Southern California Interdisciplinary Law Journal

Fall, 1993

Symposium on Default Rules and Contractual Consent

Comment

*311 THE BOUNDARIES OF LEGAL DISCOURSE AND THE DEBATE OVER DEFAULT RULES IN

CONTRACT LAW

Lawrence B. Solum [FNa1]

The doctrine of law, taken from nature's strict confines, presents an immense field for human study. But the notions of law and of justice, even after having been treated by so many illustrious authors, have not been made sufficiently clear.

Gottfried Wilhelm Von Leibniz (1693) [FN1]

This essay analyzes the connection between two debates. The first, a debate in the theory of contract law, concerns the role of default rules as a device for specifying the obligations of the parties to a contract. The second, a debate in jurisprudence, addresses the question of the nature of the boundaries of legal discourse. These debates may be connected. The correct understanding of the limits of legal discourse arguably entails the conclusion that the default rules debate is based on confusion and mistake. The thesis of this essay is that, contrary to that conclusion, the best account of the boundaries of legal discourse underwrites the cogency of most of the debate over default rules in contract.

The debate over the role and content of default rules in contract doctrine has been a central focus of recent contract scholarship. [FN2] *312 Most of the contributions to this symposium exemplify the ongoing debate. [FN3] However, Dennis Patterson's contribution takes a somewhat different tack; he argues that the default rules debate is a pseudodebate, based on a fundamental misunderstanding of the nature of legal discourse and its relationship to other discourses such as politics, philosophy, and economics. [FN4]

Professor Patterson would concede that the notion of default rules is unobjectionable, insofar as it describes the rules of their contract. While the parties to a contract have the power to override default rules, they also lack the ability to alter the mandatory rules of contract. [FN5] However, this is not the sense in which Patterson objects to the debate over default rules in contract scholarship. Rather, Patterson objects to the idea that whenever a court is faced with a gap--missing terms--in a contract, the court should apply or devise a default rule drawn from economics, philosophy, political theory, or some other discipline. [FN6]

*313 Thus, contract doctrine employs the idea of default rules for two distinct purposes. First, the notion of default rules simply distinguishes those rules which are mandatory from those which can be altered by the parties. Second, the notion can be used as the hook upon which to hang a more general theory of contract law. Patterson's argument against the latter more ambitious and theoretical use of the default rule notion relies on a distinction between "propositions of law" and "propositions about law." [FN7] The former are legitimate tools in legal discourse while the latter are illegitimate justifications for a legal decision. "Any given instance of legal justification is legitimate if it proceeds through the use of legal forms of argument, otherwise it lacks legitimacy. A conclusion that some proposition of law is true cannot be legitimate if the conclusion is reached through the use of nonlegal forms of argument." [FN8]

Patterson argues that normative propositions of economics, political theory, or philosophy that evaluate legal rules are not propositions of law; they are propositions about law. The attempt to use propositions from outside law to support conclusions inside law involves a sort of category mistake--the attempt to make moves within the conventions of legal discourse based on the conventions of economics, politics or philosophy. [FN9] It is as if one were to try to win a *314 baseball game by throwing a touchdown pass. We now turn to the assumptions which underlie Patterson's argument.

I. THE BOUNDARIES OF LEGAL DISCOURSE

This Part begins by specifying the issue raised by Patterson's attack on the default rules debate and the nature of the boundaries of legal discourse. We then examine three models of the boundaries: the hermetic model, the monadic model, and the holistic model. Finally, the conclusions drawn from this examination of the models are applied to Patterson's critique of the default rules debate.

A. THE ISSUE: WHAT IS THE NATURE OF THE BOUNDARIES OF LEGAL DISCOURSE?

Under Patterson's broad conception of the practice of law, which includes lawyering and judging, law is a bounded realm of discourse. Thus, the justification for a proposition of law must come from within the conventions and practices that formally constitute legal discourse. Patterson advances his view as follows: "Legitimacy is a matter of using the form of justification appropriate to the proposition in question. Illegitimacy in justification results from the use of forms of argument from one discourse to justify a proposition of another discourse." [FN10] This passage suggests a picture of the boundaries of legal discourse. Legal discourse has its own forms of argument, which distinguish it from other forms of human discourse.

There are, however, quite different pictures of the boundaries of legal discourse. Natural law, in its classical, Aristotelian form, may suggest one such picture. [FN11] Paraphrasing broadly in the language of contemporary moral, political, and legal philosophy, we might summarize Aristotle's account of the nature of law as follows: Laws are socially enforced rules that contribute to the flourishing of humans and their communities. Such rules can be divided into two classes. The first class consists of the rules that are required by the principles *315 of natural justice--those laws which every human society must promulgate and enforce if it is to flourish. Laws against wanton murder might be an example of rules of this first class. The second class consists of rules that concern conventional justice--those laws which can vary from society to society, but nevertheless contribute to human flourishing. Traffic rules, such as the rule requiring drivers in California to drive on the right side of the road, might be examples of rules of the second class. Given this Aristotelian view, the legal validity of a social rule depends on whether the rule contributes to human flourishing. If a given rule is contrary to the principles of natural justice, it would follow that it is not a valid law-- although it might well be a human ordinance that the existing authorities claim is a valid law.

If this Aristotelian view of the nature of law were correct, it would follow that the legitimate justification of law would not exclude the forms of argument appropriate to moral philosophy, political theory, and economics. Those discourses address the question whether a given socially enforced rule would contribute to human flourishing. It would seem, therefore, that the conclusions of political, moral, or economic arguments would bear directly on the question whether a given socially enforced rule is in fact a law.

Thus, different legal theories provide different answers to the question, What is the nature of the boundaries of legal discourse? This section of the paper articulates three models of the boundaries of legal discourse. It argues that one of those models, the holistic model, best describes these boundaries. It then applies that model to the question of the relationship of legal discourse to other forms of discourse.

B. THREE MODELS OF THE BOUNDARIES OF LEGAL DISCOURSE

Three models will be employed to describe the boundaries of legal discourse. The first model, called the hermetic model, assumes that the boundaries of legal discourse are impervious or hermetically sealed. The second model, the monadic model, assumes that each realm of human discourse contains within itself representations of every other realm of human discourse. The third and final model, the holistic model, assumes that the boundaries of legal discourse are porous and that each realm of human discourse interacts with every other. This section argues that of the three models, only the holistic model provides a plausible account of the boundaries of legal discourse.

*316 Of course, each of the following models is a vast oversimplification of a complex phenomenon. Human discourses are complex social interactions which employ language as a necessary means. A serious model of the boundaries of human discourses must account for the actual human behaviors that make up such discourses. So, for example, legal discourse is made up of the communicative behavior of lawyers, clients, judges, clerks, witnesses, deponents, scholars, and a host of others. The simple models that follow are offered only for heuristic purposes. These models are intended to illuminate a certain point about the boundaries of legal discourse and not for the more general purpose of understanding and explaining the nature of such boundaries.

TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE

1. The Hermetic Model: Impervious Boundaries

The hermetic model, represented by figure one, assumes that legal discourse is a closed practice. Arguments and ideas from other realms of human discourse cannot cross the impervious boundaries of legal discourse. Thus, no argument from outside of legal discourse can become part of that discourse. More generally, the first model assumes that each domain of human discourse is hermetically sealed from all others.

One difficulty with the hermetic model is that it fails to account for widely recognized features of legal discourse. There are many ways in which legal discourse is not hermetically sealed from other discourses. For example, legal fact-finding allows and sometimes even requires the introduction of testimony or other evidence from nonlegal *317 realms of human discourse. Thus, if the subject matter of a legal dispute includes facts about chemistry, a proposition of chemistry may be used as part of the justification for a legal decision. An example might be a dispute about the intellectual property rights in a chemical process.

A second difficulty for the hermetic model is the way in which normative discussion in legal discourse opens the door to other realms of human discourse. Much legal discourse includes normative discussion, such as argument over what the law should be. In such normative discourse, ideas from philosophy, political theory, and economics are folded into disagreement over what is or should be the rule in a particular case. Thus, discourse inantitrust law is infused with economic ideas, such as the cross-elasticity of demand. [FN12] Discourse in tort law includes the idea of the least-cost avoider of risk. [FN13] And discourse in contract law includes the notion of a default rule. [FN14] At least since Oliver Wendell Holmes's 1897 essay, The Path of the Law, American legal discourse has explicitly and unambiguously incorporated ideas from outside the narrow conventions of legal argument. [FN15]

There is a third problem for the hermetic model: the model does not seem fully coherent. To get at this point, we can begin with a question. What is the status of arguments about the boundaries of legal discourse, given the assumption that the boundaries are hermetically sealed? Answering this question requires us first to consider whether arguments about the boundaries of legal discourse are themselves generated from within the conventions of legal argument. This initial inquiry is itself complex, and I cannot do it justice here. It does seem rather difficult to make out the case that legal discourse can effectively patrol its own boundaries if it is limited to using border guards clad only in the uniforms of conventional legal discourse. Initially, one wonders how our border guards will address the invading foreign arguments. What might they say? Turn back! No economic *318 argument is permitted here. We only allow plain law talk. What if the guardians are asked for reasons by those whom they seek to exclude? The border guards of legal discourse could not answer by saying, This is our practice; we just do it this way. To say even this much would be to speak a foreign tongue. Lawyers do not make these Wittgensteinian moves.

Moreover, one worries that the whole issue is debated in language that does not seem the usual stuff of lawyers' briefs and judicial opinions. The term discourse has a disturbingly philosophical ring, and once we start reading the footnotes of articles about discourse, we run almost immediately into such monsters and apparitions as Wittgenstein [FN16] and Quine. [FN17] Are these gentlemen the companions of Learned Hand and Ruth Bader Ginsburg? I think not!

Let us assume that arguments about the boundaries of legal discourse are not themselves part of legal discourse. If the boundaries of legal discourse are hermetically sealed, then such boundary-defining arguments cannot enter into legal discourse. But if that conclusion is correct, the question becomes, How can we meaningfully say that the borders of legal discourse are sealed? Economic, political, and philosophical discourse does make its way into briefs and opinions. Law-and-economics discourse and arguments of political theory do come into judicial opinions. Arguments of political theory are raised. Rawls is cited in judicial opinions. [FN18] If there is nothing legitimate to be said by way of criticism, then in what sense would the boundaries of legal discourse be closed? No one can object to these invasions from other realms of discourse on the ground that foreigners simply are not allowed, because the objections would themselves be foreign invaders. The point is that the hermetic model of the boundaries of legal discourse lacks coherence. We cannot really form a consistent picture of legal discourse that is sealed off from other forms of human discourse after we have already begun to have legal discussions using the language of economics, political theory, and philosophy.

TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE

*319 2. The Monadic Model: Perceptive Boundaries

The hermetic model is implausible because it assumes that legal discourse is hermetically sealed and, therefore, impervious to arguments from other discourses. Can we hypothesize a model of legal discourse in which legal argument is a closed practice but also reflects arguments and ideas from other human discourses? What is required seems contradictory: we need a model of legal discourse that is closed, yet one that incorporates other realms of human discourse.

TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE

For heuristic purposes, we might consider an analogy to the monad, the fundamental unit of Leibniz's metaphysics. In his Monadology, Leibniz wrote that "every created monad represents the *320 whole universe." [FN19] By analogy, we might imagine that every domain of human discourse contains within itself a representation of the whole universe of human discourse. How does this representation occur? Leibniz's monads were perceptive; each perceived the whole of the universe. The monadic model of the boundaries of legal discourse, represented in figure two, postulates that each discourse has perceptive boundaries. Thus, each discourse perceives and represents what occurs in every other discourse.

It is difficult with the monadic model, however, to postulate a mechanism by which a realm of human discourse can be both perceptive and closed. No plausible mechanism exists which would allow legal discourse to close its boundaries, and also to represent all the other realms of human discourse. This situation is illustrated in figure three. Any mechanism which allows legal discourse to perceive other discourses also opens the door for legal discourse to interact with those discourses. If law reflects economics, political theory, and philosophy, then in what sense could we say that legal discourse is closed? Absent such a mechanism, the monadic model should be rejected.

The next step is to consider a model of legal discourse that is not closed--a model in which legal discourse interacts with and is penetrated by economics, political theory, and philosophy.

3. The Holistic Model: Porous Boundaries

The third model assumes that legal discourse has porous boundaries. Represented by figure four, the holistic model places legal discourse in the same sphere of discussion as all other human discourses. Legal discourse blends into other discourses; they interpenetrate one another and the lines between them may be blurry and jagged rather than focused and straight.

As depicted in figure four, the model represents each realm of human discourse as containing elements that are positioned in relationship to the core and the periphery. At the periphery are such relatively empirical and situation-specific matters as the calling of an individual strike in baseball discourse, the decision of an individual

TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE

*321 case in legal discourse, or the observation of a single price in economics. Closer to the core are such relatively abstract, general and unobservable matters as propositions about the spirit of the game of baseball, legal theories such as positivism, or the abstract axioms of economic theory. [FN20]

In the holistic model, arguments, ideas, and concepts can enter legal discourse from economics, philosophy, chemistry and so forth. Consider two ways in which such entry might occur. First, in any given legal dispute, argument about what the law is can escalate to a fairly high level of abstraction--the realm of jurisprudence. One picture of how such escalation might occur is offered by Ronald Dworkin in his theory of law as integrity. [FN21] The picture offered by the holistic model is that discourse at the periphery, such as an opinion in a particular case, is connected to discourse nearer the core, such as arguments of political theory. One imagines a chain of reasoning that begins at the periphery and moves toward the core, gradually moving from concrete legal argument to legal theory to political theory.

*322 There is a second way in which ideas from other discourses can enter legal discourse. In any given legal dispute, the resolution of the particular controversy may require the consideration of notions from some other discourse. One way in which this occurs involves fact-finding. For example, an intellectual property case may require the introduction of testimony about bioengineering. The outcome of a criminal case may depend directly on evidence, the admissibility of which hinges on the truth of a proposition of biochemistry. Another way in which such entry occurs involves norm specification. For example, commercial practice is considered in the formulation of a rule of contract law. The correct legal rule may depend on the discourse of a group of merchants. Under the holistic model, we might picture this sort of interaction of legal discourse with other discourses as movement around the perimeter. We move on an arc from law to chemistry, accounting, or some other realm of human discourse.

A qualification of the holistic model should be emphasized at this point. Saying that the boundaries of legal discourse are porous is not the same as saying they are nonexistent. Arguments drawn from political theory do not necessarily translate into legal arguments. The holistic model assumes that migration is possible--that there is no a priori reason to rule the arguments of political theory out of the court of legal argument. On the other hand, there may be a posteriori reasons for the exclusion of such arguments. There may be good reasons to treat a given legal rule as settled. The rule may be reasonably just, incorporated into a statute, and hence treated as settled for reasons of fairness. Under these circumstances, it may well be the case that judges will reject, out of hand as it were, arguments that question whether the rule is the best one. The judge may say in effect:

You may be right that this rule is not the most efficient one, but this rule is settled law and I won't even listen to arguments that it should be replaced, simply because a somewhat better rule is available. The costs of changing the rules every time a marginal improvement is available would far exceed the benefits.

The same message might be given in shorthand: This is the settled rule of law.

Given the holistic view, the distinction between propositions of law and propositions about law will not be hard and fast. Rather, the distinction will be drawn on the basis of pragmatic concerns. What Patterson calls "propositions about law" might become "propositions of law" when it suits human purposes to make them so. For example, when we are dealing with an unsettled area of law or applying a highly *323 abstract text, such as the Sherman Act or the Equal Protection Clause, to a novel fact situation, judges may turn directly to economics, political theory, or moral philosophy. When they do so, they will do so as judges. Sometimes they will translate economic jargon into legal language. Other times they will gloss arguments of political philosophy with analogies to the precedents. Legal discourse does have its own characteristic forms, vocabulary, and styles of argument. If one compares the forms of legal argument to border guards, one would do well to remember that border guards usually examine papers, ask some questions, stamp a passport, and then welcome the foreign visitor.

C. A HOLISTIC VIEW OF LEGITIMACY IN LEGAL DISCOURSE

What constitutes a legitimate move in legal discourse? On the hermetic model, the only legitimate arguments are those which come from within the closed practice of legal discourse. On the monadic model, arguments from any realm of human discourse are legitimate within legal discourse, but only insofar as the arguments from all human discourses are mysteriously present within the practice of legal argument.

If I am right and both of these models fail to capture the nature of the boundaries of legal discourse, the question becomes, What constitutes a legitimate move in legal argument, given a holistic picture of the relationship of legal discourse to the rest of human discourse? Two different views about legitimacy in legal discourse are suggested by the contributors to this symposium. In order to illustrate the issue, I will exaggerate the contrast between the positions taken by Professor Burton and Professor Patterson. Professor Burton argues that legitimacy requires normative justification of the rules of law. When asked, Why is this the law?, it is not a sufficient answer to say, Just because it is the law. [FN22] Professor Patterson suggests a very different view of legitimacy in legal discourse. He argues that the forms of legal argument do not require any external validation. In an extreme form, his position requires that Just because it is the law, is a valid answer to the question, Why does this rule govern my case?

The holistic model of legal discourse suggests that both of these views contain part of the truth. Sometimes, justification of the law is required, even within the practice of law. There will be cases when *324 arguments that a rule, such as separate but equal, is settled law simply will not do. Sometimes, however, no purpose would be served by offering justifications. The question about the law's legitimacy is not sufficiently compelling to warrant an answer. No practical purpose would be served by a judge writing an opinion that takes up a question that has been settled and about which the legal and political community is in accord. The point is that there are no a priori answers to the question whether the law should or will draw on other discourses for the answers to legal questions. It all depends upon the particular case. With this point made, I leave the argument over the boundaries of legal discourse and turn to the debate over default rules.

II. DEFAULT RULES

This Part addresses the question of whether the default rules debate makes sense. The particular objections explored here are quite different from those offered by Professor Patterson. I will try to show that certain moves made within legal scholarship--and largely within the wide boundaries of legal discourse--are not fully coherent for reasons based on considerations heretofore outside legal scholarship. We can begin with a recapitulation of the notion of a completely specified contract, upon which the default rules debate is founded.

A. THE DREAM: THE COMPLETELY SPECIFIED CONTRACT

Some economic theorists of contract dream a powerful dream. They dream of a completely specified contract, one that spells out the obligations of the parties for all possible "future states of the world." [FN23] As with many dreams, the completely specified contract is alluring; the notion of a fully specified contract leads to the idea that contract law supplies default rules that govern the obligations of the parties when their contract is incomplete. The search for the complete set of *325 default rules then becomes an important item on the agenda of contract scholars, and the criteria for the selection of default rules become a pressing issue in contract theory.

Let us begin by trying to make the notion of a completely specified contract more precise. Possible world [FN24] semantics will be employed to cash out the idea of a "possible future state of the world." Once we have a clear understanding of this idea, we can move to the fully specified contract itself.

We begin with the notion of a possible world. If X is possible, we say that X occurs in some possible world. We add the notion of the actual world. Actual is an indexical term that picks out this world from all possible worlds. A completely specified contract need not deal with all possible worlds, but only with the possible future states of the actual world. This limitation is expressed in possible world semantics via what is called an "accessibility" relation. Those possible worlds that share the history of the actual world up to now are called historically accessible. Of course, it is logically possible that the future states of the actual world could be just about anything you can imagine; there is no logical contradiction in a possible world which shares the history of the world up until now, but is completely empty the very next moment. We should restrict the domain of possible worlds to those that share the basic laws of nature (physics, etc.) with the actual world; these worlds are called nomologically accessible. The historically and nomologically accessible worlds, then, are those that share the history of the actual world up to now and share our laws of nature.

Finally, we need to define two more elements that enter into the notion of a completely specified contract: the state of the world and the obligations of the parties. First, we must stipulate that a state of a possible world is a unique time slice of that world. Thus, world w subz at time t subq is at state Sw subz t subq . [FN25] Second, we must stipulate that the obligations of the parties to a contract for a given state of the world will be a *326 set of descriptions, O, that contains for each party, p sub1 , p sub2 , . . . p subn , the obligation descriptions o sub1 , o sub2 , . . . o subk that pick out the action types required by the contract. Thus, party p subn 's discrete obligation o subk would be p subn o subk .

Given these definitions, a completely specified contract will take the form of a list of ordered pairs. One member of each pair will be a state of some historically and nomologically accessible possible world, and the other will be a specification of the obligations of the parties for that event type. Let the right arrow symbol, ->, represent the relationship between the two members of the ordered pair. A completely specified contract for n parties with k obligations in q time slices of z possible worlds can be represented as follows:

TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE

*327 In other words, a completely specified contract must spell out all obligations of all parties for all time slices of all nomologically and historically accessible possible worlds. If such a contract were to be drafted, it would be rather long--perhaps of infinite length. [FN26] Moreover, if contract law is to provide default rules for all situations in which an actual contract falls short of the ideal of a fully specified contract, then contract law must either contain an infinite number of rules or a finite number of rules that predetermine outcomes in an infinite number of cases. [FN27]

The dream of a completely specified contract has taken on a nightmarish quality. It is not clear that the idea of a contract that specifies the obligations of the parties for all possible future states of the world makes sense.

B. ANALYZING THE DREAM: THREE REASONS FOR GAPS

Something has gone wrong. The notion that a contract should specify the obligations of theparties for all possible future states of the world is clearly off the mark. This section diagnoses the problem by looking at three different reasons for the existence of contractual gaps.

1. Failure to Specify Background Assumptions

The first cause for a gap arises from unspecified background assumptions. For example, the contract requires payment of $10,000,000.00 on a certain date, and the obligee makes her payment by tendering 1,000,000,000 pennies. This situation arguably exposes a *328 gap. The contract does not specify what will happen in the event of payment in pennies. The gap is created by the background assumption that payment will be made by check or perhaps by $100 bills, but certainly not by pennies. Similarly, the contract may require the delivery of 1,000 widgets by a date 100 days in the future, but the widgets are delivered ten per day for 100 days. The gap is created by another background assumption--that deliveries usually occur all at once or in a few stages.

Can we really imagine a contract that is complete in the sense that it spells out all of our background assumptions? No. The attempt to do so runs squarely into the problem illustrated by the translation of a fully specified contract into possible-world semantics. The set of all our background assumptions is simply the set of all the facts we take for granted about the world. These assumptions need not be consciously entertained; they can be things we simply do not think about. It is silly to try to count these assumptions, but we might say that most of them are negative. We do not think about Martian invasions or the invention of a cheap, fusion-powered home manufactory that can make anything you wish out of energy and air. Our background assumptions cannot be spelled out by mortal humans of finite capacities. The notion of a contract that covers all possible worlds with events like Martian invasions is not what we need to solve the problems that judges and lawyers face in contract doctrine.

2. Contractual Generality and the Variability of Experience

Consider a second and related reason for so-called gaps. Contracts are written in relatively general language, but human experience is enormously variable. This point, made by Randy Barnett, [FN28] *329 draws on H.L.A. Hart, [FN29] but the ultimate source is Aristotle. Recall Aristotle's classic discussion of equity in Chapter Ten of Book Five of the Nicomachean Ethics:

[A]ll law is universal but about some things it is not possible to make a universal statement which shall be correct. In those cases, then, in which it is necessary to speak universally, but not possible to do so correctly, the law takes the usual case, though it is not ignorant of the possibility of error. And it is none the less correct; for the error is not in the law nor [ [sic] in the legislator but in the nature of the thing, since the matter of practical affairs is of this kind from the start. When the law speaks universally, then, and a case arises on it which is not covered by the universal statement, then it is right, where the legislator fails us and has erred by over-simplicity, to correct the omission--to say what the legislator himself would have said had he been present, and would have put into his law if he had known. [FN30]

Aristotle is discussing the law in general, but his point is good for the norms created by a valid contract. Such norms must be stated in relatively general and universal language, but practical affairs are particular and variable. The notion of a completely specified contract assumes that one could spell out all the rules in advance by adding an enormous number of relatively general and universal clauses to the actual contract negotiated by the parties. But this assumption is incorrect. Every attempt at a completely specified contract is doomed to fail, because the tool with which such a contract would need to be constructed--a writing containing the words and sentences of a natural language--is not up to the task.

3. Recurring Omissions of Anticipatable Contingencies

Consider a third source of contractual gaps: the omission by the parties of an anticipatable contingency. Of course, such examples are familiar from contracts casebooks. These are the garden-variety gaps with which we are all familiar, and of course the law does provide default rules for such recurring gaps. In a common law system, such gap filling by default rules is almost inevitable. The fact patterns recur, opinions are published and particular rules governing omission types begin to form. There is nothing wrong with identifying such default rules, but we must remember that the collation of the default rules that cover such gaps will not produce anything like a completely *330 specified contract. In other words, contract law, including all of its default rules, is necessarily incomplete.

C. WAKING UP: WHAT DEFAULT RULES CAN AND CANNOT DO

What can default rules do? They can fill recurring gaps of omission, but they cannot specify background assumptions or overcome the problem of contractual generality. What do we do with these remaining gaps? How should the law respond when life takes a turn that could not have been anticipated or an unarticulated background assumption proves false? One possibility is suggested by the notion of ex post tailoring of default rules. [FN31] The law should create a default rule, after the fact, and the rule should be tailored to the particular situation.

In a sense, this suggestion involves a conceptual mistake. The notion of a default rule as developed in relation to the idea of a completely specified contract does not involve the particular decision making in response to an individual case. If default rules are rules, then they are ex ante, available in advance of the case to which they are applied. In a sense, however, the suggestion of ex post tailoring is exactly right. There is a question of terminology. What should we call this notion that judges should depart from or add to the terms of a contract or other legal rule when adherence to the rule would lead to an injustice in a particular case? The word that comes to mind is "equity"--for this notion of ex post tailoring is a very old idea [FN32]--an idea that does not in any way rest on the notions of a completely specified contract or a default rule.

III. DEFAULT RULES IN HOLISTIC LEGAL DISCOURSE

The question then becomes, what role can a properly chastened theory of default rules play in legal discourse? We should answer this *331 question in light of the holistic model of the boundaries of legal discourse. The boundaries are porous or permeable to ideas and concepts developed in other discourses. We should note that in this model there is no reason to believe that legal discourse must be subordinate in some way to other discourses. That economic ideas can enter and influence legal discourse does not entail that economic ideas control legal discourse. So what role does, can, and should the notion of default rules play in contract doctrine? An answer might begin by looking at the role that the default rules notion has already played in the doctrine. Brief excerpts from the opinions in three recent cases are illustrative.

Consider the following passage from a recent court of appeals decision:

The contract language, and the structure and history of the negotiation, do not support an inference that every one of the parties was given the right to obtain, through opposition to a change, application of the VIII(C) standard. That being so, we may search for default rules. The common law has addressed a cognate problem--whether those who acquire portions of a burdened estate may sue each other to enforce an equitable servitude benefiting a third party. [FN33]

A similar use of the default rules notion appears in the following excerpt from a 1991 Third Circuit decision: "[T]he UCC provides for express and implied warranties if the seller fails to disclaim expressly those warranties. Thus, even though warranties are an important term left blank by the parties, the default rules of the UCC fill in that blank." [FN34] And finally, a recent district court decision incorporated the notion of a default rule in the following way:

Although we acknowledge that Shearson's proposition may be viewed in a limited sense as a logical corollary to the rule that allows a disclaimer of third party rights, we nevertheless question the underlying soundness of applying that rule under the particular circumstances present here, where the party alleged to have committed a breach is on notice that its own contract has its genesis in a prior contract establishing both the existence of a beneficiary and obvious reliance by the beneficiary on the services at issue. In assessing what the appropriate rule of default should be in such a *332 situation, see generally, Ayres & Gertner, "Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules," 99 Yale. [sic] L.J. 87 (1989), we do not believe that the best solution to effective fulfillment of this chain of contracts is to deny Coronet a right of action in contract simply because Coronet did not ensure by express contractual provision that it would retain a right that the law already provided to it. Such logic ignores the nature of the interrelation of contracts in this case. [FN35]

What are we to make of these passages?

First, courts clearly do make use of the notion of a default rule in the sense that Patterson would find unobjectionable, distinguishing default rules from mandatory rules. Second, courts seem to engage in a discourse which Patterson might find uncomfortable. Gaps are discovered, and default rules become the object of judicial search. Economic theories are even cited in the opinions. But these two points need to be qualified by a third point: when the default rules debate enters legal discourse it is transformed. Judges relate the debate to familiar patterns of legal argument. The U.C.C., and not economic theory, is the likely source of the default rule to govern a commercial case. This is not to say that the judicial debate could not escalate to the level of economic theory. Rather, my point is that such escalation from garden-variety law talk to economic theory is likely to be rare. It is the exception, not the rule.

Is the default-rules debate a pseudo-debate? Yes and no. On the one hand, some of the debate operates within a framework of analysis that is fundamentally misconceived. To the extent that the default rules debate rests on the foundational notion of a completely specified contract, the debate is misconceived. On the other hand, it does not follow that economic, political, or philosophical analysis has no legitimate part to play in a properly chastened discussion over the role and content of default rules in contract doctrine. Legal discourse is not like ancient China, forbidding entrance to all foreign visitors. Law talk is part of a larger whole, a family of discourses that connect to each other in myriad ways. As Holmes put it,

If your subject is law, the roads are plain to anthropology, the science of man, to political economy, the theory of legislation, ethics, and thus by several paths to your final view of life. It would be equally true of any subject. The only difference is in the ease of seeing the way. To be master of any branch of knowledge, you must *333 master those which lie next to it; and thus to know anything you must know all.[FN36]

Holmes's language is poetic, and, as is frequently the case, he overstates his point. There is, however, an important truth here. Law's empire has open borders. [FN37]

[FNa1]. Associate Dean for Academic Affairs, Professor of Law, and William M. Rains Fellow, Loyola Marymount University.

[FN1]. GOTTFRIED W. VON LEIBNIZ, Codex Iuris Gentium (Praefatio), in LEIBNIZ: POLITICAL WRITINGS 170 (Patrick Riley trans. & ed., 2d ed. 1988).

[FN2]. For a sampling of the literature, see Ian Ayres & Robert Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 Yale L.J. 87 (1989); Randy E. Barnett, The Sound of Silence: Default Rules and Contractual Consent, 78 Va. L. Rev. 821 (1992); Lucian A. Bebchuk & Steven Shavell, Information and the Scope of Liability for Breach of Contract: The Rule of Hadley v. Baxendale, 7 J.L. Econ. & Org. 284 (1991); David Charny, Hypothetical Bargains: The Normative Structure of Contract Interpretation, 89 Mich. L. Rev. 1815 (1991); Jules L. Coleman et al., A Bargaining Theory Approach to Default Provisions and Disclosure Rules in Contract Law, 12 Harv. J.L. & Pub. Pol'y 639 (1989); Richard Craswell, Contract Law, Default Rules, and the Philosophy of Promising, 88 Mich. L. Rev. 489 (1989); Clayton P. Gillette, Commercial Relationships and the Selection of Default Rules for Remote Risks, 19 J. Legal Stud. 535 (1990); Charles J. Goetz & Robert E. Scott, The Limits of Expanded Choice: An Analysis of the Interactions Between Express and Implied Contract Terms, 73 Cal. L. Rev. 261 (1985); Jason S. Johnston, Strategic Bargaining and the Economic Theory of Contract Default Rules, 100 Yale L.J. 615 (1990); Alan Schwartz, A Theory of Loan Priorities, 18 J. Legal Stud. 209 (1989); Robert E. Scott, A Relational Theory of Default Rules for Commercial Contracts, 19 J. Legal Stud. 597 (1990).

[FN3]. See Symposium, Default Rules and Contractual Consent, 3 S. CAL. INTERDISC. L.J. 1 (1993).

[FN4]. Dennis Patterson, The Pseudo-Debate Over Default Rules in Contract Law, 3 S. CAL. INTERDISC. L.J. 235 (1993).

[FN5]. One observer explains this fully:

Many, if not most, rules in contract law are default rules; they apply only if the parties fail to contract around them. For example, if the parties' contract does not mention whether or not a contract is assignable, the law provides that, as a general matter, most contracts are assignable. The parties, however, are free to contract around this rule by providing that the contract cannot be assigned. Another default rule limits damages recoverable by a party injured by contract breach to those damages that are foreseeable. The parties are generally free to contract for greater liability if they so choose.

Robert K. Rasmussen, Debtor's Choice: A Menu Approach to Corporate Bankruptcy, 71 TEX. L. REV. 51, 61-62 (1992).

[FN6]. Patterson, supra note 4, at 253. As Patterson frames the debate:

In the literature, the question takes the form, "How should the courts set default rules?" I believe this question involves a jurisprudential mistake. Courts do not set default rules. Courts interpret contracts. To interpret a contract is to decide what the parties' rights and duties are under an agreement. To do this, a court must identify the appropriate law of the jurisdiction and apply it to the case before it.

Id. at 238.

[FN7]. Id. at 239-40. Patterson phrases the distinction as follows:

When legal theorists describe the law, there are two different sorts of claims they might make. The first type involves propositions about law. For example, a moral philosopher might say that the tort law of a given jurisdiction is unfair because it fails to provide a cause of action to relatives who hear of, but do not witness, negligent conduct fatal to their close relations. The claim that the law is unfair is a proposition about the tort law of this jurisdiction.

Id. (footnote omitted); see also LORD LLOYD OF HAMPSTEAD & M.D.A. FREEMAN, LLOYD'S INTRODUCTION TO JURISPRUDENCE 810 (5th ed. 1985) (using phrase "propositions not of law, but about the law" to mark distinction between the law in force and sentences in a textbook where the law in force is stated).

[FN8]. Patterson, supra note 4, at 241.

[FN9]. Patterson's account of the nature of the mistake is based on his view of legal truth. He argues:

Perhaps owing to the fact that there are no contracts in the world in quite the same way there are cats, the means for appraising the grounds for propositions of law are different in kind from those used to assess empirical propositions. Consider again the assertion regarding the Statute of Frauds. How are we to appraise the proposition that no contract for the sale of goods with a price over $500 is enforceable unless in writing? If the rule is articulated in a statute, . . . then consultation of the authoritatively codified rule is sufficient to appraise the asserted proposition and find it true.

Id. at 265.

[FN10]. Id. at 268 (footnotes omitted).

[FN11]. See generally ARISTOTLE, Nicomachean Ethics, in 2 THE COMPLETE WORKS OF ARISTOTLE 1729 (Jonathan Barnes ed., W.D. Ross & J.D. Urmson trans., 1984). Hereinafter, page references will be to Immanuel Bekker's standard edition of the Greek text, first published in 1831. Aristotle's account of law is offered in Book V of the Nicomachean Ethics. An important account of natural law theory that emphasizes the Aristotelian view is found in ROGER A. SHINER, NORM AND NATURE (1992).

[FN12]. See Eastman Kodak Co. v. Image Technical Servs., Inc., 112 S. Ct. 2072, 2083 (1992); United States v. General Dynamics Corp., 415 U.S. 486, 490 (1974); United States v. Grinnell Corp., 384 U.S. 563, 592-93 (1966). A Westlaw search in the ALLFEDS database revealed 362 cases using the phrase "cross-elasticity of demand" as of June 9, 1993.

[FN13]. See, e.g., Insurance Co. of N. Am. v. Forty-Eight Insulations, Inc., 633 F.2d 1212, 1231 (6th Cir. 1980) (Merritt, J., dissenting); Freedman Truck Ctr., Inc., v. General Motors Corp., 784 F. Supp. 167, 172 n.1 (D.N.J. 1992); Jackson v. PKM Corp., 422 N.W.2d 657, 665 (Mich. 1988).

[FN14]. On this point, see infra text accompanying notes 33-35.

[FN15]. Recall Holmes's injunction that "every lawyer ought to seek an understanding of economics." OLIVER W. HOLMES, The Path of the Law, in COLLECTED LEGAL PAPERS 167, 195 (1920).

[FN16]. See Patterson, supra note 4, at 271-72 n.130.

[FN17]. Id. at 286 n.188.

[FN18]. For example, courts cite to JOHN RAWLS, A THEORY OF JUSTICE (1971). See, e.g., Goetz v. Crosson, 967 F.2d 29, 39 (2d Cir. 1992) (Newman, J., concurring); United States v. Bodiford, 753 F.2d 380, 382 n.10 (5th Cir. 1985); Flynt v. Leis, 574 F.2d 874, 877 n.5 (6th Cir. 1978). A Westlaw search revealed 14 federal cases and 33 state cases citing Rawls's book as of June 6, 1993.

[FN19]. GOTTFRIED W. VON LEIBNIZ, The Monadology (1714), in MONADOLOGY AND OTHER PHILOSOPHICAL ESSAYS 148, 158 (Paul Schrecker & Anne M. Schrecker trans., 1965). For another translation, see GOTTFRIED W. VON LEIBNIZ, The Monadology, in LEIBNIZ: SELECTIONS 533, 546 (Phillip P. Wiener ed., 1951).

[FN20]. This holistic model is inspired by, but not dependent upon, Willard Van Orman Quine's holistic view of meaning. See W.V. QUINE, PURSUIT OF TRUTH 13-16 (1990); W.V. QUINE, THEORIES AND THINGS 70-72 (1981); see also CHRISTOPHER HOOKWAY, QUINE: LANGUAGE, EXPERIENCE AND REALITY ch. 10 (1988). I am indebted to Tyler Burge for the diagram, which is based upon one drawn during his course on Quine, offered at UCLA in the early 1980s. Quine's metaphor of the web of belief is also related to Quine's holism. See W.V. QUINE & J.S. ULLIAN, THE WEB OF BELIEF (2d ed. 1978). Professor Burton explicitly incorporates the web of belief in his view of legal reasoning. See STEVEN J. BURTON, AN INTRODUCTION TO LAW AND LEGAL REASONING 132-36 (1985).

[FN21]. See RONALD DWORKIN, LAW'S EMPIRE (1986); RONALD DWORKIN, A MATTER OF PRINCIPLE (1985); RONALD DWORKIN, TAKING RIGHTS SERIOUSLY (1978).

[FN22]. Steven J. Burton, Comment on Professor Patterson's Pseudo-Debate Over Default Rules in Contract Law, 3 S. CAL. INTERDISC. L.J. 303, 308 (1993).

[FN23]. See Ian Ayres & Robert Gertner, Strategic Contractual Inefficiency and the Optimal Choice of Legal Rules, 101 YALE L.J. 729, 731 (1992).

Legal scholars use the term "incomplete contracting" to refer to contracts in which the obligations are not fully specified. A contract to sell a good would be "obligationally" incomplete, for example, if it failed to specify the price, quantity, or date of delivery. In contrast, a contract is obligationally complete if the obligations of the parties are fully specified for all future states of the world. A contract that failed to specify the seller's obligations in the event of a flood or the buyer's breach would thus be obligationally incomplete. Default rules respond to obligational incompleteness by filling these obligational gaps.

Id. at 730.

[FN24]. See generally DAVID LEWIS, ON THE PLURALITY OF WORLDS (1986); SAUL A. KRIPKE, NAMING AND NECESSITY (1980). The idea of possible worlds was introduced by Leibniz. See GOTTFRIED W. VON LEIBNIZ, The Theodicy (1710), in LEIBNIZ: SELECTIONS, supra note 19, at 509, 509-11. Leibniz used the idea of a possible world in answer to the argument against the existence of good from the problem of evil. The argument is not proven, Leibniz maintained, until it is shown that the actual world is not the best of all possible worlds. Id. "World" here refers to the whole universe through time and not just the planet Earth.