationalist would have none of this. I pursue this territory infra Part IV. Relational contracts exhibit this same dynamic. Like friends, parties to a relational contract ought to have no trouble with enforceability as a general matter, even if some of the deal is to be enforced by social norms rather than the law. Married partners do in fact have such trouble entering into enforceable contracts; accordingly, relational contracts are better modeled as a friendship on this dimension.

H. A Caveat or Three

I hope I have shown here why friendship rather than marriage ought to serve as the paradigmatic relational contract. Does this really mean that all contracts are friendships? I am not committed to that view. Instead, I hope to have shown through Parts II and III that relational contracts partake in the morphology of friendship (and vice-versa) and can usefully be analyzed through their similarities to the structures of friendships. Not all contracts are the same, and not all friendships are the same. But putting the two concepts side-by-side is both a useful exercise on its own and, ultimately, a more useful exercise than putting relational contracts and marriages side-by-side. Analogical and morphological reasoning does not commit me to the full-scale identity of relational contracts as friendships, only to the idea that they have relevant similarities that can help us understand each concept better. 164 See generally Lloyd L. Weinreb, Legal Reason: The Use of Analogy in Legal Argument (2005). Perhaps those who have been urging the relational-contract-as-marriage analogy would respond by saying, “We weren’t expecting you to take us so seriously! It is just an analogy for goodness’ sake.” Of course, that is a viable retort; my sur-reply would focus on why friendship is the better analogy, essentially Part III’s argument.

To be sure, many relational contracts are built upon the edifice of a friendship, and the pre-existing friendship itself is the predicate that creates the relational character to the contract. Consider Eric Posner’s observation: “[W]hat appears to be an arm’s length contract between two anonymous firms is often the result of negotiations between two friends who belong to the same social club or sit on the board of the same charitable organization. . . . Contracting parties are often friends.” 165 See Posner, supra note 5, at 150–51; accord Posner, supra note 50, at 757 (“Business is almost always conducted in a highly social manner.”). But see id. at 761–62 (recognizing that these business friendships are not “real” friendships); Moore, supra note 122, at 727 (calling business friendships “fictive”); Macaulay et al., supra note 23, at 230 n.3 (“A lawyer who deals with contracts and fails to understand the power and the limits of trust and the social sanctions flowing from ‘fictive friendships’ is incompetent.”). And sometimes friendship is not the predicate for the relational contract but is a result of the relationship triggered under the contract. Posner further observes that “[f]riendships arise not as the natural byproduct of time spent together and mutual interest; on the contrary, parties spend a great deal of effort, time, and money trying to make friends” to further the business arrangement. 166Posner, supra note 50, at 756–57.

These overlaps between friendships (whether “fictive” or “real”) 167See supra note 165. and relational contracts are interesting in their own right and are often overlooked by commentators and courts. 168However, at least some business school professors have not missed the boat here. See Ingram & Roberts, supra note 10, at 387; Ingram & Zou, supra note 10, at 167; DiMaggio & Louch, supra note 10, at 619; Perrone et al., supra note 10, at 422; Uzzi, Financial Capital, supra note 10, at 481; Uzzi, Sources and Consequences, supra note 10, at 674. The arguments offered thus far might very well help guide thinking about how to handle such overlaps. However, some parties to relational contracts do not have pre-existing or consequent friendships attendant to their contractual relationship. Yet I still hope the perspectives offered here advance our thinking about contract theory and doctrine. In Part IV, I hope to furnish some of the intellectual payoffs of thinking about the world of relational contracting as similar to friendships—both for relational contract theory and for the law of contracts. But first some words of caution.

Nothing that I have said thus far should prevent us from concluding that friendships can encompass more than contracts and that relational contracts can be more or less than full-scale friendships. Of course, parties in most relational contracts are motivated by financial gain in a way most parties to friendships are not, at least in the paradigmatic case of friendship. And that is probably the best reason there is to keep these concepts in separate spheres, if one is so inclined. 169Indeed, this is the strategy of Ellman, supra note 37, discussed in Wightman, supra note 4, at 105–06. Ellman concludes that the contract concept cannot be applied between intimates (his particular concern is co-habiting lovers, not friends) because contracts are about financial gain, but personal relationships are not. Or, put another way, in intimate relationships, the relationship is itself the goal, so there are no external standards to apply to the relationship when a dispute arises. From the perspective of relational contract theory (which Ellman simply dismisses with reference to Eisenberg’s inadequate critiques), this is much too simplified a story. Yet, I have endeavored here to complicate the “separate spheres” story, 170For more on this story and reasons to reject it in its simplified form, see Zelizer, supra note 22. showing some of the transactional underpinnings of friendships and some of the friendship-like underpinnings of relational contracts, which may appear at first to be mere commercial enterprises. Relational contract theory itself encourages us to stop fetishizing the separation of contracts and social life—and I hope to have shown here how it can be useful and illuminating to both contracts and “private” social life to understand how these categories do not so comfortably inhabit separate spheres. We can continue to find ways to preserve something unique and special about each sphere. However, I fear we cannot do so by pretending that these spheres do not interact in absolutely central ways and by creating artificial distinctions between contracts as the jurisdiction of the law and social friendship as existing somewhere always and necessarily outside the law.

Certainly, just as there are very close friendships that do not fit the contract model (but do not destroy the usefulness of thinking of friendship as a relational contract), there are types of contracts that do not fit the friendship model. But this turns out not to be a problem for a relational contract theory that recognizes that many contracts will have very few, if any, relational characteristics. The less a contractual relationship resembles a friendship in the way I have sketched it here, the less reason anyone has to treat that relationship the way one would treat a relational contract.

Finally, I must say something about the seed of doubt I planted at the end of Part II. If I am serious about the morphological similarities between friendships and relational contracts, it should give me some pause that relational contracts are often between organizations and between natural persons and organizations, while the moral seriousness of friendship makes sense only between natural persons. Although I might try to rebuff such a critique by emphasizing that even friendships between organizations are routinely the product of friendships between natural persons—and, after all, organizations are just collectives of natural persons who are perfectly capable of the moral seriousness of friendship—my instinct is not to push this analogical and morphological project too far. As Daniel Markovits suggests in connection with the limited reach of his own “collaborative” view of contract, which similarly cannot be easily extended to cover organizations 171Indeed, in my own analysis of Markovits’s theory of contract, I focus on this shortcoming to his view. See generally Leib, supra note 29 (elaborating on why Markovits’s theory of contract does not have a neat application to very core cases of contracts that involve organizations).   :

Individual persons have reasons to respect and seek community with each other because individual persons’ moral status commands that they be treated never merely as means but always as ends in themselves. Organizations, by contrast, have no comparable moral status, even when they are treated, artificially, as persons at law. Quite to the contrary, organizations should be treated precisely as mere means, and someone who treats an organization as an end in itself makes, at least presumptively, a moral error. 172Markovits, supra note 19, at 1465.

Mutatis mutandis, thinking about relational contracts as friendships may trap me into a similar moral error. Friendship is a special case of partiality, where we are assumed to treat our friends as ends in themselves; an economic organization should be treated only as a means. Although we can have an emotional vulnerability to an organization and value it—alumni contributions to educational institutions seem to reflect this relationship—it would be very bizarre to call our ongoing relationships with these organizations friendships. Certainly, firms as entities barely can be thought to have emotions at all. To say they do is not as much a moral error as it is just bewildering.

Although I concede that this difference between “types” of contracts is important in developing a general theory of contract that applies to all contracts (though it may be a hybrid theory that actually treats organizations under a different set of principles within contract law), 173This idea of a “second” contract law (or a “third,” for all three types of contracts to which I alluded above) is suggested as a possibility by Markovits. See generally id. (examining whether organizations and individuals require their own type of contract law). It is even developed beyond mere suggestion in Meir Dan-Cohen, Rights, Persons, and Organizations: A Legal Theory for Bureaucratic Society 82–84, 96–102 (1986). For my discussion about this potential approach, see Leib, supra note 29, at 11–12. I am a little less worried about the distinction for my much more modest ambitions here. After all, I am not building a general theory of contract. I am simply observing some counterintuitive similarities and overlaps between the concepts, suggesting that these analogies and overlaps bear some useful insights about contract law and theory. Since I do not focus on the value of the practice of contract in friendship (as Markovits does in respectful communities of collaboration), I do not need to worry that many core cases of contract will not share in the moral value of friendships (the way many core cases of contract cannot share in the moral value of respectful communities of collaboration). For my purposes, a comparison of relational contracts to friendships that proves useful in analyzing the practice of contract is sufficient.

IV. Friendship, Legality, Formality, and Social Norms

In the final analysis, an analogical argument is only as powerful as what it can illuminate. One can point out thirty-five similarities between two concepts, but if there are no similarities that illuminate important questions, it is hard to justify writing an article about those comparisons. Revealing the analogies between relational contracts and friendship does, however, shed some new light on old problems.

A. The Legal Enforceability of Friendship?

As counterintuitive as it may seem, viewing relational contracts as friendships (and friendships as relational contracts) supports the idea that some duties that emerge from within friendship are legally enforceable. Friends can generate legal obligations through what may seem to be acts of friendship and private promises. Because viewing relational contracts as friendships and friendships as relational contracts reveals the substantial verisimilitude and overlapping of the concepts, we cannot neatly reserve friendship for outside the boundaries of law. This does not mean, however, that all such duties are legally enforceable. Figuring out which promises or duties are enforceable is no easy task for the law, but the law cannot and does not embrace an oversimplified exclusion of all legal enforceability between friends.

Let us consider several examples. The easiest case for enforceability seems to be a deal between good friends that meets all of contract law’s classical requirements: offer, acceptance, consideration, and the like. Very few would suggest that the law must stay out of friendship completely to protect friends from the law. It would certainly surprise many friends in the business world who have transacted based on the assumption that law would enforce their transactions, only to learn that the law will leave them to their social norms because the law is supposed to stay out of friendships. Even if the underlying friendship is explicit and serves as the trust predicate for the deal itself (that is, without the friendship, no deal would have been reached in the first place), legal enforceability is not likely to be denied. Recognizing that many relational contracts overlap with friendships helps us see the necessity of some degree of enforceability, even in relational contracts that are purportedly more reliant on social norms than the law. That the parties see their obligations as relying on friendship at formation is no bar. If the friends meet legal requirements, the law will enforce their agreements, regardless of what the parties themselves think is the source of their obligations to each other. We might say that through the use of formalities, friends “choose” the apparatus of law, but that is not necessarily true. Usually, nothing about the threshold question of enforceability in the standard business deal with formalities depends on why friends are motivated to deal with one another.

But change the example slightly. Imagine two friends transacting with all of contract law’s requirements, but one says to the other prior to formation: “Buddy, I need you to sign this form contract because my boss requires it. The form claims to be our total deal on all points, but you know full well that I’m going to ignore those provisions that don’t really sum up the deal we’ve been talking about. We’re old friends and I’m not going to screw you.” Formal application of the parol evidence rule would render the side deal unenforceable; the writing explicitly states that it controls. Yet, from the perspective of relational contract theory, at least some of the internal norms of the relationship should be given legal effect. Accordingly, the real deal rather than the paper deal should prevail; friendship’s duty of good faith becomes enforceable against a form contract. 174This approach was taken in Bates v. Southgate, 31 N.E.2d 551, 558 (Mass. 1941), where the court found that an underlying friendship between contracting parties may excuse a party from carefully investigating a form contract for exculpatory clauses because of the good will of a friendship, and Estes v. Magee, 109 P.2d 631, 633–34 (Idaho 1940), where the court found that a doctor’s release form was inapplicable against a patient who signed only because he relied on his doctor’s friendship. A more general “right to rely” on a friend that can be enforced as a matter of contract law is supported in Spiess v. Brandt, 41 N.W.2d 561, 566–67 (Minn. 1950); Bank Leumi Trust Co. v. Luckey Platt Ctr. Assocs., 665 N.Y.S.2d 976, 978 (N.Y. App. Div. 1997); Callahan v. Callahan, 514 N.Y.S.2d 819, 821–22 (N.Y. App. Div. 1987); Liebergesell v. Evans, 613 P.2d 1170, 1176 (Wash. 1980); and Graff v. Geisel, 234 P.2d 884, 890 (Wash. 1951). So it turns out that my twin models here are actually consistent with law as applied in a range of cases. Quite generally, the more relational a contract—the more it involves or is like a “real” friendship—the more we are justified in giving legal effect to friendship’s duties of good faith, best efforts, and reasonable adjustments, and the more the failure to abide by these standards might constitute an independent breach. Rigorous application of the law’s formal requirements in this class of cases seems less appropriate because of the underlying relationship.

Or consider a friend who gives her counterpart a good deal on renting a house she owns. 175Thanks to Ira Ellman for his suggestions regarding this hypothetical. Because of the relationship, the owner charges her friend a much cheaper rent than could be had on the open market. Because of the friendship and a desire to avoid awkwardness, the owner does not bother with a formal lease agreement. Still, the friends exchange a few e-mails with basic terms about price, duration, and general expectations. Imagine that, although the e-mails are sufficient to meet the statute of frauds requirements, nothing is said in the e-mails about subletting, and the default rule in the jurisdiction is that all leases are freely alienable. 176 See, e.g., Cal. Civ. Code §§ 1995.210–.270 (West 2009). Can the renter sublease the property to a third party at market rate and keep the windfall? If the owner sues, what should the result be?

From one perspective, a court might say that the default rule should prevail—one of the risks of dealing as friends is that one can be taken advantage of. That, after all, is the fragility and risk of friendship. Without that fragility and risk, there may not be a way to forge a friendship. Indeed, friendship norms require forgiveness, 177 See generally Kim Atkins, Friendship, Trust, and Forgiveness, 29 Philosophia 111, 111 (2002) (arguing for a conception of forgiveness as the outcome of a process of mutuality rather than something brought about solely by the actions of an individual). and the owner’s forgiveness might be imputed by a court stuck with the task of intervening in the relationship (or fashioning a remedy upon a rupture). But from another perspective, the renter is clearly being unjustly enriched and is taking advantage of the friendship. Why should a contract law designed to be true to the real deal between the parties not police this opportunistic conduct of the renter?

Notice that whichever perspective one takes here, however, it is ultimately relational—and either approach vindicates a relational approach to both friendship and contract law. Does this mean that “relational contract as friendship” and “friendship as relational contract” produces no answer about what a court should do since it could do either and still be relationalist? I do not think so; that both considerations should be within the sensitivity of the court does not put the court in equipoise. Rather, more detail about the motivations of the owner’s “good deal,” the renter’s opportunism, and the dealings between them could ultimately decide the dispute. If the renter had to go to another state to take care of an ailing family member or friend and needed the sublease’s windfall to cover costs in the other state, it might make more sense to require the owner to take the loss and adhere to the default rule. By contrast, if it is a simple case of opportunism and preying on a friend, it is much easier within contract law to force the renter to disgorge or share the ill-gotten profits.

What about the realm of private promise, favor, gift, or duty, when friends do not seem to be self-consciously entering the legal or economic sphere as perhaps they did in the previous cases? Here too, the twin perspectives of relational contracts as friendships and friendships as relational contracts prove illuminating—and recommend legal enforceability as a matter of contract law some of the time. But that it recommends enforceability some of the time is itself an innovation given the propensity of the scholarship surrounding “gratuitous promises” to counsel for a general policy of unenforceability in this class of cases. 178 See, e.g., Melvin Eisenberg, The World of Contract and the World of Gift, 85 Cal. L. Rev. 821, 847–49 (1997).

When I casually promise a friend to pick her up from an airport or promise a friend to take in her newspaper when she is away, my friend and I rarely contemplate legal enforceability. Indeed, one would think a friend absurd, or not much of a friend, for demanding legal enforcement of such favors; most would probably think a judge ridiculous for granting it. But one of the lessons of relational contract theory more generally is that parties to relational contracts will only rarely contemplate legal enforceability themselves. 179 See, e.g., Keyes & Burns, supra note 161, at 587 (“Relational researchers have demonstrated that in business relationships . . . ‘cooperation without reference to legal entitlements is normal.’” (quoting Hedley, supra note 159, at 396)). Even the modern Restatement of Contracts makes clear that “[n]either real nor apparent intention that a promise be legally binding is essential to the formation of a contract.” 180 Restatement (Second) of Contracts § 21 (1981). Thus, something other than mere contemplation upon formation must serve as the relevant test for legal enforceability. Although such a test may be sufficient to exclude some very private cases of friends doing small favors for one another, 181 See Macneil, RCT: C & Q, supra note 16, at 902 n.96 (suggesting that Eisenberg has in mind a concern about the legal enforcement of “tacit household patterns about who takes out the garbage”). Whether one needs a clear test to carve out these trivial cases is a matter of debate. The perspective of relational contract as friendship supports Feinman’s response on this front that not all social bonds are legally enforceable, but we can only find the enforceable ones if we are willing to look at the bonds and duties in the first place. See Feinman, supra note 15, at 748. it is an inadequate test for contracts more generally. Relational contract theory teaches that many business deals recognized by the law might not satisfy an “intent to form legal relations” test. Even if we adopt the relationalists’ preference for the incorporation of social norms by the law, the law here would have to recognize that the social norm itself counsels against legal enforceability some of the time—but not all the time.

Does the analysis change when we enter the financial realm? Can a “commercial purpose” or “economic activity” test work? Perhaps. Think about a simple informal loan to a friend. Suppose a friend loans you $25,000 because you loaned her about that much when she was in financial trouble. 182Assume, of course, that you have ignored Shakespeare’s (or Antonio’s) advice:If thou wilt lend this money, lend it notAs to thy friends; for when did friendship takeA breed for barren metal of his friend?But lend it rather to thine enemy,Who, if he break, thou mayst with better faceExact the penalty.William Shakespeare, The Merchant of Venice act 1, sc. 3, ll. 133–38 . No formal papers are drawn up, and she would never have loaned that much to a stranger without collateral of some kind. Can she enforce the debt when you delay repayment beyond a reasonable amount of time? Or was the loan a mere “act of friendship” that should be shielded from legal enforceability and treated as a gift? The perspectives of friendship as relational contract and relational contract as friendship help us see this outlay as part of an overall exchange relationship, allowing parts of contract law to make it enforceable. To be sure, you might have longer than usual to repay the loan 183 See Scott & Scott, supra note 6, at 1266 (arguing that friendship allows longer time horizons for debt repayment). —and might even get a favorable “friendship” interest rate (if you had not already agreed on one). Perhaps you would also get an easy discharge in a bankruptcy situation. But, most generally, it would be quite hard to argue that the obligation to repay the loan should not be enforceable at all. 184Consider Virgin Money’s recent effort to develop financial products to structure loans to friends. See Virgin Money, http://www.virginmoneyus.com (last visited Sept. 21, 2009). It calls some of its services “Handshake Basic” and “Handshake Plus.” The former comes “smiles included,” and the latter “lowers chance of default” and “raises chance of glee.” Virgin Money, Social Loans Borrowers, Lenders, and Curious Inquirers, Welcome, http://www.virginmoneyus.com/PersonalLoans/tabid/54/Default.aspx (last visited Sept. 21, 2009). Virgin Money’s motto is “to provide customers . . . a top-notch private loan experience that does good things for both [the] pocket and [the] relationship.” Washington Diamond, Virgin Money, http://www.washingtondiamond.com/Financing-Options_2 (last visited Nov. 24, 2009). It appears that Virgin bought Circle Lending, which was a company exclusively devoted to these sorts of financial products, in which the default rate is unusually high (perhaps because people assume their friends won’t sue them on the default). For a story on Circle Lending (and the high default rate among friends), see Avoiding the Pitfalls of Family Borrowing (NPR radio broadcast Oct. 6, 2006), available at http://www.npr.org/templates/story/story.php?storyId=6208227&sc=emaf.

Ultimately, there is reason for equivocation about the realm of the financial, highlighting that friendship duties do not become enforceable merely by entering the financial sphere; it is not the case that anything that counts as a commercial activity would qualify for enforcement. 185The proposal that contract law should enforce any promise with a commercial purpose or in furtherance of economic activity is usually attributed to Daniel A. Farber & John H. Matheson, Beyond Promissory Estoppel: Contract Law and the “Invisible Handshake, 52 U. Chi. L. Rev. 903, 905, 937–38 (1985). Farber and Matheson acknowledge that their argument might support extending enforceability into intimate contexts. When friends buy each other rounds of drinks at a local bar for several years and then one skips out on his responsibility before leaving town for good, does the stiffed friend have a legal cause of action? There is undoubtedly a commercial dimension to the relationship, but it seems reasonable here to think a friend odd for considering such a lawsuit. Yet why is the friend odd? Not quite because a duty never developed in the first place; on the contrary, the reciprocation of the gifts over years can be seen as an implicit promise, leading to reasonable reliance for reciprocation. But we would still think the friend crazy mostly because the lawsuit would cost more than the drink. The law should not be used to enforce this duty precisely because the law is designed to be too expensive a remedy for such minor derelictions of duties. 186Indeed, this reality serves as a useful counterpoint to one of the arguments—given expression in Balfour—offered by non-relationalists against the legal incorporation of social obligations: that the floodgates of litigation will be forced wide open, burdening the courts. Given the cost of litigation, this just is not much of a serious concern and helps reinforce the idea that private promises will tend to stay private whatever position the law takes about enforceability. See Keyes & Burns, supra note 161, at 585 (“Existing procedural mechanisms and costs . . . deter individuals from litigating unworthy claims.”). It is not that we can rely on a simple test of what the parties contemplated (though perhaps here parties did contemplate the risk of being stiffed; that is also constitutive of friendship). Nor can we rely on a test of commercial or financial activity, for that test is met here, but no legal enforceability should follow. 187What counts as “commerce,” in any case, has long been a puzzle for the law. The Commerce Clause jurisprudence of the Supreme Court is as good a proof as any that definitions of commerce are ever-changing and quite hard to pin down. See U.S. Const. art. I, § 8, cl. 3; United States v. Lopez, 514 U.S. 549 (1995).

But when we up the ante in the reciprocal gift exchange (or implied promise) context, it begins to seem harder to deny enforceability. If every year friends took turns buying each other season tickets for the local ballet company—an expense of several thousands of dollars—a stiffed friend might well pursue legal sanctions without being deemed absurd. 188Some evidence that litigation over season tickets is not unusual can be found in the following line of cases: Lipton v. Donnenfeld, 773 N.Y.S.2d 82 (N.Y. App. Div. 2004); Brand v. Lipton, 711 N.Y.S.2d 486 (N.Y. App. Div. 2000); Kane v. Berken, 600 N.W.2d 55 (Wis. Ct. App. 1999) (unreported table decision); Kully v. Goldman, 305 N.W.2d 800 (Neb. 1981); Davey v. King, 595 A.2d 999 (D.C. 1991); and Tauber v. Jacobson, 293 A.2d 861 (D.C. 1972). Thanks to Joe Perillo for triggering my interest in this litigation context. What is more, a court would have good cause to force a friend to make good on an implied promise to provide such season tickets so long as there was an inequity without enforcement. 189There is a certain parallel here with the line of cases that prevents employers from easily firing salespeople without paying them earned commissions, finding such termination practices breaches of the implied duty of good faith. See, e.g., Fortune v. Nat’l Cash Register Co., 364 N.E.2d 1251 (Mass. 1977); Caton v. Leach Corp., 896 F.2d 939, 946 (5th Cir. 1990). This line of cases is cited and discussed in Steven J. Burton & Eric G. Andersen, Contractual Good Faith: Formation, Performance, Breach, Enforcement §§ 3.2.2, 3.4.1 (1995). If the stiffed friend had already received as many tickets from his friend as he had purchased for his friend, no cause of action should lie. The court’s job is not to keep the friendship going—that is, after all, quite clearly the realm of the non-legal social sanction; 190I will discuss this limitation to the enforcement of friendship and relational contracts more generally infra Part IV.B, since this claim contravenes some relationalists’ normative conclusions about how to handle relational contracts. the court’s role is only to prevent and remediate serious derelictions of duties that result in substantial injury. What this example reveals is that an “intention to form legal relations” and tests for enforceability having to do with “commercial activity” only get us so far when it comes to determining the legal enforceability of acts and promises predicated on friendship; context and common sense must prevail, as relationalists generally urge.

Consider a slightly different example, revisiting the area of “noncommercial” private promise. My good friend calls me from the airport to let me know she is in town. She needs to get downtown quickly for an important meeting, and I tell her that I will be there in fifteen minutes to drive her. She tells me that if I cannot get there within twenty minutes, she will take public transportation. Although she would like to see me, her being on time is “of the essence” because the meeting is with an important prospective client. She relies on my insistence that I can help and she waits me out. She misses the meeting, it turns out, which renders the trip a total loss. She also loses the potential to develop a working relationship with the company she was supposed to meet. Am I legally obligated to make her whole?

From the perspective of those who wish to keep all private duties unenforceable, this is an easy case. If my friend sues me, we might very well think her odd. But it can also be analogized to the season ticket case because there might be enough money at stake to make a lawsuit financially worthwhile. Yet, it is hard to be sure what is appropriate for the law to do without knowing more. Did I just flake out? Was there unforeseeable traffic? Did I have an emergency with my child? Did I deliberately make her late because I am actually competing with her business and wanted the same client she was going to meet downtown?

In some of these scenarios—like the child emergency or unforeseeable traffic—it is plausible to say that a hypothetical lawsuit should be dismissed. Indeed, a relational contract theory that urges courts to apply internal social norms to the relationship would likely find that within friendship there is an implicit right to veer from strict compliance with obligations for a range of good faith reasons. 191 See Fiske & Tetlock, supra note 116, at 280 (suggesting that some obligations in the social sphere carry with them an assumption that they can be deferred or canceled upon certain changes of circumstance). But the closer I get to being negligent or intentionally malfeasant, the less it makes sense to refuse legal enforcement of the underlying duty. In some measure, our pre-existing friendship serves as the source of the norms that determine whether I owe my friend any remediation. If I acted in good faith, that is all our friendship required. If I violated my good faith obligation to her, it is easier to say that my opportunism and betrayal is legally cognizable.

One could reasonably ask, however, whether the duties discussed here, when enforced, are contractual duties. Isn’t the offer to pick up my friend a gift? What is the friend’s consideration? 192The consideration doctrine would indeed prevent many promises within a friendship from becoming legally enforceable. See, e.g., Kim v. Son, No. G039818, 2009 WL 597232 (Cal. Ct. App. Mar. 9, 2009) (finding a promise literally written in blood between friends to repay money lost in a bad investment unenforceable because the promise wasn’t bargained-for). But it isn’t clear that the consideration doctrine ought to be used this way—especially in cases where it seems easy to say that the evidentiary, channeling, and cautionary functions of the doctrine have been met. See generally Lon L. Fuller, Consideration and Form, 41 Colum. L. Rev. 799 (1941) (discussing the rationale of legal formalities and an examination of the policies of the common law consideration doctrine). What is the bargain? 193 See Ellman, supra note 37, at 1375 (“Mutual gifting arising from mutual concern and affection is not the same as bargained-for exchange.”). Is it her forbearance from taking public transportation or a cab from the airport? That seems thin, though perhaps it constitutes sufficient reliance for courts to apply a “promissory estoppel” theory to get around the consideration requirement. 194 See generally Restatement (Second) of Contracts § 90 (1981). Notably, in developing a remedy in this sort of context, we probably cannot be more precise than § 90(1) already is: “[t]he remedy granted for breach may be limited as justice requires.” After all, the reliance evidenced by not taking alternative transportation is exactly what would be expected to flow from the promise to make a gift (if it is a gift) in that context.

Relational contract as friendship and friendship as relational contract provide an alternative answer, which does not require hemming such scenarios into doctrines that are not designed to do the work that they would have to bear in cases like the airport example or the season ticket hypothetical. These examples highlight that in practices of reciprocal gift-giving, upon which friendship is predicated, the relationship itself can generate duties that can be recognized through contract. This is not to say that all promises to make gifts become enforceable between friends or relational contracting partners. Nor is the occasional enforceability of friendship’s duties between friends predicated on mere reliance or restitution, though (reasonable) reliance and unjust enrichment may be preconditions for enforceability too. Rather, some duties stem from the relationship itself and only make sense in that context. Under a relational contract theory, however, nothing prevents those “status-based” duties from being thought of and captured by contract. 195This argument draws from Wightman, supra note 4, at 125 (discussing non-bargain contract cases and limiting them to intimate associations in which gift giving is reciprocal). One of the benefits of relational contract theory (to those who embrace it) is precisely its willingness to admit promissory, reliance, restitutionary, and status-oriented modalities of obligation into its overall structure. To be sure, a relational contract theory will account for the possibility that some “socially valuable” gifts “derive[] [their] value from [their] role in nonlegal relationships, and therefore efforts to regulate it with the law [c]ould reduce its value.” 196Posner, supra note 78, at 567. However, not all gifts work this way, and a pattern of reciprocal gift giving is a signal that exchange can be enforced without degrading its value. 197For a related argument about the role of reciprocity in the enforcement of intimate obligations in the will context, see Melanie B. Leslie, Enforcing Family Promises: Reliance, Reciprocity, and Relational Contract, 77 N.C. L. Rev. 551 (1999). To be sure, we cannot lose sight of the different ways in which gifts are used in different relationships and how their value changes depending on the context. Frank Flynn’s work pursues this nuance, demonstrating that we should not treat all gift exchanges the same way, even within friendships. See, e.g., Ames, Flynn & Weber, supra note 77 (explaining that gift givers may have different reasons for giving gifts such as affect, role, or cost–benefit calculations). As I suggested above, even “traditional” contract law can reach this result through what is known as promissory estoppel—but relational contract theory puts this type of liability as a central rather than marginal case. It also provides a framework of reasons for enforcement and a careful method for weighing the equities, even if courts want to squeeze these sorts of cases into the promissory estoppel box, as might be expected.

Take one final example. Consider the practice of some groups of friends who contribute to a pool of funds as a savings device or as a financial instrument. Often organized within ethnic groups (of friends), the rotating credit group is a way to incentivize and enforce savings in tight-knit communities. 198The practice is described and discussed in Posner, supra note 5, at 151, and in Eric A. Posner, The Regulation of Groups: The Influence of Legal and Nonlegal Sanctions on Collective Action, 63 U. Chi. L. Rev. 133 (1996). It works like this:

[A] rotating credit group typically consists of a small number of people . . . who periodically contribute money to a pot. At the beginning of each period, one member takes the pot. Members determine the recipient by lottery or bidding. Failure to make timely payments and other breaches result in nonlegal sanctions such as criticism that, carried along the channels of gossip, injures the defaulter’s reputation and may lead to social ostracism. When everyone has taken one pot, the group dissolves.

. . . [P]articipation in a rotating credit group either reduces the time necessary to save up to buy an indivisible good or earns interest. 199Posner, supra note 197, at 169–70. For a classic analysis of these practices, see Clifford Geertz, The Rotating Credit Association: A “Middle Rung” in Development, 10 Econ. Dev. & Cultural Change 241 (1962).

Some of these groups may formalize their deals, making this look like the first, easy case of clear enforceability (when the groups do not otherwise run afoul of securities, tax, lottery, or other laws, which they frequently do). But imagine a scenario in which no deal was formalized, as is often the case. If friends engage in this practice for a long time without formalities, can contract law infer a legally enforceable deal? 200One bankruptcy court has held that obligations under a rotating credit agreement are legally enforceable. In re Ray, 51 B.R. 454, 460–61 (Bankr. D. Haw. 1985). Despite the fact that these agreements are quite common in the United States, there is remarkably little reported litigation to guide courts when confronting such obligations. Perhaps this is evidence that non-legal sanctions are functioning effectively. But perhaps not; fear of legal sanctions or the legal system may also be keeping these disputes out of court. Or, because informal, non-legal sanctions are what the parties really contemplate for enforcement, should the law leave the parties as it finds them upon a party’s breach of the underlying deal? 201For analysis of one such case, see Posner, supra note 198, at 173–75 (highlighting a case in which a court opted to dismiss a claim trying to enforce a rotating credit agreement on the ground that it was an illegal lottery). More generally, though, it is probably hard to say with any degree of certainty that parties contemplate only non-legal sanctions upon default. See id. at 170 (“[A]mong West Indian immigrants in New York City [participating in credit circles], ‘so rare was default that when queried, most organizers did not know how they would have dealt with such a situation.’” (quoting Aubrey W. Bonnett, Institutional Adaptation of West Indian Immigrants to America: An Analysis of Rotating Credit Associations 63–64 (1981))). That the relational-contract-as-friendship and friendship-as-relational-contract perspectives answer with a weak “it depends” should not be surprising, since any court faced with such a case would need to conduct a careful study of the internal norms and expectations of the participants in the credit group.

To say, as the relational-contract-as-friendship perspective would, that some relational duties are legally enforceable does not mean that all are, and courts need some way to figure out which are and which are not. This is where relational contract theory is weakest, 202 See Barnett, supra note 8, at 1190–91 (arguing that what differentiates a contract theory from a social theory is an account of when an exchange relationship merits legal protection). but the relational-contract-as-friendship model offers slightly more illumination. Utilizing an “intent to form legal relations test” as courts have in other contexts, 203 See Balfour v. Balfour, (1919) 2 Eng. Rep. 571, 579 (K.B.) (Atkin, L.J.) (discussing the intent of a husband and wife to form a binding contract where the husband merely promised to give his wife a monthly stipend and observing that “the promise here was not intended by either party to be attended by legal consequences”). or a “commercial purpose test” as suggested by Farber and Matheson, 204 See Farber & Matheson, supra note 185, at 904–05 (“[A]ny promise made in furtherance of an economic activity is enforceable.”). will too often leave friends without recourse when they find themselves betrayed and will too often ensnare friends into legal obligations they never wanted in the first place. Yet some inquiry into the intent of the underlying deal (if one is ascertainable), an estimate of the nature and extent of the breach, and the good faith or opportunism of the breaching party should guide the exercise of common sense. Promise and reliance play their part, but a fine-grained commonsensical analysis of the status of the parties and the practice in which they are engaged is also necessary. Courts may not be perfect at the task and perhaps should try to shuttle these cases to mediators, arbitrators, and agencies where possible. 205 See Macneil, RCT: C & Q, supra note 16, at 906. Many disputes between friends probably find themselves in small claims court, where mediation and arbitration are quite common dispute resolution techniques. But there is no question that the law has jurisdiction and that the courthouse doors are open for friends to sue one another. Nevertheless, courts are not so obviously error-prone, nor are friendship duties so clearly isolated in the social sphere, that courts should refuse enforcement altogether by drawing a line in the sand.

Contractual obligations emerge from a multiplicity of places, whether from consent, promise, reliance, benefits-conferred, or status-based sources. 206Indeed, although a fuller discussion is beyond the scope of this Article, it suffices to say that viewing relational contract as a friendship and friendship as a relational contract may help us see why “contract thinking” was not a “fatal flaw” at all in the cohabitation cases like Marvin v. Marvin, 134 Cal. Rptr. 815 (Cal. 1976). Contra Ellman, supra note 37, at 1365, 1367. Ellman criticizes cases like Marvin for failing to apply more status-oriented obligations to generate the liabilities the courts find in cohabitation cases through implied contracts. Id. at 1375–77. Through the lens of relational contract theory—especially the form developed here—it is easy to see how the theory incorporates precisely the type of status-based liabilities Ellman prefers. His conception of contract is so narrow that he fails to see how capaciously one can plausibly and convincingly draw “contract thinking.” The twin models here (friendship as relational contract and relational contract as friendship) help us see this in salient relief—and further support those who would build a contract theory around several loci of obligation. In any given context, “contract law” may prefer to focus liability on one of these sources of obligation, but there is no singular organizing principle that accounts for all contracts. We do contract law no great service by building the edifice of contract around only one of these sources of liability, suggesting that the others are merely exceptional cases. Relational contract theory highlights just how misleading such models are. 207It is probably worth noting here that friendship can play numerous roles in establishing liabilities. The friendship can serve an evidentiary function (friendship proves there is a best efforts clause, which does not otherwise appear in the paper deal); it can sometimes create the obligation itself through implication and conduct (the friendship underwrites a finding of breach because friends would not rely upon or defend their actions on the basis of caveat emptor); and it can sometimes be relevant in fashioning a remedy (because friends betrayed one another, a specific performance order is especially unlikely to result in successful repair).

B. Revisiting Relational Contract Law

The window into relational contract theory developed here supports many of the theory’s normative prescriptions for standard relational contracting in business. If relational contracts are like friendships insofar as they both leave many features of deals unspecified, and they both breed interdependence, it might make sense to expect a higher degree of good faith, best efforts, and reasonable adjustments between parties to such contracts, as we would with all friendships. 208There is some circularity here, of course. I used the requirements of good faith in both friendship and relational contracts as a reason to adopt the analogy in the first place, so it is a form of cheating to use the analogy on the back end as a reason to embrace the good faith requirement in relational contracting. Although I concede this point of circularity, it is worth noting that since the analogy also works on so many other levels (and on levels that create the need for the good faith requirement), it seems fair to say that the twin models ultimately “support” this normative dimension of relational contract law. In any case, when I say the models “support” a normative prescription, I do not necessarily mean that they furnish independent reasons to embrace the normative prescription, but rather that they are consistent with the prescription in ways that help us understand it better as a feature of relational contract theory. Although these dimensions of the relationship will not be legally enforceable all the time (as I just suggested above in the discussion of relational contract of friendship itself), courts should carefully decide when and how to incorporate these social norms into the parties’ legal relationships. Since these dimensions of the relationship will often be predicates for the anticipatory trust in the contractual relationship (and violations of such norms will be reliably opportunistic), refusing to enforce them legally would have the perverse consequence of eroding the possibility for the relationship in the first place.

The discussion here also can support the relationalists’ case to relax rigid requirements of offer, acceptance, definiteness, and consideration within relational contracts. One of the most illuminating features of putting relational contracts and friendships side by side is seeing how absolutely central informality is in the constitution of these relationships. Just as we might degrade or deter friendship by requiring parties to engage in formalities, so too would we risk undermining relational contracts and their internal efficiency by refusing to recognize their need for informality, reciprocal gift giving as exchange, and incompleteness. By requiring formalities to get enforceability, we are clipping the wings of relational contracts. Formalism about formation increases contracting costs so substantially that we may never get these contracts (or friendships) off the ground in the first place. 209 See Ingram & Zou, supra note 10, at 170 (“In a business context, the benefit of relational exchange is to lower transaction costs, and related, to enable some transactions which would be otherwise impossible.”). But see Cross, supra note 22, at 1501 (arguing that sometimes resort to legal formalities and sanctions is necessary to get relational contracts off the ground in a commercial context outside of close-knit groups). Even in cases in which relational contracts can only get off the ground with some resort to legal formalities, there is no reason to require them (as non-relationalists would) when parties otherwise do not want them because, presumptively, the parties view formalities as undermining the relationship. The discussion here only further supports the arguments of those relationalists who argue against adherence to contract’s “classical” restrictions in the case of relational contracts. “Planning for litigation [through formalities] becomes a self-fulfilling prophecy.” 210Larry E. Ribstein, Are Partners Fiduciaries?, 2005 U. Ill. L. Rev. 209, 235. This is, admittedly, an ironic use of Ribstein.

As with formation, so with interpretation. The relational-contract-as-friendship model also supports the relationalists’ commitment to deeply contextual interpretation over formalistic modes of interpretation that would only give effect to a paper deal. When friendship is concerned, there are equitable reasons to enable judges to look behind the paper deals to the real deals. Seeing a relational contract as partaking in the morphology of friendship helps support the case for this equitable peek at the real deal in relational contracts too. Relational contracts, like friendships, are complex relationships that need to keep some obligations incomplete to function properly; those agreements adjust over time. But just because those obligations develop over time without strict formalities does not make them any less obligatory—and does not make their violation any less a breach of the parties’ real deal. Courts need the freedom to enforce those implicit dimensions of contracts to keep those relationships from being abused. Non-legal sanctions help, but they may not be strong enough to deter serious cases of opportunism. Moreover, formalism in interpretation also enhances contracting costs substantially and risks threatening the entire enterprise.

Ultimately, formalism just does not make sense for relational contracts—whether from the coherentist impulse that we ought to have a law that reasonably tracks how people actually do business, the economic standpoint that we ought not force people into contracting costs that they need not undertake (after all, most contracts succeed and parties rarely have to worry about court intervention), the economic standpoint that forcing ex ante bargaining risks rupturing the trust and solidarity the relationship already establishes, or the normative standpoint that the cost of formalism is routine injustice and there is no benefit substantial enough to outweigh it. Even if formalism is possible and desirable in discrete transactions in which there is no friendship or relationship to speak of (or in which such a description is plainly too attenuated to take seriously), such a technique cannot and should not be used to marginalize the real deal that parties enter into and rely on within relational contracts. Thus, the discussion here further underscores relationalists’ lack of sympathy for formalism in relational contract disputes.

Yet, not all of the relationalists’ normative prescriptions withstand scrutiny when viewed from within the models explored here. For example, to the extent that some relationalists have urged that relational contracts should be treated like partnerships and that courts should “adapt or revise the terms of ongoing relational contracts in such a way that an unexpected loss that would otherwise fall on one party will be shared by reducing the other party’s profits,” 211Eisenberg, supra note 21, at 817–18. those prescriptions are supported by the relational-contract-as-marriage model much better than they are by the relational-contract-as-friendship model. Although friendship has a basis in equality, nothing about friendship suggests that people in it are only selfless and other-regarding. I have argued elsewhere that very close friends might be required to serve as fiduciaries for one another, with attendant duties of loyalty or unselfishness. 212 See generally Leib, supra note 22 (arguing that courts should sometimes apply a tailored law of fiduciaries to enforce legal duties between friends). However, the average case of friendship does not require full-scale loyalty and a lack of self-interest. While marriage is generally such a partnership, in which egalitarian sharing is a worthy aspiration, friendship does not present the same case for legally-enforced equality. Accordingly, the model of relational-contract-as-friendship provides a basis for revising this normative dimension of relational contract theory. Friendship certainly envisions the parties acting in good faith, but good faith does not mean partnership in the vast majority of cases.

Furthermore, the relational-contract-as-friendship model suggests a reason to be skeptical about the desire of some relationalists to use the law to keep relationships and relational contracts together. 213One can see this dimension of relational contract theory on display in Speidel, supra note 8, at 839, 846. Although fighting opportunism within relational contracts is a desirable goal from the standpoint of any relational contract theory, the fixation on relationship preservation evident in much relational theory does not resonate as well with the relational-contract-as-friendship model. Indeed, the relationship-preservation element of some relational contract theory may very well be traceable to the relational-contract-as-marriage model, where a relationship-preservation posture is not uncommon. Within friendships we tend to make exit easier than within marriages, and contract law that is animated by an account of relational-contract-as-friendship would likely focus on enabling exit more than preserving a relationship. 214To be sure, Speidel highlights that even using law to preserve the relationship does not really deprive parties of their underlying right to agree to terminate by negotiating their way out of obligation. Id. at 846. But from the perspective of relational contract as friendship, a relational contract law should allow parties to break up easily if that is what they want to do—so long as they are no longer in debt to one another. This does not mean that parties should be free to shirk a vested obligation—like the hypothetical party who gets free season tickets under the assumption that he will pay for the round of season tickets next year but exits the relationship before his turn to pay comes due. But once the vested obligation is discharged, the law should not waste any resources disabling free exit. 215There is some inchoate research in business schools and sociology departments that investigates the deterioration of business friendships (though focused largely on workplace friendships)—and such research may help courts and scholars develop protocols for dealing with contractual breaches in light of deteriorating business friendships. See Ingram & Zou, supra note 10, at 181 (citing Patricia M. Sias et al., Narratives of Workplace Friendship Deterioration, 21 J. Soc. Pers. Relationships 321 (2004); Patricia M. Sias & Tara Perry, Disengaging from Workplace Relationships—A Research Note, 30 Human Comm. Res. 589 (2004)).

C. The Debasement/Crowding Thesis: Social Norms and Legal Norms Playing Nicely Together

The perspectives afforded by the friendship-as-relational-contract and relational-contract-as-friendship models also shed some new light on an old critique of relational contract theory’s normative suggestions. In short, non-relationalists routinely reject relational contract theory’s normative suggestion that the law incorporate social norms because doing so would risk threatening the very foundation and value of the relationships themselves. To give legal effect to social or relational norms debases the social norms and the relationships themselves because they are intended to and always should remain outside the sphere of law. 216For a nice approach to the debasement thesis from within the philosophy of friendship, see Neera K. Badhwar, Friendship and Commercial Societies, 7 Pol. Phil. & Econ. 301 (2008). Although Badhwar is concerned with a somewhat different theoretical concern—that market societies generally debase friendship—her response to it (that market relationships and friendships actually have much in common) is pertinent here.

This longstanding claim of non-relationalists cannot be taken lightly, and it cannot be definitively resolved here. It is, however, a version of the “crowding” thesis 217The “crowding” thesis is explored in, for example, Bruno S. Frey, Not Just for the Money (1997); Iris Bohnet et al., More Order with Less Law: On Contract Enforcement, Trust, and Crowding, 95 Am. Pol. Sci. Rev. 131 (2001); and Larry E. Ribstein, Law v. Trust, 81 B.U. L. Rev. 553, 568–71 (2001). See also sources cited supra note 22. —that law crowds out trust relationships—about which much ink has been spilled. Here is one way of putting the crowding critique: What if legal regulation of the “strong form” 218 See generally Jay B. Barney & Mark H. Hansen, Trustworthiness as a Source of Competitive Advantage, 15 Strategic Mgmt. J. 175 (1994) (developing the concept of “strong form trust”); see also Ribstein, supra note 217, at 558–68. trust in friendships actually served to crowd it out altogether? If law replaces trust by intervening and enforcing social norms, a regime of trust enforcement through incorporation of social norms would actually undermine the very important brand of trust it was seeking to protect by incorporating it into the law. The effort to protect friendships (mostly by punishing bad and false friends) would have the perverse effect of discouraging them.

The models erected here, however, make a modest contribution to the debate about these issues. They support the claims of those who are generally suspicious of the crowding thesis, and they tend to provide reasons to reject the “debasement” critique offered by non-relationalists. In particular, the friendship-as-relational-contract and relational-contract-as-friendship models reveal the supplementary rather than substitutional nature of law to social norms. 219Admittedly, business school professors already seem to know that formal contracts and informal norms can complement rather than substitute for one another. See supra note 64. It is not fully clear why there is virtually no uptake of this work within the legal academy. Because the domains of relational contract and friendship so often overlap and because they are so similarly structured, it is much easier to see that the admixture of these spheres (in actual practical terms and in morphological similarity) disables us from treating trust and intimate relations as divorced from law completely. 220 See generally Zelizer, supra note 22 (undermining the “separate spheres” story). Quite the opposite turns out to be true. Friendship and the trust afforded therein is a routine predicate for entering into legal relations, and legal relations serve to develop trust upon which intimate relations can develop. Accordingly, it is altogether too simple to assume that the law’s occasional incorporation of friendship’s social norms will disable them from doing their work in promoting voluntary cooperation. As the discussions here reveal, voluntary cooperation relies on a complex mix of legal and social norms, and the occasional incorporation of some social norms into legal norms will not serve to undermine them altogether. Indeed, friendships occasionally will need the support of legal institutions to give potential friends and relational contracting partners the social trust necessary to enter these complex relationships at the outset. None of this is to say that legal and social norms do not interact, only that they are not either/or propositions.

Those who would press the crowding thesis further might suggest that legal enforcement of the duties of friendship would deter people from entering into these relationships. 221 But see Scott & Scott, supra note 6, at 1247 (“[C]ontract and commitment are quite compatible.”); id. at 1330 (“[L]egal enforcement of long term commitments is a particularly valuable method of bolstering extralegal norms.”). And we would be much less enriched in a world with fewer friendships. Yet, assuming friendship has both intrinsic value and some instrumental value, friendship provides so many of its own incentives for entry (and it is so easy to exit when the costs grow too great) that it is hard to imagine that occasional legal enforcement of a friendship duty would actually deter people from establishing and developing their friendships. The enforcement will be rare, in any case, because both social norms and legal costs will keep litigants out of court most of the time. It is unusual that a duty has sufficient value to make it worth suing to enforce it in the first place, and given reasonable limitations—in particular, that the law should intervene only upon a substantial default from a social norm in which legal enforcement is not specifically or impliedly negated—it is fair to conclude that the law will do little to prevent people from forming and enjoying friendships. 222This paragraph (and some of what follows) draws upon my discussion of the crowding thesis in Leib, supra note 22, at 726–32. In that work, I also develop an argument about legal enforcement’s “information-forcing” dimension within friendships. We burden family all the time with special legal duties, but this does not tend to crowd out trust in that context either.

A second way to think about the crowding thesis is that legal enforcement of the social norms of trust and friendship would interfere with friendship itself. This is another way of thinking about the “debasement” thesis that is not a “just-so” claim about what friendship needs to retain its value. The “just-so” debasement thesis fails because it is predicated on a conception of legal enforcement of contracts that includes only contracts between strangers and a conception of intimate relations that is altogether too pure or Pollyannaish. 223For a discussion of the assumption underlying the debasement thesis, see Bagchi, supra note 21, at 3–5 and Kimel, supra note 3, at 253–55. As Ribstein puts the refined thesis, “In short, legal coercion does not help develop norms of trustworthiness or of trust. Moreover, . . . regulation impedes development of trust norms by interfering with opportunities to be genuinely trusting or trustworthy.” 224Ribstein, supra note 214, at 567.

But viewing friendship-as-relational-contract and relational-contract-as-friendship further undermines this line of argumentation. The argument seems predicated on the assumption that one can have either strong-form trust that operates outside the law or legally-policed trust, and that in any given context they must be substitutes rather than complements. But that assumption—explicit in Ribstein’s work 225 Id. at 568 (“[L]aw must be regarded as a substitute for rather than complement of social capital because it undermines the institutions that create it.”). —virtually guarantees the crowding conclusion. In any case, the premise is flawed. It simply is not true that “the fact that there is some reliance on the threat of sanctions mean[s] that there will be no room for trust.” 226 Annette C. Baier, Trust and Antitrust, in Moral Prejudices: Essays on Ethics 118, 139 (1994). The discussions here help show why that assumption is false: trust relationships and legal relationships are mutually reinforcing in many cases. Indeed, as Larry Mitchell has argued, “certain preconditions, like a legal system sustaining the values of trust, are necessary for trust to flourish.” 227Lawrence E. Mitchell, Trust. Contract. Process., in Progressive Corporate Law 185, 204 (Lawrence E. Mitchell ed., 1995). Law makes trust possible, in part—even the strong form necessary for open-ended relationships like relational contracts and friendships. Accordingly, it is unlikely that Eric Posner is right that “law only matters when a lawsuit occurs.” 228 Posner, supra note 5, at 63.

A related observation: Those who write about social norms and legal norms routinely fail to recognize that these norms cannot be so neatly divided into separate spheres. Indeed, at least some of the “debasement” thesis requires a strict separation. But consider when we can say that “the law” intervenes. It is often present as a back-stop and only rarely finally disposes of a dispute. Even when a dispute arises that seems to be addressed through non-legal sanctions, an implicit calculation not to use legal options may be in the background. Or what about when a dispute arises and one party tells the other that she thinks she has some legal rights in play? Is a decision by the threatened party to relent and reach a negotiated settlement obviously a legal one, traceable to the use of legal rhetoric? Or do social norms supplement this situation, proving, again, that the social and legal work in tandem all the time? What about when one party sends a lawyer’s letter to the other? Is that a legal sanction, even though the law is not applied directly and coercively? When a dispute is arbitrated after such a missive, can we say for certain which norm accomplished the compliance and settlement? Does a mere whiff of law mentioned in a letter replace all the work social norms do to structure the parties’ negotiations? Or what about after a complaint is filed but before either party spends money on discovery? Is a settlement triggered by the filing of a law suit—rather than a disposition of law itself by a black-robed individual—a clear case of settling merely a legal dispute? Or, were social norms of settlement and reputational concerns also in play? In short, to the extent that non-relationalists fear that social norms will be crowded out once the law has taken an interest in them, they would do well to remember how porous these categories are and how often the two types of norms mix to produce compliance. The mixture is likely to reveal itself early and often in relational contracts and friendships.

There is yet a further reason to be skeptical of the crowding thesis. Crowding is most likely to result—and most likely to serve as a substitute rather than a complement—only when there is an environment of perfect enforcement, in which legal norms perfectly incorporate all social norms. That does not happen for reasons we have already seen (for example, only some social norms are appropriate for the law to enforce; the law is costly; and there is stigma associated with suing our friends and relational contract partners so that we are very unlikely to use the law even when legal options are available) and for reasons I am about to develop. In the legal landscape of very imperfect enforcement and regulation, as in contract law, trust is far from crowded out. Thus, to the extent there is any crowding, law will only make small incursions into our friendships’ strong-form trust; the law will help police only major and costly defections but will not intrude too heavily on social norms, which will continue to function.

Finally, there is, perhaps, a more philosophical rendering of the debasement thesis that can also be shown to be erroneous with the help of the concepts developed here. This form of the argument emphasizes that the very value of friendship’s duties inheres in their unenforceability. Once those duties are backed by sanctions, the normative force and communicative potential of compliance becomes eroded. 229 See generally Melvin A. Eisenberg, The Theory of Contracts, in The Theory of Contract Law 206 (Peter Benson ed., 2001) (arguing for a form of the debasement thesis); Bagchi, supra note 21, at 19 (citing Meir Dan-Cohen, In Defense of Defiance, 23 Phil. & Pub. Aff. 24 (1994)); Eisenberg, supra note 178, at 850. The concern seems to be that friendship itself ceases to be an independent reason for compliance once legal enforceability enters the picture.

But there are several reasons to reject this line of argumentation. First, it is too simplistic to assume that the value of friendship inheres in “voluntary” compliance. People comply with their friendship duties for all sorts of reasons. A sense of obligation in friendship is common, whether that sense comes from, at the extreme, a fear of being sued or, more usually, a fear of a non-legal sanction. In short, although it is true that the type of sanction may have some quantum of effect on the communicative potential of compliance (and I just endeavored to show that the line between social and legal sanctions is probably overdrawn, in any case), sanctions are available even if the law does not show its hand. This means that if we insist that the value of friendship comes from “voluntary” compliance, we are likely to find that friendship is not valuable very often at all. This is not a likely conclusion.

Second, no matter what your contract theory or theory of friendship, many contracts between friends will be enforceable. Relational contract theory has revealed that a great deal of contracting occurs among friends—and virtually all agree that contracts between friends will be legally enforceable. If that is right, any contract that exists between friends will kill the friendship unless we insist on a policy of non-enforcement for all such contracts. Even those who embrace the debasement thesis likely would reject this conclusion, notwithstanding the reality that friends will never know if compliance was achieved through the threat of legal sanctions or simply because of the friendship. 230Bellia makes a similar argument, relying on friendship as the quintessential interpersonal relationship. See, e.g., Bellia, supra note 22, at 36 n.38.

Third, parties to friendships routinely engage in both altruistic and selfish reasoning all the time; friends do not expect anything different and only hope for generosity when it most matters. Friendship as a possible motivation for action is not any less independent when other instrumental factors are considered side by side. Choosing certain instrumental reasons as motivations can risk diluting friendship, to be sure, but the existence of different reasons does not always challenge the realm of friendship. Motivations from friendship that compete with other motivations need not be seen to erode interpersonal relationships at all. Moreover, in light of the under-compensatory nature of possible legal sanctions (which I shall address presently), in most cases there is plenty of room for the “independent reason” of friendship to prove itself as the prime motivation for undertaking compliance.

D. The Under-compensatory Nature of Contractual Remedies

The window into relational contracts and friendships developed here may also shed some light upon the question of why remedies in contract are dramatically under-compensatory. Generally speaking, one cannot recover in a contract suit for emotional distress, for punitive damages, upon “penalty” clauses, for uncertain or speculative damages, for unforeseeable damages, or for attorneys’ fees. Indeed, it is generally quite difficult to get specific performance, which is the performance you are actually owed. 231These are not controversial claims requiring elaborate citation. See generally Posner, supra note 5, at 164 (showcasing the ways that contract law’s remedies are under-compensatory). In short, an aggrieved party to a case is almost never made whole irrespective of the measure of damages awarded. Even an award of specific performance does not really make the aggrieved party whole, since she had to sue to get it and cannot get her attorneys’ fees back. 232For this reason, although I am largely sympathetic to Stephen A. Smith’s effort to see contract damages as still leaving room for parties to act on non-legal reasons, see Stephen A. Smith, Performance, Punishment and the Nature of Contractual Obligation, 60 Mod. L. Rev. 360, 362–63, 368, 370 (1997), I do not think specific performance needs to be seen as an exceptional remedy from this perspective. Specific performance is also under-compensatory, leaving room for non-legal sanctions to do their work.

Multiple theories have, of course, been offered to explain these pervasive policies within contract law from the perspective of classical and neo-classical contract theory. And, indeed, most relationalists concede that the law as it stands is not perfectly explained by relational theory, which self-consciously highlights how much the law is driven by an assumption of contracts between strangers. 233 But see Kimel, supra note 3, at 235–43 (specifying an “empirical–doctrinal” argument from within the relationalist camp, which aims to show that “contract law already recognizes and is significantly informed by the relational nature of contracting”). For additional articles developing this line of argumentation, see for example, David Campbell, The Relational Constitution of the Discrete Contract, in Contract and Economic Organisation 40 (David Campbell & Peter Vincent-Jones eds., 1996) and Campbell & Collins, supra note 55, at 25–32. Nevertheless, the perspective developed here might contribute to understanding this dimension of contract law.

The view of relational contracts as friendships and friendships as relational contracts in part suggests why contract law is designed this way. That is, contract law is under-compensatory because it is structured to minimize incursions into private ordering, especially when a thick set of social norms is likely to govern transactions. One can see this embedded design feature of contract law as further evidence that we should not worry too much about the “crowding” thesis. The law does not crowd out trust precisely because its interventions leave plenty of room for non-legal sanctions to do their work. That is, even if the “crowding” thesis were true—a point not conceded here—it is substantially mitigated when the legal norms are themselves largely under-compensatory. This underscores their supplementary rather than substitutional nature. Perhaps if the law perfectly tracked “moral” non-legal obligations, it is possible that people would no longer be able to distinguish their friendship motivations from their business motivations (though having two motivations hardly kills the possibility for being a good friend, as I explored above). By leaving room open for social norms to function even when a party invokes the law, parties can continue to act upon and defy social norms.

However, this does highlight yet another point about relational contract theory. Its insistence on “incorporating” social norms is actually doubly misleading and leads to many mistaken characterizations of it. I am hopeful that my discussions here help clarify these confusions. First, as we have already seen, sometimes the relevant social norm is precisely to preclude legal enforcement altogether. This does not undermine relational theory’s incorporation thesis, but it can be missed when it is, as it is so often, caricatured. Second, and more relevant here, although social norms are meant to be translated into contracts by finding that contracts have formed and interpreting them in accordance with the real deals that the parties contemplated, we cannot forget that when it comes to remedies, the law always remains under-compensatory, whatever the social norms. That is for good reason, too. This will leave some room for the non-legal sanctions and recognize that the predication of the relationship outside the law should have some force. 234Accordingly, Bagchi’s recent effort to reserve only reliance-based recovery for those situations in which seemingly social promises are deemed enforceable is an unnecessary and unwelcome overlay where the law of remedies is concerned. Bagchi, supra note 21, at 5. As I have endeavored to explain here, all relational contracts will have a complex matrix of social promises and more formal ones. The law cannot and should not build an artificial separation between these spheres (though it will have to sort out with common sense which promises and duties to enforce and which not to), whether through the law of remedies or otherwise. Since the law of remedies leaves social obligations to remedy themselves in part in all cases—even cases in which the law intervenes—those who want to have an untouchable symbolic space for non-legal relationships already have it in the law of remedies. Forcing these obligations into a reliance box within the law of remedies is unnecessary and pays too much homage to a promise-based account of contract. Moreover, the urge to keep expectancy damages “pure” of promises that occur between intimate friends misconstrues expectancy as much more fully compensatory than it actually is. As we saw in the previous section, although the crowding thesis is probably false in its most extreme form, we have to worry about it most in an environment of perfect enforcement. Yet contract law perhaps already incorporates a defense mechanism against crowding.

In any case, even standard contract damages might not veer all that far from what we might expect friends to owe one another upon a breach. Consider Raz: “Normally our responsibility to make good harm we cause to others depends on fault . . . . Friends have a no-fault obligation to each other though normally it does not require full compensation.” 235 Raz, supra note 116, at 211. Thus, from the perspective of relational contract as friendship, we can begin to understand why remedies are triggered without fault as a general matter and why they do not require full compensation; under-compensation suffices. Still, none of this is to suggest that fault is not relevant in all sorts of ways when assessing contractual liability and its magnitude; 236 See generally Richard Craswell, When Is Willful Breach ‘Willful’? The Link Between Definitions and Damages, 107 Mich. L. Rev. 1501 (2009) (discussing bad faith as a basis for a remedy in contractual breaches and as a basis for liability in establishing breach). nor is it to suggest that friendship will not often demand more than the law of remedies will ever provide.

Yet the general argument here is that it is not necessarily a bad thing for friendship to demand more than contract law will ever give. Indeed, this very divergence helps alleviate some of the recent concern expressed by Seana Shiffrin, 237 See Shiffrin, supra note 21, at 709–13. who finds the divergence between promissory norms in private life (say, between friends) and contract norms disabling for the moral agent. Quite the reverse may be true: Contract law’s divergence from social norms is a good thing that provides a realm of freedom for social norms. The under-compensatory nature of contractual remedies makes that possible without forcing the vulnerable and those who are victims of another party’s opportunism to be wholly without recourse. This divergence may, after all, be just what an honest relational theory would predict—internal social norms would enforce substantial breaches with relatively modest remedies so that non-breaching parties are not left too badly off, but would rely on non-legal sanctions to enforce most of the deal. Even if this balance of legal and non-legal sanctions is not contemplated by every relational contract, it is a decent approximation of most, helps explain why the law is the way it is, and dulls some of the bite of the “crowding” thesis.

Conclusion

Little in this Article is likely to have convinced a non-relationalist to jump aboard the relationalist bandwagon—or minivan, really—in contract theory. Indeed, it may confirm that relationalists tend to go astray in just the ways opponents most fear: toward the judicialization of intimate relations. Still, my ambitions here were relatively modest. I have only tried to give relational theory some new life, sharpening some of the theory’s claims against its competitors by refracting the relational theory of contract through an analogy to friendship. I have also offered a provocation about friendship’s contractual structure that should be less threatening once it is seen as a relational contract of sorts. Finally, I have argued that the relational-contract-as-marriage model, which has currency among relationalists, should be replaced with a model of relational contract as friendship. The new models developed here (friendship as relational contract and relational contract as friendship) are more honest to relational contract theory, to marriage, and to friendship, and they help relational contract theory produce some new insights, support old ones, and revise some of its normative agenda.

Footnotes

*Visiting Associate Professor of Law, University of California—Berkeley Law School (Spring 2010); Associate Professor of Law, University of California—Hastings College of the Law; Scholar-in-Residence, Columbia University School of Law (Spring 2009). Thanks to the 1066 Foundation and the Roger Traynor Scholarly Publication Award for financial support; to Verity Winship, Steve Vladeck, Bob Scott, Alice Ristroph, Eric Posner, Michael Pratt, Michael O’Hear, Stewart Macaulay, Dan Markel, Julian Ku, Katy Kuh, Adam Kolber, Paul Ingram, Adil Haque, Jay Feinman, Ira Ellman, Bill Dodge, Reza Dibadj, I. Bennett Capers, Dave Campbell, Mike Cahill, and Curtis Bridgeman for comments on earlier drafts; to the participants of a workshop at Hofstra Law School (“Prawfsfest!”) for feedback; to the participants of the Bay Area Forum for Law & Ethics (BAFFLE) at Berkeley; and to Doug Ecks for wonderful research assistance.

1For a maximalist version of this thesis, see, for example, Neil G. Williams, What To Do When There’s No ‘I Do’: A Model for Awarding Damages Under Promissory Estoppel, 70 Wash. L. Rev. 1019, 1047 (1995) (“Modern marriages are . . . tantamount to contractual relationships.”).

2This group is usually taken to include Ian Macneil, Stewart Macaulay, Richard Speidel, and Jay Feinman. This Article largely addresses the work of those relational contract theorists who are essentially legal scholars, not theorists of relational contracting in economics, organizational theory, history, and sociology, though some aspects of these approaches do inform several of my arguments here. For some examples of the work in this latter group, see Naomi R. Lamoreaux et al., Beyond Markets and Hierarchies: Toward a New Synthesis of American Business History, 108 Am. Hist. Rev. 404 (2003) (explaining how relational contracting can be a useful—and causal—response to changing economic organization); Charles F. Sabel & Jonathan Zeitlin, Neither Modularity nor Relational Contracting: Inter-firm Collaboration in the New Economy, 5 Enterprise & Soc’y 388 (2004) (criticizing some ways of writing business history and some ways of thinking about informality within relational contracting); George Baker et al., Contracting for Control (Mar. 21, 2006) (unpublished manuscript), available at http://www.law.columbia.edu/null/Contracts+Conf+-+April+7-8,+2006+-+Paper+-+Gibbons?exclusive=filemgr.download&file_id=961193&showthumb=0 (arguing that although many firms contract with informality in relational contracting environments, firms are not indifferent to enforcement mechanisms and will try to contract to control temptations to renege); Oliver Hart & John Moore, Contracts as Reference Points (Harvard Law Sch., John M. Olin Ctr. for Law, Econ. & Bus., Discussion Paper No. 572, 2006), available at http://www.law.harvard.edu/programs/olin_center/papers/pdf/Hart_Moore_572.pdf (exploring some of the challenges of developing efficient long-term contracts in light of elements of such contracts that simply cannot be enforced by courts). Thanks to Bob Scott for some guidance in this literature.

3For a recent effort to specify the commitments of relational contract theory clearly, see Dori Kimel, The Choice of Paradigm for Theory of Contract: Reflections on the Relational Model, 27 Oxford J. Legal Stud. 233 (2007).

4 See, e.g., Ian R. Macneil, The New Social Contract 15, 22, 29, 68, 91 (1980); Robert W. Gordon, Macaulay, Macneil, and the Discovery of Solidarity and Power in Contract Law, 1985 Wis. L. Rev. 565, 569 (“[P]arties [to relational contracts] treat their contracts . . . like marriages . . . .”); Kimel, supra note 3, at 245; John Wightman, Intimate Relationships, Relational Contract Theory, and the Reach of Contract, 8 Feminist Legal Stud. 93, 100 n.23, 102 (2000); see also Gillian K. Hadfield, Problematic Relations: Franchising and the Law of Incomplete Contracts, 42 Stan. L. Rev. 927, 964 (1990) (exploring how franchising relationships—quintessentially treated as relational contracts—are like marriages); Stewart Macaulay, Non-contractual Relations in Business: A Preliminary Study, 28 Am. Soc. Rev. 55, 65 (1963) (“A breach of contract law suit may settle a particular dispute, but such an action often results in a ‘divorce’ ending the ‘marriage’ between the two businesses.”); Marjorie Maguire Shultz, Contractual Ordering of Marriage: A New Model for State Policy, 70 Cal. L. Rev. 204 (1982) (exploring how marriages are like relational contracts).

5 Eric A. Posner, Law and Social Norms 160 (2000) (“A contract is a kind of marriage.”).

6Elizabeth S. Scott & Robert E. Scott, Marriage as Relational Contract, 84 Va. L. Rev. 1225 (1998).

7 See, e.g., Dori Kimel, From Promise to Contract: Towards a Liberal Theory of Contract 80–81 (2003) (noting that relational contract theory literature “does not speak in one voice”).

8Although often credited to Robert E. Scott, The Case for Formalism in Relational Contract, 94 Nw. U. L. Rev. 847, 852 (2000), the phrase appeared earlier in a 1992 essay by Randy E. Barnett. Randy E. Barnett, Conflicting Visions: A Critique of Ian Macneil’s Relational Theory of Contract, 78 Va. L. Rev. 1175, 1200 (1992); see also Richard E. Speidel, The Characteristics and Challenges of Relational Contracts, 94 Nw. U. L. Rev. 823, 845 n.86 (2000) (“Scott now states that ‘[w]e are all relationalists.’ Nonsense. Those advocating the virulent strain of the new formalists are . . . [not] relationalists and the sooner we say so the better.” (citation omitted)).

9This insight is often traced to the pioneering work of Stewart Macaulay. See, e.g., Stewart Macaulay, Changing a Continuing Relationship Between a Large Corporation and Those Who Deal with It: Automobile Manufacturers, Their Dealers, and the Legal System, 1965 Wis. L. Rev. 483; Stewart Macaulay, Elegant Models, Empirical Pictures, and the Complexities of Contract, 11 Law & Soc’y Rev. 507 (1977); Macaulay, supra note 4. People also like to cite Bob Ellickson’s important book here, along with an early piece by Lisa Bernstein. See Robert C. Ellickson, Order Without Law: How Neighbors Settle Disputes (1991); Lisa Bernstein, Opting Out of the Legal System: Extralegal Contractual Relations in the Diamond Industry, 21 J. Legal Stud. 115 (1992); see also David Charny, Nonlegal Sanctions in Commercial Relationships, 104 Harv. L. Rev. 375 (1990).It should be noted, however, that some have found that even in relational contracts, legal and formal contracting serve a core function of forging trust. See Simon Deakin et al., Contract Law, Trust Relations, and Incentives for Cooperation: A Comparative Study, in Contracts, Co-operation, and Competition 110 (Simon Deakin & Jonathan Michie eds., 1997).

10For some recent evidence of the affective nature of many business relationships, see Paul Ingram & Xi Zou, Business Friendships, 28 Res. Org. Behav. 167 (2008); Paul Ingram & Arik Lifschitz, Kinship in the Shadow of the Corporation: The Interbuilder Network in Clyde River Shipbuilding, 1711-1990, 71 Am. Soc. Rev. 334 (2006); and Paul Ingram & Peter W. Roberts, Friendships Among Competitors in the Sydney Hotel Industry, 106 Am. J. Soc. 387 (2000). Ingram & Zou observe that “[b]usiness friendships represent substantial benefits to careers and organizational performance, and they are becoming more common.” Ingram & Zou, supra at 182. They are, however, clear that this fact is not without costs: “[B]usiness friendships also represent notable psychic, social and cultural tensions . . . .” Id. Still, they are also clear that “the option of segregating economic and social worlds is not available to contemporary business people.” Id.It is worth noting that much of Ingram’s work is devoted to horizontal ties in business—affective relationships among competitors—rather than vertical ties among buyers and sellers, who are likely to be engaged in contractual relations as well. For some useful work on vertical ties in business, see Paul DiMaggio & Hugh Louch, Socially Embedded Consumer Transactions: For What Kinds of Purchases Do People Most Often Use Networks?, 63 Am. Soc. Rev. 619 (1998); Vincenzo Perrone et al., Free to Be Trusted? Organizational Constraints on Trust in Boundary Spanners, 14 Org. Sci. 422 (2003); Brian Uzzi, Embeddedness in the Making of Financial Capital: How Social Relations and Networks Benefit Firms Seeking Financing, 64 Am. Soc. Rev. 481 (1999) [hereinafter Uzzi, Financial Capital]; and Brian Uzzi, The Sources and Consequences of Embeddedness for the Economic Performance of Organizations: The Network Effect, 61 Am. Soc. Rev. 674 (1996) [hereinafter Uzzi, Sources and Consequences].

11The best source on the implicit dimensions of contract broadly understood is Implicit Dimensions of Contract: Discrete, Relational, and Network Contracts (David Campbell et al. eds., 2003) [hereinafter Implicit Dimensions of Contract].

12Stewart Macaulay, The Real and the Paper Deal: Empirical Pictures of Relationships, Complexity and the Urge for Transparent Simple Rules, in Implicit Dimensions of Contract, supra note 11, at 51.

13Even now, some contract theorists cannot resist challenging the empirical claim. See Kimel, supra note 7, at 81 n.50. In a recent article on relational contract theory, however, Kimel backs off the challenge to the modest form of the empirical claim. See Kimel, supra note 3, at 234.

14Some relationalists are committed to the view that exchange itself is deeply relational and that all contractual intercourse is shaped in meaningful ways by relational features because contracts could not get off the ground without sufficient social solidarity. Ironically, this view may be too moderate to be at the core of relational contract theory. In short, once all contracts get folded into this form of relationalism, the claim is no longer strong enough to warrant the disparate treatment of different forms of contracts, which seems central to the normative thrust of relational contract theory. Indeed, non-relationalists do not bother opposing this form of relationalism. Non-relationalists are only too happy to concede that all contracts should be treated under the same rules: if all contracts are relational, all contracts are subject to the same law. Nevertheless, as the text immediately following this note reveals, there is still room for differential treatment because of the possibility of a continuum of relational factors.

15For an analysis of how the empirical claim was absorbed into mainstream “neo-classical” contract law, see generally Jay M. Feinman, Relational Contract Theory in Context, 94 Nw. U. L. Rev. 737 (2000). Some of the ways the UCC can be read to absorb relationalism in laws governing performance, modification, interpretation, and formation are discussed infra note 34.

16In a recent clarification, Macneil changed terminology, hoping to get a “relational/as-if-discrete” continuum off the ground to replace the relational-discrete continuum. See Ian R. Macneil, Relational Contract Theory: Challenges and Queries, 94 Nw. U. L. Rev. 877, 894–900 (2000) [hereinafter Macneil, RCT: C & Q]. Macneil has always been fairly clear that “discreteness” is a “relative” property from the perspective of a relational approach. See, e.g., Ian R. Macneil, Relational Contract: What We Do and Do Not Know, 1985 Wis. L. Rev. 483, 485; Ian R. Macneil, Values in Contract: Internal and External, 78 Nw. U. L. Rev. 340 (1983). However, Macneil’s recent clarification demonstrates his ultimate commitment to the “strongest” form of the empirical claim that all contracts are relational. The problem with this form of the empirical claim, perhaps, is that it seems to dilute what is uniquely special about “relational contracts” since many relationships forged by and underwriting long-term relational contracts are rather thin. See Kimel, supra note 3, at 239 (“[T]he further we move towards [the empirical claim’s] most extreme form—all contracts are relational—the more it depends for its plausibility on an understanding of ‘relationship’ so watered-down as to make the debate entirely uninteresting.”).Macneil has also quite recently tried to change his most basic terminology, calling his theory “essential contract theory,” rather than “relational contract theory.” Macneil, RCT: C & Q, supra at 877, 881. That re-branding has not been successful, and I prefer to stick with the old terminology and its thirty years of history and development.

17There is a useful cut at the empirical question in DiMaggio & Louch, supra note 10, at 620, 622. In their close look at the 1996 General Social Survey economic sociology module, they find that “people buy and sell goods and services from friends and kin” and that “substantial percentages of major transactions take place between friends, relatives, or compound ties.” Id. More specifically, “[a]lmost one-half of all used-car purchases from individuals (46.0 percent) and home purchases where no agent is used (46.8 percent) are transactions between relatives, friends, or acquaintances.” Id. at 622–23. Approximately one-fourth of purchases of legal services and home maintenance services are similarly within a social network. Id. at 623.Does this evidence resolve the central empirical claim of relational contract theory? Of course not. But it does at least help focus attention on contractual contexts where consumer transactions tend to involve the use of strong social networks.

18An example of this type of argument was once explained by Robert Gordon:Some theorists have even advanced a bold hypothesis of a specific historical causal relationship between the fantasy world of political-legal ideological discourse about contracts and the social world of contracting: they contend that encouraging people to deal with one another as strangers progressively erodes the underlying relations of solidarity, reciprocity and trust upon which capitalist economies essentially depend.Gordon, supra note 4, at 578–79; see also Mark Vernon, The Philosophy of Friendship (2005) (arguing that friendship and capitalism can sit in tension). Brian Uzzi asks: “What modern institutions and . . . arrangements need exist if embedded [i.e., relational] exchange systems are to arise and prosper in a society?” Uzzi, Sources and Consequences, supra note 10, at 695. It may be that a more thoroughly relational contract law could be the answer. Indeed, DiMaggio & Louch generally find a high degree of satisfaction in embedded transactions, which is perhaps a reason for society to encourage them. DiMaggio & Louch, supra note 10, at 633–34.

19Daniel Markovits’s contract theory of “contract as collaboration,” for example, argues for orienting contract theory along the dimension of collaboration. Daniel Markovits, Contracts and Collaboration, 113 Yale L.J. 1417 (2004). However, he is not really a relationalist because he seems to think that parties generally come to contracts as strangers, id. at 1435, and that it is the contract itself that establishes the cooperative relation between the parties, one that has no affective dimension, id. at 1432, 1451. Add to that the fact that he explicitly claims distance from Ian Macneil (though the position he ascribes to Macneil and rejects—that contracts are not “freestanding” sources of obligation—is not obviously one Macneil endorses). Id. at 1450 n.68.

20I have explained the role of “strangership” in Ethan J. Leib, Friendship & the Law, 54 UCLA L. Rev. 631, 663−64 (2007), drawing upon two articles by Allan Silver. See Allan Silver, “Two Different Sorts of Commerce”—Friendship and Strangership in Civil Society, in Public and Private in Thought and Practice: Perspectives on a Grand Dichotomy 43 (Jeff Weintraub & Krishan Kumar eds., 1997); Allan Silver, Friendship in Commercial Society: Eighteenth-Century Social Theory and Modern Sociology, 95 Am. J. Soc. 1474 (1990).

21Forms of this argument against the analytic claim are actually quite common among non-relationalists. See generally Kimel, supra note 7 (arguing that contract facilitates detachment from personal relations); Melvin A. Eisenberg, Why There Is No Law of Relational Contracts, 94 Nw. U. L. Rev. 805, 820 (2000) (“This massive contractualization of human relationships [which is effected by relational contract theory] undesirably obscures critical differences between economic and affective relationships, between explicit and tacit reciprocity, between relationships that should be enforceable by both law and social norms and relationships that should be enforceable only by social norms.”); Kimel, supra note 3; Aditi Bagchi, Contract v. Promise (Univ. Pa. Law Sch., Public Law and Legal Theory Research Paper Series, Paper No. 07-35, 2007), available at http://ssrn.com/abstract=1012150; see also Gunther Teubner, Expertise as Social Institution: Internalising Third Parties into the Contract, in Implicit Dimensions of Contract, supra note 11, at 333, 343–46 (recognizing that relational contract theory risks contractualizing social life and arguing that would be bad).It is worth noting that this type of argument, if correct, goes a long way to address Seana Shiffrin’s recent concern about the “divergence” of contract norms from moral norms of promising. See Seana Valentine Shiffrin, The Divergence of Contract and Promise, 120 Harv. L. Rev. 708 (2007). If contracts are best designed as something different than promises in “private” life, there is less reason to worry that there is norm divergence. This argument is spelled out at length by Aditi Bagchi. See Bagchi, supra.

22This argument is similar to the “crowding” thesis that because the law will tend to crowd out authentic trust relations, we should be wary of interposing the law into trust relations. In its extreme form, the argument counsels for no legal intervention where there is real trust between the parties. The merits of this argument are addressed further infra Part IV. I discussed (and rejected) the “crowding” thesis in Ethan J. Leib, Friends as Fiduciaries, 86 Wash. U. L. Rev. 665 (2009). See also Anthony J. Bellia, Jr., Promises, Trust, and Contract Law, 47 Am. J. Juris. 25 (2002) (arguing against the thesis that legal enforceability degrades the social norm of trust); Frank B. Cross, Law and Trust, 93 Geo. L.J. 1457, 1545 (2005) (“Whatever the intuitive appeal of the claims that legalization undermines trust, they cannot be sustained once they are subjected to scrutiny and empirical testing.”); Claire A. Hill & Erin Ann O’Hara, A Cognitive Theory of Trust, 84 Wash. U. L. Rev. 1717 (2006). Most generally, there is a real problem trying to separate neatly the spheres of contract and law on the one hand and intimacy on the other. See generally Viviana A. Zelizer, The Purchase of Intimacy (2005). This Article seeks to undermine the “separate spheres” story too.

23This “asking the wrong questions” objection appears in Richard Craswell, The Relational Move: Some Questions from Law and Economics, 3 S. Cal. Interdisc. L.J. 91 (1993). Craswell wants to orient contract theory toward economic analysis to focus on consequences (of price and product safety) and sees relational analysis as sociological and indifferent to consequences. Id. at 92, 98−99. It is not at all clear that this is a fair description, as many relationalists are, after all, sensitive to such consequences. See, e.g., 1 Stewart Macaulay et al., Contracts: Law in Action 643–44 (2d ed. 2003); Jay M. Feinman, Relational Contract and Default Rules, 3 S. Cal. Interdisc. L.J. 43, 55 (1993); Stewart Macaulay, Private Legislation and the Duty to Read—Business Run by IBM Machine, the Law of Contracts and Credit Cards, 19 Vand. L. Rev. 1051 (1966); Ian R. Macneil, Bureaucracy and Contracts of Adhesion, 22 Osgoode Hall L.J. 5 (1984). Nevertheless, it is likely true as a general matter that the law and economics crowd prefers ex ante reasoning in resolving contracts disputes, and the relationalist crowd tends to prefer ex post reasoning. See Speidel, supra note 8, at 844.Others in the law and economics tradition have acknowledged the pervasiveness of relational contracts but have urged an orientation to contract theory guided by value maximization of the relationship between the parties so that they can produce the most surplus. See Charles J. Goetz & Robert E. Scott, Principles of Relational Contracts, 67 Va. L. Rev. 1089 (1981); Robert E. Scott, A Relational Theory of Default Rules for Commercial Contracts, 19 J. Legal Stud. 597 (1990).

24 See Hugh Collins, The Research Agenda of Implicit Dimensions of Contract, in Implicit Dimensions of Contract, supra note 11, at 1, 22–23 (highlighting how difficult it is to see relationalists as having a singular welfarist philosophy); James W. Fox, Jr., Relational Contract Theory and Democratic Citizenship, 54 Case W. Res. L. Rev. 1, 4–5 (2003) (listing variants of the theory and their underlying political theories).

25 See generally Ian R. Macneil, Economic Analysis of Contractual Relations: Its Shortfalls and the Need for a “Rich Classificatory Apparatus, 75 Nw. U. L. Rev. 1018 (1981).

26The list changes from time to time and is not intended to be exhaustive. The standard elements in Macneil’s most recent articulation are: “(1) role integrity (requiring consistency, involving internal conflict, and being inherently complex); (2) reciprocity (the principle of getting something back for something given); (3) implementation of planning; (4) effectuation of consent; (5) flexibility; (6) contractual solidarity; (7) the restitution, reliance, and expectation interests (the ‘linking norms’); (8) creation and restraint of power (the ‘power norm’); (9) propriety of means, and (10) harmonization with the social matrix, that is, with supracontract norms.” Macneil, RCT: C & Q, supra note 16, at 879–80.

27 But see Kimel, supra note 3, at 252–53 (arguing that relational theory disables us from distinguishing between different sources of obligation and helps us organize our priorities when they conflict in interesting ways). Kimel’s critique misses the mark. To say, as relationalists do, that promissory sources of obligation and reliance sources of obligation all come within contract is not to say that we are always disabled from producing hierarchies of source of obligations to privilege in any given context. It is only to say that we cannot—as Kimel and Bagchi do, for example—build a contract model with promise at its center, allowing occasional “exceptions” for reliance-based liability. Relationalists think the practice is too capacious to be organized around any singular principle of liability.

28Relational contract theory’s commitment to contextualization and fragmentation is explained in Feinman, supra note 15.

29For my assessment of how contract theory is too unified, see Ethan J. Leib, On Collaboration, Organizations, and Conciliation in the General Theory of Contract, 24 Quinnipiac L. Rev. 1 (2005).

30Eisenberg, supra note 21, at 817–18 (citing David Campbell & Donald Harris, Flexibility in Long-Term Contractual Relationships: The Role of Co-Operation, 20 J.L. & Soc’y 166 (1993)); Robert A. Hillman, Contract Excuse and Bankruptcy Discharge, 43 Stan. L. Rev. 99, 132–33 (1990); Robert A. Hillman, Court Adjustment of Long-Term Contracts: An Analysis Under Modern Contract Law, 1987 Duke L.J. 1; Ian. R. Macneil, Contracts: Adjustment of Long-Term Relations Under Classical, Neoclassical, and Relational Contract Law, 72 Nw. U. L. Rev. 854 (1978); Richard E. Speidel, Court-Imposed Price Adjustments Under Long-Term Supply Contracts, 76 Nw. U. L. Rev. 369 (1981)); see also Richard E. Speidel, Article 2 and Relational Contracts, 26 Loy. L.A. L. Rev. 789 (1993) [hereinafter Speidel, Article 2 and Relational Contracts]; Speidel, supra note 8.

31 See Macneil, RCT: C & Q, supra note 16, at 900; see also Speidel, Article 2 and Relational Contracts, supra note 30, at 793 (arguing that relational contract theory’s normative dimension involves internalizing parties’ social norms).

32 See, e.g., Macneil, supra note 25, at 1041.

33I thank Jay Feinman for reminding me about this important dimension of relational contract theory that is often under-emphasized.

34 See, e.g., Macaulay, supra note 12, at 59; Speidel, Article 2 and Relational Contracts, supra note 30. Courses of performance, courses of dealing, and usages of trade are incorporated by, inter alia, U.C.C. § 1-102(2)(b) (1977) (establishing an “underlying purpose[] and polic[y]” of the UCC to be “the continued expansion of commercial practices through custom, usage and agreement of the parties”); id. § 1-201(3) (defining “agreement” to include these); id. § 1-205 (defining “course of dealing” and “usage of trade”); id. § 2-202 (allowing terms of agreements to be supplemented and explained by these); id. § 2-208 (allowing courts to consider these in determining the meaning of an agreement); and id. § 2-302(2) (allowing courts to consider evidence of “commercial setting” in assessing unconscionability). Commercial reasonability more generally pervades the code as a benchmark for proper behavior and is one of the code’s core gap-fillers. E.g., id. §§ 2-305, 2-309, 2-311, 2-503, 2-504, 2-602, 2-610, 2-704, 2-706, 2-709, 2-714, 2-715, 2-716, 2-718. Moreover, the UCC’s recognition of largely indefinite and vague agreements is a move in the relationalist direction. See id. §§ 2-204, 2-206 (relaxing the offer and acceptance and definiteness requirements of the common law); id. §§ 2-305, 2-306, 2-716 cmt. 2 (recognizing requirements, outputs, and open-price contracts, which might have been too vague to be enforceable under classical contract law or might have lacked consideration under classical rules). Perhaps the single largest relationalist incorporation into the code occurs in U.C.C. § 1-201(3): an agreement may be found solely “by implication,” a method of contract formation from which most formalists would recoil. Moreover, Macneil neatly demonstrates how § 2-207’s abrogation of the strict mirror-image rule for acceptance and its embrace of formation through conduct in the case of the “battle-of-the-forms” is a partial incorporation of elements of relational contract theory. See Macneil, supra note 4, at 72–75.

35Eisenberg, supra note 21, at 813.

36 Id. at 818 (“Because there is no significant distinction between contracts as a class and relational contracts, these rules, and others like them, can be separated into two broad classes: those that are good for all contracts and therefore should be general principles of contract law, and those that are not good for any contracts.”); see also id. at 817 (“There can be no special law of relational contracts, because relational contracts and contracts are virtually one and the same.”).

37Eisenberg’s critique is generally embraced by opponents of relational contract theory. See, e.g., Ira Mark Ellman, “Contract Thinking” Was Marvin’s Fatal Flaw, 76 Notre Dame L. Rev. 1365, 1369 (2001); Kimel, supra note 3, at 235–36.

38The sorites paradox was a rhetorical strategy used by the ancient Skeptics against the Stoics to prove that their confidence in the possibility of knowledge was unfounded. The argument proceeds by trying to prove that although one thinks that one knows what a “heap” is, it turns out that there are no heaps. (Sorites means heap in Greek.) One might think 10,000 grains of sand make a heap. Yet, since (1) one grain of sand is no heap and (2) for every n grains of sand that are not heaps, n+1 grains of sand will also not be a heap, there are no heaps. For more on the sorites, see Ethan J. Leib, On the Sorites: Toward a Better Understanding of Chrysippus, 21 Ancient Phil. 147 (2001). The sorites argument probably fails because proposition (2) is false. Just because a person may have some difficulty in figuring out where to draw the line between a few grains of sand and a heap does not mean we cannot distinguish the concepts at the extremes. Fuzziness of categories does not mean that they cannot be used, even when it comes to the law.

39And even if a relational contract law would apply only to a smaller class of contracts, a multi-factor test proposed by Speidel (drawing from Lewis Kornauser’s summary of Macneil’s work) is as coherent as any and raises questions about Eisenberg’s general belief in the uselessness of trying to define relational contracts:The degree to which a contract is relational depends upon the presence of some or all of the characteristics discussed below. Most notable are whether “(1) the transaction extends over time, (2) parts of the exchange cannot be measured or specified precisely [at the time of contracting], and (3) the interdependence of the parties to the exchange extends at any given moment beyond any single discrete transaction to a range of social interrelationships.”Speidel, Article 2 and Relational Contracts, supra note 30, at 792 (quoting Lewis A. Kornhauser, The Resurrection of Contract, 82 Colum. L. Rev. 184, 190 (1982)). Undoubtedly, there are difficulties in classification at the margin, but fuzziness of a concept’s boundaries does not mean that no meaningful distinctions can be drawn in clear cases. See supra note 38.

40Eisenberg, supra note 21, at 814. Eisenberg seems to contest the centrality of the spectrum approach to Macneil’s relational contract theory. Id. at 813 (“Macneil sometimes treats discreteness as an end of a spectrum rather than as a definition of a body of contracts.”). However, it really is difficult to read Macneil as ever suggesting anything other than a spectrum approach. To be fair, others (like Speidel and Hillman, for example) have tried to treat “relational contracts” as a reasonably distinct category the law could identify and treat specially and differently. In what follows, when I discuss “relational contracts,” I mean “relatively relational contracts” and embrace Macneil’s spectrum approach, which I do not think fails Eisenberg’s challenge.

41Eisenberg, supra note 21, at 814.

42 See, e.g., Macneil, supra note 4, at 17.

43 See, e.g., Barnett, supra note 8, at 1191.

44 Macneil, supra note 4, at 70.

45 But see Oliver E. Williamson, Transaction-Cost Economics: The Governance of Contractual Relations, 22 J.L. & Econ. 233, 236 (1979) (“Serious problems of recognition and application are posed by such a rich classificatory apparatus.”). Macneil replies: “A ‘rich classificatory apparatus’ of some kind is essential if contractual relations are to be both understood and subjected to successful, realistic, and reasonably consistent analysis.” Macneil, supra note 25, at 1025 n.26 (quoting Williamson, supra).

46Eisenberg, supra note 21.

47 See generally Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms, and Institutions, 99 Mich. L. Rev. 1724 (2001); Lisa Bernstein, The Questionable Empirical Basis of Article 2’s Incorporation Strategy: A Preliminary Study, 66 U. Chi. L. Rev. 710 (1999); Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search for Immanent Business Norms, 144 U. Pa. L. Rev. 1765 (1996). For a related critique about courts’ inability to incorporate norms with any degree of uniformity, see Robert E. Scott, The Uniformity Norm in Commercial Law: A Comparative Analysis of Common Law and Code Methodologies, in The Jurisprudential Foundations of Corporate and Commercial Law 149 (Jody S. Kraus & Steven D. Walt eds., 2000).

48This line of defense is on display in Stewart Macaulay, Relational Contracts Floating on a Sea of Custom? Thoughts About the Ideas of Ian Macneil and Lisa Bernstein, 94 Nw. U. L. Rev. 775, 784–804 (2000). Macaulay also analyzes Scott’s related critique. Macaulay, supra note 12, at 59–68.

49Macneil, RCT: C & Q, supra note 16, at 906.

50 See generally Eric A. Posner, A Theory of Contract Law Under Conditions of Radical Judicial Error, 94 Nw. U. L. Rev. 749 (2000).

51 Id. at 754. Although Scott abjures from signing on to the radical judicial error hypothesis, concerns about judicial competence certainly animate his argument for formalism as well. See, e.g., Scott, supra note 8, at 858–59, 861.

52 See Jeffrey E. Allen & Robert J. Staaf, The Nexus Between Usury, “Time Price” and Unconscionability in Installment Sales, 14 UCC L.J. 219 (1982).

53There are many nuances to Posner’s view, but to discuss them all is beyond the scope of this Article.

54Scott, supra note 8, at 852.

55Even Posner seems to acknowledge that if “courts can determine the parties’ intentions from context and common sense . . . . then courts should ignore form.” Posner, supra note 5, at 162. That concession also raises the possibility of a perfectly acceptable third way that might appeal to some formalists and some relationalists alike: when courts can determine parties’ intentions from context and common sense, they ought to gap-fill accordingly; when they cannot, they should endeavor to be formalistic. For his part, Macaulay would accept a different third way: he would allow courts to be formalistic only when both parties had competent counsel and clearly were able to bargain meaningfully with one another (rather than through battling of standardized forms). See Macaulay, supra note 12, at 62 (citing Binks Mfg. Co. v. Nat’l Presto Indus., Inc., 709 F.2d 1009 (7th Cir. 1983)). Campbell and Collins also develop a nuanced third way of deciding when the implicit dimensions of a contract should control in court. David Campbell & Hugh Collins, Discovering the Implicit Dimensions of Contract, in Implicit Dimensions of Contract, supra note 11, at 25, 42.

56To be sure, others in law and economics have not made such radical assumptions about judicial incompetence. See Gillian Hadfield, Judicial Competence and the Interpretation of Incomplete Contracts, 23 J. Legal Stud. 159 (1994).

57 See supra note 55. Of course, if courts or parties don’t know when judges are likely to err, it is hard to create a two-track model along the lines suggested. There is also a real cost associated with attempting to learn about party norms and expectations—that cannot be discounted fully, though the cost itself (in judicial time and energy) is not particularly easy to measure.

58Posner has emphasized that the fact-finding processes in judicial adjudications do not inspire confidence and can lead to many judicial errors. Posner, supra note 5, at 153.

59 See Macaulay, supra note 12, at 62, 77 (drawing upon “realism” to reject the possibility of formalism).

60Posner insists that although judges are incompetent at discovering and applying internal relational norms, they are perfectly capable of deciding whether parties have an intention to have legally available remedies in the first place. Posner, supra note 5, at 156; Posner, supra note 50, at 762–63, 767. This central distinction is necessary to make the argument work. But it preys on the ability of courts to stick with formalism, since Posner thinks judges can competently make these determinations through the use of formal reasoning. I am somewhat less optimistic since like most relationalists, I am a realist too—I think judges use play in the joints to manipulate formal categories. If that is correct, there will be—and must be to make his argument work—a fair bit of judicial “error” on the very threshold question about which Posner thinks judges are competent.

61Richard A. Posner, What Am I? A Potted Plant?, New Republic, Sept. 28, 1987, at 23, 23.

62 See generally Charny, supra note 9 (discussing the role of non-legal sanctions in the formation of non-formalized agreements).

63 See Ingram & Zou, supra note 10, at 170.

64There is, it should be noted, some empirical evidence that some relational contracts actually become more detailed over time. See Laura Poppo & Todd Zenger, Do Formal Contracts and Relational Governance Function as Substitutes or Complements?, 23 Strategic Mgmt. J. 707 (2002); Michael D. Ryall & Rachelle C. Sampson, Do Prior Alliances Influence Alliance Contract Structure?, in Strategic Alliances: Governance and Contracts 206 (Africa Arino & Jeffrey J. Reuer eds., 2006); Michael D. Ryall & Rachelle C. Sampson, Formal Contracts in the Presence of Relational Enforcement Mechanisms: Evidence from Technology Development Projects, 55 Mgmt. Sci. 906 (2009). Although this finding seems to stand in some tension with the idea that relational governance relies on loose and open-ended contract terms, it makes some sense too. As the relationship deepens, one can more easily get more specific about the nature of the relationship, which needed to remain more incomplete in its infancy. Macneil suggests as much in his work on relational contract theory. See Macneil, supra note 25, at 1041 (explaining that contract choice “is thus an incremental process in which parties gather increasing information and gradually agree to more and more as they proceed”). The same is probably true in friendships. As time goes on, we certainly get clearer about our expectations and responsibilities within the relationship.

65Relationalists may very well embrace formalism for relatively discrete contracts, since the norms of planning and consent are considered especially important by relational contract theory in relatively discrete contracts. See Macneil, RCT: C & Q, supra note 16, at 879–80.

66Scott & Scott, supra note 6, at 1295. I revisit this debate infra Part IV.

67 See Scott, supra note 8.

68Scott & Scott, supra note 6, at 1330. See also Posner, supra note 5, at 8 (“[M]any legal rules are best understood as efforts to harness the independent regulatory power of social norms.”). But see id. at 64 (“[J]udicial enforcement would interfere with trust relationships.”).

69Collins, supra note 24, at 1, 16; Macaulay, supra note 48, at 777–96. See generally Macaulay, supra note 12. The “resonance” idea might be said to find some support in scholarship that traces the relationship between compliance and the perceived substantive fairness of legal institutions. See generally Janice Nadler, Flouting the Law, 83 Tex. L. Rev. 1399 (2005) (arguing that harmonizing law and social norms can help to breed compliance, and the opposite can breed noncompliance); Paul H. Robinson & John M. Darley, The Utility of Desert, 91 Nw. U. L. Rev. 453 (1997). Organic solidarity is also described (with a somewhat different focus) in Gordon, supra note 4, at 569–70.

70 E.g., Scott, supra note 8, at 861.

71 See Collins, supra note 24, at 1, 8 (“[A] process of legal recognition of implicit [and relational] dimensions [of contract] is necessary and inevitable in any system of law.”); Macneil, supra note 16, at 508 (“[P]romise-centered theories [are] inadequate to deal with complex contractual relations without distortion and omission.”).

72Some relationalists would just parry this argument with realism and say there is good reason to be skeptical about judges’ capacities to adhere to truly formalistic reasoning.

73 See Collins, supra note 24, at 10.

74For my “rich classificatory apparatus” about how to define the friend, see Leib, supra note 20, at 638–53; and Ethan J. Leib, Friends and the Law: Can Public Policy Support the Institution of Friendship?, 145 Pol’y Rev. 55 (2007). I also use a particular emphasis in defining friendship in Leib, supra note 22, but limit the analysis there to very close friends. It is fair to say that the working concept of friendship in what follows is a broader category than the one I have utilized in previous work (though not as broad as the type of friendship I explore in Ethan J. Leib, The Politics of Family and Friends in Aristotle and Montaigne, 31 Interpretation: J. Pol. Phil. 165 (2004)). I am not the first, of course, to see the fiduciary relationship as a special case of the relational contract. See, e.g., Goetz & Scott, supra note 23, at 1126–28; Scott & Scott, supra note 6, at 1265–66.

75Scott & Scott, supra note 6, at 1229, 1255. It is, of course, true that the “term” and intensity can vary greatly. Short-term friends frequently pursue friendship during, say, summer camp or military training. However, so long as there is some duration and a predicted future, such relationships can qualify as real friendships for the purposes of the comparison.

76For a description of friendship that coheres with this conception, see Jeanette Kennett & Steve Matthews, What’s the Buzz? Undercover Marketing and the Corruption of Friendship, 25 J. Applied Phil. 2, 8 (2008) (“My reasons for action where [my friend] is concerned do not depend on any contingent similarity of interests, and neither are they derivative of other personal or professional commitments. In the case where my close friend’s interests diverge from mine her interests continue to have active guiding force for me, since in close friendship it is her interests as such that are important, not her interests as filtered though my assessment of my own profit or amusement.”).

77 See Jack L. Carr & Janet T. Landa, The Economics of Symbols, Clan Names, and Religion, 12 J. Legal Stud. 135, 156 (1983); Janet T. Landa, The Enigma of the Kula Ring: Gift-Exchanges and Primitive Law and Order, 3 Int’l Rev. L. & Econ. 137, 152–56 (1983); Ian R. Macneil, Exchange Revisited: Individual Utility and Social Solidarity, 96 Ethics 567, 568 (1986). Those in relational contracts display just this need to distance themselves from “mere” commercial interaction. See, e.g., Uzzi, Sources and Consequences, supra note 10, at 680 (providing evidence of how gifting works within relational exchange). Indeed, there is evidence that gifts can be valued differentially, depending on whether the recipient thinks the gift is based on affect, role, or cost-benefit analysis. See, e.g., Daniel R. Ames et al., It’s the Thought That Counts: On Perceiving How Helpers Decide to Lend a Hand, 30 Personality & Soc. Psychol. Bull. 461 (2004).

78 See Wightman, supra note 4, at 120 (arguing that although gift-giving can be associated with altruism, it is more generally reciprocal within interpersonal relationships and relational contracts, and its reciprocity complicates the story of altruism); see also Marcel Mauss, The Gift (W.D. Halls trans., Routledge 2005) (1954) (exposing the gift as a form of exchange); Posner, supra note 5, at 50, 55–56 (arguing that gift-giving is rarely deeply altruistic); Eric A. Posner, Altruism, Status, and Trust in the Law of Gifts and Gratuitous Promises, 1997 Wis. L. Rev. 567. To be fair, in some contractual settings, gifts are refused, precisely to avoid muddying the waters in a commercial exchange. See, e.g., Jordan I. Siegel et al., Egalitarianism, Cultural Distance, and FDI: A New Approach 8 (Am. Law & Econ. Ass’n Annual Meetings, Working Paper No. 133, 2008), available at http://law.bepress.com/cgi/viewcontent.cgi?article=2268&context=alea. Siegel and co-authors quote from a form letter that a “prospective customer or partner” could expect to receive if he or she gave a gift to a Motorola employee in Korea:I hereby wish to thank you for your kindness in sending a gift. While I know that your intentions were positive, in order to avoid even the appearance of anything but a business relationship built on the business value offered by our respective organizations, Motorola’s Code of Business Conduct requires that we not accept gifts from our business partners, other than low cost promotional items. Therefore, I am respectfully returning your gift. I would like to take this opportunity to emphasize that it is really unnecessary to send gifts and thank you once again for your gesture of friendship.

79Id. This is a fascinating but quite unusual business practice. As Siegel and co-authors concede, this unusual strategy forces Motorola Korea to lose business, since it is “de rigueur” within Korea to accept and give such gifts. Id. at 8.There is a large philosophical literature discussing the partiality that is constitutive of friendship. See generally Lawrence A. Blum, Friendship, Altruism, and Morality (1980); Marilyn Friedman, What Are Friends For? 192 (1993); Troy A. Jollimore, Friendship and Agent-Relative Morality (2001); David B. Annis, The Meaning, Value, and Duties of Friendship, 24 Am. Phil. Q. 349 (1987); Neera Badhwar Kapur, Why It Is Wrong to Be Always Guided by the Best: Consequentialism and Friendship, 101 Ethics 483 (1991); Marcia Baron, Impartiality and Friendship, 101 Ethics 836 (1991); Dean Cocking & Justin Oakley, Indirect Consequentialism, Friendship, and the Problem of Alienation, 106 Ethics 86 (1995); John Cottingham, Partiality, Favouritism and Morality, 36 Phil. Q. 357 (1986); Neera Kapur Badhwar, Friends as Ends in Themselves, 48 Phil. & Phenomological Res. 1 (1987); Peter Railton, Alienation, Consequentialism, and the Demands of Morality, in Consequentialism and Its Critics 93, 98–99 (Samuel Scheffler ed., 1988); William H. Wilcox, Egoists, Consequentialists, and Their Friends, 16 Phil. & Pub. Aff. 73 (1987); Susan Wolf, Morality and Partiality, 6 Phil. Perspectives 243 (1992). But see Shelly Kagan, The Limits of Morality 367 (1989) (denying that love and friendship require favoritism).

80Macaulay, supra note 12, at 78 n.65.

81One other legal scholar has pursued the morphological similarity between friendship and at least one type of relational contract: the employer-employee relationship. See Gary Chartier, Friendship, Identity, and Solidarity. An Approach to Rights in Plant Closing Cases, 16 Ratio Juris 324 (2003) (focusing upon the implicit understandings in each type of relationship).

82Goetz & Scott, supra note 23, at 1092–93.

83Scott & Scott, supra note 6, at 1248.

84 See, e.g., Goetz & Scott, supra note 23, at 1091 (arguing that contracts are relational to the extent that “parties are incapable of reducing important terms of the arrangement to well-defined obligations”). Again, it is important to draw attention to some of the new findings in the business literature, cited supra at note 64, that prior relationships can lead to more customized and specific agreements than are traditionally associated with relational contracting. The full implications of that finding must await a future date, but I do not think it undermines the morphological similarity between relational contracts and friendships; both lead to more information over time that will allow parties to fill in details of the nature of their expectations.

85Wightman has argued that the reason for or source of incompleteness in relational contracts and intimate relationships, like friendships, is different. Wightman, supra note 4, at 108−09. In relational contracts, he argues that incompleteness results from trade usages, incomplete information, trust, transaction costs, and/or information asymmetry. Id. By contrast, in intimate relationships, “leaving it open” is constitutive of the relationship. Id. I am not sure I agree that the difference is so marked since virtually all of the considerations in play in commercial contexts are likewise relevant in intimate relationships. As Wightman also notes, power inequality within both types of relationships can cause incompleteness. Id.

86These features of relational contracts are explored and explained in Goetz & Scott, supra note 23, at 1114 (describing the requirement of parties to relational contracts to maximize joint value from the relationship); id. at 1149–50 (describing the “best efforts” requirement); Scott & Scott, supra note 6, at 1265–67 (describing the “best efforts” requirement); Speidel, supra note 8, at 840–42 (exploring the heightened duty of “good faith” in connection with relational contracts); id. at 838 (arguing that courts should facilitate adjustments in relational contracts); and Wightman, supra note 4, at 103 (noting that within relational contracts, new evolutions will lead to flexible adjustment of the underlying deal). See also Hillman, Court Adjustment of Long-Term Contracts: An Analysis Under Modern Contract Law, supra note 30, at 8.

87Gordon, supra note 4, at 569.

88Scott & Scott, supra note 6, at 1269−70.

89Of course, this is generally credited as Macaulay’s insight about relational contracts in the sources cited supra note 9. Some theorists of friendship have emphasized its “voluntary” nature, suggesting that friends would find the intervention of the law into their rights and duties inappropriate. See generally Scott Feld & William C. Carter, Foci of Activity as Changing Contexts for Friendship, in Placing Friendship in Context 136 (Rebecca G. Adams & Graham Allan eds., 1998); Allan Silver, Friendship and Trust as Moral Ideals: An Historical Approach, 30 Eur. J. Soc. 274 (1989) (arguing that friendships are quintessentially voluntary). See generally Aristotle, Nicomachean Ethics bk. VII, ch. 13, ll. 1162b22–b32; bk. IX, ch. 1, ll. 1164b12–b15 (Terence Irwin trans., 1999) (arguing that friends must live by trust, not by law). I develop and critique Aristotle’s views on the subject at some length in Leib, supra note 20, at 647–53.

90Disputes between friends and ex-friends are common. See generally Leib, supra note 22, at 700−20; Leib, supra note 20, at 684−97.

91 See Scott & Scott, supra note 6, at 1298 (“Thus, the parties will agree to live under two sets of rules: a more flexible set of rules for social and relational enforcement during the marriage, and a stricter set of rules for legal enforcement upon divorce.”).

92 See Terence Daintith, The Design and Performance of Long-Term Contracts, in Contract and Organization: Legal Analysis in the Light of Economic and Social Theory 164, 187–88 (Terence Daintith & Gunther Teubner eds., 1986); Campbell & Collins, supra note 55, at 39.

93Scott & Scott, supra note 6, at 1288, 1292 (“Social sanctions include reputational harms from gossip, reduced social acceptance and ostracism.”).

94 Id. at 1289; see also Joseph Epstein, Friendship: An Exposé 69 (2006) (“Whatever else it has to do with, friendship entails obligation—sometimes ample and demanding, sometimes miniscule and subtle, but always, I believe, present.”).

95Scott & Scott, supra note 6, at 1290.

96 Id. For some work on the role of truth telling and disclosure within friendship, see Epstein, supra note 94, at 40 (“[O]ne of the things one looks for in a friend . . . is the possibility of easy candor in conversation.”); Michel de Montaigne, Of Friendship, in The Complete Essays of Montaigne 135, 136 (Donald M. Frame trans., 1958) (“Friendship feeds on communication . . . .”); Michel de Montaigne, Of the Art of Discussion, in The Complete Works of Montaigne 705 (Donald M. Frame trans., Stanford Univ. Press 1967) (arguing that friendship “delights in the sharpness and vigor [of verbal] intercourse . . . . It is not vigorous and generous enough if it is not quarrelsome, if it is civilized and artful, if it fears knocks and moves with constraint.”); Andrew Sullivan, Love Undetectable: Notes on Friendship, Sex, and Survival 204 (1998) (“[T]he more you know a friend, the more a friend he is.”); id. at 216 (“What do we tell our friends? We tell them everything.”); Dean Cocking & Jeanette Kennett, Friendship and the Self, 108 Ethics 502 (1998) (explaining, though rejecting, the “secrets” view of the friend); and Allan Silver, Friendship and Sincerity, 4 Sozialersinn 123 (2003); Silver, supra note 20, at 1477 (“Friendship . . . turns on intimacy—the confident revelation of one’s inner self to a trusted other—and on essentially expressive and consummatory behavior.”).

97We may ask people to join our list of “friends” on Friendster or Facebook with a formal invitation, but these are not generally thought of as core examples of real friendship since nothing about these lists requires an actual relationship with any intimacy. Nevertheless, these invitations do sometimes bring the invitee into a circle of confidence, in which personal details are disclosed to an “in-group.” I discuss these issues more extensively in the first chapter of my forthcoming book, Friend v. Friend: The Transformation of Friendship and What the Law Has To Do with It.

98 See, e.g., Graham A. Allan, A Sociology of Friendship and Kinship (1979); Graham Allan, Friendship: Developing a Sociological Perspective (1989) [hereinafter Allan, Sociological Perspective]; Graham Allan, Kinship and Friendship in Modern Britain (1996) [hereinafter Allan, Modern Britain]; Claude S. Fischer, To Dwell Among Friends: Personal Networks in Town and City (1982); Anthony Giddens, The Transformation of Intimacy: Sexuality, Love & Eroticism in Modern Societies (1992); Ray Pahl, On Friendship (2000); Placing Friendship in Context (Rebecca G. Adams & Graham Allan eds., 1998) [hereinafter Context]; Liz Spencer & Ray Pahl, Rethinking Friendship: Hidden Solidarities Today (2006).

99 See, e.g., The Company They Keep: Friendship in Childhood and Adolescence (William M. Bukowski et al. eds., 1996); Conversations of Friends: Speculations on Affective Development (John M. Gottman & Jeffrey G. Parker eds., 1986); The Development of Children’s Friendships (Steven R. Asher & John M. Gottman eds., 1981); Beverley Fehr, Friendship Processes (1996); William K. Rawlins, Friendship Matters: Communication, Dialectics, and the Life Course (1992); Lillian B. Rubin, Just Friends: The Role of Friendship in Our Lives (1985).

100 See, e.g., Allan Bloom, Love and Friendship (1993); Caleb Crain, American Sympathy: Men, Friendship, and Literature in the New Nation (2001); Ronald A. Sharp, Friendship and Literature: Spirit and Form (1986); Wayne C. Booth, “The Way I Loved George Eliot”: Friendship with Books as a Neglected Critical Metaphor, Kenyon Rev., Spring 1980, at 4. See generally The Norton Book of Friendship (Eudora Welty & Ronald A. Sharp eds., 1991).

101 See, e.g., Alan Bray, The Friend (2003); C.S. Lewis, The Four Loves (1988); Gilbert Meilaender, Friendship: A Study in Theological Ethics (1981); Gene Outka, Agape: An Ethical Analysis (1972).

102 See, e.g., Hauke Brunkhorst, Solidarity: From Civic Friendship to a Global Legal Community (Jeffrey Flynn trans., Mass. Inst. Tech. Press 2005) (2002); Jacques Derrida, The Politics of Friendship (George Collins trans., Verso 2005) (1994); Love and Friendship: Rethinking Politics and Affection in Modern Times (Eduardo A. Velásquez ed., 2003); James R. Martel, Love Is a Sweet Chain: Desire, Autonomy, and Friendship in Liberal Political Theory (2001); Carl Schmitt, The Concept of the Political (George Schwab trans., expanded ed., Univ. Chi. Press 2007) (1932); Jacques Derrida, The Politics of Friendship, 85 J. Phil. 632 (1988); James V. Schall, Friendship and Political Philosophy, 50 Rev. Metaphysics 121 (1996); Jason A. Scorza, Liberal Citizenship and Civic Friendship, 32 Pol. Theory 85 (2004).

103 See, e.g., Robert Brain, Friends and Lovers (1976).

104 See, e.g., David Bolotin, Plato’s Dialogue on Friendship: An Interpretation of the Lysis, with a New Translation (1979); John M. Cooper, Aristotle on Friendship, in Essays on Aristotle’s Ethics 301 (Amélie Oksenberg Rorty ed., 1980); Horst Hutter, Politics as Friendship (1978); Immanuel Kant, The Metaphysics of Morals 261–64 (Mary Gregor trans., Cambridge Univ. Press 1991) (1797); Paul Schollmeier, Other Selves: Aristotle on Personal and Political Friendship (1994); Lorraine Smith Pangle, Aristotle and the Philosophy of Friendship (2003); Suzanne Stern-Gillet, Aristotle’s Philosophy of Friendship (1995); Simon Keller, Friendship and Belief, 33 Phil. Papers 329 (2004); Anthony Kronman, Aristotle’s Idea of Political Fraternity, 24 Am. J. Juris. 114 (1979); Sibyl A. Schwartzenbach, On Civic Friendship, 107 Ethics 97 (1996); Sarah Stroud, Epistemic Partiality in Friendship, 116 Ethics 498 (2006); Jennifer E. Whiting, Impersonal Friends, 74 Monist 3 (1991); see also sources cited supra note 79.

105 Samuel Scheffler, Valuing, in Equality and Tradition (forthcoming May 2010).

106Connecting valuing to emotional vulnerability is a feature of many philosophers’ accounts of value and valuing. See, e.g., Elizabeth Anderson, Value in Ethics and Economics 1–16 (1993); Niko Kolodny, Love as Valuing a Relationship, 112 Phil. Rev. 135, 152 (2003) (“[V]aluing a relationship involves being emotionally vulnerable to it . . . .”). Still, there are also alternative accounts. See Harry G. Frankfurt, The Importance of What We Care About, in The Importance of What We Care About 80 (1998) (focusing on caring and the idea of something being important to someone); T.M. Scanlon, What We Owe to Each Other 87–107 (1998) (defining friendship as a core example of value but not linking value with emotional vulnerability).

107 But see Jeanne L. Schroeder, Pandora’s Amphora: The Ambiguity of Gifts, 46 UCLA L. Rev. 815, 827−28 (1999) (“I argue that far from being characterized solely by the cold calculation of self-interest, markets are erotic in the Hegelian–Lacanian sense that they are driven by the desire for recognition. Contract, being mutual, reflects the true love relation in which recognition is freely granted and received by equals.”).

108Even this differential form of valuing is, however, probably contingent. See generally Katharine W. Swett, “The Account Between Us”: Honor, Reciprocity and Companionship in Male Friendship in the Later Seventeenth Century, 31 Albion 1, 4 (1999) (arguing that early-modern era male friendships were characterized by “potentially conflicting values” and “contradictory elements”).

109 See, e.g., Ira Ellman, Why Making Family Law Is Hard, 35 Ariz. St. L.J. 699, 711 (2003) (“The problem with [treating intimate friendships as contracts] is that it does not acknowledge that duties can arise from relationships themselves, apart from an exchange of promises that constitutes a contract.”). The problem with Ellman’s view is that his contract theory puts explicit “exchanges of promises” at its center; relational contract theory does not, so it is more amenable to acknowledging relational obligations as contractual ones.

110For a related argument about marriage and contract, see Ellman, supra note 37, at 1367.

111 See Peter Goodrich, The Immense Rumor, 16 Yale J.L. & Human. 199, 223 (2004) (“Banishing friendship to the rites of passage or ceremonies of death keeps it hygienically private and squarely in the safety of ‘another context.’”); Silver, supra note 20, at 1476 (“[F]riendship in modern society is a quintessentially private relationship, not normatively constituted by public roles and obligations—[and] indeed, often in distinction from them. . . .”); see also Brain, supra note 103, at 75 (“Friendship should be ‘free,’ ‘pure,’ and based on moral obligations alone.”); Pahl, supra note 98, at 63 (“There are no rules and contracts to bind us to our closest friends: we simply have to trust them.”). I have tried to complicate this story about friendship’s voluntariness in earlier work. See Leib, supra note 20, at 663–67.

112 See supra Part I.B, text accompanying notes 20–22. In particular, Bagchi argues that contracts risk “suffocating personal relationships that depend on their separateness from the civil order.” Bagchi, supra note 21, at 35. I take on the “debasement” thesis more directly infra Part IV.C.

113A version of this argument is usefully spelled out in Bellia, supra note 22.

114See E. Allan Farnsworth et al., Contracts: Cases and Materials 126 (6th ed. 2001) (citing Horsley v. Chesselet, a 1978 case in which the Municipal Court of San Francisco refused to enforce an agreement to go on a date, notwithstanding substantial reliance expenses).

115The details of this argument are deferred until Part IV.

116For a discussion of the taboo, see Alan Page Fiske & Philip E. Tetlock, Taboo Trade-Offs: Reactions to Transactions that Transgress the Spheres of Justice, 18 Pol. Psych. 255 (1997). However, that something is taboo and may provoke anger or disorientation for destabilizing a presumed social order doesn’t mean it is wrong. Nor does it mean that it is avoidable. Fiske and Tetlock’s argument actually suggests that it is almost impossible to avoid making comparisons between markets and social spheres, and that we do so implicitly all the time. See id. at 282, 292–94. In any case, part of my argument here assumes that the sphere of the “market” is not as stranger-based as is traditionally assumed, so we should be less concerned about thinking of the social sphere as having some apparent market traits. And the taboo does not seem to get triggered when one suggests, as I do and will further, infra Part III, that the market sphere of relational contracting partakes in many of the value structures of friendship. See id. at 257 (“[P]eople are less disturbed by applications of Communal norms to relations that they assume should be governed by prices (such an act may even seem ‘nice’).”).To the extent we reject thinking about friendship in this way because of the taboo, the taboo that political psychologists have charted is about “pricing” the value of friendship, which is not something my approach here requires. I am dealing in analogies, not full-scale commensurability. To say that friendships can be illuminated by thinking of them as relational contracts is not to sum up the value of friendship through a contract prism.Indeed, Joseph Raz probably goes too far when he claims that “[i]t diminishes one’s potentiality as a human being to put a value on one’s friendship in terms of improved living conditions.” Joseph Raz, The Morality of Freedom 353 (1986). If one could only value friendship in instrumental terms, it may be true that one could not quite participate in friendships. But merely to notice that friendship has instrumental components that contribute in part to its value even when practiced at its best hardly limits one from having friends in the deepest sense. To concede that elements of friendship are commensurable in some ways with certain “market” relations does not mean one cannot hold on to some weaker form of the incommensurability thesis.For a useful critique of the incommensurability thesis, see Eric A. Posner, The Strategic Basis of Principled Behavior: A Critique of the Incommensurability Thesis, 146 U. Pa. L. Rev. 1185 (1998). The symposium in which Posner’s manuscript appears is a helpful introduction to the subject. See generally Matthew Adler, Symposium: Law and Incommensurability, 146 U. Pa. L. Rev. 1169 (1998).

117Richard Lempert, Norm-Making in Social Exchange: A Contract Law Model, 7 Law & Soc’y Rev. 1, 2 (1972) (citing three “exchange theory” classics: Peter M. Blau, Exchange and Power in Social Life (1964); George Caspar Homans, Social Behaviour: Its Elementary Forms (1961); and George C. Homans, Social Behavior as Exchange, 63 Am. J. Soc. 597 (1958)).

118Lempert, supra note 117, at 2–3.

119 Allan, Sociological Perspective, supra note 98, at 19; see also Fiske & Tetlock, supra note 116, at 267: “[W]hen either party merely starts to keep track of how much they give and how much they get, a Communal Sharing relationship is in trouble. The very act of keeping accounts . . . seriously undermines the relationship. . . . Even to remind the other of asymmetries is unkind and distancing. . . .” I find this a bit too precious. Of course we keep accounts, even if just approximately. We are generally generous with one another in these relationships and hope for reciprocation over a long time horizon, since we are confident in a future. But when things get out of hand and one friend is doing a lot more than the other to keep the relationship going and intimate, no one should be surprised if one friend confronts the other for an accounting—or terminates the relationship.

120 Ellickson, supra note 9, at 235.

121 See Posner, supra note 5, at 151 (“Contracting parties are often friends.”). See generally DiMaggio & Louch, supra note 10 (examining the impact of friendships on consumer transactions); Ingram & Zou, supra note 10 (exploring “business friendships”); Ingram & Roberts, supra note 10 (arguing that friendships between business competitors are beneficial); Perrone et al., supra note 10; Uzzi, Financial Capital, supra note 10 (arguing that social ties promote trust that can facilitate lending relationships); Uzzi, Sources and Consequences, supra note 10 (exploring how social ties affect economic action in the apparel industry).

122A particularly useful study of this dynamic can be found in Sally Falk Moore, Law and Social Change: The Semi-Autonomous Social Field as an Appropriate Subject of Study, 7 Law & Soc’y Rev. 719 (1973); see also Nick Rumens, Working at Intimacy: Gay Men’s Workplace Friendships, 15 Gender, Work & Org. 9, 12 (2008) (“[I]ntimacy and instrumentality are sometimes distinct but often overlapping and temporal modes of relating. There may be many pathways to intimacy, some of which may originate within or are generated from the pursuit of instrumental endeavors.”).

123 Allan, Sociological Perspective, supra note 98, at 22.

124 See supra notes 77– 78.

125For more on these dimensions of friendship, see Allan, Modern Britain, supra note 98, at 89 (“Friendship, in whatever form it takes, is defined as a relationship between equals.”); Allan, Sociological Perspective, supra note 98, at 108 (“The essence of friendship from a sociological standpoint is that it is a tie of equality.”); Brain, supra note 103, at 20 (“Equality . . . is part and parcel of friendship.”); Friedman, supra note 79, at 211 (“[V]oluntariness in friendship seems to require, overall, a measure of roughly equal and mutual adaptation, a synergism achieved through the combined and mutually interested adjustments of those who are becoming, or are already, friends.”); Pahl, supra note 98, at 162 (arguing that friendship is “egalitarian”); Graham Allan, Friendship and the Private Sphere, in Context, supra note 98, at 71, 77 (recognizing the importance of “reciprocity” in friendship); and Pat O’Connor, Women’s Friendships in a Post-Modern World, in Context, supra note 98, at 117, 127 (“[F]riendship is a relationship between equals.”).

126 Ellickson, supra note 9, at 236.

127On wishing friends well for their own sake, see Aristotle, supra note 89, at bk. VIII, ch. 3, ll. 1156b10–b12; id. at bk. VIII, ch. 2, ll. 1155b33–b34. For the literature on friends as second selves, see, for example, id. at bk. IX, ch. 9, ll. 1170b5–b8 (“[A] friend is another []self.”) and Montaigne, supra note 96, at 141, 143 (arguing that friends are of “one soul in two bodies” and maintain a “fusion of . . . wills” and are a “second self”). Sometimes, this latter dimension of friendship is described as “agreement,” “concord,” or “consensus.” See Aristotle, supra note 89, at bk. IX, ch. 6, ll. 1167a22–b15 (noting that friends have similar judgments about their common interests); Epstein, supra note 96, at 60 (citing Cicero’s suggestion that friendship requires “agreement over all things divine and human”); Goodrich, supra note 111, at 211–12 (discussing concord and agreement).

128 But see generally Markovits, supra note 19 (arguing that contracts are collaborations that unify party interests). Markovits could, however, distinguish the unification of interests in contracts and the unification of interests in friendship by highlighting the limited scope of the unification in contracts (friendship requires a broader scope of unity) and the lack of affective association in contractual collaboration.

129This is why, perhaps, as friends get closer it makes greater sense to treat them as fiduciaries, not as “mere” relational contractors. For the argument to view close friends as fiduciaries, see Leib, supra note 22.

130I defend this view in Leib, supra note 20, at 646–47 (citing Epstein, supra note 94, at 240–41; Montaigne, The Art of Discussion, supra note 96, at 705; Pahl, supra note 98, at 86; Sullivan, supra note 96, at 204; Bernard Yack, The Problems of a Political Animal: Community, Justice, and Conflict in Aristotelian Political Thought 110 (1993)).

131 See Stuart Oskamp & Daniel Perlman, Effects of Friendship and Disliking on Cooperation in a Mixed-Motive Game, 10 J. Conflict Resol. 221, 221 (1966) (suggesting that friendship can sometimes discourage parties from cooperating in a cooperation game because of competitive urges); Graham M. Vaughan et al., Bias in Reward Allocation in an Intergroup and an Interpersonal Context, 44 Soc. Psychol. Q. 37, 40 (1981) (finding that young friends will allocate as much to their friends as to an arbitrary in-group).

132One can also see this morphology through the use of Markovits’s theory of contract as collaboration. Although he overstates the distinction between contracts and interpersonal relations because he thinks contracts cannot have an affective dimension, one can see through his theory (portions of which could be adapted by relationalists, mutatis mutandis) how relational contracts can establish a similar interest-structure that prevails in friendships. See Markovits, supra note 19, at 1432, 1451.

133 See Leib, supra note 29, at 2–3, 9–20.

134Perhaps I am throwing in the towel too quickly here. As Verity Winship observed in a conversation regarding a prior draft, we do have groups of friends to whom we relate qua group—and we also have relationships with couples (whether as couples ourselves or as individuals). These sorts of friendships might, after all, serve as plausible analogies for Type 2 and Type 3 contracts. I have not worked out how that might be played out through the rest of my analysis, but I think it is a provocative suggestion. Indeed, much of the economic sociology cited supra note 10 clearly assumes that friendships among organizations can be meaningful. See Ingram & Roberts, supra note 10, at 387. And a recent article about friendships between countries similarly suggests that organizational friendship is a meaningful concept. See P.E. Digeser, Friendship Between States, 39 Brit. J. Pol. Sci. 323 (2009).

135 See sources cited supra note 4.

136 See generally Nancy D. Polikoff, Beyond (Straight and Gay) Marriage: Valuing All Families Under the Law (2008) (arguing that the law should not impute more value to marriage than to other family forms); Dan Markel, Jennifer M. Collins & Ethan J. Leib, Criminal Justice and the Challenge of Family Ties, 4 U. Ill. L. Rev. 1147, 1153 (2007) (“[F]amily ties preferences can disrupt norms of equality that should otherwise prevail in an attractive regime of liberal governance.”).

137 See Laura A. Rosenbury, Friends with Benefits?, 106 Mich. L. Rev. 189, 191 (2007) (“The gendered prerequisites of marriage are seen as unrelated to family law’s goal of achieving gender equality, thereby permitting the boundaries of the legal family to continue to perpetuate gender inequality.”).

138Katherine M. Franke, Longing for Loving, 76 Fordham L. Rev. 2685, 2705 (2008). Franke’s very recent turn toward friendship is exciting for those of us who have been working in this area. I have qualms about some of her claims in the article about what friendship is—and about how she imagines that we ought to orient the law towards friendship. However, I am very sympathetic to her project of re-centering our lives and law toward friendship to displace the overwhelming role of marriage.

139 See Ariela R. Dubler, In the Shadow of Marriage: Single Women and the Legal Construction of the Family and the State, 112 Yale L.J. 1641, 1712 (2003) (“[M]arriage continues to regulate the terrain outside of its formal borders, preserving its legal and ideological supremacy as a normative model for all intimate relations and as an arbiter of which relationships deserve legal recognition and protection.”); Franke, supra note 138, at 2698 (“To borrow a term Janet Halley has recently put to much use, all of us are conscripted into the cause of ‘carrying a brief for’ marriage whether or not we so wish.” (quoting Janet Halley, Split Decisions: How and Why to Take a Break from Feminism 17 (2006)).

140On relational obligations more generally, see Samuel Scheffler, Relationships and Responsibilities, 26 Phil. & Pub. Aff. 189 (1997).

141I discuss gender and friendship in Leib, supra note 20, at 667–69. There, I make clear that it is actually controversial to see friendship as gender-neutral. Still, unlike marriage under federal law and the laws of most states, friendship is not gendered by law.

142 David D. Friedman, Law’s Order: What Economics Has to Do with Law and Why It Matters 172 (2000).

143Indeed, unlike marriage, there is rarely a moment of divorce in friendship. In the standard case, friends don’t really break up; it is a slow process of developing new, more intimate ties elsewhere. Still, although it is easy to point out when a marriage ends officially, the actual process of breaking up and growing apart over time does seem to resemble the exiting process of friendships. Unless there is a massive betrayal in either case, it is hard to say when things have turned irrevocably against the relationship’s durability.

144To be sure, they may have to turn to lawyers and may be prevented from free exit because they have to pay damages first—in both relational contracts and friendships. But the analogy with marriage doesn’t hold; no one needs the state’s permission to terminate these other types of relationships. Additionally, many relational contracts will be like friendships in yet another way: there will be no agreement to terminate, just a walking away, a growing apart, a fizzle. And that type of ending will also rarely trigger legal involvement. This is not so in marriages, where even amicable dissolutions require elaborate legal intervention.

145Scott & Scott, supra note 6, at 1255.

146 Posner, supra note 5, at 77.

147I discuss the potentially dyadic nature of friendship in Leib, supra note 20, at 644 n.50. More generally, in my earlier work I conclude that friendship must be “exclusive,” but that exclusivity is of a different type than the one that must prevail in marriage. See id. After all, infidelity in marriage is actually criminalized in many states and the District of Columbia. Jennifer M. Collins, Ethan J. Leib & Dan Markel, Punishing Family Status, 88 B.U. L. Rev. 1327, 1411 (2008) (citing Elizabeth F. Emens, Monogamy’s Law: Compulsory Monogamy and Polyamorous Existence, 29 N.Y.U. Rev. L. & Soc. Change 277, 290 n.49).

148For more on this feature of marriage, see Carolyn J. Frantz & Hanoch Dagan, Properties of Marriage, 104 Colum. L. Rev. 75, 81–98 (2004).

149For sources on equality in friendship, see supra note 125.

150 See Scott & Scott, supra note 6, at 1270–71.

151 See, e.g., Posner, supra note 5, at 79.

152 See generally Leib, supra note 22 (arguing that close friends can be thought of as fiduciaries). It should be noted that there is a danger in having most or many contracts trigger the fiduciary duties we tend to associate with spouses; we might cause a “stifling despotism of virtue” within contractual relations. See Roberto Mangabeira Unger, The Critical Legal Studies Movement, 96 Harv. L. Rev. 561, 641 (1983). This is part of why it is generally best to see relational contracts as “mere” friendships (and vice-versa)—and only very close friends as fiduciaries.

153Common law marriage is clearly the exception to this rule. But common law marriage is rare (it exists in nine states and the District of Columbia) and getting rarer still even though five states have grandfathered couples under earlier, and now repealed, common law marriage laws. See Nat’l Conference of State Legislatures, Common Law Marriage (2009), http://www.ncsl.org/programs/cyf/commonlaw.htm. Moreover, even if a couple “marries” through a common law marriage, the parties must make formal showings to receive treatment as a married couple (by, for example, changing their last names to match one another’s, filing joint tax returns, or holding themselves out to the public as married), and may only dissolve their relationship through a formal divorce.

154 See Eisenberg, supra note 21, at 817–18.

155 Compare Posner, supra note 5, at 69–70, 160 (embracing, from the point of view of a legal economist, the model of relational contract as marriage), and Scott & Scott, supra note 6 (discussing how marriage can be thought of as a relational contract), with Posner, supra note 50 (endorsing formalism from the same point of view of a legal economist who embraces the relational-contract-as-marriage model), and Scott, supra note 8 (embracing formalism from the perspective of a legal economist who embraces the relational contract-as-marriage model). I should distinguish formalism about formation from formalism about interpretation. Although these formalists embrace formalism across the board, here I am only talking about formalism about formation.

156 But see Haw. Rev. Stat. §§ 572c-1 to -7 (2009) (illustrating a “reciprocal–beneficiary” statute allowing parties to contract into a form of friendship); Vt. Stat. Ann. tit. 15, §§ 1301–1306 (2002) (also a “reciprocal–beneficiary” statute); Brain, supra note 103, 106–07 (urging more ceremony surrounding friendship formation); David L. Chambers, For the Best of Friends and for Lovers of All Sorts, A Status Other than Marriage, 76 Notre Dame L. Rev. 1347, 1348 (2001) (arguing for a “designated friend” registry, enabling the state to keep track of who is entering friendships); Leib, supra note 20, at 683 n.267 (explaining the German custom of Brüderschaft trinken, which is a friendship ceremony of sorts and considering the value of establishing more ceremonies for entering friendships). Under the Hawaii statute cited above, “reciprocal beneficiary status [is] available to heterosexual couples unable to marry, such as brother-and-sister, father-and-daughter, and mother-and-son couples, as well as to same-sex couples. . . . [and provides] many benefits available to married couples including . . . the right to hold property by tenancy of the entireties, eligibility for health insurance coverage as family members, wrongful death and loss of consortium claims, [and] family and funeral employment leave. . . .” Lynn D. Wardle, Counting the Costs of Civil Unions: Some Potential Detrimental Effects on Family Law, 11 Widner J. Pub. L. 401, 415 (2002) (footnote omitted). The Vermont statute allows “couples related by blood or adoption to register in a new nonmarital status and receive nominal marital rights and benefits, mostly relating to health care and final medical decisions.” Id. at 417. See also Rosenbury, supra note 137 at 223 (“The status of designated friends could therefore be described as a ‘marriage–lite’ approach.” (quoting J. Thomas Oldham, Lessons from Jerry Hall v. Mick Jagger Regarding U.S. Regulation of Heterosexual Cohabitants, or, Can’t Get No Satisfaction, 76 Notre Dame L. Rev. 1409, 1430-33 (2001)).

157 See Posner, supra note 5, at 77.

158 See generally Friedman, supra note 79, at 208–09 (“Friendship receives far less formal recognition in our rituals and conventions . . . than do familial relationships and is maintained by fewer of those formalities.”). Although I have argued, see Leib, supra note 20, at 685–94, and will argue infra Part IV, for establishing some legal obligations that emanate from friendship, the ones I have in mind here to establish the analogy are only ethical obligations.

159 See, e.g., Balfour v. Balfour, (1919) 2 Eng. Rep. 571, 575 (K.B.) (finding agreement by separated husband and wife for support unenforceable); Miller v. Miller, 35 N.W. 464, 464 (Iowa 1887) (finding written agreement between husband and wife unenforceable). For a largely supportive argument in favor of Balfour, see Stephen Hedley, Keeping Contract in Its Place—Balfour v. Balfour and the Enforceability of Informal Agreements, 5 O.J.L.S. 391 (1985). For other (not quite relational) approaches to Balfour, see Michael Freeman, Contracting in the Haven: Balfour v. Balfour Revisited, in Exploring the Boundaries of Contract 68, 68–82 (Roger Halson ed., 1996); Peter Goodrich, Friends in High Places: Amity and Agreement in Alsatia, 1 Int’l J. L. Context 41, 44–46 (2005).

160 See Posner, supra note 5, at 68. However, Posner readily concedes that the “autonomy” description is an odd one—and hard to reconcile with how the law actually regulates the family. Id. at 68–69.

161 See, e.g., Wightman, supra note 4, at 94; Mary Keyes & Kylie Burns, Contract and the Family: Whither Intention?, 26 Melb. U. L. Rev. 577, 578 (2002).

162 See, e.g., Aristotle, supra note 89, at bk. VIII, ch. 13, ll. 1162b22–b32; id. at bk. IX, ch. 1, ll. 1164b12–b15 (discussing how some cities do not allow legal actions on contracts between friends); Franke, supra note 138, at 2705.

163An area of likely dispute between relationalists and non-relationalists would be whether friends can generate legally enforceable obligations to each other without formalities. See infra note 175. The relationalists might allow it because (some) social norms are meant to be incorporated by the law—and relationalists don’t fetishize promissory or formal obligation as the core of contract. The non-relationalist would have none of this. I pursue this territory infra Part IV.

164 See generally Lloyd L. Weinreb, Legal Reason: The Use of Analogy in Legal Argument (2005). Perhaps those who have been urging the relational-contract-as-marriage analogy would respond by saying, “We weren’t expecting you to take us so seriously! It is just an analogy for goodness’ sake.” Of course, that is a viable retort; my sur-reply would focus on why friendship is the better analogy, essentially Part III’s argument.

165 See Posner, supra note 5, at 150–51; accord Posner, supra note 50, at 757 (“Business is almost always conducted in a highly social manner.”). But see id. at 761–62 (recognizing that these business friendships are not “real” friendships); Moore, supra note 122, at 727 (calling business friendships “fictive”); Macaulay et al., supra note 23, at 230 n.3 (“A lawyer who deals with contracts and fails to understand the power and the limits of trust and the social sanctions flowing from ‘fictive friendships’ is incompetent.”).

166Posner, supra note 50, at 756–57.

167See supra note 165.

168However, at least some business school professors have not missed the boat here. See Ingram & Roberts, supra note 10, at 387; Ingram & Zou, supra note 10, at 167; DiMaggio & Louch, supra note 10, at 619; Perrone et al., supra note 10, at 422; Uzzi, Financial Capital, supra note 10, at 481; Uzzi, Sources and Consequences, supra note 10, at 674.

169Indeed, this is the strategy of Ellman, supra note 37, discussed in Wightman, supra note 4, at 105–06. Ellman concludes that the contract concept cannot be applied between intimates (his particular concern is co-habiting lovers, not friends) because contracts are about financial gain, but personal relationships are not. Or, put another way, in intimate relationships, the relationship is itself the goal, so there are no external standards to apply to the relationship when a dispute arises. From the perspective of relational contract theory (which Ellman simply dismisses with reference to Eisenberg’s inadequate critiques), this is much too simplified a story.

170For more on this story and reasons to reject it in its simplified form, see Zelizer, supra note 22.

171Indeed, in my own analysis of Markovits’s theory of contract, I focus on this shortcoming to his view. See generally Leib, supra note 29 (elaborating on why Markovits’s theory of contract does not have a neat application to very core cases of contracts that involve organizations).  

172Markovits, supra note 19, at 1465.

173This idea of a “second” contract law (or a “third,” for all three types of contracts to which I alluded above) is suggested as a possibility by Markovits. See generally id. (examining whether organizations and individuals require their own type of contract law). It is even developed beyond mere suggestion in Meir Dan-Cohen, Rights, Persons, and Organizations: A Legal Theory for Bureaucratic Society 82–84, 96–102 (1986). For my discussion about this potential approach, see Leib, supra note 29, at 11–12.

174This approach was taken in Bates v. Southgate, 31 N.E.2d 551, 558 (Mass. 1941), where the court found that an underlying friendship between contracting parties may excuse a party from carefully investigating a form contract for exculpatory clauses because of the good will of a friendship, and Estes v. Magee, 109 P.2d 631, 633–34 (Idaho 1940), where the court found that a doctor’s release form was inapplicable against a patient who signed only because he relied on his doctor’s friendship. A more general “right to rely” on a friend that can be enforced as a matter of contract law is supported in Spiess v. Brandt, 41 N.W.2d 561, 566–67 (Minn. 1950); Bank Leumi Trust Co. v. Luckey Platt Ctr. Assocs., 665 N.Y.S.2d 976, 978 (N.Y. App. Div. 1997); Callahan v. Callahan, 514 N.Y.S.2d 819, 821–22 (N.Y. App. Div. 1987); Liebergesell v. Evans, 613 P.2d 1170, 1176 (Wash. 1980); and Graff v. Geisel, 234 P.2d 884, 890 (Wash. 1951). So it turns out that my twin models here are actually consistent with law as applied in a range of cases.

175Thanks to Ira Ellman for his suggestions regarding this hypothetical.

176 See, e.g., Cal. Civ. Code §§ 1995.210–.270 (West 2009).

177 See generally Kim Atkins, Friendship, Trust, and Forgiveness, 29 Philosophia 111, 111 (2002) (arguing for a conception of forgiveness as the outcome of a process of mutuality rather than something brought about solely by the actions of an individual).

178 See, e.g., Melvin Eisenberg, The World of Contract and the World of Gift, 85 Cal. L. Rev. 821, 847–49 (1997).

179 See, e.g., Keyes & Burns, supra note 161, at 587 (“Relational researchers have demonstrated that in business relationships . . . ‘cooperation without reference to legal entitlements is normal.’” (quoting Hedley, supra note 159, at 396)).

180 Restatement (Second) of Contracts § 21 (1981).

181 See Macneil, RCT: C & Q, supra note 16, at 902 n.96 (suggesting that Eisenberg has in mind a concern about the legal enforcement of “tacit household patterns about who takes out the garbage”). Whether one needs a clear test to carve out these trivial cases is a matter of debate. The perspective of relational contract as friendship supports Feinman’s response on this front that not all social bonds are legally enforceable, but we can only find the enforceable ones if we are willing to look at the bonds and duties in the first place. See Feinman, supra note 15, at 748.

182Assume, of course, that you have ignored Shakespeare’s (or Antonio’s) advice:If thou wilt lend this money, lend it notAs to thy friends; for when did friendship takeA breed for barren metal of his friend?But lend it rather to thine enemy,Who, if he break, thou mayst with better faceExact the penalty.William Shakespeare, The Merchant of Venice act 1, sc. 3, ll. 133–38 .

183 See Scott & Scott, supra note 6, at 1266 (arguing that friendship allows longer time horizons for debt repayment).

184Consider Virgin Money’s recent effort to develop financial products to structure loans to friends. See Virgin Money, http://www.virginmoneyus.com (last visited Sept. 21, 2009). It calls some of its services “Handshake Basic” and “Handshake Plus.” The former comes “smiles included,” and the latter “lowers chance of default” and “raises chance of glee.” Virgin Money, Social Loans Borrowers, Lenders, and Curious Inquirers, Welcome, http://www.virginmoneyus.com/PersonalLoans/tabid/54/Default.aspx (last visited Sept. 21, 2009). Virgin Money’s motto is “to provide customers . . . a top-notch private loan experience that does good things for both [the] pocket and [the] relationship.” Washington Diamond, Virgin Money, http://www.washingtondiamond.com/Financing-Options_2 (last visited Nov. 24, 2009). It appears that Virgin bought Circle Lending, which was a company exclusively devoted to these sorts of financial products, in which the default rate is unusually high (perhaps because people assume their friends won’t sue them on the default). For a story on Circle Lending (and the high default rate among friends), see Avoiding the Pitfalls of Family Borrowing (NPR radio broadcast Oct. 6, 2006), available at http://www.npr.org/templates/story/story.php?storyId=6208227&sc=emaf.

185The proposal that contract law should enforce any promise with a commercial purpose or in furtherance of economic activity is usually attributed to Daniel A. Farber & John H. Matheson, Beyond Promissory Estoppel: Contract Law and the “Invisible Handshake, 52 U. Chi. L. Rev. 903, 905, 937–38 (1985). Farber and Matheson acknowledge that their argument might support extending enforceability into intimate contexts.

186Indeed, this reality serves as a useful counterpoint to one of the arguments—given expression in Balfour—offered by non-relationalists against the legal incorporation of social obligations: that the floodgates of litigation will be forced wide open, burdening the courts. Given the cost of litigation, this just is not much of a serious concern and helps reinforce the idea that private promises will tend to stay private whatever position the law takes about enforceability. See Keyes & Burns, supra note 161, at 585 (“Existing procedural mechanisms and costs . . . deter individuals from litigating unworthy claims.”).

187What counts as “commerce,” in any case, has long been a puzzle for the law. The Commerce Clause jurisprudence of the Supreme Court is as good a proof as any that definitions of commerce are ever-changing and quite hard to pin down. See U.S. Const. art. I, § 8, cl. 3; United States v. Lopez, 514 U.S. 549 (1995).

188Some evidence that litigation over season tickets is not unusual can be found in the following line of cases: Lipton v. Donnenfeld, 773 N.Y.S.2d 82 (N.Y. App. Div. 2004); Brand v. Lipton, 711 N.Y.S.2d 486 (N.Y. App. Div. 2000); Kane v. Berken, 600 N.W.2d 55 (Wis. Ct. App. 1999) (unreported table decision); Kully v. Goldman, 305 N.W.2d 800 (Neb. 1981); Davey v. King, 595 A.2d 999 (D.C. 1991); and Tauber v. Jacobson, 293 A.2d 861 (D.C. 1972). Thanks to Joe Perillo for triggering my interest in this litigation context.

189There is a certain parallel here with the line of cases that prevents employers from easily firing salespeople without paying them earned commissions, finding such termination practices breaches of the implied duty of good faith. See, e.g., Fortune v. Nat’l Cash Register Co., 364 N.E.2d 1251 (Mass. 1977); Caton v. Leach Corp., 896 F.2d 939, 946 (5th Cir. 1990). This line of cases is cited and discussed in Steven J. Burton & Eric G. Andersen, Contractual Good Faith: Formation, Performance, Breach, Enforcement §§ 3.2.2, 3.4.1 (1995).

190I will discuss this limitation to the enforcement of friendship and relational contracts more generally infra Part IV.B, since this claim contravenes some relationalists’ normative conclusions about how to handle relational contracts.

191 See Fiske & Tetlock, supra note 116, at 280 (suggesting that some obligations in the social sphere carry with them an assumption that they can be deferred or canceled upon certain changes of circumstance).

192The consideration doctrine would indeed prevent many promises within a friendship from becoming legally enforceable. See, e.g., Kim v. Son, No. G039818, 2009 WL 597232 (Cal. Ct. App. Mar. 9, 2009) (finding a promise literally written in blood between friends to repay money lost in a bad investment unenforceable because the promise wasn’t bargained-for). But it isn’t clear that the consideration doctrine ought to be used this way—especially in cases where it seems easy to say that the evidentiary, channeling, and cautionary functions of the doctrine have been met. See generally Lon L. Fuller, Consideration and Form, 41 Colum. L. Rev. 799 (1941) (discussing the rationale of legal formalities and an examination of the policies of the common law consideration doctrine).

193 See Ellman, supra note 37, at 1375 (“Mutual gifting arising from mutual concern and affection is not the same as bargained-for exchange.”).

194 See generally Restatement (Second) of Contracts § 90 (1981). Notably, in developing a remedy in this sort of context, we probably cannot be more precise than § 90(1) already is: “[t]he remedy granted for breach may be limited as justice requires.”

195This argument draws from Wightman, supra note 4, at 125 (discussing non-bargain contract cases and limiting them to intimate associations in which gift giving is reciprocal).

196Posner, supra note 78, at 567.

197For a related argument about the role of reciprocity in the enforcement of intimate obligations in the will context, see Melanie B. Leslie, Enforcing Family Promises: Reliance, Reciprocity, and Relational Contract, 77 N.C. L. Rev. 551 (1999). To be sure, we cannot lose sight of the different ways in which gifts are used in different relationships and how their value changes depending on the context. Frank Flynn’s work pursues this nuance, demonstrating that we should not treat all gift exchanges the same way, even within friendships. See, e.g., Ames, Flynn & Weber, supra note 77 (explaining that gift givers may have different reasons for giving gifts such as affect, role, or cost–benefit calculations).

198The practice is described and discussed in Posner, supra note 5, at 151, and in Eric A. Posner, The Regulation of Groups: The Influence of Legal and Nonlegal Sanctions on Collective Action, 63 U. Chi. L. Rev. 133 (1996).

199Posner, supra note 197, at 169–70. For a classic analysis of these practices, see Clifford Geertz, The Rotating Credit Association: A “Middle Rung” in Development, 10 Econ. Dev. & Cultural Change 241 (1962).

200One bankruptcy court has held that obligations under a rotating credit agreement are legally enforceable. In re Ray, 51 B.R. 454, 460–61 (Bankr. D. Haw. 1985). Despite the fact that these agreements are quite common in the United States, there is remarkably little reported litigation to guide courts when confronting such obligations. Perhaps this is evidence that non-legal sanctions are functioning effectively. But perhaps not; fear of legal sanctions or the legal system may also be keeping these disputes out of court.

201For analysis of one such case, see Posner, supra note 198, at 173–75 (highlighting a case in which a court opted to dismiss a claim trying to enforce a rotating credit agreement on the ground that it was an illegal lottery). More generally, though, it is probably hard to say with any degree of certainty that parties contemplate only non-legal sanctions upon default. See id. at 170 (“[A]mong West Indian immigrants in New York City [participating in credit circles], ‘so rare was default that when queried, most organizers did not know how they would have dealt with such a situation.’” (quoting Aubrey W. Bonnett, Institutional Adaptation of West Indian Immigrants to America: An Analysis of Rotating Credit Associations 63–64 (1981))).

202 See Barnett, supra note 8, at 1190–91 (arguing that what differentiates a contract theory from a social theory is an account of when an exchange relationship merits legal protection).

203 See Balfour v. Balfour, (1919) 2 Eng. Rep. 571, 579 (K.B.) (Atkin, L.J.) (discussing the intent of a husband and wife to form a binding contract where the husband merely promised to give his wife a monthly stipend and observing that “the promise here was not intended by either party to be attended by legal consequences”).

204 See Farber & Matheson, supra note 185, at 904–05 (“[A]ny promise made in furtherance of an economic activity is enforceable.”).

205 See Macneil, RCT: C & Q, supra note 16, at 906. Many disputes between friends probably find themselves in small claims court, where mediation and arbitration are quite common dispute resolution techniques. But there is no question that the law has jurisdiction and that the courthouse doors are open for friends to sue one another.

206Indeed, although a fuller discussion is beyond the scope of this Article, it suffices to say that viewing relational contract as a friendship and friendship as a relational contract may help us see why “contract thinking” was not a “fatal flaw” at all in the cohabitation cases like Marvin v. Marvin, 134 Cal. Rptr. 815 (Cal. 1976). Contra Ellman, supra note 37, at 1365, 1367. Ellman criticizes cases like Marvin for failing to apply more status-oriented obligations to generate the liabilities the courts find in cohabitation cases through implied contracts. Id. at 1375–77. Through the lens of relational contract theory—especially the form developed here—it is easy to see how the theory incorporates precisely the type of status-based liabilities Ellman prefers. His conception of contract is so narrow that he fails to see how capaciously one can plausibly and convincingly draw “contract thinking.”

207It is probably worth noting here that friendship can play numerous roles in establishing liabilities. The friendship can serve an evidentiary function (friendship proves there is a best efforts clause, which does not otherwise appear in the paper deal); it can sometimes create the obligation itself through implication and conduct (the friendship underwrites a finding of breach because friends would not rely upon or defend their actions on the basis of caveat emptor); and it can sometimes be relevant in fashioning a remedy (because friends betrayed one another, a specific performance order is especially unlikely to result in successful repair).

208There is some circularity here, of course. I used the requirements of good faith in both friendship and relational contracts as a reason to adopt the analogy in the first place, so it is a form of cheating to use the analogy on the back end as a reason to embrace the good faith requirement in relational contracting. Although I concede this point of circularity, it is worth noting that since the analogy also works on so many other levels (and on levels that create the need for the good faith requirement), it seems fair to say that the twin models ultimately “support” this normative dimension of relational contract law. In any case, when I say the models “support” a normative prescription, I do not necessarily mean that they furnish independent reasons to embrace the normative prescription, but rather that they are consistent with the prescription in ways that help us understand it better as a feature of relational contract theory.

209 See Ingram & Zou, supra note 10, at 170 (“In a business context, the benefit of relational exchange is to lower transaction costs, and related, to enable some transactions which would be otherwise impossible.”). But see Cross, supra note 22, at 1501 (arguing that sometimes resort to legal formalities and sanctions is necessary to get relational contracts off the ground in a commercial context outside of close-knit groups). Even in cases in which relational contracts can only get off the ground with some resort to legal formalities, there is no reason to require them (as non-relationalists would) when parties otherwise do not want them because, presumptively, the parties view formalities as undermining the relationship.

210Larry E. Ribstein, Are Partners Fiduciaries?, 2005 U. Ill. L. Rev. 209, 235. This is, admittedly, an ironic use of Ribstein.

211Eisenberg, supra note 21, at 817–18.

212 See generally Leib, supra note 22 (arguing that courts should sometimes apply a tailored law of fiduciaries to enforce legal duties between friends).

213One can see this dimension of relational contract theory on display in Speidel, supra note 8, at 839, 846.

214To be sure, Speidel highlights that even using law to preserve the relationship does not really deprive parties of their underlying right to agree to terminate by negotiating their way out of obligation. Id. at 846. But from the perspective of relational contract as friendship, a relational contract law should allow parties to break up easily if that is what they want to do—so long as they are no longer in debt to one another.

215There is some inchoate research in business schools and sociology departments that investigates the deterioration of business friendships (though focused largely on workplace friendships)—and such research may help courts and scholars develop protocols for dealing with contractual breaches in light of deteriorating business friendships. See Ingram & Zou, supra note 10, at 181 (citing Patricia M. Sias et al., Narratives of Workplace Friendship Deterioration, 21 J. Soc. Pers. Relationships 321 (2004); Patricia M. Sias & Tara Perry, Disengaging from Workplace Relationships—A Research Note, 30 Human Comm. Res. 589 (2004)).

216For a nice approach to the debasement thesis from within the philosophy of friendship, see Neera K. Badhwar, Friendship and Commercial Societies, 7 Pol. Phil. & Econ. 301 (2008). Although Badhwar is concerned with a somewhat different theoretical concern—that market societies generally debase friendship—her response to it (that market relationships and friendships actually have much in common) is pertinent here.

217The “crowding” thesis is explored in, for example, Bruno S. Frey, Not Just for the Money (1997); Iris Bohnet et al., More Order with Less Law: On Contract Enforcement, Trust, and Crowding, 95 Am. Pol. Sci. Rev. 131 (2001); and Larry E. Ribstein, Law v. Trust, 81 B.U. L. Rev. 553, 568–71 (2001). See also sources cited supra note 22.

218 See generally Jay B. Barney & Mark H. Hansen, Trustworthiness as a Source of Competitive Advantage, 15 Strategic Mgmt. J. 175 (1994) (developing the concept of “strong form trust”); see also Ribstein, supra note 217, at 558–68.

219Admittedly, business school professors already seem to know that formal contracts and informal norms can complement rather than substitute for one another. See supra note 64. It is not fully clear why there is virtually no uptake of this work within the legal academy.

220 See generally Zelizer, supra note 22 (undermining the “separate spheres” story).

221 But see Scott & Scott, supra note 6, at 1247 (“[C]ontract and commitment are quite compatible.”); id. at 1330 (“[L]egal enforcement of long term commitments is a particularly valuable method of bolstering extralegal norms.”).

222This paragraph (and some of what follows) draws upon my discussion of the crowding thesis in Leib, supra note 22, at 726–32. In that work, I also develop an argument about legal enforcement’s “information-forcing” dimension within friendships.

223For a discussion of the assumption underlying the debasement thesis, see Bagchi, supra note 21, at 3–5 and Kimel, supra note 3, at 253–55.

224Ribstein, supra note 214, at 567.

225 Id. at 568 (“[L]aw must be regarded as a substitute for rather than complement of social capital because it undermines the institutions that create it.”).

226 Annette C. Baier, Trust and Antitrust, in Moral Prejudices: Essays on Ethics 118, 139 (1994).

227Lawrence E. Mitchell, Trust. Contract. Process., in Progressive Corporate Law 185, 204 (Lawrence E. Mitchell ed., 1995).

228 Posner, supra note 5, at 63.

229 See generally Melvin A. Eisenberg, The Theory of Contracts, in The Theory of Contract Law 206 (Peter Benson ed., 2001) (arguing for a form of the debasement thesis); Bagchi, supra note 21, at 19 (citing Meir Dan-Cohen, In Defense of Defiance, 23 Phil. & Pub. Aff. 24 (1994)); Eisenberg, supra note 178, at 850.

230Bellia makes a similar argument, relying on friendship as the quintessential interpersonal relationship. See, e.g., Bellia, supra note 22, at 36 n.38.

231These are not controversial claims requiring elaborate citation. See generally Posner, supra note 5, at 164 (showcasing the ways that contract law’s remedies are under-compensatory).

232For this reason, although I am largely sympathetic to Stephen A. Smith’s effort to see contract damages as still leaving room for parties to act on non-legal reasons, see Stephen A. Smith, Performance, Punishment and the Nature of Contractual Obligation, 60 Mod. L. Rev. 360, 362–63, 368, 370 (1997), I do not think specific performance needs to be seen as an exceptional remedy from this perspective. Specific performance is also under-compensatory, leaving room for non-legal sanctions to do their work.

233 But see Kimel, supra note 3, at 235–43 (specifying an “empirical–doctrinal” argument from within the relationalist camp, which aims to show that “contract law already recognizes and is significantly informed by the relational nature of contracting”). For additional articles developing this line of argumentation, see for example, David Campbell, The Relational Constitution of the Discrete Contract, in Contract and Economic Organisation 40 (David Campbell & Peter Vincent-Jones eds., 1996) and Campbell & Collins, supra note 55, at 25–32.

234Accordingly, Bagchi’s recent effort to reserve only reliance-based recovery for those situations in which seemingly social promises are deemed enforceable is an unnecessary and unwelcome overlay where the law of remedies is concerned. Bagchi, supra note 21, at 5. As I have endeavored to explain here, all relational contracts will have a complex matrix of social promises and more formal ones. The law cannot and should not build an artificial separation between these spheres (though it will have to sort out with common sense which promises and duties to enforce and which not to), whether through the law of remedies or otherwise. Since the law of remedies leaves social obligations to remedy themselves in part in all cases—even cases in which the law intervenes—those who want to have an untouchable symbolic space for non-legal relationships already have it in the law of remedies. Forcing these obligations into a reliance box within the law of remedies is unnecessary and pays too much homage to a promise-based account of contract. Moreover, the urge to keep expectancy damages “pure” of promises that occur between intimate friends misconstrues expectancy as much more fully compensatory than it actually is.

235 Raz, supra note 116, at 211.

236 See generally Richard Craswell, When Is Willful Breach ‘Willful’? The Link Between Definitions and Damages, 107 Mich. L. Rev. 1501 (2009) (discussing bad faith as a basis for a remedy in contractual breaches and as a basis for liability in establishing breach).

237 See Shiffrin, supra note 21, at 709–13.