Preferred Citation: Post, Robert, editor. Law and the Order of Culture. Berkeley:  University of California Press,  c1991 1991. http://ark.cdlib.org/ark:/13030/ft9q2nb693/

Law and the Order of Culture Edited and with a New Introduction by Robert Post UNIVERSITY OF CALIFORNIA PRESS Berkeley · Los Angeles · Oxford © 1991 The Regents of the University of California

Preferred Citation: Post, Robert, editor. Law and the Order of Culture. Berkeley:  University of California Press,  c1991 1991. http://ark.cdlib.org/ark:/13030/ft9q2nb693/

Introduction:
The Relatively Autonomous Discourse of Law

We have long been accustomed to think of law as something apart. The grand ideals of justice, of impartiality and fairness, have seemed to remove law from the ordinary, disordered paths of life.^[1] For this reason efforts to unearth connections between law and culture have appeared vaguely tinged with exposé, as though the idol were revealed to have merely human feet.

In recent years, with a firmer sense of the encompassing inevitability of culture, the scandal has diminished, and the enterprise of actually tracing the uneasy relationship of law to culture has begun in earnest. There is an emerging consensus that culture ought to be regarded, in Raymond Williams's words, "as the signifying system through which necessarily (though among other means) a social order is communicated, reproduced, experienced, and explored."^[2] Culture, we have come finally to believe, is the precondition for the very possibility of human meaning, and is not merely ornamental, but engaged in the hard, practical, and important work of reproducing the social order.

The realization that social order requires the mediation of social meaning, and that social meaning arises through the operation of systems that are simultaneously symbolic and practical, has enabled the study of law to join the larger family of human sciences. It has empowered legal scholars to proceed on the assumption "that the creation of legal meaning—'jurisgenesis'—takes place always through an essentially cultural medium."^[3] It has liberated the legal academy from, in Joseph Vining's eloquent formulation, "projecting an image of law as a set of rules outside, a grid that, could you only tap it with your fingernail, would give out a hard metallic ring" (9). Legal scholarship has in consequence begun to produce a substantial body of work that, like contemporary literary criticism, explores "the social and historic dimensions of symbolic practice" and pushes "beyond the boundaries of the text, to establish links between the text and values, institutions, and practices elsewhere in the culture."^[4]

This scholarship suggests that jurisgenesis cannot be conceived merely as the reflection or reproduction of the ambient culture. As the uniquely "authorized discourse for the state," occidental law attempts to legitimize its coercive bite through distinctive aspirations toward such values as objectivity, neutrality, and rationality.^[5] This aspiration is deeply reflexive, focusing on questions of internal consistency that we have come to subsume under the concept of the "rule of law."

A fascinating dialectic is thus set in motion, in which our law uses the resources of the larger culture precisely in order to establish its own particular kind of cultural discourse.^[6] In this respect occidental law may accurately be described as a "relatively autonomous" cultural form.^[7]

To conceive the law as a relatively autonomous form of discourse is immediately to glimpse the potential for a new and stimulating agenda of research. Consider, for example, three questions. The first might be called the "boundary question," for it would concern the internal mechanisms by which the boundaries of legal discourse are maintained. Somehow legal discourse manages both to situate itself within the larger culture and at the same time to retain its own distinct discursive identity. To understand how this is accomplished we need to study not merely the interpretive practices of legal actors, but also how these practices are tied to the allocation of power within the legal system.

A second topic for research might be called the "legitimacy question," for it would address the issue of how legal discourse sustains its legitimate authority in society at large. If the boundary question focuses on the distinctive features of legal discourse, the legitimacy question shifts our attention to the functioning of legal discourse within the general culture. It invites us to conceive of legal discourse as "simply one of many competing normative disciplinary discourses, discourses of morality, religion, and social custom"^[8] and to inquire into the processes by which legal discourse retains its characteristic prestige and power. This inquiry cannot be divorced from the boundary question, for the legitimacy of the law may well depend upon the specific means by which legal discourse establishes its own relative autonomy.

Finally, there is what may be called the "political question," which would concern the practical implications of conceptualizing a world in which meaning has been dissolved into faceless and impersonal systems of discourse. Quite apart from the descriptive accounts of the law implied by the boundary and legitimacy questions, we must also confront the political and moral consequences to ourselves of inhabiting such a world. We need to ask what values the law ought in such circumstances to serve. And of course that inquiry depends upon the darker and deeper question of how political freedom and action, those most irreducibly humanistic of virtues, ought to be conceived when meaning, and therefore identity itself, has been dispersed into the empty matrices of cultural systemics.

These are difficult and pressing questions, driving some of the best and most innovative contemporary legal scholarship. They also form the framework for the essays in this volume. Although these essays are written from a wide variety of disciplinary perspectives—Carla Hesse writes as a historian, Jennifer Nedelsky as a political scientist, Kim Lane Scheppele as a sociologist, Martin Stone as a philosopher, François Ewald as a cultural theorist, and Joseph Vining and myself as legal academics—they nevertheless retain a common focus on the theoretical problematic of the law as a relatively autonomous form of cultural discourse.

They illustrate in microcosm the emerging possibilities of the interdisciplinary study of law.

Of the three inquiries I have posed, the boundary question is perhaps the most traditional, for it evokes longstanding work in the theory of legal interpretation. Nevertheless in recent decades, as discipline after discipline has taken the "interpretive turn,"^[9] legal theorists have tended to look outward and to seek to explain legal interpretation on the principles of general hermeneutic theory.^[10] While this effort has had the salutary effect of uncovering shared cultural foundations of the law, it has also, at least in its more extreme manifestations, challenged the relative autonomy of the law, for it would subsume the constitution of legal thought under an undifferentiated cultural methodology.

Joseph Vining's essay in this volume, while conceding the undoubted "convergences" of interpretative practice, nevertheless questions the underlying assumption of this effort, "that interpretation is the same phenomenon or experience whatever its setting" (1). Vining gently calls attention to various distinctive aspects of legal interpretative method: the characteristic absence of a definitive legal text to interpret until one is made visible by processes of legal interpretation; the characteristic authorlessness of various kinds of legal texts; the characteristic use of legal texts to facilitate legal writing.

Of course these facets of legal interpretation correspond to well-developed problematics within what may be called abstract interpretative theory. The "pre-interpretative" necessity of identifying a specific legal text illustrates the generic "problem of the frame—of the distinction between inside and outside" posed by the "intertextuality" of texts.^[11] The obscurity of individual legal authors highlights the influential project advanced by Foucault of analyzing the "author-function" as "a variable and complex function of discourse."^[12] Reading legal texts in order to use them exemplifies Gadamer's thesis that all interpretation rests on "application."^[13] But these general correspondences do not vitiate Vining's thesis. The relative autonomy of the law, after all, does not presuppose legal exceptionalism. It instead claims that the concrete form in which abstract methodological issues are presented, experienced, structured, and resolved is distinctive to the law. Vining's essay elegantly illuminates the possibilities of such a claim.

Vining observes the special way in which certain kinds of legal texts gain their "authority" through a form of transcendental personification, by purporting to speak "for the law or for an entity . . . that in turn speaks for the law" (6). The observation raises a central methodological issue, which is how this effort to "speak for" the law is sustained. The issue involves not merely questions of hermeneutic method, but also of the relationship between the substance of legal interpretation and that of other forms of cultural discourse. My own contribution to this volume, for example, suggests that in constitutional law the attempt to

"speak for" the Constitution necessarily entails an appeal to political rather than narrowly legal values. Judicial debates over "theories" of constitutional interpretation may most usefully be seen as conflict over the nature and use of these values. Method and substance, legal and nonlegal forms of discourse, are thus intricately and self-consciously braided together to produce what we immediately recognize as distinctively constitutional patterns of thought.

Jennifer Nedelsky's essay in this volume offers a very different perspective on the relationship between legal and other cultural discourses. She convincingly demonstrates "that the image of protective boundaries as essential to the integrity and autonomy of the self is deep and pervasive in our culture" (168) and that this image permeates and influences legal thought in such disparate areas as property, privacy, and constitutional rights. The connection between law and the general culture is thus shown to function at the level of unconscious metaphoric structures.

The dependence of that connection upon matters of institutional design is deftly illuminated in Kim Lane Scheppele's essay. In the Anglo-American legal tradition, judges are deemed responsible for findings of law, juries for findings of fact. Judges are professionals, trained in the discourse of the law; juries are composed of laypersons, who are supposed to represent the general community. Scheppele demonstrates that there is no ontological distinction between propositions of law and propositions of facts, because "[l]aw and fact are mutually constituting—not simply hard to tell apart" (62). The difference between the two, then, inheres entirely in the law's "way of allocating authority across different sets of actors" (62).

This division of power marks the boundary between professional legal discourse and general "cultural knowledge, background knowledge available in the social world" (63). Because cases must be decided on the basis of both law and facts, this boundary is paradoxically folded into legal practice itself, foreclosing any possibility of internally insulating legal discourse from the influence of the ambient culture. Yet this influence can be controlled and ultimately subordinated to a distinctive professional legal discourse through administrative manipulation of the law/fact distinction.

Scheppele's essay forcefully illustrates the relationship between the legitimacy and boundary questions, for her conclusion is that "[l]egal authority . . . rests on the ability of legal narratives to reside simultaneously in the normative universes of legal and nonlegal worlds" (65). For Scheppele, in other words, the cultural legitimacy of the law depends upon interpretative practices that maintain a certain homology between legal discourse and other forms of culture.

Nedelsky agrees, noting that "American constitutionalism has the power it does in part because its metaphoric structure resonates so powerfully with the

fears and (elusive because illusory) goals of the separative self" (181). But Nedelsky also argues for the adoption of "new symbols, myths, and metaphors to replace the old" (181). Her work thus presupposes that the law is not necessarily bound by the discourse of the general culture, but can instead become an independent force for change. In suggesting this consequence Nedelsky's article points toward the growing literature of "critical" legal sociology, which is concerned with documenting the vast extent to which "legal discourses constitute social life," in the sense that "the law and legal institutions can be seen as structures of meaning that radiate throughout social life and serve as part of the material people use to negotiate their understanding of everyday events and relationships."^[14] The clear implication is that legal legitimacy may have other foundations than mere homology.

Two other articles in this volume, those by François Ewald and Carla Hesse, reflect this same divergence of perspectives. Ewald is concerned with explicating the following suggestive passage from Foucault:

Another consequence of [the] development of bio-power was the growing importance assumed by the action of the norm, at the expense of the juridical system of the law. Law cannot help but be armed, and its arm, par excellence , is death; to those who transgress it, it replies, at least as a last resort, with that absolute menace. The law always refers to the sword. But a power whose task is to take charge of life needs continuous regulatory and corrective mechanisms. . . . Such a power has to qualify, measure, appraise, and hierarchize, rather than display itself in its murderous splendor; it does not have to draw the line that separates the enemies of the sovereign from his obedient subjects; it effects distributions around the norm. I do not mean to say that the law fades into the background or that the institutions of justice tend to disappear, but rather that the law operates more and more as a norm, and that the judicial institution is increasingly incorporated into a continuum of apparatuses (medical, administrative, and so on) whose functions are for the most part regulatory. A normalizing society is the historical outcome of a technology of power centered on life.^[15]

Foucault distinguishes the violence of the law, which for him characterizes the "juridical system of the law," from its capacity to organize, administer, and constitute forms of social life, which he views as an essentially regulatory mechanism. Regulation can occur at the specific level of the "disciplines," which involve the "anatomo-politics of the human body ," and which center "on the body as a machine: its disciplining, the optimization of its capabilities, the extortion of its forces, the parallel increase of its usefulness and its docility, its integration into systems of efficient and economic controls." Regulation can also occur at the more general level of "a bio-politics of the population ," which centers instead on "the species body, the body imbued with the mechanics of life and serving as the basis of the biological processes: propagation, births and mortality, the level of health, life expectancy and longevity, with all the conditions that can cause these to vary." For Foucault these "two poles" of the "disciplines of the body and the regulations of

the population" describe the reach of what he calls "bio-power," which is "the organization of power over life."^[16]

Bio-power functions through the operation of norms, and Ewald's fascinating article explicates the concept of the norm as "that form of the common standard produced through the group's reference to itself" (159). Norms are characterized simultaneously by positivism, relativity, and polarity (155–58). Understood in this fashion, norms make possible the regulatory apparatuses of modern life, ranging from "the panoptical logic of discipline," to "the probabilistic schema of insurance," to the "communicative logic" of industrial standardization (154). And, in particular, norms are for Ewald what make "law possible in modern societies" (155). The possibility and legitimacy of the law are thus perceived as dependent upon the congruence between legal discourse and the more general structures of bio-power.^[17]

Carla Hesse presents a quite different perspective. Her purpose is to challenge Foucault's influential hypothesis that "the modern concept of the author" emerged "historically as the cultural incarnation of a new axis in sociopolitical discourse: the inviolable relation between the rights-bearing individual and private property" (109). As developed by the subsequent work of Martha Woodmansee and Mark Rose, Foucault's hypothesis requires a particular interpretation of the emergence of copyright law as reflecting "new bourgeois socio-economic relations and cultural values" (110). Hesse questions this view. Her careful and authoritative reconstruction of the emergence of French copyright law during the Revolution demonstrates that there were in fact competing cultural understandings of the author, as private proprietor and as public servant. The law did not passively reflect one or the other of these understandings, but rather actively and creatively compromised between them. Hesse concludes that we "need a more complex view of the relationship between the law and cultural change, one that accounts for the political as well as the socio-economic forces at work in the reshaping of the legal world" (131).

The realm of the political is of course that of action and decision. By suggesting the dependence of the law upon the political, Hesse implies that we (at least in some degree) choose our legal fate. Like Nedelsky, Hesse envisions a law which does not merely reflect more general cultural domains, but rather actively affects them. But what, we may ask, is the source of the law's authority in such circumstances? Although Hesse does not explicitly address this question, her references to the Revolution make it plain enough that at least one important source is what Foucault calls the law's "murderous splendor." Behind the law's immense power of cultural constitution always lies the "armed" violence of the state.

It is important to observe that Scheppele, Nedelsky, Hesse, and Ewald are addressing the legitimacy question at vastly different levels of abstraction. Scheppele is concerned with the narrative resolution of particular cases; Hesse with the

cultural background of specific laws; Nedelsky with the generic metaphoric patterns that underlie legal reasoning; and Ewald with the connection between law and the structural components of cultural order. Yet the essentially dialectical relationship of law to culture can be found at each of these levels. Pierre Bourdieu has offered a striking rendition of that relationship:

The law is the quintessential form of "active" discourse, able by its own operation to produce effects. It would not be excessive to say that it creates the social world, but only if we remember that it is this world which first creates the law. It is important to ascertain the social conditions—and the limits—of the law's quasi-magical power. . . . [T]he schemas of perception and judgment which are at the origin of our construction of the social world are produced by collective historical labor, yet are based on the structures of this world themselves. . . . Our thought categories contribute to the production of the world, but only within the limits of their correspondence with preexisting structures. Symbolic acts of naming achieve their power of creative utterance to the extent, and only to the extent, that they propose principles of vision and division objectively adapted to the preexisting divisions of which they are the products.^[18]

The essays in this volume well illustrate this ceaseless dialectic.

Carla Hesse's tale of the creation of French copyright law relates how citizens, engaged in political controversy, used the law to embody and express their decisions and compromises. The underlying image is the familiar one of law attendant upon a politics that occurs in a realm of exchange among fully constituted subjects. This image funds most contemporary constitutional thought and scholarship.^[19] My own essay in this volume, for example, works in the tradition of this image.

The image is logically compatible with the concept of law as a relatively autonomous discourse. Cultural systems, however necessary for the creation of human meaning, can be conceptualized as a medium of political exchange and hence of legal valorization. They can be analogized to a form of language that makes possible expressions of political and legal action. What I have called the "political question," therefore,need assume no transcendent difficulties, but may merely pose the problem of politics as usual: What is to be done?

The situation is quite otherwise, however, if cultural discourse is understood to constitute and to undermine the subjectivity of the very citizens who seek to engage in politics. If "power," to use Foucault's words, "makes individuals subjects,"^[20] then, as Stephen Greenblatt has remarked, "the human subject itself [begins] to seem remarkably unfree, the ideological product of the relations of power in a particular society."^[21] The issue is not merely one of historical determinism and consequent political passivity,^[22] but, more deeply, of the possible meaning of politics for a self that is merely "a cultural artifact."^[23] It would seem

that any potential political outcome would merely reinscribe initial conditions of deprivation. Hence Foucault:

The system of right, the domain of law, are permanent agents of the relations of domination, these polymorphous techniques of subjugation. Right should be viewed, I believe, not in terms of a legitimacy to be established, but in terms of the methods of subjugation that it instigates.^[24]

In such circumstances the force of the political question strikes with a vengeance.

Foucault himself speaks of the relationship between power and freedom as "an 'agonism,'" a "relationship which is at the same time reciprocal incitation and struggle; less of a face-to-face confrontation which paralyzes both sides than a permanent provocation." There can be on this account no "essential freedom," only a perpetual conflict between "the recalcitrance of the will" and the power that makes the self possible.^[25] This agonistic conflict finds its reflection in "'oppositional' American criticism," with its emphasis "on indeterminant negativity."^[26] It can also be seen in the recent marriage of Marxism and deconstruction that proffers the possibility of "permanent revolution" and "an infinite progress in socially reconstructive action."^[27]

These perspectives are the most recent manifestation of a tradition stretching back to figures like Fichte, Schiller, Schelling, in which nonconsensual influences on the self are perceived as coercive invasions of the autonomy of the subject, to be overcome through political liberation.^[28] The longing to overcome a dehumanized world is given a distinctive post-modernist twist when coercive external structures are also and simultaneously seen as constitutive of the very self that seeks autonomy. Political emancipation therefore necessarily turns Sisyphean, aptly described in images of the "permanent" and the "indeterminant."

In the legal academy this form of politics has been most profoundly and influentially explored by Roberto Mangabeira Unger, the acknowledged intellectual leader of the important Critical Legal Studies Movement. Unger's first book powerfully explored the "antinomies" of liberal thought, contradictions between "reason and desire," "rules and values," and "theory and fact." It sought to overcome these antinomies through the "regulative ideal" of a "universal community" in which "the sense of immanent order would be brought into harmony with the capacity of criticism or transcendence" and "the antagonism of the individual with the social aspect of personality" would be dissolved in "each person's recognition of the concrete individuality of his fellows."^[29] Fully aware of the unreality of this ideal, the book closed with a famous prayer of frustration:

When philosophy has gained the truth of which it is capable, it passes into politics and prayer, politics through which the world is changed, prayer through which men ask God to complete the change of the world by carrying them into His presence and giving them what, left to themselves, they would always lack. . . .

But our days pass, and still we do not know you fully. Why then do you remain silent? Speak, God.^[30]

Unger's most recent work, an impressive trilogy entitled Politics: A Work in Constructive Social Theory ,^[31] assays a practical political and legal solution to these antinomies. In Politics , however, the force of the antinomies has been reduced to a single master contradiction: the self requires context for its substantive realization, but all context unduly limits and constrains the self.^[32] Unger's solution to this dilemma is to propose a highly elaborated form of politics and law that collapses "the distance between context-preserving routine and context-transforming conflict," thereby enabling "the assertion of will over custom and compulsion."^[33] By designing institutions that "respect and encourage our context-smashing abilities,"^[34] Unger drives "the idea of society as artifact to the hilt."^[35]

Unger's work, in short, represents the most developed institutional and legal vision of the kind of permanent "agonism," the endlessly repetitive struggle between will and power, that is so powerfully attractive for the post-modern left. It is a vision that is deeply and movingly analyzed by Martin Stone's essay in this volume. Stone's query is whether this agonism can properly be called political at all, for it urges no particular action or program other than the reiterated rehabilitation of the will. The will, however, is only what makes politics possible; the practice of politics consists in giving the will specific and concerted direction. "We might read Unger's theory," Stone therefore suggests, "as a rigorous formalism : not as a politics but as a grammar that retraces the seam, circles the rim, of the political as such " (100). Stone's disturbing implication is that the agonism of the post-modern left ought to be seen as a retreat from politics, as an anguished effort to buttress the failing "fiction" of the autonomous self that would make redemptive politics possible (94–95).

And yet we do go on. Jennifer Nedelsky's essay concludes this volume with a glimpse of a "genuine option": We could be "radically transformed if we were to come to see ourselves as 'inseparable from all other beings in the universe'" (184, 183). We could "approach the daunting task of reimagining the rhetoric of law and freedom" (182). From within the feminist tradition, Nedelsky seizes the occasion of the political question for a classically clarion call to decision and to action. Traversing the external understandings necessary for apprehending the boundary and legitimacy questions, her work nevertheless retains what Stone would surely view as a quintessentially political commitment. The possibility of that commitment does not ring falsely, at least to me. In this way the old struggle for particular values, for particular forms of politics and law, continues even in the teeth of the discursive dissolution of the self. Were it otherwise all our comprehension of the relative autonomy of the law would not much matter.

Generalization in Interpretive Theory

Joseph Vining

There are arguments at large about the nature of legal interpretation, proceeding from an implicit proposition that interpretation is the same phenomenon or experience whatever its setting. An assumption that there is one phenomenon can be found in discussions among lawyers of interpretation and in discussions among nonlawyers of legal interpretation—and as often in the work of those who would deny there is any significance to theorizing about interpretation, as of those who think persuasion to a particular theory will have the utmost consequence for law and society.

Proceeding from such a proposition, rather than toward it, raises the risk that distinctive features of legal interpretation may be overlooked. If there is to be a common understanding or theory of interpretation it should not be built upon misinterpretation of the evidence. As examples from law appear more frequently in nonlegal settings, and nonlegal examples in discussion of law, distinctive features of legal interpretation are the more easily overlooked. These features are linked, but they may be roughly divided and treated under four headings. Some are more obvious than others. Some will be more obvious to lawyers than to nonlawyers, but it is an odd characteristic of current discussion that the reverse may also be true.^[1]

Identification of the Text

The first of these features of legal interpretation has to do with what is often called identifying a text. Some, such as Steven Knapp and Walter Benn Michaels, argue that identification of marks or sounds as a text involves belief that they are a product of the intentions of an agent (using "agent" in its special philosophic sense rather than its common or legal sense).^[2] Others would identify marks as a text through their relationship to other marks in a somewhat free-floating system.^[3] But, for both, there seems to be assumed something lying before the reader, some determinate set of marks, some work, much as there is an object before one when one comes up to a piece of sculpture or listens to a piece of music that has a beginning and an end. That is not the situation in law.

Not that one never comes in law to a sentence, a page, an opinion one has to read and make sense of; of course one does, but that is at the very end of one's

interpretive work, and perhaps that is not even then the end. For in the repeated runs at analysis of a situation made possible by a hierarchical institution of argument, one piece of writing with the meaning of which one has been wrestling, one section of a statute or constitution, the law of one jurisdiction or another, may be pulled away and replaced by another sentence, another section, another book.

When a practitioner asks, "What is the law here? How is this case to be analyzed?" no hand thrusts out a text and says, "Here, this is what we are now going to read and construe." Interpretation in law is, from the beginning, of the law. The law is not to be equated with any particular set of marks or sounds nor even with the meaning of any particular set of marks or sounds. And a small methodological consequence of this may be noted, that it would not be true to the experience of interpretation in law to suppose that if one could achieve an account of the identification of and reading of one or another example of a legal text, one would then have in hand the problem of legal interpretation.^[4]

Cases do not come ready made, as teachers repeatedly emphasize to students being gradually introduced to law. The initial question in practice is of the form "How ought this situation to be thought about under the law?"—after, of course, the situation is delimited as a situation —or "What ought to be done here under the law?" In the course of analyzing a situation, one text, then another, may be invoked and become the focus of attention. Starting as an "Occupational Health and Safety Act case," a death in a factory may end as a case of corporate homicide under the general criminal law. It may become a constitutional case as well as a case of ordinary law. A car repossession case involving allocation of money and risk, seen immediately as a case of common law contract or a "Uniform Commercial Code case" between "buyer" and "seller" or "debtor" and "creditor," may in addition become a "Federal Trade Commission Act case" between "dealer" and "customer," involving federal administrative regulation of economic power. Or it may become a case of civil wrong, a tort case, rather than a contract case, or a securities law case rather than a case of either contract or tort.^[5] The applicability of particular texts, the focusing upon particular judicial opinions, evolves as analysis proceeds.

Recalling this very basic aspect of the practice and experience of legal interpretation is not to introduce what is sometimes thought most distinctive about the world of law, the clever litigator working against that which is self-evident. There is no self-evident nature of a case to be worked against. If a case comes to litigation (as it does, indeed can, in only a very tiny percentage of the situations in which a legal question is put), it may be given well-formed contours by the time it gets to an appellate court. Still, appellate judges, even appellate judges, must struggle with its nature or characterization. Once an appellate opinion is written, of course it appears the case always was what it came to be. The interpretive question seems to be how given texts are applied to particular facts—even though those particular facts are remarked upon, and corralled off from the whole, in response to

the proposed applicability of texts. And perhaps, once the appellate opinion is written, the interpretive question not only seems to be that, but is. But then that appellate opinion becomes merely one more of the many expressions of the law that legal practitioners, starting again with another situation, may or may not be led to as they go about considering what the meaning of the law is for the situation they have before them.

This evolutionary quality of the text to be interpreted, that is encountered in practice, is not simply a matter of canon formation—particular texts or parts of texts gradually becoming central in legal analysis, the focus of discussion, through hierarchical devices or otherwise. And it does not simply reflect the flight of any text, once composed, into mere candidacy for attention (though this is implicated: if authority consists in being paid attention—which is also a form of praise—every text is only a candidate for attention, and its capacity to evoke and maintain attention can never be assumed). Largely unexplored, this is in law an analogue, pale perhaps, of the formation of melody in music. It is difficult to explore, however comfortable with it thousands of practitioners appear to be in their daily work. On it, on what a case is "about" and how it takes shape, any very direct comment confronts the special problems language imbued with the static poses for discussion of the dynamic.^[6] But an eye can be kept on it, and an eye at least on the actual dynamism of legal thinking must help avoid a misleadingly photographic view of law as one in which a human being or human beings in general stare at a line carved over a doorway, or at a book in the hand, or, to take an example sometimes discussed in literary theory, tracings in the sand on a beach.^[7] Particular texts simply do not step forward when one asks, "What is the law here, how am I to interpret the law?" And even the universe of possibly relevant texts always has a quivering edge.

This fact indeed, or, if not fact, this truth of legal experience, is always in the background of "easy cases." A case easy in the hands of one may not be easy in the hands of another, and the text that records the disposition of a case as an easy case becomes just one more datum to be considered critically by a later analyst looking at a new situation. Situations are not, in contemplation of law, each unique in the sense Tolstoy took to be his own view of situations in history;^[8] but situations in contemplation of law are living, and that alone brings them, and the law, out from under the picture of matching up a rule to a set of facts.^[9]

Types of Writings

Mention of constitution, statute, contract, opinion points to the second problem in the thrust toward generality in discussions of interpretation. The fruitful pulling of experience in law toward experience in other fields has in it the danger that those reflecting upon interpretation may be tempted to induce

their sense of interpretation from the instances of it they see, including instances in law, without sufficiently taking into account that much of the writing being treated in law may be different in kind and that writings in law may differ in kind from one another. What is perceived to be going on in work with a contract, for example, is not ground for any very definite assertion about what is done in working with a statute, and neither the handling of contracts nor of statutes is more than a provocative source of speculation about the way a constitution or a set of opinions is read. Even among lawyers methodological conclusions are sometimes drawn after—and as a result of—carelessly lumping various kinds of legal texts together.

In literature the language of texts is rarely negotiated and bargained over. Particular texts rarely begin their life fashioned by more than one hand. There is Beaumont and Fletcher, there is the phenomenon of editing, there is the folk ballad built up through successive contributions of anonymous singers, and there is the writer who writes or revises a piece over a lifetime. But among the texts literary critics discuss there is not often one in the genesis of which an individual agrees to words he does not want or mean to say, through some exchange relationship with another or as a result of delegating the writing of it, or the words of which are the immediate outcome of a process of assembly in which many have been independently at work. Nor is there often one with words attributed to a human being who may never have seen them, as to a voter or to a principal "bound" by the words of an agent. In law such texts are encountered quite often, and this should give pause to literary critics working with law, and indeed to philosophers reaching out to law who are similarly schooled in and accustomed to text-focused discussion in their various fields of professional inquiry.

The differences to be found upon examination of legal texts, type by type and text by text, are pertinent to the large uncertainties of current discussion. In particular, lawyers' work with contracts, statutes, the writings of agents, and the writings of delegees does not as such point to the objectivity of language or the irrelevance of a statement's origin. To take only contracts and statutes: in the construction of many contracts the true intent of the parties is explicitly excluded as a subject of inquiry. This may illustrate only that in many cases legal analysis is not interpretation of meaning. Legal analysis of writing can proceed rather in what might be called a "tort mode." "Tort," the law of civil wrong, has as its concern action and visiting the consequences of action upon the actor, and a party may be made to pay money (or money is shifted from one stream of wealth to another) because she should have known the consequence of inserting or allowing someone acting for her to insert a mark or a set of marks into a machine called a contract, regardless of what she might have wanted those marks to express.^[10] The history of contracts is prominent in the teaching of contracts, in great degree because the emergence of contract from tort and the difficulty even now of distinguishing between tort analysis (of acts intended or unintended, and their con-

sequences) and contract analysis (if the latter suggests inquiry into the concrete meanings of words) can be thus demonstrated and emphasized.

As for statutes, often and perhaps typically the origin of a statute's words in committee processes and bargained or strategic voting forecloses inquiry into the meaning of those who have touched it. Nonetheless, statutes in general seem to be read, and not simply as vector products of the forces that made them in the legislative machinery.^[11] But this fact, this evidence of what lawyers do, is again not ground for conclusions (by lawyers or nonlawyers) about the way other kinds of texts—opinions or administrative regulations or, a fortiori, nonlegal texts—are or should be read. The reading of statutes for "their intent," the paying of close attention to nuance and form in them, may be a necessary and even desirable form of self-delusion. Self-discipline is still discipline; a society under the rule of law may allow itself some arbitrary freedoms, play a trick or so on itself. But the scope of such reading and the amount of such attention are limited by the necessity and the desirability of self-delusion.^[12] It is not to be concluded that because, in law, texts that are the product of machinery seem to be interpreted, therefore interpretive theory should pay attention generally to marks on paper and sounds that are the product of processes untouched by mind. The often noticed tendency in legal practice to pull back to and focus upon judicial opinions may be not merely a result of the conventions of legal education, evidence not merely of judge-centeredness or concentration upon litigation by sociologically uninformed lawyers, but a reflection of a belief that opinions are more likely to be texts that can be read and can be the subject of interpretation.

Distinctions between Reading and Writing

The third observation to be made about generalization in current discussions of interpretation is that it may be overlooked how much lawyers—readers of legal texts—are also and (almost) at the same time writers of legal texts. In order and association this point could as well follow discussion of the first—the difficulty of establishing what the text is that one is to interpret—as the second, the differences in the kinds of writings that are attended to in law.^[13]

While there are sensitive general accounts of the phenomenon of reading that show the reading of a major literary work and the reading of law as a responsible reconstitutive act,^[14] the widespread image of law as given to the subject—whether in Babylonian bas-relief, Weberian sociology, or modern positivism^[15] —has in it the potential to mislead, and may exert the greater influence as law ceases its intellectual isolation and excites the curiosity of those who are not lawyers. When one picks up a novel or poem the possibility of a difference between reading the novel or poem and writing it is an obvious starting point (not quite

so obvious of course in music or drama, where performance is necessary to the play or the song).^[16] But in law a difference between reading and writing is not an obvious starting point at all, not just because one cannot pick up and hold in one's hand a large and shifting mass of legal texts with indefinable edges, but because as a lawyer one reads for the purpose of oneself making a statement of law for which one is responsible. Lawyers in the schools make their statements to students or the world at large; in administration or in what is commonly designated as the practice of law lawyers make their statements to client or commission or, acting as judges or attorneys general, to the world at large.^[17] The lawyer reads in order to write, must read from the beginning in an active frame of mind and must make a statement of—and for—the law. That statement, in turn, becomes part of what is read for a time by others who themselves must make statements of law. Insofar as lawyers listen to any general account of interpretation that ignores their constant writing of the law, they are led astray; and insofar as those outside the profession seek to illumine their own experience of interpretation with an understanding of legal interpretation as passive or static, they are led astray.

Identification of the Author

How much do all these points—the difficulty of identifying the text, the variety of texts, the joinder of writing with reading—fold into and derive from that most signal feature of legal discourse, that writers of legal texts do not speak for themselves? Perhaps this is a fourth point; perhaps it is not (but it is not the same as the appearance in law of delegated writing). Certainly it should not be overlooked when looking to law.

To put it concisely though paradoxically, if writers of legal texts (again, not including contracts, which may not be speech) were to speak for themselves, they would speak without authority. They speak for the law or for an entity—the court , the agency , the legislature —that in turn speaks for the law, with the possible exception of the legislature, and even then there is an ostensible speaking for, for the "sovereign" if not "the law." The late-twentieth-century sensibility may want to ignore these others, these authors beyond, as passé figments, but if one does ignore them one does so at the price of depriving speech of authority.

It is further true that a speaker's speaking for an entity—or for the law—is never a given, but is a question always alive, to be determined by the listener. Though a situation may be disposed of for the moment, as if the speaker spoke the law, the question whether the speaker was authorized so to dispose of that situation may be examined whenever the disposition is examined; there is no foreclosing of the question: it is posed and answered in a nonfinal way again and again as time goes on.

From this general observation legislators again may possibly have to be

excepted (other than with respect to the narrow range of utterances that may be disregarded on constitutional grounds). But if the legislator is an exception, that very fact is part of the reason listeners are pressed back to the utterances of judges—that blank interposition, that wall between the listener and the author whom the listener is seeking to hear, which a legislative claim always to speak for the sovereign presents. Of course the legislature could claim to be not agent but itself sovereign, but that is not the assertion that is made. It is, possibly, an assertion that never has been made where authority has been claimed and law invoked.

Convergences

Emphasizing these features of the experience of legal interpretation does not proceed from or inevitably toward some form of nominalism. Consider again that central term in current discussion, intent. Precisely because they are writers lawyers might be expected to be especially sensitive to the complexities of intent. That they are writers is what might hold them back from embracing propositions about authorial intent made in arguments either for or against its existence or importance to interpretation—this rather than philistinism or foreignness in the work they do.

Legal writing is often not expressive. Very often it is manipulative only. But when it is expressive and there is true effort to make a statement of law, participation in it is almost as fine as participation in any kind of writing. The writer of a poem does not say what the poem means by reference to his conscious intent in the writing of it, because he cannot; and in adding one more voice to those who have made this point I may invite for company the not insubstantial number who, though they may not consider themselves poets, have been moved now and again to write a poem. If the poet could so explain himself, he might then have written differently. Much about a poem can in a real sense be explained to the writer of it without (as we say) putting words into his mouth. Listening to another expound his poem, the poet, if he is to respond at all, reflects not on the memory of his conscious thought but on why he stopped his daily business to write (daily business including taking a walk or staring at his desk) and, if the poem did not appear in full bloom, on how and why he labored over it as he did. The experience of reaching—one always reaches in a poem—includes the experience of trying to get lines and words right. And, it must be said, the writer of legal prose does the same.

He writes, and looks. The thrust of this seems inconsistent with that , the emphasis within this passage is misleading, this or that word is not just (the very metaphor just is a legal one). The writer tries to get the word, passage, or statement more satisfactory, or, in some rare cases, satisfactory. If he does not abandon it for reasons of time or fatigue, it may eventually lock into place, not be improv-

able without hurting it in some (other) way. But, as he works, molding and rewriting, he cannot say what he is doing. If he could the labor would be over. It will be allowed, and by those who dwell upon the writing of a paragraph of prose as well as those who venture to write poetry, that in the placing of words, and in all that touches upon the aesthetic of echo and multiple reference, much self-consciousness or consciousness of what one is doing leads away from the living and resonant and toward the rote and the dead. But the same is true in the first drawing of a word from the mind, when, as the saying is, a word comes to mind.^[18] Intentions are veiled, they are being expressed in the work, and writer as well as reader looks to the work as evidence. (The lawyer especially is in the position of having readers: other writers may be able to conceive of writing without readers talking to them about it, but not lawyers, who are therefore constantly confronted with the question of their own intent.)

And what the reader does, the close reader, is seek ultimately to identify with the writer who is thus reaching and trying. If such resonance and such identification turn out to be possible, the writing is good, a carrier of meaning. Hermes is without clothes, save for his traveling hat, because he assumes such a multitude of forms.

Advantages of Particularity

There are thus generalities to be achieved. It needs merely to be noted that what is critical in the achieving of them is to see from the inside those experiences that are to be merged. But particularity has its larger uses too. An attentive understanding of law might be helpful in saving practitioners in other fields from misdirection. Take as an example the field currently being most vigorously developed—with what is widely thought to be the most significant potential for large practical changes in ways of life—the design and programming of sophisticated computers.

Work on equipping computers to process human language draws on linguistics, which in one of its prominent current forms distinguishes what are proposed as rules for the construction of sentences, "syntax," from "semantics," which has to do with the meaning of the constructed sentences. One hears it said that syntax determines what constitutes a legal sentence of human language, and there is hope for machine programs that will reject illegal sentences—all without regard to meaning.^[19]

The terms legal and illegal in this setting should be a little warning, a cause for caution. Legal and illegal are powerful and attractive terms, used to make the governing distinction between syntax and semantics more vividly plausible, but also quite possibly drawn upon for the substance of the distinction itself. There is something of the same in Virginia Woolf's reference to "law" in her comment

in Three Guineas , "The Church being a spiritual profession has to give spiritual and not merely historical reasons for its actions; it has to consult the mind, not the law."^[20] If it should come to be understood that consulting the law is consulting the mind, and that what is legal or illegal is not determinable without an inquiry into meaning, the work of machine designers and programmers and speculation upon the possibilities of future human organization based upon their contributions would not be confuted; but legal practice and thought would no longer be a source or prop of work in these fields. Its powerful attractions and validations would be withdrawn. If machine designers are then left with a notion of rules, divorced from meaning and mind, that can be connected only to natural science and the laws and rules thereof, and if this should prove unsatisfactory given the source and character of a scientific law, they may just possibly be freed to proceed in new directions, and will at least be somewhat better protected from elaboration that in the end proves disappointing and discouraging, if not indeed dangerous as it is absorbed into the forms and methods of human organization.^[21]

Prospects

So too explorers from the literary and linguistic disciplines who sail into law may discover something of help to them in their own thinking, if they stay for a time—if they do not return too quickly with the news that great numbers of apparently intelligent people are reading for meaning marks on paper that no one means, or marks that are the product of forces not associated in any way with mind. If they come to law, as some literary critics do, bringing with them belief that there is an ineradicable connection between the spoken and a speaker, they may help lessen somewhat the pressure lawyers have felt so long from pushing against the long depersonalizing thrust of the modern age. Their presence may even make it more difficult for lawyers to mislead themselves, as they do, when talk turns to the Rule of Law and (trying to express what it is to live in a world in which no one including the legal analyst is above or outside the law) they end somewhat like the tragic Antigone, projecting an image of law as a set of rules outside, a grid that, could you only tap it with your fingernail, would give out a hard metallic ring. Lawyers may need nonlawyers to help them turn to what they actually do when under the Rule of Law they work responsibly with common and public texts.

Regardless of what those who are now discovering law leave behind, on their return to their own disciplines they may find it more difficult to collapse meaning and the speaking voice either into the intent of an author bound to time and place or into one or another form of system or process, a language system, the reasons of history. They may find that they are encouraged, themselves, to maintain a connection between words and intention, to keep words a gateway to intention

even for the fashioner of them. The small space within the skull of the individual is not the only retreat from the empty reaches of system and process, if it is a retreat at all. Nothing rules out another alternative a priori; lawyers cannot forsake the personal and the concrete for the impersonal objectivity of system, but they do not stop with mere particularity. After a sojourn listening to lawyers and attending to what lawyers actually do, some at least may be less uncomfortable than before with an occasional thought of a person speaking who is not an individual, an occasional thought even of a transcendental intention, less uncomfortable too with the possibility of conceiving their own activity as having, like law, something of a moral thrust toward the future and the whole and toward action—of conceiving writing and speaking, perhaps in general, as a discovery of what one did not know was in one, with others taking it up to place it against the whole and using it to make their own statements, which are then again subject to criticism against the whole by anyone who is also in responsible contact with what there is to be expressed.

Theories of Constitutional Interpretation

Robert Post

Modern democracy invites us to replace the notion of a regime founded upon laws, of a legitimate power, by the notion of a regime founded upon the legitimacy of a debate as to what is legitimate and what is illegitimate—a debate which is necessarily without any guarantor and without any end.
—Claude Lefort^[1]

In 1979 Ernest Chambers was a barber who had for nine years represented a predominantly black district of Omaha in the Nebraska Unicameral Legislature. He had been brought up in "a religious strait-jacket" in the fundamentalist Church of God and Christ, but as he had grown older he had come to renounce Christianity and all belief in God. Consequently he was uncomfortable when the chaplain hired by the legislature opened each session with prayer. In fact he felt compelled to leave the legislative chamber, so that he and the chaplain were "almost in a race to see whether" the chaplain could "get to the front before" Chambers could "get out the back door."^[2]

The chaplain of the Nebraska Legislature during that time was Robert E. Palmer, a Presbyterian clergyman who had ministered to the legislators since 1965. His prayers were short, almost perfunctory. He strove to make them nonsectarian, to reflect "just civil religion in America," which he understood to consist of "the Judeo-Christian tradition," the "kind of religious expressions that are common to the vast, overwhelming majority of most all Americans." He viewed the purpose of his prayers to be the provision of "an opportunity for Senators to be drawn closer to their understanding of God as they understand God, in order that the divine wisdom might be theirs as they conduct their business for the day." And so he would, for example, pray "in the name of Jesus—our Friend, our Saviour, our Example, our Guide," and he would "ask" that the Senators come to realize that "they are part of the team working together to win the game for the benefit of the people of this state."^[3]

Chambers attempted to convince his colleagues to end the practice of legislative prayer. When they refused, he took the characteristically American step of filing suit in federal court. His claim was elegantly simple: the payment of a state salary to the minister of a single Christian denomination for fourteen years for the purpose of offering official prayers to the state legislature was a violation of the Establishment Clause of the First Amendment to the United States Constitution. That clause provides: "Congress shall make no law respecting an establishment of religion."^[4]

The trial court held that while the payment of the chaplain's salary violated the Establishment Clause, the observance of legislative prayer did not.^[5] The appellate court went even further and declared that the whole "prayer practice" was unconstitutional.^[6] The case was then accepted for decision by the United States Supreme Court, by which time the concrete concerns of Ernest Chambers and Robert Palmer had dwindled to little or no moment. Chambers's lawsuit had become merely a medium through which the Court could ponder the legal meaning for the entire nation of the Establishment Clause. The methods by which the Court ascertains this constitutional meaning are of the utmost legal and political importance. Ernest Chambers's lawsuit would prove to be the occasion for an unusually clear and dramatic display of these methods.

I

Sometimes, although rarely, the words of the Constitution appear to speak for themselves. In such circumstances the Constitution does not seem to require interpretation. Article I, Section 3, Clause 1 of the Constitution, for example, states that "the Senate of the United States shall be composed of two Senators from each State." If a third California Senator should one day present herself for accreditation in Washington, D.C., no court in the country would think twice before disapproving of the application. From a phenomenological point of view, there would be no question of "interpreting" the constitutional language, for its meaning and application would appear clear and obvious.^[7]

The most famous expression of the experience of this clarity is by Justice Owen Roberts, who in 1936 wrote that the "judicial branch of the government has only one duty,—to lay the article of the Constitution which is involved beside the statute which is challenged and to decide whether the latter squares with the former."^[8] In legal circles, this approach is sometimes characterized as a "plain meaning" or "textualist" theory of interpretation. Yet, strictly speaking, the approach is not a theory at all; it is instead a description of what happens when constitutional meaning is not problematic.

But if for any reason that meaning has become questionable, it is no help at all to instruct a judge to follow the "plain meaning" of the constitutional text. A meaning that has ceased to be plain cannot be made so by sheer force of will.^[9] In Chambers's lawsuit, for example, either the meaning of the Establishment Clause with respect to the issue of legislative prayer is "plain," or it is not. If the latter, the question of constitutional meaning cannot be resolved by staring harder at the ten words of the clause. What is required instead is a means of interpreting the text so as to mediate between the clause and its application.^[10]

Because judges must be able to justify their decisions, they must also be able to justify the means of interpretation that they employ to reach those decisions,

particularly if their choice affects the ultimate result or significance of a case. Judges must be able to explain why they have decided to interpret the Constitution through one set of inquiries rather than another. In legal (although not in philosophical or literary) parlance, judges require and must be able to articulate a "theory" of constitutional interpretation.

Any such theory of interpretation, however, must accommodate itself to the role of judicial review within American democracy. When a court sets aside a statute as unconstitutional, it in essence deems the statute invalid in the name of the Constitution. Courts have claimed the power to do this because, in the famous words of John Marshall in Marbury v. Madison , the decision that established the institution of judicial review, "it is emphatically the province and duty of the judicial department to say what the law is."^[11] The implicit premise of this claim is that the Constitution is a form of "law," just like the law which courts ordinarily interpret and apply. Fidelity to law is a preeminent value in a nation that, as Marshall put it in Marbury , prides itself in being "a government of laws, and not of men."^[12]

But, as the force of Marshall's argument in Marbury also required him to acknowledge, the Constitution is something more than ordinary law; it is "the fundamental and paramount law of the nation." The Constitution is "fundamental" because it is the vehicle through which "the people . . . establish, for their future government, such principles as, in their opinion, shall most conduce to their own happiness." The Constitution is therefore "the basis on which the whole American fabric has been erected."^[13] The question arises, therefore, why it should be the province and duty of the federal judiciary to discern in that "American fabric" the "principles" and "opinion" of "the people," when that judiciary is not elected by and hence structurally responsible to the people. Why shouldn't that task be allocated instead to the democratically elected branches of government, which are presumptively in closer contact with the popular mind?

This question, which is sometimes termed the "counter-majoritarian" difficulty,^[14] has proved durable enough to sustain the work of generations of constitutional scholars. The question makes a powerful political point. Judicial determinations of unconstitutionality nullify the actions of democratically elected branches of government. Such determinations are for all practical purposes final; often the only formal recourse is the cumbersome and impractical process of constitutional amendment. "Who are these nine Justices," one may well have asked the Court in 1857 after the Dred Scott decision,^[15] "so definitively to instruct the nation about the 'American fabric'?"

II

The United States Supreme Court voted 6 to 3 against Ernest Chambers. If one were simply to view the Constitution as ordinary law, this outcome

would have been something of a surprise. The relevant precedents of the Court pointed unambiguously to the unconstitutionality of the practice of legislative prayer. As Justice William J. Brennan pointed out in his dissent, it is "obvious that, if the Court were to judge legislative prayer through the unsentimental eye of our settled doctrine, it would have to strike it down as a clear violation of the Establishment Clause."^[16]

In ordinary adjudication, courts follow the principle of stare decisis , which is to say that they follow the doctrinal rules laid down in controlling precedents.^[17] In American law, the principle constitutes a fundamental aspect of "the rule of law,"^[18] for it requires courts to decide cases on the basis of public and predictable rules, applied in an even-handed manner, upon which persons can rely in the conduct of their lives.^[19] In constitutional adjudication, "adherence to precedent can contribute to the important notion that the law is impersonal in character, that the Court believes itself to be following a 'law which binds [it] as well as the litigants.'"^[20] The principle of stare decisis helps to ensure that our constitutional order retains the kind of stability and continuity that are prerequisite for institutional legitimacy.

If the Court in Ernest Chambers's case had followed the principle of stare decisis , it would have deemed controlling, as did the Court of Appeals below,^[21] the three-part doctrinal test laid down in Lemon v. Kurtzman :^[22]

Every analysis in this area must begin with consideration of the cumulative criteria developed by the Court over many years. Three such tests may be gleaned from our cases. First, the statute must have a secular legislative purpose; second, its principal or primary effect must be one that neither advances nor inhibits religion; finally, the statute must not foster "an excessive government entanglement with religion."

The primary purpose of religious prayer cannot reasonably be deemed to be secular; nor can its principal effect be understood as anything other than enhancing religion. As for the potential for "excessive government entanglement with religion," it is apparent that official sponsorship of prayer necessarily entangles the state in decisions about which forms of prayer are appropriate or inappropriate. The word was passed to Reverend Palmer, for example, that Jewish senators in the Nebraska Legislature were offended by his many references to Christ.^[23] Eighty years before, when a state senator conveyed a similar message to the legislative chaplain of the State Senate of California, a local clergyman thundered that the senator's "words were those of an irreverent and godless man" and that his offense was a "crowning infamy."^[24] The point of the "entanglements" prong of the Lemon rule is to ensure that the state not be embroiled in religious quarrels of this kind.

"In sum," as Justice Brennan remarked, "I have no doubt that, if any group of law students were asked to apply the principles of Lemon to the question of legislative prayer, they would nearly unanimously find the practice to be uncon-

stitutional."^[25] A fascinating aspect of the Chambers decision, however, is that the majority neither disagreed with this assessment, nor attempted to alter the Lemon doctrine. In fact it ignored Lemon altogether, making no effort whatever to justify its decision by reference to past precedents.

Instead the Court, in an opinion written by Chief Justice Warren Burger and joined by five other Justices, focused its analysis on the fact that the "opening of sessions of legislative and other deliberative bodies with prayer is deeply embedded in the history and tradition of this country." The Court noted that most States of the Union have traditionally opened their legislative sessions with prayer, and that Congress has continuously employed chaplains to offer legislative prayer since the eighteenth century. Indeed, on 22 September 1789, three days before Congress approved the language of the First Amendment (and the Establishment Clause) and sent it to the States for ratification, Congress enacted a statute providing for the payment of congressional chaplains.^[26]

Although the Court conceded that "standing alone, historical patterns cannot justify contemporary violations of constitutional guarantees," it concluded that the evidence in Chambers's case was different, for it definitely established "not only . . . what the draftsmen intended the Establishment Clause to mean, but also . . . how they thought that Clause applied to the practice authorized by the First Congress—their actions reveal their intent." "Clearly," the Court concluded, "the men who wrote the First Amendment Religion Clauses did not view paid legislative chaplains and opening prayers as a violation of that Amendment."^[27]

The premise of the majority's opinion is thus that the meaning of the Constitution is better ascertained through strong evidence of the intent of the Framers than through fidelity to past precedents and doctrine. The reason is apparently that the intent of the Framers best embodies those "principles" which the "people" desired to instantiate in their Constitution. In the eyes of the majority, therefore, it is more important that the Constitution be interpreted in a manner which accurately expresses these principles than that it be interpreted in a manner which remains faithful to the principle of stare decisis .^[28]

The principle of stare decisis , moreover, is inconsistent with a quite different method of constitutional interpretation. William Brennan, in a dissent joined by one other Justice, also gave only cursory attention to the rules of Lemon . In fact he appeared to agree with the majority that "the path of formal doctrine . . . can only imperfectly capture the nature and importance of the issues at stake in this case."^[29] But rather than focusing on the intentions of the Framers of the First Amendment, Brennan offered instead an "account" of "the underlying function of the Establishment Clause" and of the relationship between that function and the practice of legislative prayer.

According to Brennan, the Establishment Clause embodies the twin principles of "separation between church and state" and "neutrality" as between diverse religions. These two principles, in turn, serve four purposes. They guarantee "the

individual right to conscience" by ensuring that persons are not coerced to support (through taxes or otherwise) religious practices with which they disagree; they "keep the state from interfering in the essential autonomy of religious life"; "they prevent the trivialization and degradation of religion by too close an attachment to the organs of government"; and they "help assure that essentially religious issues, precisely because of their importance and sensitivity, not become the occasion for battle in the political arena."^[30]

Brennan convincingly demonstrated that Nebraska's authorization of legislative prayer was inconsistent with each of these four purposes. Indeed, as Brennan noted, the majority said "almost nothing contrary to" this functional analysis, relying instead almost entirely on evidence of the historical intent of the Framers. The majority ultimately differed from the dissent, therefore, neither on the application of doctrine nor on the function of the Establishment Clause, but rather on the relevance of evidence of original intent for constitutional interpretation.

Brennan explicitly rejected such evidence as definitive of constitutional meaning, arguing that "the Constitution is not a static document whose meaning on every detail is fixed for all time by the life experience of the Framers." He contended that the Constitution must be understood instead as "a document meant to last for the ages," the bearer of an "inherent adaptability" that could not be cabined by any "static and lifeless" meaning. His proposed analysis of the Establishment Clause's "underlying function" was meant to illustrate how courts could discern the contemporary significance of "the majestic generalities of the Bill of Rights."^[31] His dissent pictured the Constitution as a living, evolving entity, whose full meaning could be ascertained neither by doctrinal precedent nor by evidence of original intent. The dissent, therefore, was predicated upon yet a third form of constitutional interpretation.

The outcome of Ernest Chambers's lawsuit thus appears as a triangular structure, in which three distinct theories of interpretation compete for control of the Constitution. In one corner is a form of interpretation that strives to implement the Constitution through the articulation of explicit doctrinal rules. In a second corner is a form of interpretation that attempts to construe the Constitution to reflect the original intent of its Framers. In yet a third corner is a form of interpretation that reads the Constitution in a manner designed to express the deepest contemporary purposes of the people. Each of these three theories is immediately recognizable and familiar to those who practice constitutional adjudication.

III

The purpose of constitutional adjudication is to assess the constitutional validity of state actions, like the hiring of legislative chaplains. But courts

can achieve this purpose only to the extent they have the authority to evaluate, in the name of the Constitution, the validity of otherwise perfectly legal state actions. Every act of constitutional interpretation invokes and depends upon this authority, and for this reason "constitutional interpretation is essentially about the sources of authority in American political life."^[32] What in fact distinguishes the three theories of interpretation displayed in Chambers —theories that I shall respectively call "doctrinal," "historical," and "responsive" interpretation—is that each appeals to a different conception of constitutional authority.

The Authority of Law

There is, first, the authority of the Constitution as law. The Constitution controls state actions because the Constitution is the highest law, above all merely quotidian state activity. The concept of the "constitution as hard law, law written in virtually capital letters (LAW), law as meaning reliable law," has been termed "by far the most important idea of the Constitution."^[33] Because "courts are the mere instruments of the law,"^[34] they are peculiarly fitted to interpret a Constitution whose authority lies in its character as law. It is therefore no accident that in Marbury Marshall appealed precisely to this image of constitutional authority in establishing the institution of judicial review.

If the Constitution predominates because it is law, its interpretation must be constrained by the values of the rule of law, which means that courts must construe it through a process of reasoning that is replicable, that remains fairly stable, and that is consistently applied.^[35] In American adjudication the principle of stare decisis has been an essential component of the rule of law.^[36] The principle is of particular importance on those occasions when constitutional adjudication involves vague textual referents (like "equal protection of the laws," or "due process of law"), with regard to which there is "only limited evidence of exactly how the Framers intended the [text] to apply."^[37] On these occasions the principle of stare decisis holds courts to a consistent and stable interpretation of Constitution.

Without such consistency and stability, it would be difficult to understand the Constitution as having any existence as law. Suppose, for example, that the Supreme Court were to decide one day in decision A that the practice of legislative prayer was constitutional, perhaps because in its view the Framers had so intended. Imagine that a month later the Court were to decide in decision B, without any reference to A, that the practice was unconstitutional, perhaps because the Court's view of the Framers' intent had changed. And assume that one month later the Court were to determine in decision C, without any reference to A or B, that the practice was partially constitutional, perhaps because its reading of the historical evidence had once again altered. In such circumstances state legislators would simply not know what to do; they would have no rule of law by which to decide whether or not they could constitutionally hire legislative chaplains.^[38]

It is of course implausible to suppose that the Court would so swiftly and radically change its assessment of the historical evidence. But the question is whether this implausibility derives from the unequivocal state of the historical record, or rather from the Court's implicit obligation to remain faithful to its own prior determinations. Since historical evidence is often equivocal, particularly with respect to matters of contemporary constitutional moment, it is the latter obligation, I would suggest, that plays an important role in enabling courts to create stable and predictable rules upon which persons can rely in the arrangement of their lives and institutions.^[39] This obligation receives formal acknowledgement in the principle of stare decisis .

This means that the principle of stare decisis often underlies the capacity of constitutional adjudication to generate a system of constitutional law . Thus the Chambers decision creates a rule of constitutional law only because of the implicit commitment of the Court to act in the future in ways that are consistent with the Chambers holding. Put another way, the legal implications of Chambers depend upon the implicit and necessary expectation that the Court will in the future treat Chambers in a way that it declined in Chambers to treat Lemon .^[40]

Of course the principle of stare decisis is an immensely flexible instrument, allowing courts to treat precedents on the one hand as the source of specific and binding formal rules,^[41] or on the other as an amorphous mass of material to be rendered consistent through the virtue of "integrity."^[42] What every application of the principle requires, however, is that a court focus its analysis on the doctrine which has emerged from relevant prior cases. The principle of stare decisis therefore creates a chain of cases, in which each decision is an interpretation of immediately prior decisions.

Construing the Constitution in a manner that is faithful to its authority as law thus leads to what I shall call "doctrinal" interpretation. The implication of doctrinal interpretation is that the actual text of the Constitution is remitted to one end of a growing line of precedents. Even if the very first judicial decision to interpret the Establishment Clause had concentrated its attention on the specific words of the Clause or the intentions of its Framers, the practice of doctrinal interpretation would require the second decision to focus chiefly on the meaning of the first decision, the third decision chiefly on the meaning of the second, and so forth. In this process the text of the Constitution recedes until, as one prominent commentator has put it, it comes to seem "rather like . . . a remote ancestor who came over on the Mayflower."^[43]

The vast majority of constitutional decisions rely primarily upon doctrinal interpretation. Novices are often quite struck by the relative absence of the Constitution from constitutional opinions, which seem oriented instead toward specific doctrinal "tests," like the Lemon tests, derived from prior judicial decisions. But this should be no surprise if the most powerful justification for the Constitution's authority is that it is law, and the most defensible justification for

judicial review is that it is the peculiar province and duty of the courts to expound the law.

The Authority of Consent

If doctrinal interpretation rests on the equation of constitutional authority with law, what I shall call "historical interpretation" rests instead on the equation of constitutional authority with consent. The story is simple and familiar. The Framers of the Constitution proposed a compact to limit the power of government; the people signified their agreement to that compact by their ratification of the Constitution, and that agreement is what gives the Constitution its authority. The interpretation of the Constitution should therefore be designed to give effect to the terms of that original act of agreement.

The story behind historical interpretation has enormous resonance in a liberal society like our own. It conceives of the Constitution as binding in the same way that a promise is binding, as a single voluntary act of willful self-regulation. In interpreting such a Constitution courts can portray themselves as merely the passive enforcers of the democratic will that "ordained and established" the Constitution. Thus, as former Attorney General Edwin Meese III could observe, "A Jurisprudence of Original Intention . . . reflects a deeply rooted commitment to the idea of democracy. The Constitution represents the consent of the governed to the structures and powers of the government. The Constitution is the fundamental will of the people; that is why it is the fundamental law."^[44]

Different variants of historical interpretation emphasize different forms of evidence as probative of that original exercise of "fundamental will." Thus for some commentators the constitutional "text" is a privileged form of evidence, because "the text is the intention of the authors or of the framers";^[45] whereas for others the "relevant inquiry must focus on the public understanding of the language when the Constitution was developed."^[46] By far the most common form of historical interpretation, and the one used by Chief Justice Burger in Chambers , regards the intentions of the Framers as the best evidence of the agreement represented by the Constitution.^[47]

This form of historical interpretation has become quite controversial in recent years, in part because some members of the resurgent Right have attempted to use historical interpretation as a means of constricting the discretion of supposedly liberal judges. The notion is that judges will have less room to maneuver if they are bound to the specific factual intentions of the Framers. But this notion is a vulgar misinterpretation of the principles of historical interpretation, as is elegantly illustrated by an example suggested by Paul Freund. Article I, Section 8, Clause 14 of the Constitution gives to Congress the power "to make Rules for the Government and Regulation of the land and naval forces." It can be said with complete certainty that no one in the eighteenth century had the intent to endow Congress with the power to make rules for the regulation of an

air force. But no reasonable person would conclude from this undisputed fact that Congress does not now have this power.^[48] This is because the intent behind the Clause would naturally be understood as giving Congress the power to regulate the "military" or the "armed forces," or some other such general concept.

The point of Freund's example is that the intent of the Framers cannot be understood as the kind of simple historical fact that resists interpretation. It must instead be conceived as a purpose which can be characterized in terms that are more or less general.^[49] Once this move is taken, however, historical interpretation assumes a flexibility inimical to the political purposes of the Right.

Historical interpretation is a rather uncommon phenomenon in modern constitutional adjudication. In part this may be due to the logical and evidentiary difficulties involved in the effort to unearth historical intentions. It is hard enough to ascertain the intentions of a living individual. It is harder still to determine the intention of a group of living individuals, like a legislature or a Congress. The difficulty is compounded when the group of individuals is two centuries remote in time and the evidence of their thoughts and purposes is scattered, fragmentary, ambiguous, and conflicting. And the task is made almost impossible when the relevant intentions pertain to questions which in all probability never occurred to that group of individuals and which are meaningful only in light of circumstances that would to them be inconceivable.

It is the rare case indeed that, like Chambers , seems to present strong evidence of original intention on the precise question to be adjudicated. Even in such a case, as Brennan points out in his dissent, the logic of democratic consent requires that the intentions of those who ratified the Constitution be controlling, rather than the intentions of those who merely proposed constitutional language for popular adoption.^[50] And, as Brennan cheerfully notes, "'We know practically nothing about what went on in the state legislatures' during the process of ratifying the Bill of Rights."^[51]

It is important to understand, however, that these obstacles to historical interpretation, while formidable, are not necessarily fatal. First, in any given case the available historical evidence of intent may be more or less compelling. Second, the nature of the evidence that will count as probative of intent may itself be entirely a matter of "generally accepted conventions,"^[52] and hence shaped in a manner designed to ease the course of historical inquiry. For example, historical interpretation now focuses on the intent of the Constitution's framers, rather than of its ratifiers, because the former are by common convention taken to be conclusive of the latter. Similarly, The Federalist Papers are by common convention now presumed to constitute authoritative (and convenient) evidence of the intent of the Framers, although any historian could easily demonstrate the empirical inadequacy of the presumption. This tension between the kind of evidence of intent necessary to legitimate political authority and the kind of evidence of intent necessary to persuade professional historical judgment illustrates the truth of

Nietzsche's remark that "history, so far as it serves life, serves an unhistorical power."^[53] Third, historical interpretation need not focus on the intentions of the Framers or Ratifiers at all, but may attempt instead to ascertain consent through inquiries aimed at altogether different kinds of evidence.^[54]

Ultimately, therefore, the infrequency of historical interpretation in contemporary constitutional interpretation may stem less from evidentiary difficulties than from the intrinsic limitations of any theory of interpretation resting on the authority of consent. If that authority is understood to arise at the moment of the Constitution's ratification, then in fact no living person has "consented" to the First Amendment, or indeed to most of the Constitution. Why, it may be asked, should the consent of our predecessors have authority over us?^[55] When faced with this difficulty, consent theorists often resort to notions of "implied" or "tacit" consent, notions that rapidly drain the concept of consent of its ability to legitimate authority. These notions have a stopgap, jerry-built quality that renders them ultimately unsatisfactory.^[56] In fact Hanna Pitkin has demonstrated that principled consent theorists like Locke or Joseph Tussman, when seriously pressed with the absence of actual consent, transform the issue into a question of hypothetical consent. "True authority" thus "emerges as being one to which [persons] ought to consent , quite apart from whether they have done so."^[57]

A similar transformation is visible in the arena of constitutional interpretation. Thus it is said that even if the "legitimacy" of the Constitution cannot rest upon a prior act of consent, it may nevertheless be founded on the fact that persons now ought to view it as "a good Constitution and therefore one worthy of continuing support."^[58] This is essentially the form of constitutional authority appealed to by Justice Brennan in his dissent in Chambers . Because interpretation founded on this form of authority must ultimately be accountable to contemporary concepts of value, I shall call it "responsive interpretation."

The Authority of Ethos

The classic statement of responsive interpretation is by Oliver Wendell Holmes:

When we are dealing with words that are also a constitutent act, like the Constitution of the United States, we must realize that they have called into life a being the development of which could not have been foreseen completely by the most gifted of its begetters. It was enough for them to realize or to hope that they had created an organism; it has taken a century and has cost their successors much sweat and blood to prove that they created a nation. The case before us must be considered in the light of our whole experience and not merely of what was said a hundred years ago.^[59]

For Holmes the authority of the Constitution is not exhausted in a single creative act of consent, but continues to inhere in the national "being" that the Constitution has "called into life." Hence the nature of that authority can be captured neither by rules laid down in judicial precedents, nor by notions of original inten-

tion. The authority must rather be conceived as flowing from the "whole experience" of nationhood. That experience legitimately claims our allegiance because we are necessarily included within it, and hence responsible both for what it has been and what it might become. What is authoritative is thus neither more nor less than our common commitment to the flourishing of the mutual enterprise of nationhood.

The radical and paradoxical implication of this perspective is that the Constitution explicitly loses its character as a specific document or a discrete text. It becomes instead, as Karl Llewellyn bluntly put it, a "going Constitution," a "working Constitution" which has a content that "is in good part utterly extra-Documentary," and which represents the "fundamental framework" of "the governmental machine."^[60] In this way the Constitution is transformed into what Kant might call the "regulative" idea of the enterprise of constitutional adjudication, the "imaginary focus from which the concepts" of that enterprise "seem to proceed, even though there is nothing knowable at that focus."^[61]

The Constitution as a regulative idea defines the telos and shape of constitutional interpretation: it demands a continual effort to articulate the authority of our "fundamental nature as a people" and hence concomitantly to summon "us to our powers as co-founders and to our responsibilities," in the full knowledge that "how we are able to constitute ourselves is profoundly tied to how we are already constituted by our own distinctive history."^[62] In this sense responsive interpretation requires judges to view the Constitution as a form of what Philippe Nonet and Philip Selznick have called "responsive law," law that submits to "the sovereignty of purpose" by functioning "as a facilitator of response to social needs and aspirations."^[63]

There is a tension, however, between using law to implement a succession of merely present purposes, and using law to sustain the "general ends"^[64] constitutive of our "fundamental framework" of governance. The authority of constitutional law inheres only in the latter, for it alone claims fidelity to the "whole experience" that has comprised "our distinctive history." To maintain its legitimacy, therefore, responsive interpretation must be oriented toward the kind of general ends that have been closely linked over the long run to an historical instantiation of national identity. But such ends can provide the basis for adjudication only if they can also "be made objective enough and authoritative enough to control adaptive rule making."^[65] In this regard Justice Brennan's dissent in Chambers is paradigmatic. His effort to inquire into "the underlying function of the Establishment Clause" is specific enough to engender legal consequences, but general enough to express a deep vision of the secular nature of the American state.

Although the theory of responsive interpretation sounds exotic, responsive interpretation is in fact rather common in judicial opinions (certainly much more so than historical interpretation). In the area of the First Amendment's guarantee

of freedom of speech, for example, the outcome of cases depends upon whether judges perceive the purpose of that freedom to be that of assuring an "unfettered interchange of ideas for the bringing about of political and social changes desired by the people,"^[66] or instead that of fostering "individual liberty and dignity."^[67] In the area of the constitutional right to privacy, the outcome of cases depends upon whether judges conceive the purpose of the Due Process Clause to be that of safeguarding those "liberties that are 'deeply rooted in this Nation's history and tradition,'"^[68] or instead that of protecting intimate decisions of a kind that "define one's identity."^[69]

Responsive interpretation is in fact a vast umbrella sheltering a myriad of different approaches to the Constitution. It need not have the specifically liberal cast that is visible in Brennan's dissent in Chambers . It can be used by those who stress the constitutional priority of democratic decision making and hence who emphasize judicial caution and prudence, as well as by those who stress the constitutional primacy of individual rights. It has commonly been used by judges and scholars of both the Right and the Left.

Responsive interpretation does, however, have an important vulnerability. It contains within it no particularly persuasive response to the counter-majoritarian difficulty. If doctrinal interpretation portrays courts as merely the instruments of the law, if historical interpretation portrays courts as merely the instruments of an original democratic will, responsive interpretation portrays courts instead as arbiters of the fundamental character and objectives of the nation. And why, it may be asked, should courts be entrusted to act in that capacity, particularly when in doing so they set aside alternative visions of the national character and objectives propounded by the democratically elected branches of government?

One possible response to this question, which is increasingly visible in the literature, is to stress Hans-Georg Gadamer's theory that all interpretation necessarily involves a conversation between a reader and a text, and so effects a merger between a text and a reader's own purposes and perspectives. Even if this theory is accepted, however, it does not repair the vulnerability of responsive interpretation. This is because the theory's thrust is entirely to describe the conditions that make reading possible, and hence it can offer no guidance to the judge who, having determined the original intent of the Framers to the best of his ability (and therefore in a manner necessarily influenced by his own perspective), must decide whether to be bound by that determination (like Chief Justice Burger in Chambers ), or instead to set it aside in favor of a more self-consciously responsive approach (like Justice Brennan in Chambers ). The implications of hermeneutic insights for theories of constitutional interpretation are thus quite modest, a fact that is recognized by its more sophisticated proponents.

The acknowledgment of these limitations is, for example, the point of David Hoy's distinction between the "application" of a text, which is "a prior cognitive operation where we first find the text to be saying something to us," and the

"appropriation" of a text, which is "a willful, self-conscious act": "Application is not an option and is not subjective. But appropriation (e.g., making the text seem more rather than less relevant) is an optional strategy, such that it can be used or avoided."^[70] Responsive interpretation is a matter of appropriation, and as such cannot be defended by the hermeneutic turn.

IV

We are thus in a position to connect each of the three theories of constitutional interpretation displayed in Chambers to a different conception of constitutional authority. Doctrinal interpretation, which follows the principle of stare decisis , invokes the authority of the Constitution as law. Historical interpretation, which implements an original act of will, is validated by the authority of the Constitution as consent. Responsive interpretation, which engages in an ongoing process of national self-definition, appeals to the authority of the Constitution as, for lack of a better word, ethos.

I will not make the strong claim that these are the only possible conceptions of constitutional authority (and hence the only possible theories of constitutional interpretation), but I will make the more modest descriptive claim that these three conceptions dominate the actual practice of constitutional adjudication. All the many methods of constitutional interpretation that have proliferated in the legal literature of the past decade, ranging from those that stress the values of democratic participation to those that stress the values of autonomous individualism,^[71] ultimately rest upon one or another of these three conceptions of constitutional authority.

Each of these forms of authority is by itself incomplete and incapable of sustaining the enterprise of constitutional adjudication. The authority of the Constitution as law, for example, requires the authority of either consent or ethos in order to initiate a chain of precedents. The authority of consent and that of ethos, on the other hand, each require doctrinal elaboration in order to find embodiment as law. The authority of consent anchors constitutional interpretation in the democratic principles that are necessary and desirable in a country like the United States, while the authority of ethos offers an indispensable flexibility in the interpretation of a document designed to last for the lifetime of the nation.

Yet if these three forms of authority are on one level systematically interdependent, they are at a different level potentially divergent and incompatible.^[72] As the Chambers decision illustrates, a court in a constitutional case may be called upon to decide which form of authority should govern its efforts, and its decision may determine the outcome of the case. Because this decision is most often understood to depend upon an antecedent characterization of the Constitution

(as, e.g., "law," "compact," or "ethos"), arguments about theories of interpretation commonly modulate into arguments about the inherent "nature" of the Constitution. To the extent that the three theories of constitutional interpretation are perceived as incompatible, it is due to the fact that they are seen as flowing from incompatible notions of the Constitution itself.

But this vision of constitutional authority is fundamentally flawed, for it postulates a form of constitutional authority that is external to the processes of its own interpretation. It imagines that the nature of the Constitution can somehow be determined in a manner which is independent of the practice of constitutional interpretation, and that the practice is therefore logically controlled by this antecedent determination of constitutional authority.^[73] But a better account of the practice of constitutional interpretation would situate constitutional authority instead in the relationship obtaining between participants in that practice and the Constitution. Paradoxically, then, constitutional interpretation is not merely about the Constitution but about the more radical and profound question of how we stand in connection to the Constitution.

The Nature of the Authority of Law

If we ask, for example, what it means to defer to the authority of the Constitution as law, the answer is that this authority embodies the values of stability, predictability, and reliance which are necessary to the legitimacy of any modern legal system. Not only are these values themselves important, but they are also the means by which the law orders behavior so as to achieve justice and other desired objectives. The authority of the Constitution as law flows precisely from the acknowledgment of these values.

Once this point is made clear, however, it is also evident that these values, no matter how important, may or may not be compelling in particular circumstances. The values of the rule of law are most pressing when there is agreement that the law is generally just and otherwise fulfilling its proper purposes. In such circumstances doctrinal interpretation and the principle of stare decisis hold the law steady on its course. But if there is disagreement about the justice of the law, or about its purposes, or about its effectiveness in achieving those purposes, then the balance can begin to tip away from the values of stability and predictability.^[74] At a certain point, when dissatisfaction with the status quo reaches a sufficient magnitude, we can expect to see the doctrinal chain snapped.^[75]

This means, however, that the authority of the Constitution as law does not stand outside the processes of constitutional interpretation, like an axiom in a geometrical proof, but is rather implicated within that very process. In any specific case we may question whether that authority is compelling enough to mandate a particular result. Thus it is not the antecedent "nature" of the Constitution that requires doctrinal interpretation, but rather the decision to recognize and be

bound by the values embodied in the authority of the Constitution as law. By acknowledging these values we create a certain relationship to the Constitution, one in which the authority of the rule of law becomes visible and pressing.

In the American legal system this authority appears both flexible and inevitable. It is not disabled even if in particular cases we deny its mandate and break with the principle of stare decisis . That is because when the chain of doctrine is broken and precedent is either explicitly or effectively overruled, a new decision must be announced, and for that decision itself to have any effect, it must be crafted in the form of a rule of law that will be respected according to the principle of stare decisis . This means that doctrinal interpretation is presupposed even in the moments of its repudiation. Thus although the practice of constitutional adjudication at times may and sometimes must depart from doctrinal interpretation, it is a form of interpretation to which the practice will also inevitably return.

The Symmetrical Nature of the Authority of Consent and the Authority of Ethos

If doctrinal interpretation views the Constitution only dimly at one end of a long corridor of precedents, historical and responsive interpretation each confront the Constitution, so to speak, face to face. The directness of this inquiry liberates courts from the chains of doctrine, and empowers them to alter and amend precedents. It also empowers them to uncover and articulate substantive constitutional values. For historical interpretation, this power rests on a court's claim to speak with the authority of an original act of consent. For responsive interpretation, this power rests on a court's claim to speak with the authority of our deepest national identity and commitments. Although these claims appear on their surface to be very different, as different as Burger's majority opinion in Chambers from Brennan's dissent, in fact they each share an underlying structural similarity.

The authority of consent rests on the capacity of the individual voluntarily to assume obligations. Absent special circumstances to the contrary, a person's contracts are viewed as binding and authoritative. This fact has important consequences for constitutional interpretation. Imagine the dismay you would feel, for example, if you were to have worked for and achieved the ratification of a constitutional amendment, say the Equal Rights Amendment, only to have it interpreted by a court in a manner flatly contrary to your intent and to the intent of the amendment's supporters and ratifiers. In such circumstances you would want a judge to subordinate her personal perspectives and faithfully to implement the act of consent by virtue of which the amendment had become authoritative. You would no doubt experience a judge's departure from this duty as a betrayal.

The appeal of historical interpretation trades on this experience of betrayal. It is important to understand, however, that this experience does not depend

upon anything so simple as the physical casting of a vote. Constitutional amendments are ratified, not by general elections, but by state legislatures or special state conventions. Your experience of betrayal would depend, not upon whether you personally were a member of one of these special ratifying bodies, but rather upon your identification with those who had physically signified their consent. What would count is your sense that the members of the state legislatures or conventions who had actually assented to the Equal Rights Amendment spoke "for" you.

This same identification can extend in time as well as in space. Thus when confronted with constitutional provisions that are a century or more old, historical interpretation can be understood implicitly to assert an identification, a community of interest, with the framers or ratifiers of those provisions. "Their" consent, so the implicit assertion would go, is "our" consent; they spoke "for" us. It follows that the authority of historical interpretation will in significant measure depend upon the persuasiveness of that assertion. That is why in Chambers Chief Justice Burger offers an extended discussion of "the unambiguous and unbroken history of more than two hundred years," which he claims establishes "that the practice of opening legislative sessions with prayer has become part of the fabric of our society."^[76] The power of Burger's opinion rests in the end upon a claimed continuity of identification with those who had proposed and ratified the First Amendment.

This claim, however, is neither more nor less than a characterization of the national ethos. It is a claim about our national identity and history. Thus while the debate between majority and dissent in Chambers can at one level be seen as struggle between historical and responsive interpretation, it can at a deeper level be understood as a disagreement about whether we can now identify with our ancestors, or whether we have over the centuries become so different from them, so much more secular or diverse, that we have lost any persuasive identification with the consent of those who ratified the First Amendment.^[77]

This deep symmetry between historical and responsive interpretation stems from the fact that both ultimately flow from the authority of a will that affirms its own identity.^[78] Responsive interpretation makes this authority explicit, because it openly affirms responsibility for the nature of our national ethos. While historical interpretation seemingly presents itself as a self-denying submission to the identity of past ratifiers, closer analysis reveals that that identity is authoritative only insofar as we can be persuaded to adopt it as our own.^[79] In either case, the authority of the Constitution ceases to stand apart from the processes of its interpretation. That authority does not flow from the antecedent nature of the Constitution, but rather from the particular relationship we have forged with the Constitution.

In this regard, however, responsive interpretation is unique, for it alone explicitly thematizes this relational nature of constitutional authority. Both his-

torical and doctrinal interpretation purport to submit to a Constitution whose authority is independent and fixed, either in the preexisting consent of the ratifiers or in the preexisting rules of controlling precedents. Although this submission is illusory, it is an illusion capable of disarming dissent. Responsive interpretation, however, disavows this illusion, and frankly locates constitutional authority in the relationship between the Constitution and its interpreters. As a consequence responsive interpretation generates an intense and singular kind of political dynamics.

A good example is Brown v. Board of Education .^[80] The decision did not turn on what the ratifiers of the Fourteenth Amendment thought,^[81] nor on what the Court had previously held in Plessy v. Ferguson .^[82] Instead the ideal of racial equality had become so pressing to the Court that there was no alternative but to interpret the Equal Protection Clause in light of its imperatives. But precisely because this interpretation rested upon an open avowal of a national ideal, Brown represented a courageous gamble. The Court's embrace of the value of racial equality could have been a misreading of the national ethos; indeed the Court's gamble was intensely controversial and came close to failing precisely because that ethos was in fact so divided.^[83]

By refusing to interpret the Constitution as if it were a source of external compulsion, either of past precedent or of past consent, responsive interpretation always places a court in such an exposed position, purporting to speak for the fundamental ethos of the contemporary community, but justified in the end only by the wisdom of its own insight. Under conditions of cultural division, that position can be the platform for a special form of leadership (as in Brown ), or it can be the cause of the most unhappy form of vulnerability (as in Roe v. Wade ).

Roe , which at the time of its decision stood without significant historical or precedential support,^[84] illustrates the structural vulnerability of responsive interpretation to the charge that it articulates values that are merely local and partisan, rather than general and truly constitutive of the nation. The charge is unlikely to surface when there is cultural consensus, because the invocation of contemporary values will under such circumstances be unobtrusive and perhaps even unnoticed. But in the absence of consensus the frank ambition of responsive interpretation to "speak for" the character of the nation, while expressive of the outlook of some, will necessarily constitute a hegemonic imposition upon others.^[85] Thus the enterprise of responsive interpretation can become the locus of an overt struggle for the definition of national identity. In the legal academy responsive interpretation has been profoundly controversial because of the unease generated by perceived judicial participation in such a struggle.

Paradoxically, however, the root cause of this unease is precisely responsive interpretation's explicit thematization of the relational nature of constitutional authority, a form of authority that it in fact shares with both historical and doctrinal interpretation.

The Inevitability of Responsive Interpretation

Responsive interpretation is in some respects similar to what in the contemporary legal literature is called "noninterpretivism,"^[86] which can roughly be defined as that form of constitutional interpretation which seeks "the principal stuff of constitutional judgment in one's rendition of society's fundamental values rather than in the document's broader themes."^[87] Noninterpretivism, and by extension responsive interpretation, is frequently attacked as breaking faith with a judge's obligation to interpret the Constitution rather than to enact her own desires.

Understood in a psychological sense, the attack is clearly justified. If a judge believes that the Constitution means X, but the judge decides Y because she prefers Y, the judge's decision is presumptively illegitimate. But this framing of the issue is ultimately trivial, for it proceeds on an assumption of bad faith, and it prejudges the critical question, which is the nature of that Constitution to which the judge should maintain fidelity. If noninterpretivism is defined as rendering judgment upon the basis of extraconstitutional factors, then it will of course be vulnerable, but only in an uninteresting and merely stipulative sense. Properly understood, however, responsive interpretation avoids this vulnerability by including the additional claim that our "fundamental nature as a people" is part of the legitimate authority of the Constitution.

One objection to responsive interpretation, therefore, is that it mistakes the "root premise" of American constitutionalism, which is "that the Supreme Court, like the other branches of government, is constrained by the written constitution."^[88] The point is that responsive interpretation, which explicitly dissolves the Constitution as a specific written text, rests on an unacceptable notion of the Constitution. The force of this objection, which is considerable, derives from the circumstance in which the words of the Constitution appear to speak plainly to us. Recall the case of the third California senator, which a court could settle merely by reading the text of the document. In such a case it seems as if the document itself were authoritative, as if meaning flowed naturally from that handwritten, hand-signed parchment kept under glass in the National Archives.^[89] It would appear to follow that any theory of interpretation which abandons that document is illegitimate.

This reasoning, however, proves far too much. It is true that when the document's meaning is unproblematic we feel constrained to regard its language as authoritative without further inquiry. But when for whatever reason the document's meaning does seem problematic, we are necessarily forced outside the text in search of some authority to guide our interpretation of the text. Thus every theory of constitutional interpretation is at some level inconsistent with the notion of a narrow fidelity to a written document. Doctrinal interpretation, for example, which is the sine qua non of constitutional adjudication, applies not the words of the document, but legal rules that judges have subsequently created. Most constitutional cases are decided on the basis of doctrinal "tests" that have very little

to do with the text of the parchment which resides in the National Archives. Even historical interpretation looks for authority not to the text of the written document, but rather to the consent of those who agreed to it. The charge that responsive interpretation abandons the written document, therefore, is an accusation that would disqualify virtually all forms of constitutional interpretation.

A second objection to responsive interpretation, however, is that it abandons the document in a particularly unacceptable way. Historical interpretation focuses on an original act of consent because that consent "points toward" the document and illuminates its specific meaning. Doctrinal interpretation, it might be said, focuses on rules of precedent because such rules also "point toward" the document and are attempts to elucidate its meaning. Responsive interpretation, on the other hand, turns away from the document altogether in an effort to uncover present values.

This objection captures what I take to be a major animus of the contemporary debate, and for that reason it needs to be carefully parsed. It is true that because historical interpretation looks to the consent of the ratifiers, the historical document actually ratified is central to the interpretative enterprise. But doctrinal interpretation can be said to "point toward" that document only in the most attenuated metaphorical sense, a sense in which it is equally true to say that responsive interpretation "points toward" the document.

Responsive interpretation rests on the claim that the Constitution is not "static and lifeless," to use Brennan's words in Chambers . Instead, as Holmes put it, the Constitution is understood as having "called into life a being" that, like any "organism," must grow and develop on the basis of its "experience."^[90] Thus the ambition and challenge of responsive interpretation is to determine which aspects of our contemporary ethos may be regarded as legitimate "growth from the seeds which the fathers planted," and hence as bearing "the essential content and the spirit of the Constitution."^[91] Only these aspects of the national ethos are genetically related to the document and thus may properly form the basis for responsive interpretation. In this sense responsive interpretation does indeed "point (backward) toward" the document, in at least as strong a metaphoric sense as does doctrinal interpretation.

Admittedly the organic metaphor that underlies this account of responsive interpretation is highly problematic. It is important to understand, however, that responsive interpretation could equally well rest upon other and perhaps more convincing metaphors. It could invoke, for example, the image of an evolving "tradition" that is constitutive of cultural meaning.^[92] Or it could adopt the sociological language of communitarianism, as in this passage from Philip Selznick:

A social contract is a constitutive contract. Its function is to create a political community by founding the legitimacy of government on the consent of the governed. Once the community is formed it has a logic and a dynamic of its own. Even the fundamental obligations of government and citizenry—obligations of loyalty, self-restraint, and care—flow from

the nature of the community and of its historical premises, not from the terms of an agreement.^[93]

Each of these metaphors can be used to describe a national ethos that both changes over time and yet also manages to retain a distinctive identity. Each portrays a national ethos in which we are implicated, and to which we are therefore responsible. Each is therefore capable of sustaining the enterprise of responsive interpretation.

In these heady days of postmodernism, of course, it is easy enough to deny the truth of all these metaphors, and to repudiate the very existence of any overarching national ethos. The political consequences of such a denial, however, are grim. They were in fact first systematically articulated by Thomas Hobbes, and today the premises of his work remain visible in the writings of those influenced by economics and public choice theory. A clear example can be found in the views of a constitutional scholar like Robert Bork, who argues that there is no such thing as a distinctive national ethos, but only a vast collection of individual preferences.^[94] It follows from this perspective that any attempt to interpret the Constitution on the basis of the authority of a national ethos will necessarily degenerate into an unwarranted imposition of private judicial preferences.^[95]

Two preliminary points should be made about this perspective. First, it is inconsistent with historical interpretation, with which it is sometimes associated. Historical interpretation rests on the implicit assertion that the national ethos supports an identification with the ratifiers of the Constitution. But if there are only discrete individual preferences, and if the nation does not have any national ethos, there is no reason whatever why the consent of those long dead should hold any particular authority for the present generation. Second, as the example of Hobbes illustrates, this perspective has difficulty offering a plausible account of political authority as anything other than a collective need for forceful and clear rules of conduct to save individuals from the destructive consequences of their own egoism. But this form of authority, stressing as it does the values of continuity and predictability, is compatible only with the authority of the Constitution as law, which is to say with doctrinal interpretation. The actual implication of this perspective, therefore, is that the principle of stare decisis should hold until interrupted by contemporaneous processes of constitutional amendment.

The consequences of denying the existence of a national ethos are thus dramatic, far-reaching, and singularly unattractive. It transforms the overriding concern of constitutional adjudication into the maintenance of rules (any rules), for only such rules stand between us and a chaos of individual desires. Because the primary objective of these rules will be the preservation of order, those subject to constitutional rules will necessarily be reduced "to mere objects of the administered life."^[96] The Constitution is thus ultimately converted into a form of "repressive law" that "gives short shrift to the interests of the governed."^[97]

This transformation is relevant to an assessment of the position of scholars like Bork. Although the existence or absence of a national ethos appears at first blush to be an empirical question that is independent of the perspective of a court, in fact matters are not so simple. As the example of Brown v. Board of Education illustrates, a court can, through the eloquent articulation of public ideals, actually help to solidify a national ethos. The national ethos to which responsive interpretation appeals, in other words, may in significant ways be affected by the very practice of responsive interpretation. The question facing a court, therefore, is whether it should interpret the Constitution in ways that may express or establish a national ethos, or whether it should do so in ways that may confirm its absence. I think this question answers itself, which is why constitutional interpretation has never at any time proceeded on Hobbesian premises.

There is, however, yet a fourth objection to responsive interpretation, one which exerts considerably more influence than the Hobbesian perspective. It does not deny that the nation has an ethos which forms an important component of its public life, but it contends that it is inappropriate for judges to appeal to that ethos as a form of constitutional authority, because the conservation and articulation of that ethos should be placed in the hands of democratically elected officials rather than judges. The objection, in other words, rests on an institutional analysis of how courts ought to function in a democracy. It is of course on precisely such institutional considerations that the counter-majoritarian difficulty is ultimately founded. The stubborn persistence of the difficulty suggests the presence of powerful truths that cannot be brushed aside.

They are, however, only partial truths. If the Constitution is not to degenerate into merely repressive law, authoritative only because of the need for clear and predictable rules, courts interpreting the Constitution must be allowed to speak from the authority of a national ethos, in the form of either historical or responsive interpretation. Taken to its logical conclusion, therefore, the counter-majoritarian difficulty leads to exactly the same unacceptable vision of constitutional law as that which flows from overtly Hobbesian premises.^[98] But this consequence is unacceptable to proponents of the counter-majoritarian difficulty, for the institutional considerations by which they seek to circumscribe judicial power are themselves based on a particular account of the national ethos, one that characteristically stresses the importance of majority will in the form of government created by the Constitution.^[99] Proponents of the counter-majoritarian difficulty are thus torn between their account of appropriate institutional principles and the fact that these principles, if fully implemented, would preclude judges from appealing to the very national ethos from which the principles flow.

Institutional objections to responsive interpretation are consequently riven by internal tensions. For this very reason, however, institutional objections have rarely if ever implied a simple repudiation of responsive interpretation. Instead they have characteristically generated counsels of caution, urgent recommenda-

tions that responsive interpretation be used only sparingly and in ways consonant with the underlying conception of the national ethos upon which the institutional objections are themselves based. They have led, in other words, to forms of responsive interpretation based upon a particular understanding of the national ethos as founded upon majoritarian principles.^[100]

V

The fact that identical judges use different theories of constitutional interpretation in different cases is often used as evidence of the unprincipled nature of constitutional law. And, indeed, if the choice of an interpretative theory depended on the nature of the Constitution, and if that nature were antecedently and externally given, it would be difficult to condone the ways in which judges actually use interpretative theories. But if, as I have argued, constitutional interpretation depends instead upon a relational concept of constitutional authority, judges can legitimately select a specific interpretative theory in light of the circumstances of a particular case.

Thus a court can justifiably use historical interpretation with respect to an issue in a case if it believes that the national ethos supports an identification with a past act of consent relevant to that issue. But it can justifiably use responsive interpretation if it can discern with respect to that issue the presence of a national ethos that in a pertinent way historically embodies the essential content and spirit of the Constitution, and that precludes identification with any past act of consent. Hence the choice between historical and responsive interpretation can turn on an appraisal of the national ethos.

The selection of doctrinal interpretation entails a different kind of appraisal, one that requires a court to determine whether the values of the rule of law outweigh the inadequacy of controlling precedents. An important reason why precedents may be inadequate is that they are inconsistent with the interpretation of the Constitution required by the national ethos, either in the form of historical or responsive interpretation. The striking of this balance between the rule of law and the national ethos is both necessary and legitimate.

Thus the selection of a specific theory of constitutional interpretation for a particular case can be justified in the same way that any legal decision can be justified. Justification flows not from logical compulsion but rather from the principled application of pertinent considerations. Constitutional adjudication, like all law, is in this way revealed as balanced on the human faculty of judgment. As in all human endeavor, pertinent considerations may be more or less compelling, and consequently the ability to exercise judgment more or less sustained.

Nevertheless, the pattern of judgment in constitutional law can reveal a good deal about the nature of fundamental authority in our democratic state. It sug-

gests, for example, that visions of the national ethos, and hence of a "humanly meaningful authority,"^[101] are at the core of our practice of constitutional adjudication. This is encouraging news to set against the view of those who, like Jürgen Habermas and others, perceive the tidal current of this century as flowing toward "undeviating organization,"^[102] with its concomitant conditions of bureaucracy, alienation, deracination, and instrumental rationality. But it is also cause for concern if, as appears increasingly likely to be the case for many of us, the vision of national ethos authoritatively enacted by the Supreme Court is contrary to our own. Our consolation in such circumstances is the strength that Claude Lefort identifies with modern democracy: the ever-present possibility that our protests will create a reconstituted political perspective that will in turn alter the character of future judicial appointments.

But that possibility, of course, simply pulls Ernest Chambers round full circle, back to his original efforts to convince his fellow legislators of the deep impertinence of legislative prayer.

Facing Facts in Legal Interpretation

Kim Lane Scheppele

Questions of Law and Questions of Fact

As every first-year law student soon learns, questions of law are very different from questions of fact. When one is asked, "What are the facts of this case?" it doesn't do to talk about strict liability, or consideration, or the rule against perpetuities. One needs to describe what happened in the world, what events are at issue in this lawsuit. When asked about questions of law, the field of inquiry shifts. One discusses the legal standards to be applied to the case and how they are to be understood, and one stops discussing this particular car accident or the particular words of this idiosyncratic will. Statements of fact are descriptions of events and activities in the world; statements of law are interpretations of legal texts and legal rules.

The separation of questions of law and questions of fact has an even greater role outside the classroom. Much of the institutional framework of courts hinges on the distinction. In a jury trial, the jury decides questions of fact while the judge instructs on matters of law. The division of labor between judge and jury is premised entirely on being able to sort out the two sorts of issues.^[1] By being able to decide the facts of a particular case, the jury retains a substantial check on the powers of judges, or at least so the theory goes.^[2] Keeping fact and law separate is a political strategy designed to minimize the abuse of power.

The division of labor between trial and appeals courts is also premised on a sharp distinction between law and fact. Answers to questions of fact are considered to be virtually fixed at trial, and only issues of law can be raised on appeal.^[3] Appeals courts, because they do not have the parties present and do not have an opportunity to hear all the evidence presented orally, are thought to be poorly situated to work out what happened and are supposed to limit their review to questions of law in which they have special expertise. Here, too, the law/fact distinction marks out the institutional boundaries of different legal actors, separating the role of the appeals court judge from the role of the trial judge or jury.

And the importance of the law/fact distinction does not end there. In the determination of what shall count as precedent, judgments of law carry weight for future cases; judgments of fact generally do not. The idea here is that questions of law transcend particular decisions, while questions of fact are idiosyn-

cratic and peculiar to individual disputes. Finding that John intentionally injured Lisa in one case does not mean that, when similar evidence is presented in another case, a judge or jury should find that Roy intentionally injured Hannah. Evidence may appear to be the same, but the implications and resonances may be different in the two cases, and the judgments about "what happened" need to be left to case-by-case determination. Judgments of law, on the other hand, transcend the individual case and provide a standard to be used in all similar cases. Finding that intentionality must be demonstrated in order to convict a man of battery is a matter of law, the standard to be invoked the next time a case of that sort comes along. So, once again, law and fact have different roles in the institutional setting, with legal judgments adding to the corpus of what is called law, while factual judgments simply sort out what happened in individual cases.

This much is well known. But if one leans on the distinction between law and fact, it collapses under a bit of pressure. This is not a new insight. James Bradley Thayer's 1898 discussion of the distinction between law and fact in jury trials makes clear the confusion that results from trying to separate the two as a practical matter.^[4] Nathan Isaacs argued that whether a matter was deemed to be a question of law or a question of fact was itself a matter of law, and that the classification of specific issues as law or fact depended more on the history of the problem or the psychology of the judge than on logic.^[5] A. W. B. Simpson, writing on the constitutive nature of rules, has pointed out that descriptions of fact must often use categories that the rules themselves create, and so there can be no neat separation of law and fact as an analytic matter.^[6] The existence of "mixed questions of law and fact" has always bedeviled the distinction.^[7] And the American legal realists, at pains to show the limits of logic in the law, were extraordinarily fond of pointing out the extent to which the requirements of law shaped the facts as seen by judges, so that deciding the law and deciding the facts were not different stages of a syllogistic process but were rather all mixed up together.^[8] Most writers who have taken up the question agree that the law/fact distinction is not a matter of logic but instead an institutional convention, shaped by historical experience and rooted in a tradition of practice.

At the institutional level, then, the law/fact distinction works to limit the role of the judge, to define the limited and special role of appeals courts, and to separate out what pronouncements count as law. But at the level of legal reasoning, it is widely believed that this distinction cannot do this work as a matter of logic, but must instead rest on some system of practice or sense of history. So, while the law/fact distinction is crucially important in institutional organization, it is less useful in understanding the structure and complexity of legal reasoning.^[9]

Against this background, it is quite surprising that the writers on interpretation in law have generally adopted a rather flat view of the law/fact distinction. Law is what needs to be interpreted, but facts are simply true or false.^[10] Theories of interpretation are overwhelmingly about how to read legal texts, and the var-

ious strategies of interpretation provide orienting rules in understanding these texts. The debates among the advocates of intentionalism, conventionalism, interpretivism, structuralism, literalism, and the rest proceed as though legal reasoning were merely a matter of understanding a legal text.^[11] Understanding the facts drops out as an uninteresting or unchallenging or irrelevant part of the process.

But if we have taken a fully interpretive attitude toward the law, then our very attitude about the facts of legal cases must come under scrutiny. How may we consistently believe that legal texts are open to interpretation, but that the facts of cases are not? Of course, there are obvious differences between texts and events,^[12] but these differences are minimized in trials. In the courtroom, we don't have the event before us, but instead we have verbal accounts of events (whether orally or in writing) or sometimes trace evidence from which inferences may be drawn. And though occasionally things are thought to speak for themselves in law (res ipsa loquitor ), they can only do so because they have a role in an ongoing story, providing one "obvious" interpretation of a given piece of evidence in that context.^[13] At a trial, actions and harms are converted into accounts and claims. Events are made into texts. And these are texts that, like legal texts, are open to interpretation.

This "textualization" of social life at trial is magnified on appeal, where judges rarely see the parties to the lawsuit and have only a written record to consult for the evidence. Though they may question counsel in oral argument, judges usually learn the facts of cases in the same way that they learn the law—from reading written documents and working through plausible interpretations of those written texts. Why should we think that judges interpret the law but take the facts as merely given? Why should we think that we need theories of interpretation to guide and justify the activities of judges in saying what the law means, but that we have no need of interpretive accounts when judges are saying what the facts mean?^[14]

Strikingly, in the jurisprudence of interpretation, facts have generally not been taken to be interesting or problematic. Debates on legal interpretation proceed as if only the law side of an assumed law/fact divide needed serious interpretive work. In this paper, I will show just how much interpretive flexibility judges have in the construction of facts and how flexibility in the interpretation of facts makes problematic any theory of legal interpretation that considers only legal texts. To do this, I will work through one case in some detail. That case is Riggs v. Palmer , one of the classic leading cases on legal interpretation. Once I have shown that legal interpretation necessarily constructs accounts of law and fact simultaneously, I will return to the question of the law/fact distinction in the institutional design of American courts and ask whether the flexibility judges have in interpreting facts undermines the legitimacy of legal institutions.

The Interpretation of Law:
Riggs v. Palmer Revisited

Elmer Palmer poisoned his grandfather. The motive was clear. If Elmer didn't act quickly, he stood to lose a large inheritance.

Francis Palmer had written a will in which he left a small part of his estate to his two daughters, Mrs. Riggs and Mrs. Preston, and the rest to his grandson, Elmer. If Elmer died first or died without children who could inherit this estate, the entire estate would revert to the daughters.

But two years after the elder Palmer made his will, he married one Mrs. Bresee. The couple drafted an antenuptial contract, agreeing that if Mrs. Bresee survived Francis Palmer, she would be supported on Palmer's farm. Elmer thought that his grandfather would soon change his will to benefit the new wife and exclude him completely. Seeing that his own position was threatened, Elmer killed his grandfather in order to inherit while the inheriting was good.

The New York Court of Appeals, hearing the case in 1889, was asked: Given that Elmer killed his grandfather, should he be allowed to inherit under his grandfather's will? Francis Palmer's will was valid; at the time of Francis's death, Elmer was still listed as chief beneficiary. Nothing in the statute explicitly prevented murderers from going on to inherit from those they murdered. But it didn't seem right to the majority of judges on the New York Court of Appeals that a murderer should be allowed to gain from his dastardly deed.^[15]

The decision in the case was noteworthy for its explicit attention to interpretive style. How the decision ought to be arrived at was given as much, if not more, consideration than the reasoning or the outcome itself. Judge Earl, writing for the majority, admitted that

it is quite true that statutes regulating the making, proof and effect of wills, and the devolution of property, if literally construed, and if their force and effect can in no way and under no circumstances be controlled or modified, give the property to the murderer.^[16]

But Judge Earl went on to say that not the words but the intentions of the lawmakers in drafting the statute should be controlling here under a theory of "rational interpretation" or "equitable construction":

It could never have been their [the lawmakers'] intention that a donee who murdered the testator to make the will operative should have any benefit under it. If such a case had been present to their minds, and it had been supposed necessary to make some provision of law to meet it, it cannot be doubted that they would have provided for it. It is a familiar canon of construction that a thing which is within the intention of the makers of a statute is as much within the statute as if it were within the letter; and a thing which is within the letter of the statute is not within the statute, unless it be within the intention of the makers.^[17]

On this view, judges can ignore the words and go straight to the intentions of the lawmakers, assuming that lawmakers could not have intended absurd things.^[18] With this general method in mind, Judge Earl wrote, "We need not . . . be much troubled by the general language contained in the laws." Instead, judges should rely on "fundamental maxims of the common law," such as "No one shall be permitted . . . to take advantage of his own wrong."^[19] Such maxims "are dictated by public policy, have their foundation in universal law administered in all civilized countries and have nowhere been superceded by statute."^[20] Elmer was not allowed to inherit.

The dissent by Judge Gray used a different interpretive approach and reached the opposite conclusion, though he clearly felt himself pulled on grounds of conscience toward the majority position. But law was law:

If I believed that the decision . . . could be affected by considerations of an equitable nature, I should not hesitate to assent to views which commend themselves to the conscience. But the matter does not lie within the domain of conscience. We are bound by rigid rules of law.^[21]

The statute in question stated that "no will in writing, except in the cases hereinafter mentioned, nor any part thereof, shall be revoked or altered otherwise."^[22] And no explicit provision thereinafter mentioned covered the facts raised in Riggs . So, there was nothing for it, Gray reasoned, but to conclude that Francis Palmer's will should be carried out as written, even if the effect were to give his murderer most of his estate. To do otherwise would be to act illegitimately where the laws were silent and to impose a penalty above and beyond the punishment already meted out in the criminal trial.

Whether Elmer inherits or not seems to rest entirely on the theory of interpretation that a judge adopts. Both Earl and Gray claim to be following the law, though each uses different strategies in divining what it says. As a result, what Gray giveth, Earl taketh away. Is one wrong? Or is the law sufficiently flexible to allow for opposing outcomes? If the law is that flexible, then what, if anything, is meant to be excluded? And if the outcome rests so completely on the theory of interpretation the judge selects, then isn't the theory of interpretation doing at least as much work as the law in determining the outcome?^[23]

These questions place a high premium on being able to justify the interpretive strategy of the judge. And on this question, commentators have been as varied as they have been numerous. Some, like Roscoe Pound, side with Judge Gray and argue that filling in defective statutes with judicial creativity may solve the immediate problem, and may even produce justice, but that this "judicial speculation" goes beyond what a judge should do. If the judge makes law rather than merely interprets it, then the judge "puts a meaning into the text as a juggler puts coins, or what not, into a dummy's hair, to be pulled forth presently with an

air of discovery. It is essentially a legislative, not a judicial, process."^[24] Pound blasts the majority opinion in Riggs as the most conspicuous example of "spurious interpretation" outside the area of constitutional law. The appropriate strategy of interpretation in a hard case like Riggs , Pound asserts, is for the court to read the statute as it stands and to wait for the legislature to change the statute to deal better with the awkward results.

Of course, reading a statute as it stands itself invokes complicated interpretive ideas. Stanley Fish argues that the Riggs case does not really pit a literal interpretation against a nonliteral one, but rather frames the same problem against two different assumed purposes:

No reading is the literal reading in the sense that it is available apart from any purpose whatsoever. If it is assumed that the purpose of probate is to ensure the orderly devolution of property at all costs, then the statute in this case will have the plain meaning urged by the defendant; but if it is assumed that no law ever operates in favor of someone who would profit by his crime, then the "same" statute will have a meaning that is different, but no less plain. In either case, the statute will have been literally construed, and what the court will have done is prefer one literal construction to another by invoking one purpose (assumed background) rather than another.^[25]

Fish undermines Pound's solution by showing how any reading is going to presuppose exactly what Pound says a judge should not use: a background theory about the larger purpose of the statute that cannot be directly drawn from the statute's words.

Ronald Dworkin (who has long used Riggs as a stalking horse for the idea that judges do and should draw on principles for deciding hard cases)^[26] argues that the judge should read a statute in the best possible light against the backdrop of a set of coherent principles in the law. Dworkin makes the case in his early work that principles are as much a part of the law as rules are, and so provide appropriate materials for judges to draw on in deciding hard cases. Riggs is, for Dworkin, a perfect example of this strategy put to good effect. In his later work, Dworkin provides a much more complicated account of why this should be the case. The goal of interpretation is to weave a seamless web of principle. Appealing to the coherence of law, seeing each case as an instance of a broader practice of principled judgment, the judge will be directed to better answers, on Dworkin's view. Although Dworkin does not explicitly work through what a good answer would look like in Riggs , one deeply suspects that he still favors Judge Earl's view. But any judge, on Dworkin's analysis, must take into account a great complexity of things to come up with a meaningfully coherent way of making the best of the principles in the vicinity of this case.

Principle is not such an easy matter even in the Riggs case, which seems on the surface to pit nasty mechanical judging against morally inspired, principled interpretation. When Benjamin Cardozo turned his hand to writing about this case, he noticed that there were important principles on both sides:

There was the principle of the binding force of a will disposing of the estate of a testator in conformity with law. That principle, pushed to the limits of its logic, seemed to uphold the title of the murderer. There was the principle that civil courts may not add to the pains and penalties of crimes. That, pushed to the limits of its logic, seemed again to uphold his title. But over against these was another principle, of greater generality, its roots deeply fastened in universal sentiments of justice, the principle that no man should profit from his own iniquity or take advantage of his own wrong. The logic of this principle prevailed over the logic of the others.^[27]

And why did this latter principle prevail? Cardozo answered that Elmer lost his legacy "because the social interest served by refusing to permit the criminal to profit by his crime is greater than that served by the preservation and enforcement of legal rights of ownership."^[28] Just how and why the standard of greater social interest enters to trump all the others is not clear on Cardozo's account (and this is where Dworkin's thoughtful attention to justifying the principled basis of the law represents an improvement over Cardozo's analysis), but the problem is clear. Principles of interpretation need a complex justification of their own. And the justification for a particular interpretive strategy enters judging as though it were part of the law itself, adding more principles for the courts to use.

These different theories of interpretation and of justification compete for ascendancy so that "interpretations struggle side by side with litigants before the bar."^[29] Debates about theories of interpretation rage over cases like Riggs precisely because these cases make so clear how very much is at stake in the choice of method for reading a statute.

All of these authors writing on Riggs have assumed that deciding on the strategy of legal interpretation decides the case. Once the general approach has been justified, the result of the case follows without difficulty. With Pound's theory, Elmer inherits. With Dworkin's and Cardozo's theory, Elmer loses the estate. Fish leaves the choice open, but argues that the choice of background context determines the result. All agree that once the judge has decided how to interpret the law, the outcome of the case is clear. What makes Riggs a hard case is that different outcomes are reached with different theories of interpretation. And debate is joined over how the legal text ought to be read. But an important premise is buried in this debate.

Arguments over the interpretation of the law assume that the facts have come into an appeals court as given. In one sense, this is certainly correct. The New York Court of Appeals is not going to deny that Elmer poisoned his grandfather or that he stood to inherit as a result. But in another sense, a case on appeal always presents the opportunity for redescription or recharacterization of the facts that have been found at trial. And the characterization of facts can make all the difference in the legal result, even when the question of interpretive strategy has been answered.

What might have happened if the Riggs court took a different view of the

facts? It's not easy to see how facts can be recharacterized, once the opinions have been written. Though, as Karl Llewellyn wrote, "The facts are hardy weeds. They will not down,"^[30] opinion writing tends to obscure alternative versions, to make the version of the facts as presented seem to meet the Walter Cronkite test: "And that's the way it is." But behind every description of facts, there are many other versions, equally true but differently organized. Changes in emphasis, alternative points of view, different symbolic contexts, varying background assumptions all have their effects on which version of a particular story seems the most compelling. And there is, in Riggs , a particular alternative version that was available to the court but not chosen, a version that incorporated a legal fiction. To see this, let's examine the legal context in which Riggs arose.

The Road Not Taken:
Riggs and the Fiction of Civil Death

Arriving in the Court of Appeals in 1889, Riggs confronted the New York courts for the first time with a murdering heir. The only other reported case raising a similar point before then was Owens v. Owens , decided by the North Carolina Supreme Court a year earlier.^[31] These are late dates in American history for state courts to be wrestling for the first time with cases like this. In over a century of American legal development, no other case raising the issue of a murderer's right to inherit from his victim had been brought in the law reports.^[32]

Something else must have been happening with cases of murdering heirs before then. It is unlikely that there were no such offenses before 1888, temptation being what it is and undoubtedly was. Moreover, all of a sudden in the late 1880s there was a flood of such cases.^[33] Was there a sudden run on rich relatives? Why did this period abruptly produce such a wealth of devious heirs? One factor contributing to the increase in these cases may have been the rise of life insurance, which may have provided more for heirs to inherit.^[34] In England, the cases of murdering heirs in the nineteenth century involved various claims on insurance policies.^[35] These cases could be decided on grounds of fraud in contract (since the person claiming under the policy was generally the one who had negotiated the contract and also was responsible for the death), so the courts didn't have to reach the questions raised in Riggs . But many of the American cases that came up at this time, Riggs in particular, did not turn on insurance questions.^[36] It was the inheritance plain and simple that was at issue, and the increased availability of insurance seems to have little to do with these claims. Heirs must have been murdering all along in order to benefit under their victims' wills, even in advance of the practice of insuring lives, but for some reason the courts that produced written records were not called upon to rule whether they could inherit under statutes of wills—at least not until the 1880s.

One obvious possibility is that there was some other legal provision covering these cases before the 1880s. We wouldn't expect to see cases of murdering heirs arising under the statutes of wills if this "provision X" operated in such a way that the cases were resolved on some prior issue, never getting as far as invoking the statutes of wills with their difficult interpretive problems. Moreover, provision X must have involved rules so clear that cases did not need to be brought to appellate courts to test them at all, or murdering heirs would have turned up under some other doctrinal heading. If this is true, though, provision X must have changed in the latter part of the nineteenth century, making what was formerly clear unclear enough to bring a deluge of cases into the law reports.

There was just such a provision X in operation in the early days of American law: the fiction of civil death.^[37] If a person were convicted of a serious crime, the law would consider the person to be civilly dead, incapable of existing at law. And if a person were dead to the law, then this would surely affect the ability to inherit.^[38]

Civil death has a long history in the common law. Upon conviction for treason or felony in the ancient common law, a convict would be placed by law in a state of attainder. Dire consequences followed on this: convicted of treason, the convict forfeited all his property to the king; convicted of a felony, the convict was held to have "corrupted blood" and so the descent of property was blocked and the property then escheated to the lord;^[39] convicted of treason or felony, the convict was deprived of all civil rights. This latter deprivation was known as civil death.

The provisions for forfeiture and corruption of blood that existed in English law were not generally incorporated into American law. The federal Constitution provided that "no Attainder of Treason shall work Corruption of Blood, or Forfeiture except during the Life of the Person attainted."^[40] Laws in most states (some constitutional, some statutory) explicitly forbid forfeiture and corruption of blood altogether, but these changes in the old concept of attainder did not prevent a convict from being deemed civilly dead upon conviction of a felony. In pronouncing a convict civilly dead, the states did not usually seize the convict's property, as they would in the case of forfeiture, and they did not prevent the passage of the property to heirs, as they would in the case of corruption of blood. But civil death often had the effect of removing property from a convict's control, by allowing it to pass on to heirs as though the convict were already dead.

What else did civil death involve? As this fiction was developed in America, the convict was held to be dead for only some purposes in law, and so civil death did not bring with it all the consequences of natural death. A convict could not bring suit to enforce a contract, but could have a contract suit brought against him. A convict could not bring an action to collect a debt, but his own possessions were open to his creditors. Generally, a convict was barred from bringing an action, but he could be named as a defendant.^[41] A convict's marriage could be instantly dissolved by law and the convict's spouse was free to remarry without having to

go through divorce proceedings, but the unconvicted spouse might elect to keep the marriage intact.^[42] Convicts' political rights were stripped away, and so convicts could not vote, hold office, serve on juries, or bear arms. And although most states did not take a convict's property or require invocation of the convict's will when he was civilly dead, many states did impose restrictions on what convicts could do with their old property and on how they could acquire new property.^[43] More generally, the practice of depriving someone of all of the benefits of law accompanied the declaration of civil death in the nineteenth century, and elements of such deprivations are with us still.^[44]

Because the convict could not benefit from the law, people deemed civilly dead could not inherit, regardless of whether the inheritance came from someone they murdered or from anyone else. There are no cases in New York on this point,^[45] but Wright v. Wright , decided by the Court of Chancery of South Carolina in 1804,^[46] indicates that this was not controversial. Catharine Wright died (of natural causes, as far as we can tell) and Alexander Wright claimed part of her estate. Alexander and Catharine had been married, but Alexander had been banished and declared civilly dead.^[47] The court had to address two questions: 1) what Catharine Wright's status was after her husband had been deemed dead and 2) whether her husband could inherit from her in his legally altered state. The court thought this was an easy case:

Both these points are extremely clear: the first, that she is considered as a feme sole , in every point of view; may contract debts, and acquire property, which she may again dispose of by deed or will. 2d, that the husband being banished, is considered as civiliter mortuus , and such rights as would have survived to him on the death of his wife, are extinct and gone with him; and the estate must consequently go to those to whom it is limited over.^[48]

Alexander Wright could not inherit from his wife (and in fact, Catharine was arguably not his wife at all anymore) because he was dead in law, even if not dead in fact.

We can begin to see now what all this has to do with Riggs . Under this fiction, Elmer would have been considered civilly dead. He had been tried for the poisoning of his grandfather, convicted of murder in the second degree, and was serving out his sentence in the state reformatory.^[49] The New York State Penal Code that went into effect in 1882 explicitly stated: "A person sentenced to imprisonment for life is thereafter deemed civilly dead."^[50]

Why then did Elmer's lawyer argue on behalf of his client that "it cannot be successfully urged that Elmer E. Palmer is civilly dead"?^[51] And why did the Riggs court leave out any mention of civil death in its opinions?

The doctrine of civil death was significantly changed in the second half of the nineteenth century. A statute, passed in England in 1870,^[52] abolished forfeitures and corruption of blood, and also laid out in one clear statement what befell someone upon conviction. Though the English statute did not use the language

of civil death, it clearly listed many things that were associated with the state of civil death in America. If a convict was in the military or held another public position, he was removed from office. If he got a pension or other payment from government, he was cut off. He could no longer vote. He could not bring suit to enforce any old obligations owed him prior to his conviction, nor could he undertake any new obligations while serving his sentence. His property was not to be forfeited any longer—and this was an important alteration in the English law—but it would instead be put under the control of an administrator who was to manage this property as the administrator saw fit. Under this new statute, convicts were allowed to retain ownership of their property, even though it fell under the control of someone else. But the statute was silent about an important matter: it did not say explicitly whether a convict could inherit or not. It even implied that the convict could now inherit, by referring to real and personal property that might come into the convict's possession after his conviction, which, since the convict could no longer contract, was most likely to occur through inheritance. For American courts that still looked to English law as a source of inspiration for American practices and standards, these new views about convicts' property represented a major change.

The New York Court of Appeals considered the English statute and the issue of civil death at the first opportunity, in Avery v. Everett .^[53] Just one year before Riggs was brought to the same court, Avery arose on the question of the status of a convict's property. The case did not present the same drama as Riggs ; in Avery , Charles Southwick came into his father's estate nonviolently, following the succession provided for in his father's will after his mother died in 1869. Charles's father, John Southwick, had left all his property to his wife and, after her death, to his son. John's will specified that if his son died with no children, the property was to be passed on to John's nephew, Augustus Southwick, and his heirs.

In 1875, six years after his mother's death and already possessed of the estate, Charles was convicted of second degree murder, having killed someone who had nothing to do with the matter of the inheritance. With no children, Charles passed his interest on to one George Everett, who was promptly sued by John Avery, the heir of Augustus Southwick. The Court of Appeals was called upon to decide whether Charles was to be considered civilly dead following his conviction, and whether his property was to pass on as though he were naturally dead. If Charles were naturally dead, Avery would get the property. But if Charles were legally alive, he could then give the property to Everett.

Judge Andrews, writing for the majority, said that the New York statute was clear: Charles was civilly dead. But, Andrews wrote, that didn't settle the point about property. Did Charles's father intend that his son's civil death be treated as a natural death? Charles might be pardoned and go on to have children, the opinion reasoned, and that would create heirs who ought to be able to benefit under the will. Treating Charles as though he were naturally dead would place any children he might eventually have in a most peculiar legal situation. "We have

found no authority upon the construction of the word 'death' in a will as applied to circumstances like these,"^[54] Andrews wrote. Treating this as a case of first impression,^[55] he delved deeply into the history of civil death to find out how other jurisdictions had handled the question of convicts' property rights.

Andrews found that even in the ancient common law, forfeiture of property upon conviction was not automatic but instead went into effect only if the king or lord who could claim the property actually did. In the absence of such a claim, the property was retained by the convict. With the passage of the 1870 statute in England, Andrews argued, all forfeiture claims were swept away, and convicts in England did not have to give up their property. So it was under the laws of New York also, Andrews believed.^[56] He found no cases in New York where the will of a convict was carried out upon the testator's civil death, and with the force of the English statute behind him, Andrews concluded: "The inference seems almost irresistible that the doctrine that civil death, consequent upon a life sentence, divests the criminal of his estate, has no foundation in our law."^[57] In Andrews's majority opinion, Charles was allowed to keep his property, even though he was civilly dead.

But there was an impassioned dissent in Avery from none other than Judge Earl, who wrote for the majority in Riggs . For him, it seemed obvious, and good policy, for civil death to separate a convict from his property. The statute pertaining to civil death

brought the law into harmony with our social organization and governmental system. As the convict could no longer discharge any of his obligations to society, he was to possess no civil rights whatever. As he could not discharge any of the duties of husband or father, the family ties were severed. As he could have no use for property and no power to manage or possess the same, that was to pass away from him. He became civilly dead in law, and the law ceased to know or to take any notice of him. He no longer possessed any rights growing out of organized society or depending upon or given by law.^[58]

Arguing that "any wise system of laws" would hold that a convict's property "be devolved upon his successor or heirs or next of kin," Judge Earl continued,

The life convict was not declared civilly dead in the law simply to deprive him of the right to vote, to sit as a juror, to bear arms, to marry and hold office, because his physical conditions were such that he could do none of those things. What, then, according to the conclusion reached by my brethren, is meant in the statute by 'civilly dead'?^[59]

With the majority decision in Avery , civil death no longer meant that a convict was stripped of his property. And this is why Elmer, the murdering heir, put the court in such an awkward position the following year. Under the new interpretation of civil death, Elmer could keep his property. And that meant that he could hang on to anything that was his (arguably including his inheritance)—so for the first time a murdering heir could benefit from his crime.

It didn't have to be that way. Avery v. Everett did not actually touch the question of inheritance, and so could easily have been distinguished from Riggs . Depriving

people of property they already have in their possession is a very different matter from restricting the ways in which they can come by property in the first place. Limitations on forfeiture and corruption of blood, then, did not have to affect courts' rulings on the inheritance of murderers. In a decision about a decade later, the Supreme Court of California, interpreting a civil death statute with wording identical to the one in New York, found, "Civil death imports a deprivation of all rights whose exercise or enjoyment depends upon some provision of positive law." And since "the right of inheritance is a civil right existing only by virtue of the law," someone convicted of a crime could not inherit.^[60] Explicitly distinguishing Avery , the California court ruled that taking away a convict's property was one thing, amounting to forfeiture; preventing him from inheriting more property was quite another matter. And civil death, in the California court's view, clearly prevented convicts from benefiting from inheritance.

The New York Court of Appeals in Riggs could have done the same. Distinguishing Avery was an easy matter; relying on their own case law on civil death would have provided ample justification for a decision finding Elmer to have, in law at least, passed on. What would the New York Court of Appeals opinions have looked like if they had gone the way of the California court?

Reconstructing the Facts in Riggs

Imagine that the Riggs court had been willing to use the fiction of civil death. By deeming Elmer Palmer to be dead, the facts could be described differently: Elmer murdered his grandfather in order to inherit property under his grandfather's will. In doing so, Elmer killed himself also.

But, certainly, one might object: Elmer didn't really kill himself. Civil death is not a factual description in the sense that it is a true representation of a state of affairs. Elmer could be deemed dead by a court, but this is merely a legal fiction, a game of "let's pretend" played out to reach a particular legal result. Legal fictions, the objection goes, are not legal facts. They are legal lies. If the court were to have called Elmer civilly dead, the court would have been using a falsehood to achieve a particular legal result.

Or, alternatively, one might object: the description of Elmer as civilly dead is not a statement of fact at all; it is simply the application of a legal standard to the ordinary nonfictional facts of the case at hand. Saying that Elmer is civilly dead, on this objection, is a matter of law , not of fact . Civil death is a shorthand description of a legal rule that specifies the disabilities that are to follow upon conviction for felony and is not a description of Elmer's factual state. When the court calls Elmer Palmer civilly dead, it is engaged in applying law, not describing facts.

To begin to address these objections, let's think for a moment about what makes something a legal fiction. A fiction is a legal device that relies for its oper-

ation on deeming things to be something when ordinary description would say that they are something else.^[61] The "deeming" creates a more complicated form of legal rule. A simple legal rule has the form: "If A, then B," where A is the set of legally relevant facts that must be found in order for B, the legal consequences, to be ascertained. So, for example, the simple traffic law "If you are driving above the speed limit, you will be fined" associates one legally relevant fact, driving above the speed limit, with a clear legal consequence, being fined. In a fiction, the form is: "If A₁ , then A₂ , then B," where A₁ is the set of legally relevant facts that must be found in order for a new description, A₂ , to be applied. And once the new, now fictional, description A₂ is invoked, then the legal consequences B follow.^[62] So, in the Riggs case, A₁ would include a) that Elmer Palmer murdered his grandfather and b) that he was sentenced to prison for life. Given these two facts, the fictional description A₂ , that Elmer is now civilly dead, can be invoked. And from this fictional description A₂ , the legal consequence B, that one cannot inherit if one is civilly dead, can follow. Because A₂ has a role intermediate between the first-order legally relevant facts and the ultimate legal judgment, it partakes of both. It appears to be a legal judgment, the concluding part of a simple legal rule. After all, if one has murdered someone and been sentenced to prison for life, then one can be deemed civilly dead. Civil death looks more like a legal conclusion than a statement of a fact at this stage. But in the second part of this complicated form of a rule, the fiction takes on the role of a fact necessary for the determination of the ultimate legal consequences. So, if one is civilly dead, then one cannot inherit—and the fiction functions like a fact. Whether a fiction appears to be a legal judgment or a factual statement, then, will depend on which part of the rule one is examining.^[63] Its role with respect to the ultimate legal judgment in a particular case, however, will be as a fact.

When fictions function as facts, they trade on a gap between the use of a description for the purposes of legal reasoning and the use of a description for routine communication. In other words, fictions can operate because the audience and purpose for specialized legal discourse is different from the audience and purpose for the nonlegal descriptions of events. The audience for specialized legal discourse needs to know not only the description of a particular event but also the legal significance of the description in order to work out what legal result should follow. Many legal terms operate the way that fictions do, in embodying both a legal conclusion and a factual premise for further reasoning. This process of progressive elaboration of meaning can produce a gap between the meaning of a term in ordinary description and its meaning in a court of law. Fictions are not different from other terms in having both descriptive and prescriptive functions in the law (and so they seem to be both part law and part fact), though they are different from these other terms in the extent to which they coincide with ordinary description.

To see how hard it is to separate legal description from the application of

legal rules, let's take part of the statement of the facts as it was printed in the majority opinion of Riggs :

He [Elmer] knew of the provisions made in his favor in the will, and, that he might prevent his grandfather from revoking such provisions . . . and to obtain the speedy enjoyment and immediate possession of his property, he willfully murdered him by poisoning him.^[64]

This apparently straightforward factual description includes many terms that already are infused with technical, legal meaning—for example, murder, possession , and will . Observation alone would not generate these terms; they "are all descriptions that go beyond physics."^[65] And using these terms correctly in ordinary language alone does not make them appropriate in a legal description, where these words are used in their technical senses.^[66] The only difference between words like murder, possession , and will and the fiction of civil death is that the fiction contradicts the ordinary meaning while these other terms do not.

Murder , as a term in law, carries a set of conclusions about the causal relation between a particular act of Elmer's and a particular outcome for Francis. Elmer's act had to have been found to have caused Francis's death. Moreover, Elmer had to have intended it; Judge Earl's use of the term willfully captures this legal requirement. Now willfulness and causation are not observable on the surface of social action, nor are they necessarily implied in all uses of the term murder . When Macbeth was thought to be murdering sleep, no one would think this was cause for immediate indictment.^[67] This was a nonlegal metaphorical use and did not require demonstration either of the same sort of causality or the same degree of willfulness, nor would this sort of metaphorical use call for legal punishment. It was not a meaningful legal use of the term. To use the term murdered in a legal description is to have already made a legal judgment and to have identified the range of official actions and disabilities that should follow. Use of the term murder in the Riggs statement of facts shows that the description of the facts is already mixed up with the law. The legal rule about murder operates on a statement of facts by using the term murder as a substitute for a description of detailed particulars. And so the term can be both a description of a state of affairs and a shorthand term for the judgment made under a legal rule at the same time.^[68]

Possession is another legally infused term, used to describe a particular sort of relation to property (itself a creature of the law). Elmer was living with his grandfather, so in one sense he was already in possession of the property he was to inherit. He could walk in and out of the house at will and handle the objects in it. He could probably use the house and its contents in such a way that it would be appropriate in ordinary description to say that he possessed them. Observation alone would not indicate whether Elmer had legal possession of the estate or not. But to use possession in a legal description, one must already have answered questions of law. Because the legal and nonlegal senses of possession overlap so much

in daily life, the gap between the legal and nonlegal senses of the word is not as great as it is in the case of fictions. But there are opportunities for gaps to open up here as in any field where technical language and everyday language use the same words. Some of Elmer's friends may well have thought him to possess the contents of the house before the dastardly deed was done, but Elmer's possession when used in a legal description indicates far more than his physical relation to the objects. Here again, the term is both descriptive of his factual condition and indicative of the application of legal rules.

A will , too, is a legal construction, not a description prior to judgments of law. A piece of paper may declare it's a will but it might not be. It might be a prop in a play or the purported will of a fictitious person. Another piece of paper may not seem to be a will but it is. If the document isn't signed in the proper way, it is not a will, regardless of the surface appearance. And if the document has the right characteristics but doesn't look very will-like, it could still be a will in law. Calling something a will, in short, already involves a legal judgment. And the judgment that Francis's will was valid (and hence properly called a will) was a necessary element of the present case involving Elmer's right to inherit under it. But perhaps the document that Elmer thought was his grandfather's will had already been superceded by another duly signed and witnessed statement made while the grandfather was able to think clearly about what he was doing. If Elmer had been wrong about the legal status of the will, it would not have affected the murder conviction—and the use of the term will in the murder trial would have been accompanied by a semantic modification like purported or by a clear indication that the will was a will only in Elmer's mind. But for the purposes of Riggs , the will has to have been valid, or else the question of Elmer's inheritance would not arise. And so, will , used in this description, must embody prior legal judgments about the way this object is to be described.

So, even the initial, straightforward description of the facts used by the court in Riggs is full of law before any fictions are invoked. It would be impossible to construct a description using no legal terms of this sort that would still count as a description of relevant facts for legal purposes. Suppose we were to say: "Elmer Palmer unscrewed the cap from a bottle of liquid and poured this liquid into Francis Palmer's coffee. After drinking this coffee, Francis Palmer fell over and lay on the floor." We wouldn't have the relevant sort of information necessary to make legal sense out of the event for the purposes of the Riggs case. We need to know whether the event described is a murder or not, and categorizing an event as a murder is a judgment invoking legal standards. Without these judgments already embedded in the description, the description may make sense to ordinary folk, but it doesn't make sense at law. If one could construct a description that included no legally relevant terms (a description that would count as an explanation in physics might be one option here), the account would be legally unrec-

ognizable.^[69] One would not be able to begin to figure out what to make of such a description in law. Law is not about descriptions that do not go beyond physics. All legal descriptions will contain terms invoking legal categories, which mediate between the facts of ordinary description and legal judgments.

We have seen, then, that fictions are not wholly legal judgments but not wholly factual descriptions either. For the purposes of ultimate legal judgments, they operate as facts. But they have already been processed in accordance with legal rules about when legal categories can be invoked to describe particular facts. And in this respect, fictions are no different from many other apparently factual descriptions that use legal categories.

But what about the "let's pretend" point about fictions? Courts are not simply exercising unconstrained linguistic lassitude or latitude in using fictions, lying when the results demand it. Instead, fictions are a highly stylized form of description that trades on a gap between an ordinary meaning and a technical, legal meaning. The fiction of civil death deems someone to be dead "in law" (for a legal audience) when he is not dead "in fact" (for a nonlegal audience). By saying that death in law turns on whether someone is sentenced to prison for life, the fiction infuses death , a particular term in ordinary language, with a precise, technical meaning, creating a new state of civil death . The fiction creates a new, accurate description for a particular community of interpretation; describing someone as civilly dead is no different from describing the person as bald or angry or smiling. Each of these terms has a meaning known in advance within a particular community of discourse, where knowing the meaning means knowing when to use the term appropriately. Once we know that civil death is the state someone is in when he is sentenced to life imprisonment, we can use the word appropriately and convey to others who share the field of discourse in this community just what happened.^[70] What makes fictions different from other descriptive terms is that the technical, legal description created by the fiction overrides another ordinary description that is directly contrary to the fictional one. The civilly dead person is actually alive. And we're inclined to believe that, given that the two contradictory descriptions can't both be right, the ordinary description is true and the fiction is false. So, it looks as though the law is lying.

If lying is the forwarding of self-disbelieved statements with the intention others take them for true,^[71] however, legal fictions are not lies. For one thing, courts announce when they are using fictions, often reasoning their way through to the fiction from the ordinary description, making it clear that the intention is not to mislead. For another, once a fiction takes hold, the fictional description becomes the appropriate way in law to describe particular states of affairs. It takes on the life of an ordinary nonfictional term of reference.^[72] Finally, courts are quick to point out what they themselves are doing. Legal fictions almost always come with a semantic marker that alerts the reader to the special, fictional use of the terms in the narrative.^[73] We encounter civil death rather than death on its own

and we see implied invitation,^[74]constructive discharge,^[75]juridical bay,^[76]implied contract,^[77] and quasi derelict.^[78]

If the Riggs court had used the legal fiction, it would simply be doing more of what it had already done with the other abstract nouns in its description: using specialized legal language to describe an event. The fiction may have contradicted the nonlegal meaning of death, while the other elements of the legal description would have more or less coincided with ordinary description. But for a description to have legal relevance, it must use the terms of discourse of that specialized community. Fictions do just that. Just how they do that, we'll see in a moment. But first, let's examine what would have happened in Riggs if the court had used the fiction of civil death.

Reinterpreting the Law in Riggs

If the New York Court of Appeals had taken advantage of this fiction, which it easily could have, the theories of interpretation used by Judges Earl and Gray would have produced somewhat different results.

Take Judge Gray's opinion. If Elmer were deemed dead, Judge Gray would take his inheritance away. No fancy reading, no deviation from Gray's sense of literal interpretation, would be needed to reach the conclusion that the explicit provision of Francis's valid will should be honored. Since Francis's will explicitly indicated that his estate should revert to his daughters if Elmer died, Judge Gray would say in his Riggs opinion that the property should go to Mrs. Preston and Mrs. Riggs upon Elmer's civil death. Gray, of course, originally ruled that the property should pass to Elmer, but this change in the characterization of facts would allow Gray's "literal" reading of the statute to result in the opposite conclusion. Since the will explicitly provided for the transmission of property to Francis's daughters if Elmer had already died, Judge Gray would have no trouble finding that Elmer should not now inherit.

Judge Earl, himself intent on finding legislative intent, would not have had to stretch as far as he did to rule against Elmer. The statute of wills was undoubtedly drafted against a background where the civil death fiction was assumed to be part of the existing law. There would be no need to add a provision about murderers to the statute of wills if all murderers were already prevented from taking anything under a standard fiction. Once the ground seemed to be cut out from under the fiction in Avery , however, the statute of wills was left vulnerable to the sort of argument raised in Riggs .^[79] But an examination of legislative intent in drafting the statute would reveal that the lawmakers probably assumed such cases would never arise under the statute of wills because they were settled under another legal heading. It would be a cumbersome business indeed if legislators had to make every statute free-standing, without being able to count on the rest

of the law serving as a stable base. Earl could invoke the provision of civil death as part of a legislative intent argument without having to turn to the controversial maxim that opened his opinion to charges of judicial lawmaking.

Earl's opinion, then, could easily turn on a quite straightforward view of legislative intent, without relying at all on the general unwritten principles of the common law. And since it was the invocation of an unwritten maxim, not the appeal to legislative intent, that upset the critics, Earl's opinion could have used the statutorily created fiction to rule against Elmer. This would come closer to applying the law by appealing to the more obvious strategies for working out what a statute required rather than uncovering the principles lurking in the shadows. In fact, judging from the vigor of Earl's dissent in Avery , he might well have preferred to decide the case on civil death grounds anyway.

Recharacterizing the facts and creatively interpreting the law are alternative strategies for accomplishing the same results, as this extended discussion of Riggs reveals. A court can reach the same outcome by adopting a fiction and reading the statute "literally" as it would get by using a less strained characterization of the facts along with a more creative strategy for interpreting the law. Similarly, a single theory of interpretation of a legal text can yield different results when one factual description is substituted for another.

Realizing this, we can see that the debate over strategies of interpretation has missed a critical point. Suppose that we were successful in arguing that either legislative intent or plain words or an appeal to moral principle should be used to interpret statutes whenever they are unclear. Suppose we were agreed on the strategy of interpretation that judges should use to make their decisions the best they can be. As long as judges still have the flexibility to characterize the facts of cases, a theory of interpretation of legal texts alone will fail to provide determinate answers. Law does not live by doctrine alone. Legal rules and legal facts are mutually constituting. Legal descriptions always reflect legal judgment, just as legal texts always operate from a particular strategy of framing facts.

The Law/Fact Distinction Revisited

We are now in a position to see how misleading the usual view of the distinction between law and fact can be. There are two important lessons to be learned from our detailed consideration of this case.

First, courts don't first settle on an interpretation of the facts and then figure out what the law means. The practice of judging simultaneously engages both in an ongoing project of meaning-making, producing a single opinion in which fact and law are woven together in one coherent whole. Constructing the facts and constructing the law are not two separate enterprises, but are mutually implicated in the same project.^[80] Describing Palmer as civilly dead already involves reference to statutory provisions that create the status of civil death in the first place; the

interpretation of what the law allows on that point leads to a construction of the facts of the case. But looking to the statute on civil death was already prompted by a reading of the facts at the start, which in turn depended on knowing basic legal categories to sort out legally relevant facts from legally irrelevant ones. And so on. Both interpretations go on at once, and each presupposes that one has done the other first. There is no place to start. There are just ways to go on.

Second, the statements of the facts of cases in legal reasoning don't have a life of their own apart from legal interpretation, nor do legal texts have a meaning outside a context of facts. There are no "pure" descriptions of the facts or "abstract" accounts of legal rules. Statements of fact come already infused with legal judgment and legal rules come already fact-laden. As we have seen, rules generally take the form "If A, then B." And A in these formulations will always specify some statement of fact that cannot be separated from the rule. It is part of the rule, which itself wouldn't be a rule without it. Similarly, descriptions always use categories whose meaning is given by reference to some normative system of correct use. Legal descriptions already invoke legal standards, just as legal standards always have as part of their specification ideas about facts.

For example, the simple statutory provision "A person sentenced to imprisonment for life is thereafter deemed civilly dead" requires complex understandings of fact to be understood as a rule. Knowing when someone is civilly dead requires understanding what a person is, what a sentence is, and what imprisonment for a life term is. Understanding civil death as a rule (whatever that means, since the term civil death itself draws by metaphorical association on strong views about death as a factual state) is impossible without some associated statement about the circumstances that would activate it. Statements of fact are always going to be integral parts of rules. And judgment according to legal standards is always going to be central to legal description.

Our exploration of Riggs should convince us that interpretations of facts and interpretations of law cannot be separated. But we started down this long and winding road by noticing that the law/fact distinction is crucial for the organization of legal institutions. So, while legal descriptions may not be separable from legal standards, a great deal of the organization of American legal institutions is premised on the idea that such a separation can be made. How can we reconcile these two different pictures?

Legal Authority

The division of labor between judge and jury and between trial and appeals courts is based on a distinction that we have just seen is problematic.^[81] What should we make of this? We might shrug and say that there is a fuzzy boundary between the two or that the concepts are simply ambiguous in some circumstances—but certainly this isn't enough to disrupt a whole institutional

structure, especially when it has been operating for centuries as if the difference between fact and law were quite clear.

If we thought that the only conclusion to draw here were one about fuzzy boundaries or ambiguity, we would be making two mistakes. First, the criticism of the law/fact distinction presented here is far more radical than talk of fuzzy boundaries or ambiguity would suggest. Descriptions of facts are legal all the way down, just as legal rules always have as a crucial element a statement of the facts to which they are to apply. Law and fact are mutually constituting—not simply hard to tell apart.

Second, the institutional distinction may capture something that the separation of law and fact in legal reasoning does not. Institutional design is a way of allocating authority across different sets of actors. Dividing responsibility between judge and jury as well as between trial and appeals courts creates separate spheres of authoritative decision making. When juries (as opposed to judges) and trial courts (as opposed to appeals courts) are authorized to carry out a particular function that is denied to others, it gives them some measure of control over the outcome of a case, or at least over parts of that outcome.

Now we might say that this control is only an illusion. If judges on appeals courts can recharacterize the facts and completely change the legal outcome, how can we say that juries and trial courts have made any difference?

Juries make the first attempt at constructing the narrative of a particular case.^[82] They don't just determine the accuracy of the evidence presented at trial; they also reach a judgment by applying the legal standards in which they are instructed by the judge. So, juries in some way perform the same sort of operation that a judge does, but without the density of specialized knowledge that comes with legal training. Juries bring a different background knowledge to the task of judging, a background knowledge that is no less detailed and infused with judgment in describing facts than legal discourse. And this is where we can see how questions of fact appear different to juries than they do to those who have legal training.

When juries are instructed to find the facts of the case, they don't respond by saying, "What do you mean by facts?" The idea of fact is itself unproblematic in the background knowledge that the jurors bring with them. Jurors correctly perceive that they are being asked to figure out, with the knowledge they have acquired from much social experience, "what happened." And though the construction of "what happened" is infused with implicit assumptions and detailed knowledge that "goes without saying" all the way down, the task of figuring out "what happened" relies on social skills ordinary folks possess. Each of us does this all the time, working out accounts to make sense of the social world around us.

Even though juries are instructed on the law on point, they are expected to use also their general knowledge of how things work in the world. Their conclusions have authority because jurors have linked ordinary background knowledge and specialized evidence in the particular case with legal standards.^[83] And this is

where we can see the difference between facts as found by juries and facts as constructed by courts, particularly appeals courts. Jurors use cultural knowledge, background knowledge available in the social world, to orient and guide their narratives about what happened. This knowledge includes understandings of how people interact and function under difficult circumstances, and knowledge about what litigants may be assumed to know and inferences they can be assumed to have made, among other things. Judges and other legally trained officials use their legal knowledge to orient their narratives. This is not just knowledge of legal rules but also knowledge of the craft of fitting particularized descriptions into generalized legal categories.^[84] The different background knowledges and conventions of description might produce quite different outcomes, but the operating rules of courts provide that juries are to be given deference over judges and trial courts over appeals courts when it comes to narratives. As judges attempt to construct legal narratives, then, they are supposed to take into account those conventions of description that hold outside of legal practice as well as those that allow ordinary descriptive terms to be used in legal practice.^[85]

Legal narratives, then, are supposed to be constructed in such a way that they are recognizable both as stories that resonate in the general culture and as stories that work as legal accounts. Saying that Elmer Palmer murdered his grandfather in order to be able to inherit under his will is comprehensible as a story for the ordinary listener without legal training and as a story for those knowledgeable about legal concepts and categories. The narrative works in both cultural spheres at the same time. Now this is not to separate facts from background values again, after we have painstakingly showed how dependent narrative accounts must be on judgment-laden organizing concepts. But it is to point out that a narrative may be simultaneously informed by more than one set of background values. And American law is institutionally arranged so that multiple value orientations can infuse each account.

Legal fictions, on this view, stand out as problematic. Legal stories break apart from the organizing conventions of ordinary discourse when we call living people dead (even if only civilly) and use other descriptive terms in places where they do not seem to be warranted by evidence. We have seen that legal fictions are no different from other legal concepts in that they are given specialized meaning within a community of practice. But legal fictions override other ordinary meanings of a term, and so become the critical place to examine what happens when the two narrative cultures represented in law come apart.

Fictions, as we already noted, seem at first glance to be falsehoods. We can say that Elmer Palmer wasn't "really" dead because we have a very strong sense of what it would mean for someone to be dead, and Elmer Palmer didn't fit that description. But this isn't always the way that we use words, tediously matching them against the evidence to see whether their use is warranted. We also use words metaphorically, to highlight similarities between two fields of experience. We can imagine two coworkers greeting each other on Monday morning. Jane

says to Jim, "You look dead." And Jim replies, "I feel dead." Now, have they been lying? Or have they been drawing on the idea of death to describe a bit of life? Using death as a metaphor, Jane and Jim have evoked a range of associations to be carried over to the new context. But not all elements of death can be transferred in this way in this context. Jane's response to Jim's comment should not be to call an undertaker or to begin searching to fill his job. Instead, Jim's remark puts Jane on notice that, though he will be physically present, he may not be able to do the things customarily expected of him. The context makes it clear that the metaphor picks out incapacity rather than the permanence of the condition as the relevant feature of death for purposes of understanding the remark.

The fiction of civil death also picks out selected aspects of death to be transferred over to those who are living. When a convict has been sentenced to prison for life, he is in many ways like someone who is dead. He is permanently absent from the community and no longer part of the collective social life. Those who want to interact with him find him out of touch for reasons not of his choosing, nor of theirs. He cannot be, in relevant ways, a husband, a father, a friend, or a neighbor. By calling up associations with death, the legal fiction highlights just how profound is the social transformation of someone who has been sentenced to prison never to return. Just as with other metaphorical uses of terms, not all the features of the metaphorical description carry over into the new field of experience. It would be inappropriate to bury the convict, because the relevant feature of death for those purposes (cessation of heartbeat or brain waves) has not occurred. But social death is a powerful metaphor that conveys the seriousness of the transformation.^[86]

Legal fictions are metaphors, not lies. But if they were to be taken literally, they might produce an inaccurate impression, and the gap between ordinary discourse and legal discourse would widen, creating problems for legal authority. The linguistic markers that generally come with fictions alert the reader to the tropological use of the terms in the narrative. Civil death tells us that this isn't death in its usual form. Constructive escape from prison can mean the convict is still behind bars but has escaped the rules of the prison in crucial ways. General Motors may be a juristic person, but it can be present in court only through its agents.

But flagging the metaphorical term is not enough. The metaphor must be one that is understood in ordinary discourse. And if it isn't understood, it must be explained before it can be legitimately used in a legal narrative. This may account for why judges often go to some lengths to show why an X should be treated as a Y, pointing out the common features that X and Y share. For example, when a husband filed for divorce after his wife locked him out of their marital home, the judge found that the wife had "constructively deserted" her husband. If she stayed and he left, it strains credulity to describe her as having deserted him. But the judge explained that desertion provided grounds for

divorce because one spouse made it impossible for the other spouse to live with him or her. Locking out a spouse, on this logic, amounted to the same thing as deserting him, for in either event the husband could not live in the same household with the wife who locked or left.^[87] So while the use of the term desertion seems directly contrary to experience, the explanation reveals it to be a clumsy metaphor, picking out some common features in the two fields of experience and transferring the term of description to capture the relevant similarities. As metaphors get more and more strained, though, the discourse gap opens wider, and legal narrative starts to lose its authoritative force.

Why is authority attenuated in this way? The criticism that has generally attended fictions reveals that something seems quite amiss in the practice.^[88] It's not that the outcomes of cases involving fictions are generally thought to be wrong.^[89] But the use of fictions reveals just how far the description of "what happened" in a particular instance can get from the description that those close to the event would use. And if courts are supposed to be deciding this particular case and not some other, then fictions look like an abuse of judicial authority. Judges appear to be making up cases, and then deciding those, rather than deciding on the cases actually presented to them in court. And this undermines the authoritative nature of judicial decisions because the judges appear to have overstepped the boundaries of judicial discretion.

This should show us that legal authority is not simply internal to legal culture, but that it pertains to the relationship between legal culture and the culture of the world into which law is an intervention. Legal decisions have authority to the extent that the stories judges tell resonate both in the world from which the disputes and conflicts come and in the specialized world of legal discourse. Without this dual appeal, legal judgments are cut adrift from the larger world or lay stories cannot find a legal connection.

Legal authority, then, rests on the ability of legal narratives to reside simultaneously in the normative universes of legal and nonlegal worlds. In each case, the descriptions of facts of disputes are not separate from or prior to normative judgments, but are integrated with them from the start. The challenge in facing facts in legal interpretation is to realize the power of narrative in the construction of normative worlds and to see how the moral force of description operates all the way down.

Postscript:
The Murdering Heir after Riggs

The New York Court of Appeals had been bold in Riggs . Judge Earl's majority opinion struck out on new territory, ignoring the convenient fiction of civil death and proclaiming instead that no man shall profit from his wrong. This

one would have thought, would have made the New York doctrine clear. It was only seven years after Riggs , however, when another murdering heir came before the Court of Appeals and gave that court an opportunity to reconsider and refine what it had done.

Elizabeth Westcott poisoned her husband, Munroe Westcott, and stood to inherit everything under his will. Catharine Ellerson, Munroe's sister, thought she and Munroe's nieces and nephews should be able to inherit his property instead. Catharine and the other plaintiffs claimed that the will was void because Elizabeth had killed her husband, and they relied on the authority of Riggs .

Andrews, now chief judge, wrote the opinion in Ellerson v. Westcott.^[90]Riggs didn't make the will void, Andrews argued. The will was valid in Riggs , as in the present case, but

the court in a proper action will, by forbidding enforcement of a legal right, prevent her [Elizabeth] from enjoying the fruits of her iniquity. It will not and cannot set aside the will. That is valid, but it will act upon facts arising subsequent to its execution and deprive her of the use of the property.^[91]

What Andrews created in this case was another fiction: a constructive trust. Elizabeth could inherit the property under the will, but she was to hold it in trust for her heirs and she couldn't use it herself. The "trust" here wasn't real; Elizabeth hadn't gone through the usual legal procedures to establish such an institution. Moreover, Elizabeth couldn't administer the trust. Elizabeth inherited, but could not touch the ill-gotten gains. The fiction of the constructive trust operated to bar her access to the property she had gone to such great lengths to inherit.^[92]

The doctrine of constructive trust for cases like this caught on. Judge Benjamin Cardozo, encountering such a case years later, commented,

A constructive trust is the formula through which the conscience of equity finds expression. When property has been acquired in such circumstances that the holder of the legal title may not in good conscience retain the beneficial interest, equity converts him into a trustee.^[93]

Those who murdered to inherit went from being deemed dead to being deemed a trustee. That may sound like an improvement, but the result was largely the same: neither the dead person nor the trustee got to do what they wanted with the property they received.

This is not quite the end of the story. Many states have adopted the constructive trust formulation and are using that fiction actively in their decisions on the subject.^[94] The Uniform Probate Code, however, takes a different approach:

A surviving spouse, heir or devisee who feloniously and intentionally kills the decedent is not entitled to any benefits under the will or under this Article, and the estate of the decedent passes as if the killer had predeceased the decedent .^[95]

The more things change, the more they remain the same.

The Placement of Politics in Roberto Unger's Politics

Martin Stone

Self-consciousness had not yet advanced to the abstraction of subjectivity, not even so far as to understand that, when a decision is to be made, an 'I will' must be pronounced by man himself. This 'I will' constitutes the great difference between the ancient world and the modern, and in the great edifice of the state it must therefore have its appropriate objective existence.
—Hegel^[1]

Preliminary:
"Superliberalism"

The critique of liberalism has become a growth enterprise, even though in "liberalism"—in the texts and institutions canonized under this name—there is no single, unified doctrine to be upheld, only a family of ideas, loosely related. Insofar as it is part of the promise of liberalism to keep itself open to criticism from within, it seems that to think in general about law or politics just is , at least in the Anglo-American world, to stake out a position in this family, hence to engage in a practice that is, in an anodyne sense—one that leaves open the twin possibilities of conservation and criticism—liberalism's critique. Perhaps this suggestion will be heard as unduly assimilating by a group of contemporary legal scholars—the Critical Legal Studies Movement—for whom the demystification of "liberal-legalism" through the exhibition of its "contradictions" has become a programmatic idea.^[2] Roberto Unger is the most eloquent and wide-ranging voice in this movement. His recent three-volume work, Politics: A Work in Constructive Social Theory (1987), which succeeds his well-known Knowledge and Politics (1975), is billed by its publishers as a radical alternative to both Marxism and social democracy. In the course of the work Unger in fact makes several concrete political proposals, fostering the impression that the work's ambitious theoretical arguments determine a practical political program. A guiding aspiration of my essay is to bring Unger's picture of the relation between the theoretical and the practical dimensions of his undertaking into question, and therefore to question whether his engagement with liberal political theory suffices to articulate a politics at all. My quarrel with Unger takes place, however, on the ground of my gratitude to his new work for affording an exemplary opportunity to begin thinking about what it might mean for us (and who are "we"?) to confront liberalism tout court .

Unger's name for his program—"Superliberalism"—in fact accommodates a tension: on the one hand, an affirmation of liberalism, an intensification of its commitments; on the other hand, a movement beyond the tradition, its over-

coming. Can this tension be represented in the terms familiar to the daily discourse of liberal social democracy: an endorsement of liberal concepts coupled with an indictment of the institutions (or more specific conceptions ) that are thought to realize them? From this perspective the difficulty would appear to be not with liberalism's truth but with its reality, as if liberalism has been only imperfectly understood and applied. Yet this hypothesis (which also informs such philosophically mainstream interpretations of liberalism as John Rawls's A Theory of Justice ) is bound to elicit surprise, not simply because it moves against the grain of the political sensibility of Critical Legal Studies, but because Unger's own earlier Knowledge and Politics mounted a sustained attack on what he was prepared to call "the liberal doctrine" (28). Briefly, in this work "liberalism" stands for a social formation structured by certain deep-rooted "antinomies"—fact/value, reason/desire, public/private (I will not pause here to ponder the intellectual stakes of labeling these much belabored dichotomies as "antinomies"). Unger's aim—curiously declared in the final pages to be unrealizable without the help of Providence!—is to replace liberalism by the successor formation that the resolution of the "antinomies" requires. (The book's last breath is given in prayer: "Speak, God.")

Unger's recent work appears to enact a reversal. It attempts to spell out a politics of modernism—a politics, we might say, appropriate to the historical experience that Nietzsche calls "nihilism" and identifies with the annihilation of all traditional values together with their regrounding in human will. In the opening to Passion: An Essay on Personality (1984), a work that thematically anticipates Politics , Unger announces a principle that apparently guides all of his subsequent writing: "Thought speaks with authority about who we are and how we should live only when it puts our ideals and self-understandings through the skeptic's flame, risking nihilism for the sake of insight" (viii–ix). The main concern of Unger's new work, accordingly, is the nature and possibility of political justification when transcendent sanctions for human ends have come crashing down, so to speak, some time after the death of God. Holding itself responsible to this problematic of "nihilism," Politics conducts a conversation with the disappointed theorist of Knowledge and Politics , offering the promise of his satisfaction.

In examining Superliberalism, then, we might take our bearings by tracing this apparent shift within Unger's position. I have organized my account of it under two familiar legal headings, "procedure" and "substance"—but only to indicate, finally, how Unger now undertakes to elide this distinction, to articulate the substance of his program as a perfection of the procedures of political decision, a sort of technology of social will. Consider a representative passage from False Necessity , the second volume of Politics :

The conception of democracy presented here merges into the idea of the next best thing to the absolute or natural context that we can never possess. The second best is a set of artificial arrangements that . . . better ensure their own availability to challenge. (FN , 362; emphasis added)

Unger's many concrete proposals (the multiplication of branches of government, the institution of a rotating capital fund, and so on) are designed to "ensure" this heightened artificiality, plasticity, or disposability of all social arrangements, their pervasive "availability to challenge"—a condition he heralds with enthusiasm yet which is remarkably, we might note in passing, like the one Martin Heidegger describes darkly as the culmination of Western metaphysics in the will to power.^[3] The passage cited above also sounds a curious, somewhat opaque note in that it forces the following question upon us: Is "the conception of democracy presented here" merely remedial ? How is the ideal of social plasticity related to the thought, which this passage seems to engage, of dis-possession —of a something "we can never possess," a deficiency pertaining to human being? Many commentators have wished to qualify Unger's new work as "modern" or "postmodern";^[4] presumably, this is to understand it as detaching itself from something that came before, a gesture Drucilla Cornell, in an informative essay, calls Unger's "critique of nostalgia."^[5] She is not, on my view, exactly wrong. Yet I wonder whether she shouldn't be more curious to know what exactly it is that Unger believes "we can never possess": "the absolute or natural context." The absolute context?

Procedural Matters:
The Constraints of "Social Context"

In a postscript to the second edition (1984) of Knowledge and Politics , Unger says that the problems that interest him have changed less than his "ways of going about them" (337). What he might mean by this can be easily surmised. Knowledge and Politics stakes its political hopes on a method it calls "total criticism" (1–18). A total criticism of liberal society is presented as a requisite of achieving the political vision that would enable us to institute a successor to liberalism, one in which, as Unger seems to suggest, the distinctions between fact and value, reason and desire, public and private, and so on would no longer structure our lives.^[6] Yet as the book ends, it turns out that this vision, and so for the time being liberalism's promised successor, is unavailable—at least within what Unger calls, with intended contrast , "earthly life," "history," and "the world" (294, 260, 295). We might call this contrast (whatever the unrepresentable contrasting items might be) metaphysical dualism, and what philosopher of our century has not wanted to turn the page of this philosophy? Even if we were prepared to accept its seemingly immobilizing structure in Unger's early work, we are made to wonder whether it is intelligible to speak of a view as unavailable if it is not at least sometimes—available .

This is related to another, somewhat less intractable question: Is it explanatory or otherwise useful to speak theoretically of canonical types of social structure (i.e., "liberalism," "capitalism," "democracy," "communism," "feudalism"), each

unified by a closed set of necessary elements?^[7] A virtue of Unger's newest work is its rejection of the more sophisticated variations of this picture at a time when even many who write out of a deep sense of inheritance from Marx have abandoned it as well.^[8] But the theorist of Knowledge and Politics is no Marxist, and it is not this problem—the possibility of discovering empirical laws of social structure—that worries him. Indeed, he thinks of the "liberal doctrine" as unified only in a loose, unfixed way—by nothing more, it would appear, than the conceptuality that the "antinomies" describe.^[9] "Liberalism" thus betokens nothing less than political and philosophical modernity, the world as it has come to pass since, say, the time Descartes found himself thinking. But parallel to this, Unger's demand for total criticism would seem to arise not from any tractable concern about our ability to understand our social structure (something we can improve by going to school, or reading poetry, or studying logic) but—is this a displacement?—from a worry about the way in which criticism is conditioned by the concepts it inherits, controlled by its social or historical "context."

Evidently, a desire to say something like this with the word context , however unclear we may be about what is being said (as if the fact of our saying it, repeating it, were ironically an instance of our socially dependent condition), is widespread today. We may hear it in no less a theorist than Michael Walzer describing "the problem of critical force." (His figure for this problem, interestingly enough, will be that of a parasite , a person whose will—or a form of life whose life—is not quite properly their own.) "Given that every interpretation is parasitic on its 'text,'" Walzer asks, "how can it ever constitute an adequate criticism of the text?"^[10] Can we begin answering this question except by questioning whether we understand what it is asking? (Every interpretation —how is this known? And if every , what sense is to be given to a nonparasitic interpretation, or hence, by contrast, to a parasitic one? Then what exactly is the term adequate criticism looking for?) Doesn't Walzer's question—which seems often to present itself in the form of an historicism that has turned its back on "metaphysics"—inevitably give rise to and reinforce a familiar dialectic, refiguring an old page of philosophy?

In Knowledge and Politics , Unger's version of Walzer's question about the autonomy (or parasitism) of criticism borrows a metaphor from Kant's Critique of Pure Reason : "The methods of proof and argument are part of the theory to be criticized. In what form then is total criticism to bring its suit, and by what law are its claims to be judged?" (12). Is this meant to politicize our conceptuality, as if the difficulty of "proof" or "argument" were the effect of a "law" that is to be overturned or transcended? If a similar notion, perhaps more carefully controlled, seems facilitating (or seductive) in the writing of, say, Michel Foucault (doesn't his phrase "regimes of truth" call for, and at the same time overwhelm, politics?), are we instructed to reconsider this metaphor in Kant—the categories as "rules," their "deduction," the establishment of a "clear legal title"; the world as "legislated"; the "sceptics" as "nomads" breaking up "all civil society"; the critique as the "insti-

tution of a tribunal" settling "lawful claims" to disputed "titles" and "territories," establishing "a land of truth" against "opposing claims"; the "philosopher" as the "lawgiver of human reason," and so on? If political figurations are part of the weave of this text, their obscurity, as we shall see, is both unknown and controlling in Unger's own.^[11]

The commonsense reassurance that might be given to queries about "critical leverage"—through, for instance, Otto Neurath's image of a boat we rebuild on the open sea, one plank at a time using the materials at hand^[12] —seems unlikely to satisfy a theorist who, like Unger, presses "total" into the question or who speaks (oddly using Heideggerian terms) of "the danger that our views of the social ideal will turn out to be a projection of the particular society into which we happen to have been thrown" (ST , 23, 41). Could it have happened otherwise? That nothing in this world will satisfy him seems to be part of the aura and the very expression of his worry, as if he experiences the necessity of social particularity itself as an exile , a fate. Thus, if we have been reading attentively, it shouldn't come as a surprise when Knowledge and Politics ends with a prayer asking God "to complete the change of the world" by giving us what, left to ourselves, we "would always lack." This at last, and in a vocabulary that is unmistakable to us, sends the total criticism Unger believes he requires skittering across the transcendental horizon. One does not make up this lack by going to school.

A picture holds us captive in Knowledge and Politics —the picture of intellectual authority as distance . To criticize society radically one must stand beyond it, exit the cave, leave the city, go to the woods, speak from a high place in an unlawyerly voice. Henry David Thoreau presents for our American consideration an apparently empowering version of this picture when in Civil Disobedience he complains of the "statesmen and legislators" who, "standing so completely within the institution never distinctly and nakedly behold it . . . speak of moving society, but have no resting-place without it"—though it belongs to his complexity to have declared himself in the opening of Walden a "sojourner in civilized life again," thus reporting that the sought-after foundation is not to be found (or founded) where this picture tempts us to look for it.^[13] Where then?^[14] It characterizes the progress of Unger's recent work, its methodological departure from Knowledge and Politics , to see the picture of a social totality, outside of which the radical must try to stand,^[15] as producing the illusion of an unhappy choice: "It suggests that all changes short of total revolution must amount to mere conservative tinkering. It thereby induces in its adepts a fatal oscillation between unjustified confidence and equally unjustified prostration" (ST , 15). What could have been more confident, we are invited to ask here, than Unger's own earlier call to replace "the liberal doctrine," or more prostrate than the appeal beyond politics to the God who does not reveal?

Unger's new work, as I have indicated, gives itself (in contrast to what appears

in Knowledge and Politics to be a political deferral) license to undertake a wide-ranging political program, one that he urges us, as citizens and legislators, to follow into practice. Politics indeed engages new "ways of going about" the problem of social change as well as what might be called issues of social perfection. These new ways are apparently more commendable to a modern, secular sensibility, as the recent academic commentary on Unger indicates. Richard Rorty, for example, contrasts Politics with European metaphysics, analogizing the timbre of Unger's writing to the sound that Walt Whitman's "Democratic Vistas" might have had for a nineteenth-century European educated by reading Hegel and lauding Unger for theorizing about politics in a way that rejects the philosophical foundations (evidently not "resting places") that Roty regards as among the less enabling moments of our tradition.^[16] This reading would apparently establish a contrast with Unger's first book that could not be greater.

Rorty is helpfully attentive to something that is no doubt there to be heard in Unger's pages—something for which an index heading to the first of the three new volumes might well serve as the emblem: "Foundationalism, foundations (and living without them)" (248).^[17]Politics indeed rejects, in all the gestures by which it understands itself, the metaphysical tradition of thought about law and politics that reaches an extreme point in a text such as Hegel's Philosophy of Right where (for example) the entire regime of legality is to be grounded in what is indispensable to thought about action and will. In his programmatic The Critical Legal Studies Movement (1983), Unger announces his "overriding concern" to overcome "formalism" in legal thought, "a belief in the possibility of a method of legal justification that contrasts with open-ended disputes about the basic terms of social life" (1), with "ideology . . . and political prophecy" (11).^[18] Yet despite such pronouncements, I find ample occasion to wonder whether Rorty isn't too ready to allow Unger to mislead him. Indeed, a question that arises for me in reading Unger's new work is whether one can succeed in overcoming a philosophy by simply rejecting it, or in exiting from a metaphysics of legality by becoming what Critical Legal Studies calls, in an apparent inflation, political . When Unger writes that he will show how the "nihilist claim that it is all politics permits a deeper insight into social life and . . . a better grasp of constraints upon transformative action" (ST , 10; emphasis added), I become tempted to read his new work as a test case for these matters.

An obvious point of contact with Knowledge and Politics are the words context and constraint , which are found together throughout Unger's new work (in Passion contextuality names one of the two fundamental "problems" of human existence [5]; we shall come soon to solidarity , the other). Unger still laments the social conditions of criticism, announcing the "bad news" that "we must live and think most of the time in a context" (ST , 169); he never hesitates to describe such contexts as

"imprisoning."^[19] Yet now, in his recent work, this recurring thought is coupled with a political prescription to loosen contextual constraint: "We never overcome context-dependence. But we may loosen it" (ST , 21); "Conditionality is never overcome. But it may be loosened" (Passion , 10). This raises from the outset a difficulty about what "the social context" must be thought to be if its "constraints" are something we "loosen" by political action.

It might be helpful here to consider how it is sometimes affirmed (or denied) in discussions of literary theory that texts constrain our interpretation of them. It is unclear that the sense in which this might (or must) be said amounts to anything more than the thought that not everything said about a text is what we should be willing to call an interpretation of it.^[20] Is the idea of this "constraint," then, different from—lacking the externality of—what happens when, say, a goldfish inhabits a bowl? Do we inhabit a social context? Is "loosening conditionality" like increasing the size of the bowl, a matter of more (psychological, emotional, political, interpersonal) room to swim about? If so, Unger will presumably tell us how to measure this room. A metric of "loosening context-dependence" (permitting us to say of a particular context that it is looser than another) would help us see what Unger means by saying that we are "imprisoned" (a strong word, implying the operation of what I suppose Foucault wishes to call power ) by every social context.

My sense of the difficulty in construing Unger arises from typical passages in Politics such as this, in which, after stipulating that a social context includes "criteria of validity, verification, or sense" (ST , 18), he goes on to say:

A discovery of yours may be impossible to verify, validate, or even make sense of within the available forms of explanation and discourse; or it may conflict with the fundamental pictures of reality embodied in these forms. It may nevertheless be true. (ST , 20; Passion , 8)

Perhaps my asking how social contexts "constrain" stands in some internal relation to wondering in what sense forms of discourse are "available" to us. Such availability would not, presumably, be the same as the way new-model cars become "available" each year. Unger, however, does not pause to explain these most basic of his thoughts in Politics . He writes out of an apparent sense that his readers already understand them, a practice of writing that seems to define a reader as out of it who might wish to ask why something that it is impossible for us to explain or make sense of is not just nonsense . What could be meant by its "discovery"? Presumably, the answer has to do with what it is for "forms of discourse" to be "available." Do remarks like the following help?

An insight may enter into conflict with established criteria of validity, verification, and sense, or with a settled conception of fundamental reality. But if it tells a truth [How magical! Falsehoods remain incomprehensible!] then there will be criteria that can be retrospectively constructed with the aim of preserving it. (ST , 21)

In the contest between the incongruous insight and the established context, the context may go under, and the proponents of the insight may discover retrospectively the terms that justify the forbidden idea. (ST , 20; Passion , 8)

As if in response to my question, Unger speaks of a possible reconstruction—clarifying this modality will be crucial—of the social context, one that converts a "forbidden idea" into a lawful one. (Notice that in the second passage quoted we can read an explicit political allegory: a "contest," an "established context," "proponents," "justification," something "forbidden"—the elements of institutionalized legality.) But what sort of possibility is this? If the possibility of being "retrospectively constructed" means that it is always plausible or reasonable to modify our "forms of discourse," the claim seems difficult to accept on Unger's own account: one would expect those "contextual" factors that, according to Unger, determine a discovery as "forbidden" to make its own preservation unreasonable. A weaker sense of possibility seems as unobjectionable as it is trivial. Of course, the "form of discourse" might change, but if this bare thought of openness is intelligible it seems to carry little assurance that we could in fact imagine or even make sense of a change. (Isn't what is imaginable by us a question of what it makes sense to imagine?) If, then, there is no such thing as rolling back the bounds of sense by ordinary legislative or judicial action—or is this what is thought by the "legislation" of the poets? Then by what light are we citizens to read?^[21] —how could there be, in the anemic sense of possibility appropriate for Unger's statements, a politics of context?^[22]

Perhaps it will be said at this point that the problem we have in seeing how something which makes no sense to us nonetheless could (in a different "form of discourse") tell a truth indicates just what it means to be constrained by our present "form of discourse." But this seems misleading. If we can't "see" what it would be to stand beyond our "forms of discourse," it seems empty to say that we are constrained by them, for the contrast that would give this word its point has withered away. Admittedly, to come to feel such emptiness in one's words is not, as I take Stanley Fish to suggest, to dispel the worry that one tried to word; it may even invite new worries, say about one's temptation toward this emptiness, its motivation.^[23] What is taking place in talk of "constraint" by a "social context," by "available forms of discourse"? The impression might arise that these tropes, so widespread today, refigure in a new secular historicist version that which (lacking a name) we might call a severe experience of the human, one for which there is no such thing as confirmation. This historicism represents "the forbidden idea" as the presence of a sense that—given our "social practices," our "historical forms of discourse"—is unavailable to us. A senseless sense, unreasonable reason—can we not, then, banish these phantom doublings, send them away by the voice of philosophical correctness?^[24]

This voice will remind us that our "constraint" by a social context, by forms

of discourse, is not a significant hypothesis but a philosophical platitude, a form of representation; that contextuality as such is innocuous, its denial not false but puzzling.^[25] This voice will make a joke, asking perhaps whether it does any good against the constraint of "forms of discourse" to build up one's vocabulary or take a writing course. In Wittgenstein's Philosophical Investigations , with its powerful ear for the double as the philosophical ghost that hovers slightly off the ground, the voice sounds like this:

When a sentence is called senseless, it is not as it were its sense that is senseless. But a combination of words is being excluded from the language, withdrawn from circulation.^[26]

"Is being excluded [ausschliessen ]," "withdrawn from circulation [aus dem Verkehr ziehen ]"? How is this? There is clearly volition in these words. Who is doing the excluding? The answer is that "we" are—when we call a sentence out of bounds. Bounds ? Are they like constraints ? (And who are "we"?) In these words of Wittgenstein's, the question I have begun to put to Unger might take the following form: Can we comply with Unger in thinking this "exclusion," a condition of the shared social world (or of, say, a polis ), as the operation of a political will, as a regime so to speak of discourse? Or does the "exclusion" and "withdrawal" of which Wittgenstein speaks require the operation of a will prior to politics, one that, though it is present in all political and social arrangements, opens the space of the political as such : a will prior to wills ? Assuming that I have not misrepresented Unger by emphasizing difficulties that he might regard as inessential, these questions will be capable of being rederived from the programmatic core of Politics .

What is the place of the political?

Substantive Matters:
Context-Transcendence As a Social Goal

Whether one wants to reach a goal or to avoid all goals (as, e.g., the philosopher does who smells a boundary, a nook, a prison, a stupidity in every goal) . . .
—Nietzsche^[27]

Alongside "contextuality," the "problem of solidarity" is the second fundamental "problem" that runs through all of Unger's work (Passion , 20). This expression refers to the practical conflicts occasioned by the fact that the community of others is both an indespensable condition of, and threat to, the individual's self-assertion. Economic and social attachments bring goods no one could attain alone; they protect us and enlarge our self-identity. Yet in attachment lurk possibilities of domination and destruction. For not entirely perspicuous reasons, Unger calls this a "paradox" (Passion , 20); his Critical Legal Studies colleague Duncan Kennedy calls it "the fundamental contradiction."^[28]

According to Kennedy, liberal legality depends on this contradiction, the cases that judges decide being variations on the need of each person to cooperate

with and be protected from others—an unremarkable account, except for Kennedy's word contradiction . That a condition of the possibility of law is that persons have separate wills and thus are capable, in principle, of both cooperation and conflict, friendship and enmity, is hardly news; if a "contradiction" must be found here, we might as well say that law involves a "contradiction" as that our "contradictory" relations with each other can be brought under law—though saying either one is likely to produce confusion.^[29] One way the "paradox" of solidarity seems to acquire some traction, however, involves the thought, variations of which appear in writers as different as Hume and Marx, that if natural, social, or psychological circumstances were different the value of or need for legality would diminish or disappear. "Encrease to a sufficient degree the benevolence of men, or the bounty of nature," Hume writes, "and you render justice useless, by supplying its place with much nobler virtues, and more favourable blessings."^[30] On such advice, Knowledge and Politics orchestrates a solution to the problem of solidarity in an antilegalistic key, rejecting "the peculiar legal mentality that animates liberal political thought" (74) for a regulative ideal of "sympathy," a perfected community in which

it would be possible to view others as complementary rather than opposing wills; furtherance of their ends would mean the advancement of one's own. The conflict between the demands of individuality and of sociability would disappear. Each person, secure in his individuality, would be able to recognize his own humanity in other persons. (KP , 220–21)

It is worth asking whether we understand how the situation described here could be supposed to render legality otiose or even "peculiar." A familiar objection is that legal forms will remain necessary to structure cooperation. A more serious problem is symptomatized by an evident non sequitur: Isn't it already "possible to view others as complementary rather than opposing wills"? Since this possibility hardly entails the disappearance of social conflict, perhaps Unger meant to write that it would be impossible to view others as opposing rather than complementary wills. But can we still understand this as a description of distinct persons—much less of, say, friendship? A difficulty with the (even today widespread) idea of the circumstances of law or right withering away, pictured as communitarian harmony, could thus be diagnosed as a slippage of modality: to the extent that we think law as mediating the possibility , not simply the actuality, of conflict with a social enemy, legality would seem to be at least as enduring a notion as persons or friends. This thought may become more present to us when we ask whether—on any model we know of what interpersonal "harmony" involves—what makes it possible for me to be in "harmony" with the other, where I am neither his master nor his slave, does not also require the possibility of my being his enemy as well.^[31] (Nietzsche seems to place even the friendships of which we are not yet capable still under the aspect of law: "You should honour even the enemy in your friend. Can you go near to your friend without going over to him?")^[32] Unger, it is safe to say, does not pose the question of human relationship at this (still very abstract) level

of specificity, being content to allow the "resolution" of the problem of solidarity to remain as unconceived as what is "peculiar" about legality itself.^[33] "Unable to offer either a blueprint of social order or a prescription for social change," he wishes to point "toward an image that lies beyond [the] field of vision" (262)—an unimaginable image.

I mention the depoliticizing thrust of Knowledge and Politics for the last time only to set off the unmistakable claim Unger makes when he writes that his new theory "instantiates [a] mode of politics" (CLSM , 109), "informs [a] group of transformative possibilities" (FN , 324).^[34] Apparently, Unger now feels able to work in the law school library, so to speak, without lowering the level of critical resolution. He speaks of following a path of "internal development" according to which "you start from the conflicts between the available ideals of social life in your own social world or legal tradition. . . . You imagine the actualizations transformed, or you transform them in fact" (CLSM , 18).^[35] Indeed. But to say only this would be to remain at the level of a formalism. As Unger himself often points out in his discussion of legal adjudication, "Any tradition is so rich in ambiguity that persuasive arguments can be offered for developing it in alternative directions" (CLSM , 91). If this is right, should we speak of a tradition, or even a society at all?^[36]

Suppose we take some help here from Rawls, who attempts to formulate the sense in which we inherit a political tradition while finding conflicting articulations of it reasonable as a distinction between a "concept" and the various contestable "conceptions" that spell out a "basic structure of society."^[37] Rawls is determining in a particular way the problematic around which we anticipated that Unger's Politics will revolve. For if it is asked in what a conception's falling under a concept consists , the answer might presumably be along the same anodyne lines as what it is for a text to "constrain" its interpretation—a matter, whatever this means, of "our" practices, of how "we" go on, excluding, calling out of bounds.^[38] The opening pages of Rawls's A Theory of Justice are, we could say, a placement of the political (and hence of political theory) according to which no confrontation with the political tradition as such is envisioned. When Rawls later spells this out in replies to critics, his title will be: "Justice as Fairness: Political Not Metaphysical."^[39] From this (unmetaphysical) perspective, a work that claims the title Politics and proposes an "internal development" must justify some particular "actualization" or extension of the tradition, setting out a conception of "the basic structure of society"—a conception Rawls calls (when it includes more than a conception of justice) "a social ideal" (9).

God remains silent, but it is tempting to say that Unger now speaks —and in a voice that will be audible, perhaps above all, to lawyers:

The program . . . is neither just another variant of the mythic, antiliberal republic nor much less some preposterous synthesis of the established democracies with their imaginary

opposite. Instead, it represents a superliberalism . It pushes the liberal premises about state and society, about freedom from dependence and governance of social relations by the will, to the point at which they merge into a large ambition: the building of a social world less alien to a self that can always violate the generative rules of its own mental or social constructs and put other rules and other constructs in their place. (CLSM , 41)

What can the lawyer hear in this? Not simply that every domain of political life is to be opened to volitional conflict (the system of "checks and balances" was itself meant as an institutional hypostatization of the Other's potentially conflicting will), and still less that liberalism is to be grounded in a property of "a self" (a self, let us notice, for which the "social world" can be apparently only more or "less alien"). The lawyer hears that the important thing about a social structure is its capacity to be replaced . The lawyer will hear this because it is homologous with a proceduralism that is already familiar to a profession largely doubtful (witness the "legal process" and "legal realism" schools of the previous generations) about the possibility of justifying what recent American constitutional thought calls "fundamental values." Proposing to "risk nihilism," to put "our ideals and self-understandings through the skeptic's flame" (Passion , ix), Unger now discovers, in the idea of a social world that "makes available . . . the instruments of its own revision" (CLSM , 105), a way to bring the severe thought of alienated personality (a sojourner, so to speak, through disenchanted civilizations) to "a successful conclusion" (ST , 149).^[40] Rejecting a "definite enacted vision of social life" (FN , 574), a "definite institutional order" (FN , 572)—skeptically rejecting, it might seem, any positivities of political structure (or would this be an ideal limit?)—Unger's new politics legislates—what? Leaving the question open, the motto for these politics might be: "Constitutionalize sojourning and there is a Home everywhere."

To begin to understand Unger's program, it might be useful to recall John Ely's procedural interpretation of judicial review, Democracy and Distrust (1980), a standard in the constitutional theory curriculum. Few are more skeptical than Ely about the possibility of justifying Supreme Court decisions that involve strong, controversial views about what democratic majorities may constitutionally put into legislation. Although these decisions often present themselves as readings of the constitutional text or its history, Ely thinks any justification must ultimately have the form of an appeal to "fundamental values," practically a code word with him for that which cannot be justified. Yet having reached this skeptical conclusion, Ely goes on to propose a constructive advance that is supposed to be faithful and responsive to it: a policy of increased political corrigibility through judicial surveillance of the mechanisms of "democratic" (Ely doesn't explore this concept; its paradigm for him seems to be our current legislative institutions) change.

Ely's argument is at least formally analogous to one we can find in Politics when Unger announces that he wants to "incorporate  . . . the insights proclaimed

by . . . skepticism about values and ideals" (FN , 360–61; emphasis added). This gambit likewise has a negative and a positive aspect. Negatively, it declares all ideals to be contestable; then contestability is said to "require"—note the word—plastic social arrangements (ST , 45). According to Unger, "objectivity" (what the skeptic is said to contest) is not achieved, as even the theorist of Knowledge and Politics had supposed, by creating a "perfected community" (FN , 560–63) but rather "by rendering the structure insubstantial —by turning it, increasingly, into the structure of no structure" (ST , 45; emphasis added). Taking the acceptance of an "insubstantial" constitutional structure to be consistent with the rejection of all ideals of "perfection," Unger bills his program as one even a skeptic can accept (FN , 572).

But to schematize the argument in this way is to bring out its incompleteness. First: does the freedom to revise ("destabilize," Unger says) social arrangements reflexively extend in scope to Superliberalism, the scheme of freedom itself? Can this consistently be withheld? The question presents an analogue to a familiar objection against Ely, according to which the judicial "reinforcement" of democratic representation stands on the same level as other substantive choices, and cannot, therefore, survive the skeptical gambit.^[41] In fact, the problem would seem to be even more difficult for Unger since, in declining to take any particular structure—in the way Ely takes the American constitutional structure—as given, his skepticism has a broader range. Unger cannot respond by claiming that deliberative political conflict is a higher-order value implicated in the Constitution itself. Or rather, to do so—to distinguish levels or types of value—would amount to nothing less than explaining why the value of freedom that Superliberalism manifestly privileges is excused or exempt from the skepticism the argument aims to "incorporate."

Second: how exactly does one get from contestability to revisability, from the unjustifiability of particular social arrangements to the practical imperative of devising social contexts affording opportunities for self-disruption? Even if it were maintained against the first objection that the skeptic could consistently commit himself to context-disruption, it is hard to see that he is "required" (ST , 45) to do so. Unger apparently supposes that the skeptic who "disbelieves all large theoretical arguments for particular social ideals . . . for that very reason wishes to weaken the tyranny of the present over the future" (FN , 572; emphasis added). For what very reason? If we avow the impossibility of justifying any social arrangement, do we thereby give anyone a reason to weaken the influence of power and privilege? (Tell this to Thrasymachus.)^[42]

Both of these objections point toward the same gap, one that we sense when Unger speaks of how skeptical lessons "start to become constructive insights" (CLSM , 113). This seems ill-advised unless some independent justification is also provided for context-disruption, the key—dare I say "structural"?—element in his program. Can we tell from reading the three volumes what this justifica-

tion is? Can we tell, that is, what determines Unger's conception of liberalism as a thorough-going proceduralism?

The Instrumental Reading of Superliberalism

Two distinct strands in Unger's writing need to be sorted out at this point. One (the other will be discussed in the following section) claims instrumental value for greater plasticity ("disentrenchment," "revisability"), enumerating the "practical advantages" (FN , 297, 244) and "benefits" (ST , 156, 198, 7) it makes available.^[43] These include wealth and power (PP , 1), increased productivity (CLSM , 93; FN , 279), exploitation of technological opportunities (FN , 297), diminution of personal dependencies (FN , 291), and perhaps above all a partial resolution of the problem solidarity (ST , 210; FN , 290). The sum of these "material and intangible benefits" (FN , 279) is called "negative capability" (FN , 229; ST , 155; CLSM , 93), an expression Unger did not make up. Keats makes it the prerogative of the literary, but as Unger uses the term it is meant to be capable of explaining the historical emergence of new social structures (FN , 249, 280)—either quasi-intentionally (societies aim for these advantages), or functionally (though not aimed at, they work in an explanatory loop).^[44]

Though straightforward, an argument for plasticity based upon its instrumental advantages faces four related difficulties that may be summarized briefly in ascending order of scope and importance:

A Low-Grade Inconsistency

Disentrenchment of social structure is said to address the problem of solidarity by weakening personal dependency. Superliberalism's new man—its "Overman"—will be "haughty, high-spirited, and even reckless . . . secure in an inviolable independence" (FN , 134). Unger acknowledges that the program is in this way "not neutral":

It seems to value the development of individual and collective capabilities more than the continuance of particular loyalties to individuals and groups. It slights the customary practices in which such loyalties are inevitably embedded. (FN , 592)

Oddly enough, however, a centerpiece of Unger's new program of legal rights involves a revision of modern contract law to incorporate obligations arising from personal dependence and group loyalty.^[45] Indeed, Unger wishes in general to give legal protection to "expectations that arise from . . . relations of mutual reliance and vulnerability" (CLSM , 40)—thus apparently valuing and reinforcing (not "slighting" or disrupting) entrenched patterns of community and social relation. What are we to make of this conflict? Both policies—the reinforcement and the disruption of social relations—might plausibly be thought to address the problem of solidarity. (Why suppose, after all, that persons will feel less threatened in a world given over to "pitiless" reconstitution [PP , 208]?) Yet, taken

together, these policies simply reiterate the problem in a particular form. Or is Unger undecided between alternative solutions?

The Inconsistency Generalized

If both policies can plausibly be thought of as instruments for improving solidarity, perhaps this only signals an inconvenient fact about the instrumental justification of increased plasticity (disentrenchment): its need for a rich empirical account of the social fields and networks over which it ranges. More generally, that increasing plasticity sometimes has value does not mean that any increase will be valuable across the whole range of social circumstances and goals. Will marginal increases in plasticity never have off-setting consequences?^[46] Juxtaposition of these alternatives—disentrenchment versus protection of established social relations—shows very little by itself, of course. It reminds us, however, that if increased plasticity is instrumentally valuable, it is not an unqualified good.

Increased Plasticity:
Relativity to Substantive Ends

Though Unger acknowledges the worry of competing goods (FN , 280, 303), he ignores the difficulties of commensurability and information familiar to consequentialist ethics. Indeed, he appears strangely willing to concede that he has no "proposal about how to reconcile freedom with other ends" (CLSM , 105). An explanation of this concession will come fully into view only as we investigate a different justification of plasticity (see the following section), but provisional diagnosis may be useful. Unger must be supposing that since increased plasticity allows a society more easily to vary its goals, it is beneficial as a matter of form —regardless of the goal. If this were true, plasticity would not really be a "goal" itself, or it would represent a different order of goal (a metagoal), immune to competition with other goals. Unger seems to endorse such a view when he writes: "The aim is not to exalt a discrete value at the cost of others" (FN , 354; cf. 351). Or again, he denies that the "proposed social ideal and its programmatic development . . . amount . . . to a choice of one among several . . . social ideals of the same kind " (CLSM , 105; emphasis added). These claims, taking plasticity or disentrenchment out of potential conflict with other values, are difficult to understand, however. It may be wondered in particular whether, lacking reference to some social ideal, plasticity remains a useful concept at all.

This question—whether plasticity must be, as it were, in the same space as other values—bears upon a noticeable difficulty in Unger's exposition of the instrumental argument. Unger repeatedly says that levels of entrenchment are "variable" (ST , 154; FN , 278), that "contexts can be placed on a spectrum of disentrenchment" (FN , 36, 306, 307, 325), that we can conceive of "degrees" of "closure to politics" (ST , 49), that social frameworks are higher or lower on a "scale" of disentrenchment (FN , 279), at "a particular level of disentrenchment" (FN , 311), that "societies differ in the extent to which they lay themselves open to self-revision" (CLSM , 25), and that we should "move toward greater revisability" (ST , 156; emphasis added). Yet we find no precise metric , no means of determining

plasticity to inform such rankings. To grasp differences of revisability, "it is enough," we even hear at one point, "to compare the liberal democracies . . . with the societies that preceded them" (CLSM , 25).

Does an instrumental justification of social plasticity require no more precision than this? Does an intuitive contrast between modern and premodern societies reliably tell us what to count as an increase in plasticity? The problem, it must be emphasized, concerns the basis upon which judgments of plasticity are to be made, not their potential vagueness. Suppose the present American Constitution were amended to require a re-convention every twenty-five years or so: would this represent for Unger an increase in plasticity, a framework more open to revision? Doubtless there are aims to which the altered framework would be less resistant, and it could be said that relative to these possible aims the framework is more open. But could a framework be more open to all aims? Is there absolute openness? In this example we might already spot a familiar interest—say "protecting considered political judgments from the whims of temporary majorities"—that would be frustrated. Could we really, without reference to our ideals or goals, follow Unger's injunction to create more open contexts? (Wouldn't this be like being told to arrange the library's books in three progressively more interesting ways?)^[47]

In fact, Unger admits the point: "The entrenchment of an institutional system can be measured by the extent to which it makes some transformations of itself harder to achieve than others" (FN , 229).^[48] This seems right, but to set up a metric by favoring specific transformations would seem to involve the very normative closure Unger hopes to avoid. Either there is absolute openness or our way of ranking contexts will be sensitive to our ideals and interests; Unger seems to benefit from leaving this tension unresolved. Perhaps this is why he says both that a "context must be placed on a spectrum of entrenchment and disentrenchment," at "a particular level of disentrenchment," as well as that

the arrangements that are more or less denaturalized than their modern-day counterparts do not represent extreme points on a finite spectrum of emancipation from false necessity; no such well-informed spectrum exists . (FN , 165)

Skepticism Revisited

We can bring these remarks together by returning to their skeptical starting point. Promised an "insubstantial" alternative to any "definite, institutional order" (FN , 572), we found reason to think that the alternative—revisability—1) is in tension with the loyalties valorized in Unger's reconstruction of contract law; 2) is not an unqualified good; 3) requires, in its application, substantive determinations eluding the formality the skeptic desires. Stepping back, however, substantive contestability is just the problem we might have expected for an instrumental justification of revisability. Unger's talk of "benefits" and "advantages" locates revisability in the same space as other possible goals and ends. Wouldn't those essaying to live out the requirements of a faith or commu-

nity, for example, prefer never to meet Superliberalism's "reckless" type?^[49] Our point, however, is not to challenge revisability but only to take note that its defense as one among other possible goals is neither self-evident nor especially skeptical. What does Zarathustra say? "Hitherto there have been a thousand goals, for there have been a thousand peoples. . . . The one goal is still lacking."^[50]

The Expressive Reading of Superliberalism

The One Goal

Alongside the promise of "benefits" and "advantages" we find, curiously, an entirely different rhetoric, one relating revisability to self-assertion not hypothetically but in virtue of the "meaning of self-assertion" (FN , 571; emphasis added). Self-revising contexts, on this account, "display" a "truth about human freedom" (CLSM , 104; ST , 23); are more "respectful" (ST , 8) of the "context-revising agents we really are"; "do justice to a being whose most remarkable quality is . . . the power to overcome and revise" (CLSM , 23), a "strange freedom from any given finite structure" (ST , 23); enable it to "live out" its "context-transcending identity" (FN , 592), to "experience . . . its true freedom" (CLSM , 105). Superliberalism has its "origin . . . in an awareness of the infinite quality of the personality . . . the power of the self eternally to transcend the limited . . . worlds that it constructs" (CLSM , 26, 105; FN , 12), in a "conception of human activity" (ST , 22–23; FN , 33), of "personality . . . as both dependent upon context and strengthened through context smashing" (FN , 515); and it "may be justified directly" as a requirement of an "image of personality" (CLSM , 27; cf. FN , 572), "grounded" in a "foundational view of human activity" (FN , 579, 347) and "described" by the "attempt to imagine the activities of a free being." Now the language of "displaying" ("respecting," "living out") a "truth" about "personality" closes down the logical space for an external connection between independent properties: revisability and self-assertion. On this account, context-revision is the self-expression of a being whose essence it is to be free in a special sense: perpetually overcoming its context. But how does this comport with Unger's parallel assertion that this being is always constrained by its context?

This "contradiction"—so Hegel calls it, thinking of the infinite, universal ego, the subject of legal rights, as also "this person . . . throughout whatever details you like"^[51] —has in fact a considerable tradition, which Unger is following. Leaving theology aside (at least for the moment), and starting with what seems furthest away, one could take up a familiar story about a theoretical crisis within empiricism, audible already in Berkeley's need to say that the perceiving subject is not itself a perception and fully manifest in Hume's despair over how to account in terms of "impressions and ideas" for the "fiction" of a self that has impressions, and acts on them, picking out regularities of impressions over time.^[52] A "fiction"

becomes—enter Kant—a "necessity" of thought, a condition presupposed by any experience. But if a constituting, legislating, structure-creating (note again the political resonances entering or not entering these words: this marks our problem ) subject must be presupposed on pain of failing to have a world present at all, it is apparently not to be encountered in the world. So Kant invites us to think of ourselves under a fateful doubling: as something in the world that we experience, known under the same conditions that any part of nature is known, and prescinding from this, as—what ? Kant calls it, fully aware how "problematically," "autonomous," "free." This is more than the invention in philosophy of what will become, through a political inflection, Unger's "contextuality." By a roughly traceable inheritance, the structure is refigured in all the modernist discourse on personality: in Marx's critique of alienation, in which "Man . . . is not immediately identical to any of his characterizations"; in Nietzsche's "as yet undetermined animal"; in Freud's unconscious, "the true psychical reality . . . as much unknown to us as the reality of the external world"; in Lacan's "real that eludes us . . . the encounter forever missed."^[53] What is the originality of such refigurations? Can we calculate their cultural accomplishments? If we could, in what discipline or language? More philosophy? Politics?

"Contextuality" is both a necessity and a fiction perhaps, for what could more suitably describe the necessary structure of personality than a necessity personality itself made up—as if such refigurations were not so much (true or false) descriptions of personality as its historical enactment in the theater of its own thought. Has this anything to do with politics? What can Unger's grounding of "revisability" in personality's self-enactment hope to accomplish? Eventually, this question will mean for us: Is Unger attempting a political translation of personality, as it has enacted itself since Kant (or perhaps since Descartes first staged a scene of thought in order to prove his existence)? For the moment, let's take up the question in the terms in which it appears in Unger's work and describe—under three headings—how his expressive rhetoric, according to which "the program may be justified directly [without other premises?] as an interpretation of what a particular social ideal and its corresponding image of personality require" (CLSM , 27), seems to contrast with and overcome the problems that instrumentalism faces.

Value Independence

The derivation of revisability through personality seems to go through independently of anything that may be considered advantageous or disadvantageous about it. For any calculation of "advantages" must refer to the particularity of the situated person, to his or her desires and needs. But—Hegel supplies a helpful word here—"in personality . . . everything which depends on particularity is . . . a matter of indifference ."^[54] Is not "indifference" to particularity (hence to any particular social goals, any particular sociality) just what appears to be needed if "revisability" is to be considered a "metagoal," immune to competition with other social goals? Unger's claim, puzzling on the instrumentalist

account, that his program does not "exalt a discrete value at the cost of others" (FN , 354), that it is not a "choice of one among several ideals of the same kind" (CLSM , 105), might then at least have the following sense: revisability as the expression of personality is not compromised by other goods because, abstracting personality from context, there is nothing (in particular) to which the person's good can refer. Each of the endlessly self-displacing contexts that Unger's program requires are matters of "indifference" to personality as such .

Skepticism Incorporated

Can we see also how revisability as an expression of personality might be "insubstantial" enough to please the value-skeptic? The answer would appear to be not that, starting from the contestability of all values, one is "required" to affirm revisability, but that in affirming any values at all one presupposes personality (and therefore revisability) as their ground. Value presupposes evaluation. Then even though there are "a thousand goals," they come under the unity of a single form: "I will." It is this which Unger's principle of revisability intends to affirm—not a particular substantive value but value's "insubstantial" ground, the will that evaluates, decides, revises. Unger proposes, under the threat of nihilism, the empowerment of power itself.

Rights

Finally, revisability as an expression of personality complements Unger's articulation of the "destabilization rights" that, he says, "reveal the ruling intentions of the entire program" (CLSM , 43). We might bring out the connection by recalling an argument John Mackie made for a "right-based moral theory":

A plausible goal, or good for man . . . would be in the category of activity . It could not be just an end, a possession, a termination of pursuit. But Aristotle went wrong in thinking that moral philosophy could determine that a particular sort of activity constitutes the good of man in general. . . . People differ radically about the kinds of life that they choose to pursue. Even this way of putting it is misleading: in general people do not and cannot make an overall choice of a total plan of life. . . . I suggest that if we set out to formulate a goal-based moral theory, but in identifying the goal try to take adequate account of these factors, namely that the 'goal' must belong to the category of activity, that there is not one goal but indefinitely many diverse goals, and that they are the objects of progressive (not once for all or conclusive) choices, then our theory will change insensibly into a right-based one. We shall have to take as central the right of persons progressively to choose how they shall live.^[55]

Curiously, the skeletal structure of Unger's argument is present here in the prose of Oxford philosophy. Activity, the form of every goal, is that which the positing of any particular goal presupposes. Then under the threat of skepticism about the Good (Aristotelian biology won't do) we "insensibly" retreat from substance to form, from goals to the volitional activity that is their ground.^[56] Yet Mackie's argument—because it sets out initially "to formulate a goal-based theory"—goes wrong in exactly the way that the instrumental interpretation of Superliberalism went wrong. If the importance of having rights derives from the various goals we want to pursue, it seems inevitable to wonder whether we might not want to

pursue some goals that outweigh or conflict with the higher-order "goal" of protecting choice. (Might not some of the collective social activities that confer value upon choice be unsupportable within the regime of liberty rights Mackie seems to have in mind?)^[57] These questions seem inescapable whenever a so-called "right-based theory" is actually derived from a theory of goals. The change, as Mackie says, is "insensible," but it leaves a formal trace: the pressure to place inverted commas midway through around "goal" (thus suspending or negating its sense) concentrates, in a single gesture, the entire problem.

To preserve the centrality of rights, no less than to defend a genuinely right-based moral theory, we seem to require a goal-independent (noninstrumental) foundation for rights—a transcendental property of personality, for example.^[58] Rights to the disruption of social structure would then apparently be the "direct" expression of the person's capacity to "transcend" the limited, constructed world (CLSM , 26), and hence of its freedom from the will of the socially other. All persons being equal (since there is no particularity about them as persons to which their inequality might refer), the regime of right is a structure of interaction between free and equal beings, marking out the space in which personality can be pursued and expanded. As liberals we have heard this before. But Superliberalism radicalizes the formula by rigorously tracking it independently of value to its ground in personality or will.^[59] Recalling the utopian antilegalism of Knowledge and Politics , it might seem that Unger's new nihilism of ideals and values organizes a positive reversal.

Fracturing Politics

When one moves toward a goal it seems impossible that "goal-lessness as such" is the principle of our faith.
—Nietzsche^[60]

Having come this far, having questioned instrumentalism as a justification for Superliberalism but also noted Unger's desire to fund the "incorporation" of skepticism from a different account; and having presented his difficult claim not to "exalt a discrete value at the cost of others" (FN , 354), not to choose "one among several ideals of the same kind" (CLSM , 105), as the wish to ground the program of revisability in the self-enactment of personality as the prior condition of evaluation and political choice—can we not already begin to discern an aporia lying ahead? For although, on the expressive interpretation, context-transcendence represents a fundamental fact about us, it is also supposed to have prescriptive force, directing us to devise institutions "more fully respectful of context-revising freedom" (ST , 8); though basic and unavoidable it is something we are to "pursue" (CLSM , 94).^[61] How does the ontological ground of all value become, for Unger, the definitive political value?

Admittedly, the sense of this question is not very clear. But this unclarity would seem to constitute a problem for Unger—not for the question. Unger himself appears to notice the trouble when he repeatedly poses a question about whether it is legitimate for him to be "attributing normative force to conceptions of personality"—as if the problem could be grasped in the form of the so-called gap between fact and value.^[62] Unger writes apparently for readers who already understand what this distinction is about (a group in which I cannot include myself), and he claims to be doing philosophical work on it. But even if he could help me to grasp the distinction better than I do, a dilemma can now be seen that is capable of being formulated independently of it. If revisability is proffered as a goal we should choose, it would be in the same space as other social goals and may be qualified or outweighed by them. Revisability would be open to the question: What's so good about a social world "pitilessly" open to self-disruption? (Does one need to look "indifferently" upon particular realizations of sociality to feel what is so good about it?) An answer, if one could be given, would be unskeptical, to say the least. On the other hand, if a privileged place is claimed for revisability as an expression of a "truth" about personality, then what it means to choose it (as if it were an external political end) becomes obscure—or, what it is to express it becomes essentially indeterminate since "it," this "truth" about personality, is presupposed in any self-expression at all. How can we translate something always presupposed, the ground of all goals in personality, into an external political goal? Or is Unger asking us to take as our political goal simply the "insubstantial" form of all choice? The poles of this dilemma are apparently not fact and value but the instrumental and expressive interpretations of Superliberalism itself.

Supposing the problem were one of fact and value, of moving from a "description" of personality to a social "prescription," why couldn't it be remedied through an innocuous "ought" premise, one to which Unger might assume our assent? Absolved of normative transgression, the justification of Superliberalism based on a "truth" about personality would then go like this:

1. We ought to create a social world in which we are less alien.

2. A social world that expresses the truth about us is one in which we are less alien.

3. Context-transcendence is the truth about us.

4. A social world expresses our context-transcendence in being completely revisable.

5. We ought to create a social world that is completely revisable ("a social world less alien to a self that can always violate the generative rules of its own mental and social constructs and put other rules and other constructs in their place"; CLSM , 41).

However silly this may sound, it may help us to feel something of the misdescription or displacement occurring in Unger's worry about "attributing normative force to conceptions of personality." As I see it, he should also worry about something else: What is it for a social world, hence for a politics, to express or to articulate personality? What is it for personality to express itself ? Unger's answer, represented here by premise 4, indicates only the place of an answer. For no

matter how one might understand this premise, and however Unger proposes to resolve the meaning of "revisability" through a set of practical political proposals, his argument requires that no answer to this question be represented under the ordinary concept of a goal or end —something independent, something chosen; otherwise personality could no longer recognize something of its own. Nor does it help, following Mackie (who perhaps senses the aporia), to write "goal"; the quotation marks only mark the problem as a sort of suspension or irresolution of sense.

In this difficulty of resolving itself except as the will to revise itself, Superliberalism displays its indebtedness to the Kantian problematic of "autonomy"—the will that takes no end outside itself but wills itself as its own end—and to Nietzsche's inheritance of this tradition as "self-overcoming."^[63] If Unger's attempt to translate this modernist problematic politically as "revisability" straddles a gap, it is between content and form, as if "the form of volition in general"—so Kant describes autonomy, self-determination—is to determine particular volitions, as if political will must take as its content only the formal procedures of political choice. In a similar vein, Unger writes: "Short of some transcendent reality [but who writes these words meaning to engage a viable alternative], the only thing to which personality can give final authority is itself" (CLSM , 104). If "personality" is now elected by Unger to the vacant position of God, it remains only for it to "transcend" the desacralized, "imprisoning" contexts it inhabits for it to acquire as its political vocation the creation of a social world out of its own endless (perpetual, goal-less) willing. Calling for a society "broken open to everyone's will" (FN , 404), one that "would provide constant occasions to disrupt any fixed structure of power and coordination in social life" (CLSM , 30–31), Unger arrives at the conclusion that "the only structure that can be allowed to subsist is one that offers the fewest obstacles to [the] principle of pitiless recombination" (PP , 208). This privileged "principle" determines a pure proceduralism as the combination of two theses—the first tolerating nothing unwilled, the second marking an internal limit, a residuum of the dead, practico-inert: for if "recombination" is to be political recombination, the present political "I will" does not will the body politic but presupposes it. Unger's endlessly revisable "context" is thus ultimately the thought of a social constitution that wills the form (the procedures, the technology) of decision making as its only affirmative content.^[64]

Hasn't Hegel already described this Superliberalism for us: the political will that, rejecting the given, positive, and entrenched, flies "from every content as from a restriction"?^[65] And hasn't he already rigorously demonstrated within all such political translations of "the metaphysics of subjectivity" an essential nonarticulation, hence, in his historical moment, a terror of perpetual destruction?^[66] To ask these questions is not to say that Unger is Robespierre or that—one hears this said amidst the polemics that surround Critical Legal Studies—the application of Unger's program would be destructive or fanatical. On the contrary, it is

to question whether there is any such thing as an "application" of the program that remains faithful or responsible to Unger's most general theoretical account of it. It is to ask Unger: Does it matter politically whether, as in Knowledge and Politics , you take your transcendence seriously or whether, as in Politics , you present it as (some might call this antifoundationalism!) the self-enactment through endless "context-revision" of human personality?^[67] Can we calculate the difference? And it is to suggest that we might read Unger's theory as a rigorous formalism : not as a politics but as a grammar that retraces the seam, circles the rim, of the political as such . On this reading, what begins as an "overriding concern" to criticize legal formalism in the name of political conflict (CLSM , 1–14) ironically ends up grasping the political only through the resource of formalism itself, a political formalism that renders—and this is just Unger's complaint about law—all content indeterminate.^[68]

The social will: an indispensable, self-attuned "we" that turns together in language, that calls out of bounds, excludes, and withdraws from the circulation of the polis . A law—or call it a legitimacy—prior to all legislation, indeed, to all states and all constitutions written by a people who repronounce and reenact themselves as "we." (Can we question our politics, our constitution, in the name of this law?) A prior volition that structures and organizes all moments of discourse, even this one now in which we venture the question of our political constitution. Are we—and who are "we"?—making any sense ? Or do we, in encountering these words through the movement of a dialectic, misencounter ourselves? What is the therapy for this misencounter?^[69]

What happens when we attempt, as Unger attempts, to conceive or to organize this "we" politically ? Is it a displacement of politics that only announces the place of politics, a politics marked today by a confused ("liberal"?) pluralism of struggles, along with the conflicting rhetorics—instrumental and expressive—that Unger's work repeats without resolving? Or could we think of this "we" not just as a prior sociality but as the promise of a social life to come, a promise as yet unknown in our tradition? Perhaps this is the sort of question, more like a prayer than a question, about which some "must remain silent" while others must permit themselves to speak, pronouncing this promise no longer as God but as the self-overcoming of man: "O my friend, man is something that must be overcome." (Are these necessities, these commands—silence and overcoming—different or the same? If the same, as commands or as silences?)^[70] If these questions begin to indicate some of the obscurity of the problem of how modernist philosophical writing can be political , in Unger, for whom modernism means "it's all politics," they imply a fracture of his title Politics in two: on the one hand, a set of goal-based, practical proposals; on the other hand, the elaboration of personality as their receding theoretical ground. If in Knowledge and Politics these hands come

together as prayer, in Politics a severe split settles within and divides Unger's concept of the political itself.^[71]

Theory and practice thus come apart in Unger's work. But they do so in one additional way as well—perhaps finally the only way, I am inclined to think, that theoretically and practically matters. I am referring to the practice of writing (hence of relationship to others) through which Unger's work enacts itself. Although its ideals are solidarity and sympathy, Unger's writing is notably impatient and dismissive, obliterating where it should criticize or respond. It speaks of the radical potential in personal encounter, but it overpowers the other by netting him in the many "theses" and "alternative views" that rigidly structure its intellectual space. Wanting plasticity, it freezes and fixates—this is common enough in law journal prose, but one might have expected something different from the champion of disentrenchment, fluidity, experimentation, self-disruption, trashing the script , and negative capability . Radical in his ideals, Unger seems, by his own standards, reactionary in the practice of his argument—a practice that may well have its place, but not as a promising model for civic relations. But why would this be? Is it enough to say that Unger is in a political hurry? In my puzzlement, I find help in a passage from the end of his address to the Conference on Critical Legal Studies:

Finally, there is the disparity between our intentions and the archaic social form that they assume: a joint endeavor undertaken by discontented, factious intellectuals in the high style of nineteenth-century bourgeois radicalism. For all who participate in such an undertaking, the disharmony between intent and presence must be a cause of rage. (CLSM , 118)

Unger is on to something important here, but why do those for whom this passage speaks continue to take "factious intellectuals" or "the high style of nineteenth-century bourgeois radicalism" as their exemplar of what criticism should sound like? What is responsible for the entrenchment of this "archaic social form"? Perhaps—reversing the connection—it is an original rage, one that is certainly to be felt in Unger's prose, that must be the cause, not just the effect, of the disparity between intent and presence. Since I share some of his hopes I know this rage. Since the "I" that thinks is the "I" that is encountered in the world, I know this gap between intent and presence. But all this puts me in mind of the difficult irony to which our political constitution, in its very conception , must be subject: for it seems at times that it is only by encountering—or do I mean misencountering?—itself in a dead and sedimented form (call it "rhetoric") that our will registers under its own name.

Enlightenment Epistemology and the Laws of Authorship in Revolutionary France, 1777–1793

Carla Hesse

What is the relationship between authorship and the law, between the central mechanism of representation and the dominant system of discipline in the modern period? Though the history of modern authorship has long been a subject of interest to historians and literary critics, no single essay has been more influential in current investigations and discussions of it than Michel Foucault's "What Is an Author?"^[1] This essay has formed part of the poststructuralist effort to problematize the modern concept of the author as the "single center of truth to which all representation refers," that is, as the unique originator and sole determinator of his own meanings, by challenging the epistemological ground upon which this notion of "the author" stands.^[2] However, historical investigation of how the most important political event of the modern era, the French Revolution of 1789–99, reshaped the legal and political identity of the author suggests that the meaning of modern authorship has yet to be fully explored.

Foucault writes: "The coming into being of the notion of the 'author' constitutes a privileged moment of individualization in the history of ideas. . . . It would be worth examining how the author became individualized in a culture like ours." And, he continues, "Since the eighteenth century, the author has played the role of the regulator of the fictive, a role quite characteristic of our era . . . of individualism and private property."^[3] Foucault thus establishes a parallelism. The relation between the "author" and the "text," he suggests, emerged historically as the cultural incarnation of a new axis in sociopolitical discourse: the inviolable relation between the rights-bearing individual and private property. Thus, according to Foucault, this "privileged moment of individualization" was also characterized as a moment of privatization of knowledge claims as property, in which the individual author came to be the exclusive principle by which meanings are composed, manipulated, and determined, or, to use his word, "regulated." Because of the historical nature of his claims, Foucault's characterization of the appearance and function of the modern concept of the author has provided poststructuralists a target for their theoretical critique. These historical claims bear further scrutiny.

Two recent historical investigations have responded to Foucault's invitation

to "examin[e] how the author became individualized." Both Martha Woodmansee's study of the development of the concept of the author in eighteenth-century Germany and Mark Rose's exploration of the legal debates on authorship and literary property in eighteenth-century England tend to confirm Foucault's thesis. They conclude that the concept of the individual author as proprietor of his work emerged in the eighteenth century as a result of the commercialization of the book trade. Thus Woodmansee writes:

The "author" in the modern sense . . . is the product of the rise in the eighteenth century of a new group of individuals: writers who sought to earn their livelihood from the sale of their writing to a new and rapidly expanding reading public.^[4]

Similarly, Rose observes:

The emergence of the mass market for books, the valorization of original genius, and the development of the Lockean discourse of possessive individualism . . . occurred in the same period as the long legal and commercial struggle over copyright. Indeed, it was in the course of that struggle under the particular pressures of the requirements of legal argumentation that the blending of Lockean discourse and the aesthetic discourse of originality occurred and the modern conception of the author as proprietor was formed.^[5]

For both these scholars, the advent of the modern cultural system can be marked by the translation of these new bourgeois socio-economic relations and cultural values into laws of copyright (in England in 1709 and in Prussia in 1794), that is, by their institutionalization as a system of legal regulation.

Does the legal history of French authorship and copyright further confirm the picture presented by Foucault and his successors? Discussions of "authors' rights" among French and American legal theorists would tend to answer this question in the affirmative. Citing the majesterial work of the French legal philosopher Henri Desbois, a contemporary specialist in Franco-American copyright law has recently summarized the prevailing legal view of the French author in the following manner: "The French law, allegedly imbued from its revolutionary inception with natural rights philosophy, is said to enshrine the author: exclusive rights flow from one's (preferred) status as a creator."^[6]

Yet, despite the theoretical assertions of literary critics and legal theorists, we know surprisingly little about the history of authorship and publishing during the period of the French Revolution, the particular moment in which the modern notion of the author crystallized into a legal identity in France. While there have been several important studies of individual publishing enterprises during the Revolution, recent historical interest in publishing and authorship has focused almost exclusively on the prerevolutionary period.^[7] The multivolume Histoire de l'édition française (1983–), for example, devotes fewer than thirty pages to the revolutionary period.^[8] More surprising, there has been to date no single volume treating the particular question of authorship during the Revolution.

Like England and Germany, eighteenth-century France witnessed an extraor-

dinary expansion of the commercial book trade, which opened up unprecedented opportunities for writers and publishers and which posed unprecedented challenges to public authorities.^[9] But there were crucial differences that set the history of authorship in France apart from the rest of Europe in the eighteenth century: first, the presence of an absolute monarchy of unparalleled size and strength, and second, the explosion of the first great democratic revolution. The French Revolution offered the occasion for an unusually explicit and fascinating debate in France about the identity and role of the author in modern life, a debate whose legal resolution continues to provide the foundation for French copyright law. A close examination of this debate and its resolution throws into question some of the historical claims of Foucault and his successors, and thus their characterization of modern cultural life as well.

The Invention of the Author

The French Revolution did not invent the legal notion of the author. The first legal recognition of the author in France can be dated quite precisely to the six royal decrees on the book trade of 30 August 1777.^[10] Prior to this date there was no formal legal recognition of the author or his relation to his text. The decrees of 1777 represented the culmination of a fifty-year debate between the Crown and the Paris Corporation of Printers and Publishers concerning the nature and duration of royal "privileges" on texts.

According to the royal Code de la librairie , established to regulate the Parisian publishing world in 1723 and extended to the entire nation in 1744, there was no "property" in ideas or texts nor any legal recognition of authors.^[11] Employing the doctrine of divine revelation and a long tradition of medieval thought, the king's ministers argued that ideas were a gift from God, revealed through the writer. They were not owned by the author and could not be sold by him. The power to determine what was truly God's knowledge, and who could enjoy the "privilege" of its "enjoyment" (literally, "jouissance"), belonged not to the author but to God's first representative on earth, the king, and by extension his administration.^[12] Put another way, the king alone, by his "grace" and "pleasure," had the exclusive power to determine what would be permitted to be published, by whom, and for what period of time. Publishers were required to submit all books for royal censorship prior to publication in order to receive from the king a "privilege," which was at once an official approbation of the work, a permission to publish, and a kind of copyright, in that a "privilege" assured its holder a legal exclusivity on the publication of the work.

Authors were not allowed to publish their manuscripts. Only members of the royal guild of publishers and printers were permitted to engage in the printed publication of what was royally deemed to be God's knowledge. Thus, despite the

increasing power of "the author" as a new figure in the sociocultural landscape of late seventeenth- and early eighteenth-century France, the author had yet to receive legal consecration of his relation to the texts he produced. From a strictly legal point of view there was no authorial claim upon a text. Authors sold their manuscripts to publishers who then submitted them to the Crown for "privileges." Authors without independent fortunes survived by royal or aristocratic patronage rather than by commercial profits from the publication and sale of their works. The profits in "privileges" were the legal monopoly of the guild publishers.^[13] Because there was no legal recognition of authorship, royal "privileges" to publish made no legal distinction between works by living or dead authors, anonymous, collective, or unattributable works.

While in principle the Crown could revoke or extend "privileges" at will, by the end of the seventeenth century the centralizing and corporatistic royal administration had in practice encouraged the consolidation of a monopoly on the lion's share of the literary inheritance of France by the Paris Publishers' and Printers' Guild.^[14] It did this by conferring extensive "privileges" to publish both individual texts and whole areas of knowledge to particular Paris publishers and then renewing these "privileges" automatically over generations. One of the most notorious examples of this practice was the Crown's decision in 1670 to grant an exclusive "privilege" for fifty years to an editor of the Old and New Testaments.^[15]

To protect their monopoly against the protests of excluded publishers, the Paris publishers began in the late seventeenth century to evolve their own interpretation of the meaning of the "literary privilege." In 1726, the Publishers' Guild commissioned the jurist Louis d'Héricourt to write a legal brief arguing that a "privilege" was not a royal "grace" to be conferred or revoked at the king's pleasure but rather a royal confirmation of an anterior property right. Invoking John Locke's notion of the origins of property in appropriation, d'Héricourt argued that the property in ideas is derived from labor: "It is the fruits of one's own labor, which one should have the freedom to dispose of at one's will."^[16] According to the guild, then, "privileges" were nothing more than the legal recognition of an existing property right founded in the author's labor and transmitted to the publisher through a contract. Ironically, therefore, the argument that ideas were the property of the individual author was first advanced in defense of the monopoly of the Paris Publishers' Guild on texts whose authors were long since dead.^[17]

By the middle of the eighteenth century the issue of literary "privileges" became caught up in a more general movement by enlightened royal officials to deregulate commercial life, including commerce in ideas, by dismantling the corporate monopolies created in the seventeenth century by Louis XIV's minister, Jean-Baptiste Colbert. Heeding the advice and counsel of successive officers of the Royal Administration of the Book Trade, the Crown announced in 1777 an important shift in its cultural policy, which resulted in the first legal recognition of the author in France.^[18] The king's Council of State issued a series of decrees

intended definitively to refute the Guild's interpretation of literary "privileges" as confirmations of property claims and to decentralize the publishing world by breaking up the Parisian monopoly on the nation's cultural inheritance. It achieved these ends by a double-edged stroke that revealed its interest in encouraging and empowering both the individual author and the public at large at the expense of the corporate monopoly of the Paris Publishers' and Printers' Guild. In the 1777 decrees the king's council made its interpretation of "privilege" explicitly clear:

His majoesty has recognized that a privilege for a text is a grace founded in Justice. . . . The perfection of the work requires that the publisher be allowed to enjoy this exclusive claim during the lifetime of the author . . . but to grant a longer term than this, would be to convert the enjoyment of a grace into a property right.^[19]

The Crown thus reaffirmed the absolutist interpretation of royal law as an emanation of the king's grace alone.

The decrees created two categories of literary "privilege." Recognizing the author for the first time, the decree created "privilèges d'auteur" to be granted to authors in recompense for their labor and to be held by them and their heirs in perpetuity , unless sold to a third party. Authors were for the first time permitted, indeed encouraged, to hold on to their manuscripts and to engage in publication, rather than to sell their manuscripts to publishers. Publishers' "privileges" ("privilèges en librairie"), by contrast, were to be limited to the lifetime of the author and nonrenewable. After the author's death, these texts returned to the "public domain" to be enjoyed by any royally licensed publisher, with the king's permission.

The Crown's decrees thus signaled a new effort to consolidate the power to disseminate ideas in individual authors rather than in corporate publishers. This end was to be achieved, not through the recognition of property rights, but by a revision of the system of privileges. The first aim of the Crown was to individualize knowledge by creating the author as a privileged site of regulation—in both senses of the term, politically and legally. It was also the aim of the Crown to deprivatize texts whose authors were dead, to remove them from the private (property) claims of the Publishers' Guild.

Foucault's statement that the first legal recognition of the author occurred in the eighteenth century can be held to be true in the French case. But in France the author was a creation of the absolutist police state, not the liberal bourgeois revolution. Foucault is also correct in asserting that "the author" became the absolute principle by which fictions would be regulated. The "author's privilege" was granted from the king in perpetuity, thus permitting the author to control his ideas forever, unless he alienated them. Like the king's will, the author's will was to operate not only during his lifetime but indefinitely over time as well, as a patrimony or a family trust. The author was thus conceived as a little mirror of

the king, the regulator of his fictions: they only fell back into the king's domain if and when they were wilfully alienated (like any other feudal tenure).

The creation of the author by the absolutist state was the product of a political initiative within the royal administration rather than a result of commercial protest, and it had the explicit purpose of consolidating state control over the form, content, and means of disseminating knowledge by removing the publisher as intermediary between the state and the author. Now the author would be directly accountable to the Crown and its laws for the publication of knowledge.

The Enlightenment Engages

The eighteenth-century dispute between the Paris Publishers' and Printers' Guild and the Royal Administration that led to the formulation of the royal decrees of 1777 drew key Englightenment figures directly into the debate on literary property. As a consequence, the middle of the century witnessed the appearance of several systematic efforts to reground the discussion of the origins and nature of claims upon knowledge in the terms of Enlightenment epistemology. Two distinct positions emerged within enlightened circles.

In 1763, the philosopher-novelist Denis Diderot was hired by André-François LeBreton, the chief officer of the Paris Publishers' and Printers' Guild and the publisher of the Encyclopédie , to write a treatise to be presented to the new director of the Royal Administration of the Book Trade, Antoine-Raymond-Jean-Gaulbert-Gabriel de Sartine, defending the guild's view of their "privileges" as a form of property. In his Lettre historique et politique sur le commerce de la librairie , Diderot argued that ideas are the most inviolable form of property because they spring directly from the individual mind, because they are a creation of the mind, indeed the very substance of the mind, the means by which it constitutes itself. Thus he writes:

What form of wealth could belong to a man, if not a work of the mind . . . if not his own thoughts . . . the most precious part of himself, that will never perish, that will immortalize him? What comparison could there be between a man, the very substance of man, his soul, and a field, a tree, a vine, that nature has offered in the beginning equally to all, and that an individual has only appropriated through cultivating it?^[20]

Diderot made an even more individualistic argument for authorial claims than had the jurist d'Héricourt in 1726. Though Diderot's epistemological stance, as well as his ideas on aesthetics, are rife with complexities, in this 1763 Lettre he depicted ideas as emerging sui generis from the mind rather than as a result of the labor of combining sensations emanating from nature. In contrast to d'Héricourt, Diderot argued that property in land is merely a social claim, based upon appropriation through labor and thus susceptible to social mediation. Ideas, originating

in the individual mind, and not acquired through appropriation or labor, are the most natural and inviolable form of property.^[21]

Diderot thus argued that privileges for the texts of living authors should be recognized as legal confirmations of a perpetual property right. Similarly, works whose private lineage could no longer be traced should also be considered the perpetual property of the privilege holder, justified by the theory of "right of first use." In contrast, then, to his general condemnation of commercial privileges, Diderot made an exception for those conferred upon texts, arguing that the protection of exclusive claims, rather than free-market competition, was the best guarantee of the progress of knowledge and the spread of enlightenment.^[22]

The legal implications of Diderot's line of reasoning were cogently formulated by the lawyer-journalist Simon-Nicolas-Henri Linguet in his Mémoire sur les propriétés et privilèges exclusifs de la librairie, présenté en 1774 :

What is a literary privilege? It is a recognition made by public authority of the property of the author or of those to whom he has ceded it. It is the literary equivalent of a notorial act which . . . assures the rights of citizens. . . . The privilege is a seal that guarantees peaceful enjoyment; but it is not the source of that enjoyment. . . . A privilege grants nothing to the author, it only ensures protection.^[23]

These arguments, both philosophical and legal, advanced by two key figures of the French Enlightenment are consistent with, indeed they would seem further to confirm, the assertions of Foucault and others that the modern author was first conceived as individual property owner.

There was, however, a second Enlightenment position on the issue of literary "privileges" and property, articulated in France by the marquis de Condorcet in 1776, on the eve of the new Royal Regulations of the Book Trade, in a pamphlet entitled Fragments sur la liberté de la presse . Although the original context of the composition of this pamphlet remains obscure, it undoubtedly formed part of Condorcet's efforts to assist the French Minister of Finance, Anne-Robert-Jacques Turgot, in his efforts to liberalize French commerce by suppressing the monopolies of the royal guilds.^[24] The pamphlet was clearly intended to present a complete revision of the current royal code organizing and regulating the book trade.^[25] Like all Old Regime codes on the book trade, Condorcet's pamphlet treats the issues of censorship, liability, commercial regulation, protection, and policing together, as interconnected elements of a single regulatory vision. The lion's share of the document was devoted to a refutation of prepublication censorship and the commercial monopolies of the guild. He proposed replacing these regulations with liberal laws on sedition and libel and freedom of commerce in the printing, publishing, and bookselling trades. Under such laws, the "author of the publication," not the author of the work, would be held legally accountable and liable for its public consequences.^[26]

The issue of accountability led Condorcet logically to a sustained discussion

of the problem of authorial claims and literary "privileges" as well. Drawing upon arguments formulated within sensationalist epistemology, Condorcet attacked both the royal theory of literary "privileges" and the theories of authorial property rights advanced by Diderot and the lawyers for the Publishers' and Printers' Guild.^[27] He asserted that there was, formally speaking, no property in ideas. Thus he wrote:

There can be no relationship between property in ideas and that in a field, which can serve only one man. [Literary property] is not a property derived from the natural order and defended by social force, it is a property founded in society itself. It is not a true right, it is a privilege.^[28]

Unlike a piece of land, an idea can be discovered, inhabited, and used by an infinite number of people at the same time. Ideas are not the creation of individual minds, be it through revelation, appropriation, or cognition. Rather, they inhere in nature, and hence are equally and simultaneously accessible through the senses to all. They therefore can belong to no single individual.

Further, Condorcet rejected the social value of any individual claims on ideas. Since true knowledge was objective, particular individual claims on ideas could consecrate and protect nothing more than the style, the individual form, rather than the substance of an idea. Far from viewing originality as the hallmark of the modern bourgeois author, Condorcet condemned particularities of style as attributes of aristocratic culture. He argued that any "privileges" that might be derived from these attributes should be abolished; style distorts nature's truths and thus to privilege it encouraged the production of pleasant fictions and personal gain rather than the pursuit of useful knowledge and the public good.

It is thus uniquely for expressions, for phrases, that privileges exist. It is not for the substance of things [les choses ], for ideas; it is for words [les mots ], for the name of the author.^[29]

Further, legal privileges derived from individual style inhibit the spread of ideas by restricting access to them.

Privileges of this sort, like all others, are inconveniences that diminish activity by concentrating it in a small number of hands. . . . They are neither necessary, nor useful, and as we have seen, they are unjust.^[30]

Condorcet argued that there should be no individual claims upon knowledge as either property or privilege. He imagined an authorless world of free manipulation and circulation of information and ideas. These observations led him to conclude that a commercial publishing industry which sold ideas rather than authors, substance rather than style, could be organized according to the principles of periodical rather than book publishing, as was the publication of the proceedings of the Académie des sciences or the Encyclopédie : through reader subscriptions to a genre of knowledge rather than through the marketing of unique works.^[31]

The debate between Condorcet and Diderot played out a tension inherent in Englightenment epistemology (in Locke's Essay on Human Understanding itself) concerning the origins of ideas and hence the kinds of claims that could be made upon them. Did knowledge inhere in the world or in the mind? To what extent was it discovered and to what extent invented? Condorcet argued that knowledge was objective, inhering in nature, and thus fundamentally social in character, belonging to all. Diderot viewed ideas as inherently subjective and individual, originating in the individual mind and thus constituting the most inviolable form of private property.

Thus Foucault's thesis needs to be revised: eighteenth-century France witnessed the emergence not of one modern position on the nature of the author and his relation to the text (i.e., the property-bearing individual) but rather of a modern tension between Diderot's conception of the author as the original creator and hence inviolable proprietor of his works and Condorcet's depiction of the ideal author as the passive midwife to the disclosure of objective knowledge.^[32]

The Revolution

Between 4 and 11 August 1789, the newly constituted National Assembly abolished all "privileges" of the Old Regime, and a few weeks later, on 26 August, declared "freedom of the press" to be an inviolable right of man.^[33] As I have demonstrated elsewhere, the years 1789 to 1791 witnessed a systematic effort to liberate thought and spread enlightenment by dismantling the entire infrastructure of licit publishing under the Old Regime.^[34] The Royal Administration of the Book Trade, and its censors and inspectors, were formally suppressed in August 1790.^[35] Then a decree of March 1791 ended the commercial monopoly of the Publishers' and Printers' Guild.^[36] The revolutionaries wanted to free the minds of citizens from censorship and to liberate the means of spreading and exchanging thoughts—literally, the presses and bookshops.

But upon what basis was republican publishing to be founded? When the Revolution overthrew the absolutist monarchy, was it to consecrate or dethrone the "absolute author"? In formulating a resolution regarding the legal status of authors and publishers in relation to the texts they published, the revolutionaries could not escape the tension between the competing visions of modern cultural life presented by Diderot and Condorcet.

The First Initiative (1789–1791)

The revolutionary debate unfolded in two distinct stages. The first legislative effort to define the legal standing of claims upon ideas appeared as a subsection of a comprehensive law on sedition and libel that was presented to the National Assembly by Emmanuel Sieyès on behalf of the Committee on the Constitution, on 20 January 1790.^[37] The law was born out of

a convergence of the commercial interests of book publishers and the political imperatives of the National Assembly.

With the collapse of the royal institutions that regulated the printed word and the constitutional challenge to the notion of literary "privileges," pirate publishing ventures flourished. One after another, the publishing elites of the Old Regime were driven into bankruptcy by the collapse of their monopoly on the printed word.^[38] Thus the Paris publisher Jean-François Royer lamented in an avis of 1789 that "pirate editions are one of the principal reasons for the losses in the publishing business."^[39] And so too a Parisian police commissioner observed: "There is no author who will consecrate his efforts to the instruction of his century if pirating is made legal."^[40] Observations like these soon percolated upward into the discussions of the National Assembly. Publishers sent testimony that they were being driven to produce seditious and libelous material in order to stay afloat. Thus in the session of 12 January 1790 the deputy Charles de Lameth testified that "a Paris publisher has just reported to me that, unable to make any profit printing good books, he is being forced to go into the business of printing and selling libelous matter. . . . There are few printers in Paris who can afford not to."^[41] Hoping to gain the ear of the assembly, publishers thus linked the economic issue of literary property and its protection to the political questions of sedition, libel, and authorial accountability.

The National Assembly itself was in the throes of a conservative backlash against the collapse of all regulation of the printed word. In the face of a flood of anonymous, libelous, and seditious pamphlet literature, the assembly heard repeated demands for laws requiring authors to sign published works and holding authors accountable for their publications. Thus the economic complaints from publishers converged with the political outcry from Jacques-André d'Emeri for "a law on the freedom of the press" to outlaw seditious publications, or from the deputy Louis-Marie, marquis d'Estourmel for a law requiring authors, publishers, and printers to sign, and thus lay claim to, the works they produced as a means of holding them accountable.^[42] As a consequence of this agitation, the assembly moved that "the Committee on the Constitution will be charged to present forthwith, a proposal for a law regulating the freedom of the press."^[43] Eight days later, on 20 January 1790, a proposal for a law on sedition, libel, and literary property was presented to the assembly by Sieyès on behalf of the committee.^[44]

The initiation of the Sieyès proposal was part of the effort by moderates in Paris and in the assembly to restore order and check the radicalization of the Revolution in the wake of the popular revolt that swept the cities and countryside after the fall of the Bastille.^[45] By the end of 1789, Sieyès was meeting with the group of moderates who had splintered off from the Jacobin Club, including Condorcet, Lafayette, the duc de la Rochefoucauld, the duc de Liancourt, and Dupont de Nemours. In early January 1790 these men officially founded the

Society of 1789.^[46] There can be little doubt that this group, and in particular Condorcet, played a crucial role in drafting the National Assembly's first legislative effort to regulate the printed word.

Significantly, the proposal was publicly attributed to Condorcet as well as Sieyès.^[47] There is good evidence to support this attribution. In both form and content the proposal presented by Sieyès in 1790 bears a striking resemblance to the Fragments sur la liberté de la press that Condorcet had composed in 1776.^[48] Indeed, close comparison of the two texts suggests that the Sieyès projet was drafted directly from Condorcet's pamphlet. The two documents share virtually the same organizational structure, and the substantive parallels between the two texts are equally striking. Ironically, then, the pamphlet Condorcet had initially circulated as a radical indictment of the inquisitorial institutions of the Old Regime he in turn revived in 1790 to serve as a conservative check on the flood of ideas unleashed by the collapse of those very institutions, by proposing a law that would hold authors, publishers, and printers legally accountable for their publications.

This conservative turn is further disclosed in the one substantive change made in Condorcet's earlier pamphlet as reworked for the proposal presented by Sieyès. This was the section on "privileges" and literary property. In 1776 Condorcet had argued that ideas were social rather than individual in origin and that as a consequence they could not be considered a form of private property to be protected as a natural right. Further, he had argued that "privileges," as private claims upon texts, inhibited rather than aided the spread of enlightenment. By 1790, Condorcet had evidently reconsidered his position in light of recent events, because nothing could have stood in sharper contrast to this position than the clauses on literary property that replaced this passage in the Sieyès projet . Instead of denouncing literary property as a privilege, they claimed instead that "the progress of enlightenment, and consequently the public good united with notions of distributive justice to necessitate that the property of a work should be guaranteed to the author by law."^[49] They went on to specify, however, that this property right was to be limited to the author's life and ten years—the length of time deemed necessary to complete and sell an edition. The Sieyès proposal thus consecrated the notion of property in ideas, but in a limited form. The notion of limiting of authors' property rights reflected the continuing influence of Condorcet's original concern that the "progress of enlightenment" depended upon public access, rather than private claims to ideas.

Article 21 concluded the section on literary property with an effort to smooth the transition from the Old Regime of "privileges" to the new regime of property: "Publishers or others who at present have acquired for any work a privilege for a fixed term, will continue to enjoy this privilege for its entire duration."^[50] In direct contrast to Condorcet's original position, the Sieyès proposal thus argued that the spread of enlightenment was best achieved, not by liberating ideas from

particular claims entirely, but rather by ensuring the viability of the book as a legally defined and protected commodity.

The main concern of the debates on the freedom of the press in the assembly, and of the Sieyès proposal itself, was to stem the flood of libels and seditious pamphlets that poured forth after the collapse of the systems of censorship and surveillance of the Old Regime. The projet was primarily an effort to determine the limits of what could be said in print and to establish the legal accountability of authors, printers, and booksellers for what they made public. Consequently, it is the repressive aspects of the projet that have received attention from historians.^[51] The assembly, the Committee on the Constitution, and Sieyès had taken Lameth's comments about the state of the book trade seriously. The connection between the crisis in book publishing and the boom in periodical and ephemeral literature was not lost upon them. If the commercial insecurity of book publishing was driving printers and publishers into ephemeral printed matter, then the flood of ephemeral matter (i.e., seditious and libelous pamphlets) might abate if book publishing could be restored to a commercially secure and profitable footing. It is only in light of these political concerns that we can understand why the National Assembly's first legislative effort to define and protect literary property emerged within a law on sedition and libel, and why Condorcet's original position had been reversed.

The first revolutionary effort to give legal recognition to the author's claim on the text, then, was not a grant of freedom to the author, but the imposition of accountability and responsibility. Politically, it formed part of a conservative pro-order move, a police measure. The law made the author legally accountable for the text by defining it as his property.

Nor were the commercial motivations behind the law any more concerned with enhancing the power of the author over the text. In comparison with the royal decrees of 1777, authors were being dealt a rather poor deal by the revolutionary legislators. The proposal rejected the crown's grant of authors' claims in perpetuity. Instead, it argued for limiting authorial claims to ten years after the author's death in the interest of the "progress of enlightenment" and "the public good." Thus, while declaring that texts are authors' property the law in fact severely diminished the author's power to determine the fate of his texts, and put an end to the perpetual private claims (privileges) granted by the crown upon the literary inheritance of the nation. True to the spirit of Condorcet's original pamphlet, he and Sieyès wanted to free those texts for the use of all citizens. This was no theoretical matter. By advancing the notion of "limited property," the two men were proposing that the entire literary inheritance of the nation pour forth from the hands of private publishers and the heirs of authors into the public domain: Rousseau and Voltaire, as well as Racine and Molière, had all been dead for well over ten years. They would now be freely publishable, in any form, by all citizens.

While the proposal rendered publishers, authors, and heirs equal before the law, it failed to address or resolve the philosophical issue at the heart of the late-eighteenth-century debate. If property rights were inviolable natural rights, as the Declaration of the Rights of Man and the Citizen had recently proclaimed, what power did the state have to limit them or regulate their distribution? If claims upon property were instead socially constituted, were they not then just "privileges" by another name? The proposed law appeared incoherent and arbitrary: on the one hand it recognized a property right, and on the other, by defining it as noninheritable, it also advanced an instrumentalist notion of the public good that flew in the face of natural rights theory and explicitly undermined the actual power of individuals to exercise their constitutionally guaranteed right. It was an attempt at a compromise between the two epistemological stances, between individual and collective claims on ideas.

The Condorcet/Sieyès proposal of 1790 broke on the shoals of its own contradictions. While applauded upon presentation in the assembly, the proposal suffered such virulent criticism from so many quarters that it was never even brought to a vote. Much of this criticism focused on the issues of libel and sedition, and, in particular, on the articles that proposed that authors and printers could be held accountable for the seditious and criminal actions that their works could be construed as intending to incite.^[52] Radicals were quick to detect the repressive and conservative character of the entire proposal. The militant journalist Elysée Loustallot decried any law that limited the exercise of a natural right: "The patriotic public does not ask for a law granting freedom of the press. . . . The only true limits of freedom are those in the nature of things themselves."^[53] There should be no civil laws, according to Loustallot, limiting or regulating what could be said in print—even if it seemed libelous or seditious to the men in power.

Apart from the issues of sedition and libel, the measure was also criticized for its treatment of the issues of literary privileges and property. This section of the law was attacked from three different positions from 1790 to 1791. The first line of attack came from pamphleteers who took up Condorcet's position of 1776 specifically to denounce any measure that would reimpose private claims upon ideas. Thus, the comte de Kéralio attacked the very notion of property in ideas as a threat to freedom of thought and, consequently, to the progress of enlightenment:

[The National Assembly] has negated all privileges as destroyers of liberty. . . . And as liberty cannot be maintained without enlightenment and knowledge, a wise legislator will guard himself from conserving even the smallest of privileges, which, by limiting the freedom of the press, restrain freedom of thought and inhibit the expansion of human knowledge.^[54]

He viewed the Condorcet/Sieyès proposal as an unprincipled and misguided effort to translate a regime of privilege into a rhetoric of property. According to

Kéralio, the cause of "authors' property rights" was no more than a political smokescreen, serving to conceal the commercial interests of publishers.^[55] The commercial publishing world offered ample evidence to support these charges. Like Condorcet in 1776, Kéralio believed knowledge should be freely accessible to all—even to print, publish, and sell.

A second line of attack on the Condorcet/Sieyès proposal was advanced by the old corporate monopolists of the Paris Publishers' and Printers' Guild and the royally privileged theater directors, who revived and deployed Diderot's arguments in a campaign to have their "privileges" recognized as unlimited property rights. They mounted a lobbying effort to kill the proposal in committee. The playwright Jean François de LaHarpe protested to the National Assembly on 24 August 1790:

Your decrees have pronounced the abolition of all privileges. Having enjoyed such privileges for over one hundred years, the directors of the Comédie française . . . claim that all the plays that they have been given the exclusive privilege to perform since the establishment of their theater are now their eternal and inviolable property.^[56]

A few weeks later on 6 September the officers of the Paris Publishers' and Printers' Guild joined in this corporatist reaction and presented a mémoire to the assembly in which they proposed to "put before the eyes of the Committee on the Constitution, the Code for the Publishing and Printing Trades edited by the great d'Aguesseau [i.e., the code of 1723]."^[57] It was the code of 1723 that had provided the basis for arguments that publishers' "privileges" were automatically and perpetually renewable and hence actually the confirmation of an anterior property right.^[58] How could a revolution that had declared property a natural and inalienable right now take steps to limit that right?

These arguments were forcefully reiterated in 1791 as the forces of cultural reaction mobilized with greater intensity to advance their corporatistic cause under the guise of defending "authors' rights." The Committee on Agriculture and Commerce reported on a letter from the Keeper of the Seals in which he testified that the lack of a law guaranteeing literary property was ruining French letters:

It is impossible to doubt that the vigilant and active protection that the government has always accorded the property of authors and those to whom they cede their works was one of the principal reasons that literature has flourished in France more than with any other modern people. There can be no doubt that pirates of our best books will incessantly inundate the kingdom, ruin proprietors, intimidate those who are in a position to purchase manuscripts, and exhaust, in a word, the most precious branch of our national industry.^[59]

The Keeper of the Seals recommended that the Committees on the Constitution, Agriculture and Commerce, and Research meet together to resolve this question. The Committee on Agriculture and Commerce took the initiative to convene the

three committees "to propose a law on these issues which are crucial to the book trade and to literature."^[60]

The formal abolition of the Publishers' and Printers' Guild in March 1791 dealt a severe blow to the corporate lobby.^[61] But individual publishers of the old guild, as well as public officials, continued to agitate for the protection of literary property. On 22 May the Minister of Justice (formerly the Keeper of the Seals) wrote to the Committee on Agriculture and Commerce urging action.^[62] The following day François Hell, a member of the committee, received a letter from

MM. Jean-Marie Bruyset and Pierre-Marie Bruyset and son, printers of Lyon, requesting a law that will assure authors the property in their works and prohibit pirating. . . . They state that this law is urgently needed, as at this moment someone has pirated a fifteen-volume edition of Valmont's Dictionnaire which only appeared fifteen days ago, which cost 500,000 livres to produce, and the loss of which would reduce the author and printers to the state of beggars.^[63]

The Committees on Agriculture and Commerce, and on the Constitution, resolved as a consequence of this report to charge Hell with the drafting of a new law.

The Hell projet , published by order of the National Assembly sometime in the summer of 1791, gave legislative embodiment to the principles long advocated by the Paris Publishers' Guild and Diderot. Thus Hell announced to the National Assembly:

The first of all properties is that of thought; it is independent, it is anterior to all laws. . . . All other forms of property are nothing but conventions, social concessions, those of the mind and of genius are gifts from nature, they ought to be beyond any restriction. . . . The Old Regime named the act by which one guaranteed literary property a "literary privilege." A privilege! What a gross abuse of words. You have destroyed the word . . . now you can consecrate the thing.^[64]

Nothing could have been further from the views presented a year earlier by Sieyès and Condorcet. The specific clauses of the Hell projet de loi upheld all former "privileges" on the entire literary inheritance of France, which had been accrued by publishers in consequence of the code of 1723, as titles of property. Literary property was to be inheritable and transmissible in perpetuity like any other form of property. The law was to be printed at the end of every publication "replacing the text of the former privilege."^[65] This was precisely the interpretation that the Comédie française and the Paris Publishers' and Printers' Guild had long hoped to advance.

A third line of attack on the Condorcet/Sieyès proposal was voiced from within the camp of those who actually sympathized with the principle of a limited property right. Three days after the Sieyès proposal was presented to the assembly, Charles-Joseph Panckoucke, the wealthiest publisher in Paris, published the first of two articles in the Mercure de France exposing his own views on

how the crisis in book publishing could be resolved.^[66] Like Sieyès and Condorcet, he expressed ideological concerns about unlimited exclusive claims on ideas: "An author or a publisher who would be the eternal proprietors of their books, would necessarily be monopolists."^[67] Panckoucke did not believe in monopolies in ideas. He shared Condorcet's concern that perpetual monopolies on texts left the fate of public enlightenment and the spread of enlightened ideas totally in the hands of private individuals. Limits on private claims, he insisted, were justified by public interest.

He felt, however, that the limits on private claims proposed by Sieyès and Condorcet were too severe. Instead, he proposed that France adopt the model put into place by the English in 1774:

Every author enjoys at first a fourteen-year claim upon his work. If he survives that term, he obtains another fourteen years of enjoyment of his claim. At the expiration of that term, the book belongs to the public. The [English] nation has thought, with reason, that this is the proper means of reconciling private interest with the public good, and that as good books contribute to its enlightenment . . . it is just to favor its writers with these dispositions.^[68]

As a publisher of multi-authored, multivolume works, Panckoucke sought to extend the legal definition of the unit of time deemed necessary to complete an edition from ten to fourteen years. Nonetheless, the arguments of even this large commercial publisher rested not upon the inviolability of property rights but, rather, upon the ideal of an enlightened nation. Writers merited special favor not as property holders but because they were the source of the "good books" through which the public received enlightenment. Once they had received their compensation, the public good dictated that these texts belong to all.

By 1791, then, the mid-century debate between Diderot and Condorcet had resurfaced within the Revolution itself. Caught between their interest in liberating public circulation of ideas from the inquisitorial and monopolistic institutions of the Old Regime and their fear of the political consequences of the cultural anarchy that had ensued from the "freeing of the press," Condorcet and Sieyès had advanced the notion of a "limited property right" in an attempt to effect a legislative compromise between private interests and public enlightenment. But cultural libertarians, like Kéralio, took up Condorcet's arguments of 1776 and protested violently against any private claims on ideas. Alternately, the Publishers' and Printers' Guild, along with the directors of the Comédie française, mobilized a corporate lobby to argue for the inviolability of authors' property rights. Even men like Panckoucke, who agreed with the basic premise of a "limited right," found the particular stipulations of the proposal unacceptable. The Condorcet/Sieyès proposal of 1790 foundered in a sea of criticism. There was to be no law regulating claims upon ideas until 1793.

The Second Initiative (1791–1793)

The revolutionary law of 19 July 1793, which defined the legal limits and powers of the author and laid the foundation for republican publishing, has served as the basis for French publishing to this date. It is still the first standard citation in French law school textbooks on literary property.^[69] In order to understand how a law finally succeeded and why it took the form it did, critical changes in the revolutionary context between 1790 and 1793 must be considered.

In 1791 there was a crucial shift in the balance of forces for and against the notion of a "limited property right." The suppression of the Publishers' and Printers' Guild in March 1791 had dealt a severe blow to the pro-property corporate lobby. A distinct law on libel and sedition was incorporated into the Constitution in September 1791, leaving the property question to be resolved independently of the issue of censorship. This separation significantly depoliticized the property issue. The Hell proposal, which circulated for public discussion in those uncertain months of the summer of 1791, appears never to have reached the floor of the assembly for a vote. By the fall of 1791, it had become clear that the advocates of perpetual private property in ideas had wasted their energies by courting the wrong legislative committee.

The transfer of power from the Constituent to the Legislative Assembly on 1 October 1791 was accompanied by a reorganization of the structure of the assembly's committees. With this reorganization, jurisdiction over the question of literary property passed from the Committee on Agriculture and Commerce to the newly formed Committee on Public Instruction under the presidency of Condorcet.^[70] He was joined on the committee by, among others, Sieyès.^[71] Thus the question of literary claims, first raised in 1790 as part of a repressive police measure, and then as a commercial interest, was, by virtue of changing circumstances, recontextualized as a question of education and the encouragement of knowledge.

By 1791, moreover, the results of a second wave of agitation for authors' rights reached legislative formulation. This agitation came, not from corporate interests, but rather from authors for the theater protesting the monopoly of the Comédie française on dramatic works. Since the founding of the Comédie française in 1680, it was only theater directors, not playwrights, who could legally receive "privileges" to present and publish theatrical works.^[72] This monopoly had not been affected by the reforms of 1777. The agitation of "unprivileged" playwrights was therefore crucial in disassociating the cause of "authors' rights" from a rear-guard defense of old-regime privileges and realigning it politically within the prorevolutionary attack on privileged interests.

Theater authors began their agitation in 1790 with the creation of a committee led by the playwright Pierre-Augustin Caron de Beaumarchais in order to assert the rights of dramatic authors to their own works and to call for the aboli-

tion of the privileges of the Comédie française. A petition of protest bearing the signatures of twenty-one writers was presented to the National Assembly by LaHarpe on 24 August 1790. This petition was essentially an effort to reintroduce into the assembly the clauses of the Sieyès proposal that had pertained to the theater and property in dramatic works. Anyone, they argued, should be free to open a theater. The works of authors dead for more than five years should be considered "public property," but no one should be allowed to represent or publish the dramatic works of living authors without their written consent.^[73] The petition was sent to the Committee on the Constitution.^[74]

LaHarpe's plea did not fall upon deaf ears. In fact, supporters of the Comédie française charged that the petition drive had been instigated by a key member of the very committee to which it was submitted: "It's chez M. de Mirabeau . . . that this petition was cooked up."^[75] Whether true or not, there can be little doubt that Honoré-Gabriel de Mirabeau helped to advance the cause of the petitioners.^[76] Less than a month later, on 13 January 1791, the Committee on the Constitution presented a projet de loi drafted by Mirabeau on behalf of the petitioners to the National Assembly.^[77]

The Mirabeau proposal, presented by Isaac-René-Guy Le Chapelier on behalf of the committee, was essentially a redrafting of the articles of the Sieyès proposal pertaining to literary property, but this time on behalf of theater authors alone. In contrast to the Sieyès proposal, however, the preamble of the new proposal laid stress, not on authors' rights, but rather upon the rights of the public. Thus Le Chapelier argued:

In soliciting for authors . . . exclusive property rights during their lifetime and five years after their death, authors acknowledge, even invoke, the rights of the public, and they do not hesitate to swear that after a period of five years the author's works are public property. . . . The public ought to have the property of great works. . . . But despotism invaded that communal property and carved it up into exclusive privileges.^[78]

The authors represented themselves as servants of the public good, of its enlightenment, in opposition to the private interests of publishers and theater directors. Thus the authors themselves rejected the Diderotist argument for unlimited and absolute claims upon their texts and, reviving the compromise position of Sieyès and Condorcet, presented themselves as contributors to "public property" and guardians of the public claim to the nation's cultural commons. The author was now depicted as a hero of public enlightenment, rather than as a selfish property owner. Unlike the Sieyès proposal, that of Le Chapelier was passed into law, on 13 January 1791. The law abolished all past "privileges" and recognized the theater author's claims as exclusive property rights until five years after the author's death, at which point they would become part of the public domain.

This law, however, did not cover the work of authors in genres other than the theater. The initiative to define the legal status of all authors now passed to the

newly formed Committee on Public Instruction. Ironically, it was the recently empowered authors of dramatic works who again brought the issue to the attention of the committee. On 6 December 1791 the Committee on Public Instruction received a request from a deputation of authors headed by Beaumarchais to hear their charges against the directors of the theaters for noncompliance with the law of 13 January 1791. The theater directors had chosen to interpret this law to apply only to future works, leaving them free to present any work, even by a living author, which had already been printed or published. Further, they claimed publication rights on any work contracted by their companies prior to the law.^[79]

As the result of a series of meetings, the committee drafted a projet de loi that was presented by Gilbert Romme on its behalf and passed in the Legislative Assembly on 30 August 1792.^[80] The law represented a victory for the theater directors. It upheld all contracts between authors and the theaters prior to the passage of the law of 13 January 1791 and sustained the exclusive right of the theaters to stage any work performed prior to the passage of the law. Needless to say the law met with vociferous protest from authors. This time it was the playwright Marie-Joseph Chénier who headed up the protest with a letter and petition to the Committee on Public Instruction on 18 September 1792.^[81] The law, Chénier argued, had been slipped through by Romme without the support of the majority of the committee members.^[82] The committee reopened the question as a consequence of this protest.^[83]

It was not, however, just writers of dramatic works who expressed discontent over the course of 1792. On 2 January 1792 the committee received a petition from thirty authors and editors of music, not covered by the law of 1791, in which they begged the "National Assembly, in all its wisdom, to find a means to protect their property and prevent pirating."^[84] The novelist Jean-Baptiste Louvet wrote to the convention as well, requesting permission to present a petition "calling for a law against priaters, who are destroying the book trade and bringing me to ruin."^[85] These appeals were not to go unnoticed. On 20 February 1793 the Committee on Public Instruction finally assigned Chénier the task of drafting a general law against priate editions in all genres.^[86] News of the forthcoming Chénier proposal was announced in the Moniteur in April, but Chénier did not succeed in getting the floor of the convention during the troubled spring of 1793.^[87]

After the "revolution" of 31 May to 2 June 1793, which purged the Girondin faction from the convention, Condorcet ceased appearing at committee meetings. A month later he was in hiding.^[88] Sieyès took over the presidency of the committee on 23 May, but he and Chénier both soon withdrew as well.^[89] Denounced as Girondins, all three were formally excluded from the committee on 6 October 1793.^[90] It is ironic that the Girondin law that founded the basis of modern French publishing should emerge precisely at the moment of the Jacobin victory that suppressed its authors. Indeed, it was the Jacobin consolidation of power that made it possible to pass the law. On 19 July 1793 the convention at last heard

Chénier's proposal presented on behalf of the Committee on Public Instruction by Joseph Lakanal.^[91] It was passed with no recorded discussion.^[92]

The decree adopted on 19 July 1793 amounted to yet another version of the Condorect/Sieyès proposal of 1790. No longer perceived as a "Girondin" police measure intended to ensure the accountability of authors, nor as a commercial regulation to protect the private property interests of publishers, it was now presented by the Committee on Public Instruction as a mechanism for promoting and ensuring public enlightenment by encouraging and recompensing intellectual activity—that is, by granting limited property rights to authors:

Citizens, of all the forms of property, the least susceptible to contest, whose growth cannot harm republican equality, nor cast doubt upon liberty, is property in the productions of the genius. . . . By what fatality is it necessary that the man of genius, who consecrates his efforts to the instruction of his fellow citizens, should have nothing to promise himself but a sterile glory and should be deprived of his claim to legitimate recompense for his noble labors?^[93]

Like the Sieyès proposal three years earlier, this law guaranteed authors, or those to whom they ceded the text by contract, an exclusive claim upon the publication of the text for the lifetime of the author plus an additional ten years for heirs and publishers. The Royal Administration of the Book Trade, which had registered the literary "privileges" of the Old Regime, was to be replaced by a national legal deposit at the Bibliothèque nationale, where all property claims were to be legally registered. The decree differed from the Sieyès proposal in one crucial respect: it gave no retroactive protection to the former holders of "privilèges en librairie" or "privilèges d'auteur." With the law of 19 July 1793, then, the cultural capital of the Old Regime was definitively remanded from the private hands of heirs and publishers into the public domain. Rousseau and Voltaire, like Corneille, Racine, and LaFontaine, had now too been dead for well over ten years. Thus, as Condorcet had dreamed, the authors of the Enlightenment, as well as those of the classical age, became the inheritance of all.

The severing of the clauses on literary property from their original context in the Sieyès proposal on sedition and libel, the deletion of the clause reaffirming current "privileges," the mobilization of authors, and the new stress on public enlightenment significantly transformed the political meaning and impact of the law. Initially part of a concerted moderate effort to reregulate and police the printed word and ensure publishers profits, the recontextualized clauses came to be viewed as a "declaration of rights," presented as a Jacobin effort to abolish the vested interests of inherited privileges, to consecrate the bearers of enlightenment, and to enhance public access to the ideas of the Enlightenment.

But the law did not resolve the epistemological tension between Condorcet and Diderot. Rather, it produced an unstable synthesis between the two positions. It drew upon a Diderotist rhetoric of the sanctity of individual creativity as an

Figure 1.
The Revolution celebrated the author as a hero of public enlightenment rather than
as a private individual creator. A. Duplessis,  La Révolution française , c. 1790, detail. 
Photo: Musée de la Révolution française, Vizille, France.

inviolable right, but it did not rigorously respect the conclusions Diderot drew from this position. In contrast to the "privilège d'auteur" of 1777, the law did not recognize the author's claim beyond his lifetime but consecrated the notion that the only true heir to an author's work was the nation as a whole. This notion of a "public domain," of democratic access to a common cultural inheritance upon which no particular claim could be made, bore the traces, not of Diderot, but of Condorcet's faith that truths were given in nature and, though mediated through individual minds, ultimately belonged to all. Progress in human understanding depended not on private knowledge claims but rather on free and equal access to enlightenment. Authors' property rights were conceived as a recompense for the author's service as an agent of enlightenment through the publication of his ideas. The law of 1793 accomplished this task of synthesis through political negotiation rather than philosophical reasoning, that is, through a refashioning of the political identity of the author in the first few years of the Revolution, from a privileged creature of the absolutist police state into a servant of public enlightenment.

Conclusion

This understanding of the history of the formation of the legal identity of the author returns us to Foucault's original question: What is an author? The author as a legal instrument for the regulation of knowledge was created by the absolutist monarchy in 1777, not by the liberal bourgeois democracy inaugurated in 1789. The author was created by a royal regime that exercised power through privilege rather than by a constitutional regime committed to ensuring the protection of the individual as a private property owner. The revolutionary legislation did redefine the author's "privilege" as property, but not as an absolute right. The intention and the result of this redefinition of the author's claim to his text as property was not to enhance the author's power to control or determine the uses and meanings of the text. In fact, it was quite the opposite.

What the revolutionaries acknowledged and sought to ensure was not the individual dictation of meanings and truths but their maximum exchange, conflict, and social negotiation. Progress in understanding, they believed, occurred through enhanced access and exchange. In reshaping the cultural regime the revolutionaries sought to make a world appropriate not to an absolutist police state but to a liberal state founded on conflict and negotiation. The democratic bourgeois revolution did not mark a further step in the progressive consolidation of the notion of the author. Rather, the revolutionaries explicitly intended to dethrone the absolute author, a creature of privilege, and recast him, not as a private individual (the absolute bourgeois), but rather as a public servant, as the model citizen. This civic ideal of authorship was shared not just by liberal statesmen such as Condorcet and Sieyès but by even the most capitalistically inclined publishers like Panckoucke as well. Concern for the public good explains why the author's regulative powers were not further consolidated, but rather eroded and destabilized by the revolutionary legislation.

Contrary to Foucault's Diderotist interpretation, the revolutionary legislation actually reflected not one but both sides of the Enlightenment debate, effecting an epistemologically impure and unstable legal synthesis that combined an instrumentalist notion of the public good with a theory of authorship based upon natural rights. Precisely because of this legal instability, the regime by which the public exchange of ideas was regulated would be challenged and renegotiated repeatedly over the course of the revolutionary period. The legal history of French authorship thus suggests that Foucault's essay requires a historical revision: the central mechanism of the modern regime of knowledge, as it emerged from the epistemology of the French Enlightenment, was unstable from its very beginnings.

If the Old Regime first accorded Voltaire and Rousseau the possibility of legal status as privileged authors with perpetual private lineages for their texts, the Revolution relocated these figures in the public domain, the legal parallel to the

civic rituals that reposed their bodily remains in the public temple of the Pantheon. By inventing the "public domain" the French revolutionary laws on authorship shifted the problem of determining the meaning of the text away from its source, the author, and toward its destination, its re-presentation and reception by the editor and reader.^[94] In so doing the revolutionaries inaugurated a new cultural regime preoccupied as much with distinctions between different editions of texts as with epistemological debates about the origins of ideas.^[95] The questions "Who is Voltaire?" and "What is Rousseau?" were thus joined by a new set of legal, and also literary, preoccupations: whose Voltaire? which Rousseau?

The French revolutionary laws on authorship also suggest that literary historians and critics may need a more complex view of the relationship between the law and cultural change, one that accounts for the political as well as the socio-economic forces at work in the reshaping of the legal world. Though there can be little doubt, as both Rose and Woodmansee suggest, that the expansion of commerce in the printed word in the eighteenth century put unprecedented pressures upon public authorities toward the legal recognition of authorial property, the legal responses of both prerevolutionary and revolutionary authorities did not merely reflect these socio-economic changes. Politics, and a concern for public life, mediated the successive negotiations between the private interests of authors and publishers and the concerns of legal authorities. As a consequence, the revolutionary legislators produced a legal conception of authorial identity that not only consecrated but also limited the author's power of self-determination for the sake of the public good.

Norms, Discipline, and the Law

François Ewald

In the tableau that concludes the first volume of The History of Sexuality , "Right of Death and Power over Life," Michel Foucault develops the hypothesis that, ever since antiquity, new mechanisms for the exercise of "bio-power"—disciplines of the body and attempts to regulate the population—have developed in Western societies.^[1] The juridical mode of governance, characterized by forcible seizure, abduction, or repression and usually culminating in death, is increasingly replaced by bio-power, "which aims to produce, develop, and order social strength," a power that exerts a more positive influence on life, undertaking to administer it, multiply it, and impose upon it a system of regulations and precise inspection. Having noted that this transformation in the mechanisms of power signifies "nothing less than the entry of life into history," Foucault concludes by suggesting that "another consequence of this development of bio-power was the growing importance assumed by the action of the norm, at the expense of the juridical system of the law."^[2]

Foucault does not mean to suggest here that the development of bio-power is accompanied by a decline of law. His further commentary makes it clear that the formation of a normalizing society in no way diminished the power of law or caused judicial institutions to disappear. In fact, normalization tends to be accompanied by an astonishing proliferation of legislation. Practically speaking, legislators never expressed themselves as freely or as extensively as in the age of bio-power. The norm, then, is opposed not to law itself but to what Foucault would call "the juridical": the institution of law as the expression of a sovereign's power. If, as Foucault puts it, "the law cannot help but be armed," and if its weapon par excellence is death, this equation of law and death does not derive from the essential character of the law. Law can also function by formulating norms, thus becoming part of a different sort of power that "has to qualify, measure, appraise, and hierarchize rather than display itself in its murderous splendor."^[3] In the age of bio-power, the juridical , which characterized monarchical law, can readily be opposed to the normative , which comes to the fore most typically in constitutions, legal codes, and the constant and clamorous activity of the legislature."^[4]

Foucault's ideas have a dual consequence for the philosophy of law. They encourage us to distinguish law and its formal expression from the juridical. The

juridical served as a "code" that enabled monarchical power to constitute itself, formalize its structure, and reflect upon its own workings. However, such a code is not the only possible form the law can take. Neither the "regression of the juridical," which accompanies the rise of bio-power, nor the fact that the most typical mechanisms of juridical power can no longer be represented in legal form, necessarily signals the disappearance of the law.

We can and must imagine a history of law that would give meaning and function to the law's varying modes of formal expression. Foucault also compels us to reconsider what we mean by norm , which he places among the arts of judgment. Undoubtedly the norm is related to power, but it is characterized less by the use of force or violence than by an implicit logic that allows power to reflect upon its own strategies and clearly define its objects. This logic is at once the force that enables us to imagine life and the living as objects of power and the power that can take "life" in hand, creating the sphere of the bio-political.

Thus, in opposing the "action of the norm" to "the juridical system of law," Foucault suggests two possible paths of inquiry. The first, to borrow Foucault's terminology, is "ontological" and concerns modernity. It asks: What is modernity if we understand it as participating in the logic of the norm? What can we learn about the modern by approaching it in terms of the norm and the practices of power and knowledge organized around the norm? The second concerns the shift in the relationship between knowledge and power and its influence on the status and function of legal thought in modern societies. Within the framework of "the regresson of the juridical," what is the place of law? Is a theory or practice of law articulated around the norm possible? If so, what form would such theory or practice take, and what would be the risks and possibilities associated with them?

Georges Canguilhem has noted that etymology holds certain surprises and disappointments for our contemporary understanding of the word norm : "When we know that norm is the Latin word for T-square and that normalis means perpendicular, we know almost all that must be known about the area in which the meaning of the terms norm and normal originated."^[5] Joachim Ritter's Historisches Wörterbuch der Philosophie recalls the technical origin of the term. Vitruvius used it in his treatise On Architecture to indicate the instrument used to draw right angles. Through metaphor, the term would be taken up to designate the rule of law. Cicero, in particular, relies on the Stoic reference to the architectural regularity of nature, speaking of nature as the "norm of the law" (norma legis ). The norm had a long career as a synonym for the rule. Jean Calvin, for example, writes in his Institutes of the Christian Religion : "God has determined by His laws what is good and right, and by this means has meant to hold men to a certain norm."

However, at the beginning of the nineteenth century, there is a radical change in the relationship between the rule and the norm. Norm can no longer stand simply as another name for rule; rather, it comes to designate both a particular variety of rules and a way of producing them and, perhaps most significantly of all, a principle of valorization. Of course, the norm still refers to a standard measure that allows us to distinguish what is in conformity with the rule from what is not, but this distinction is no longer directly linked to the notion of rectitude. Its essential reference is no longer to the square but to the average; the norm now refers to the play of oppositions between the normal and the abnormal or pathological.

The vocabulary associated with the term expands as well: in French, normal is no longer the only word to derive from norme . It is joined by normalité (1834), normatif (1868), and normalisation (1920). This remarkable extension of the norm's domain will affect a wide variety of fields concerned with economics and technology. It will also have a major influence on the moral, juridical, and political sciences, which at the close of the nineteenth century will establish themselves (particularly in Germany) as "normative" sciences.

Thus, two centuries ago the word norm led a quiet, unremarkable existence, whereas today, along with its panoply of derivations and associated terms, it has become one of the most used and abused terms of our contemporary vocabulary, whether we speak colloquially or as social scientists. We are intimidated by norms and contemplate them suspiciously, feeling ashamed to consider ourselves simply normal. Psychologists and sociologists have made persistent efforts to establish norms whose constraining effects can be felt everywhere—even where we imagine our behavior to be least susceptible to determination. In a sense, virtue has become normalized: the virtuous individual can delude himself or herself into believing that he or she acts out of a sense of duty while in reality simply making his or her behavior conform to a particular norm. Similarly, health can be envisioned as the absence of illness, while in actuality it is merely a sign of normal organic functioning. Even taste, which appears to be a product of purely subjective aesthetic judgments, simply repeats internalized norms in the regularity of its assessments. Public hygiene, urban planning, safety measures against pollution or nuclear contamination, and quality control have all come about as the result of normative decisions of one sort or another. What is the significance of this extension of the normative, and what risks and potential benefits does it hold for the future?

One set of normative practices we might wish to examine in this context is what Foucault has described as "disciplinary society." In Discipline and Punish , Foucault suggests that the prison is in some sense the purest expression of the disciplinary order. But this is not to say that he believed disciplinary society to be based on generalized confinement. In fact, for Foucault the gradual spread of various disciplines (to the factory, the school, the hospital, or the barracks) indicates that

discipline is not primarily concerned with confinement, nor with the segregation of its subject population. Rather, discipline tends not to divide or compartmentalize society but works instead to create a homogeneous social space.

The norm is the principle that allows discipline to develop from a simple set of constraints into a mechanism; it serves as the matrix that transforms the negative restraints of the juridical into the more positive controls of normalization and helps to produce the generalization of discipline. The norm is also the means through which the disciplinary society communicates with itself. The norm relates the disciplinary institutions of production—knowledge, wealth, and finance—to one another in such a way that they become truly interdisciplinary ; it provides a common language for these various disciplines and makes it possible to translate from one disciplinary idiom into another.

In Discipline and Punish , Foucault returns again and again to the idea that discipline "produces" individuals. It not only manages them and makes use of them but actively constitutes them as its object. Within the disciplinary framework, the norm participates in this logic of individualization while also serving as the force that joins together the individuals created by discipline and allows them to communicate with one another.

It is essential to avoid confusing the terms norm and discipline . Disciplines are concerned with the body and its training, while the norm is a measurement and a means of producing a common standard. Discipline is not necessarily normative. According to Foucault, modernity coincides with the coming of a normative age. The normalization of the various disciplines and the shift from discipline as constraint to discipline as a regulatory mechanism are symptomatic of this change, as is the formation of a disciplinary society founded on a new kind of social space that is supple, flexible, homogeneous, and entirely self-contained.

Within the disciplinary order, the influence of the norm is primarily local; norms remain attached to specific practices and institutions. With the appearance of insurance, the norm will serve as a means of managing different kinds of actuarial populations,^[6] while with the institution of a Social Security system it will become a way to manage the entire population of a given state. The shift here is from the level of the micro-instrumental to that of the bio-political. Risk plays the same role in insurance that the norm does in the constitution of disciplinary strategies. The conceptual category of risk, which makes insurance possible, is the precise homologue of the disciplinary norm.

Insurance is an equivocal term that comprises 1) a technique for estimating risk in actuarial terms; 2) the practices of restitution and indemnification of damages that set this technique in motion; and 3) the institutions that structure public and private insurance schemes. I intend to discuss only the first—the techniques of risk.

What is risk ? In common parlance, the term is a synonym for danger, peril, or the unexpected misfortunes that might happen to anyone; it also implies an

objective threat of some sort. In insurance, risk refers neither to a specific occurrence nor to a kind of event that might take place but instead to a way of treating certain events that might happen to a particular group of individuals (a population). Nothing in itself is a risk—risks have no real existence. By an inverse logic, anything can be a risk—everything depends on the way the danger is analyzed and the potential event is evaluated. To adopt Kantian terminology, the category of risk is a category of understanding; it cannot be derived from intuition or sensibility. As a technology of risk, insurance is first of all a rational outline, a means of disassembling, reconstructing, and organizing certain elements of reality.

Insurance is the practice of a specific type of rationality that formalizes the calculation of probability. This explains why one can only be insured against risks, and why these can be as various as death, accident, hail, illness, childbirth, military service, a business failure, or litigation.^[7] The insurer does not passively make note of actual risks in order to insure people against them. Instead, he produces risks by making them visible and comprehensible as such in situations where the individual would ordinarily see only the unpredictable hazards of his or her particular fate.

Risk, then, is a principle of objectification. It confers a certain objective status on the events of private, professional, or commercial life: death, accident, injury, loss, or hazard. The task of insurance is to constitute a particular kind of objectivity; providing various familiar events with a real existence that changes their character. Insurance creates its own world; it confronts the world of lived experience (and all of its terrors) with the more neutral and predictable world of risk. When the first insurers boasted about the liberating effects of their statistical models, or explained that the dangers we fear are really nothing but risks we can take steps to protect ourselves against, they were, of course, speaking as advertisers. Still, their arguments rested on the idea of a very fundamental transformation of the world.

Risk is both objective and objectifying. This arises through the exercise of a rigorously positivistic attitude. Insurance has two bases: first, the statistical table or graph that testifies to the regular occurrence of certain events; second, the calculation of probabilities that are then applied to these statistics so that one can evaluate the possibility of these same events. The insurance view of the world is firmly grounded in probability and statistics. It is generally admitted that modern science took its definitive form at the time of the Scientific Revolution in the seventeenth century. One might also speak of an analogous probabilistic revolution of the nineteenth century, one that radically transformed contemporary notions of such familiar ideas as "fact," "law," and "cause."^[8] Like the Galilean revolution before it, the probabilistic revolution was received with its fair share of resistance, debate, and utter incomprehension. Some of the best-known examples of resis-

tance were Auguste Comte's or Jean-Baptiste Say's opposition to the use of probability in the social sciences. Other examples include the general philosophical condemnation of probability on the grounds that it would introduce into history an element of determinism that was altogether incompatible with liberty, or the endless legal debates during the nineteenth century over the relative merits of fault and risk as causes of responsiblity.

By the standards of an earlier world (in which we still live, at least to some extent), the insurer, like the statistician, is most remarkable for his rigorous suspension of judgment. For him, events are facts with distinct boundaries in space and time—they are complete in themselves and have no cause, or past, or future. They are individuals, pure atoms that persistently leave their trace on the surface of the world. They do not signify; they simply are . They can barely be described, and their identity is reduced to the numerical quality that allows one to tabulate them as a point or a unit.

The statistician must begin by bracketing the usual systems of signification and should remain instead at that unclear boundary where a coherent vision of the world threatens to disappear beneath the infinite residues of facts and events. Similarly, the insurer who initially notes the fact of an accident or a death is altogether indifferent to its cause. It matters little that a specific accident might have been avoidable, or that a particular individual will bear historical responsibility for a given event. The important thing about events is that they occur, or rather that their occurrence is repetitive, multiple, and regular. They become purely accidental, and are rendered objective by comparison with themselves. For the purposes of statistics, they remain without victims and without a cause, at least initially.

To put the matter sceptically, what is at issue here is the possibility of freeing oneself from the usual play of signification by concentrating on the pure factuality of facts, the pure recording of occurrences. In Kantian terms, the task for the insurer and the statistician alike is to restrict himself or herself to a single level of intuition, locating and comprehending facts exclusively in terms of their temporal and spatial situation, without appealing to a more comprehensive system of understanding. The world as perceived for statistical ends makes no sense . It is reduced to a pure accumulation of facts, data that accumulate randomly with no prospect of ever signifying as individual bits of information. To the extent that the usual system of signification has been suspended, all facts, even the most insignificant, are worthy of note.

Only the science of probability allows these data to signify. In the logic of probability, sense can emerge from this undifferentiated mass of data without any need for reference to a world outside that of pure surfaces and pure factuality, where pieces of information of indeterminate value repeat and accumulate. For statistical thought, numbers by themselves create meaning. The notion

of mass replaces such evaluative concepts as nature or essence. In the social and moral world, presumably the sphere of free agency, there are observable regularities, constants of social life (marriage, crime, suicide, and so on) whose causes remain obscure. Before the triumph of statistics, it was still possible to invoke the workings of divine Providence, or to seek some other sufficient cause that would explain otherwise inexplicable phenomena, as though the regularity of certain events could be explained by recourse to some invisible logic of causality. But probabilistic thinking makes this specular doubling altogether unnecessary; in statistics, facts lose their status as natural signs or indices of some higher meaning. They refer back to nothing but themselves. The visible world is no longer a translation of an invisible world of essences. Only the repetition of a particular social fact, its multiple occurrences, can give it meaning. According to this logic, the more frequently a particular sort of event occurs statistically, the more real it becomes. The weight and number of occurrences bring social facts into existence. Inversely, a single exceptional event counts for less in statistical terms because it occurs so rarely. The calculation of probability, then, functions as a ruse of reason: even though causes remain unknown and unknowable, they do translate into effect. By seizing upon effects, this kind of calculation allows us to determine the laws that govern the recurrence of events without ever grasping the causes behind them.

Facts are still organized in categories with distinct names: birth, death, accident, suicide, size. However, this is a particularly nominalist use of the category, for these categories make no reference to any explanatory principle. They are simply sets of groupings—open-ended collections of randomly occurring facts that are never identical to one another. The statistical category brings together diverse variables on the basis of their resemblance or potential equivalence; it serves as a principle of classification rather than as an identifying denomination.

According to the logic of statistics, then, an accident is no longer a simple misfortune that happens to someone; instead, it takes on a real existence of its own. Similarly, man no longer exists as an entity that can be explained in terms of human nature, nor does he exist anywhere within the multiplicity of living men. Rather, "man" appears in the qualities that can be attributed to him, which have taken on lives of their own: size, weight, or strength. The characteristics of a particular individual are lost in the midst of those of many other individuals. In a sense, the particular individual with a specific size and weight no longer exists. Only the standard size and weight of a population of individuals who constitute a pool of human qualities continues to have a real existence.

The meaning and import of this strange blending of traits, this peculiar statistical surgery is perhaps most evident in Alphonse Quetelet's attempts to construct a theory of the "average man," which he formulates in at least three places in his work. The project grows directly out of a sense of the significance of averages:

By gathering together a number of individuals of the same age and sex and taking the average of a set of their constant measurements, one obtains a series of constant figures that I would attribute to a fictional entity I call the average man for this group. If we were to note, for example, the size of every twenty-five-year-old Frenchman and find the average, the number we obtained would be the size of the average twenty-five-year-old man.^[9]

Thus defined, the average man is a "fictional entity": there is no actual twenty-five-year-old Frenchman who could be the average man. The average man can also be a typical example of man at a particular moment in time in a specific place:

Everything occurs as if there existed in nature particular types suited to a given country and the environmental circumstances in which they find themselves. Variants on these types come into being accidentally with equal frequency as augmentations or diminutions of the essential characteristics of a type. Suppose that we have a sufficiently large sample population: the average man for each age would find himself flanked on both sides by equal numbers of individuals, some larger and some smaller. Moreover, the groups would be distributed regularly in order of size. The largest groups would be composed of those who were closest to the mean, while the smallest groups would be those most distant from the mean. The further one gets from the average, the smaller the groups that represent this difference, and, at the extreme limits of the distribution, giants, like dwarves, are quite rare. However, these extreme cases are not anomalous—in fact, they are necessary to complete the ascending and descending series determined by the law of randomness. Each group has its own specific value and place. Thus when men are thrust together in society and their various sizes come together in the most unlikely combinations, there is between them a mysterious link that allows us to consider each individual as a necessary part of a whole which has no physical existence and escapes us in the individual instance, and which can only be perceived through the eyes of science.^[10]

We may note that in this second version, too, the average man remains a fiction: "Everything occurs as if . . . ," writes Quetelet. Of course, the law of randomness makes it apparent that something corresponds to this fiction: not a real individual who incarnates the social mean but the typical man for that society; not a model or original that serves as the standard for all men but the reference point common to them all. This point of reference provides them with a kind of "natural" identity and suggests that laws of man do exist. Finally, according to Quetelet,

The man I am considering is, in society, the analogue of the center of gravity within a body; he is the mean around which various social elements move. He is a fictional being for whom all things occur in accordance with the average expectations for the society in question. . . . This determination of the average man is not merely an idle pursuit; knowledge of social averages can serve an important purpose for the human and social sciences. The study of averages is a necessary precursor to any research into social physics, for it serves as the foundation of such study. . . . Only by taking [the average man] into account can we truly appreciate the phenomena of social equilibrium and movement.^[11]

The average man, then, is not an individual whose place in society is indeterminate or uncertain; rather, he is society itself as it sees itself objectified in the

mirror of probability and statistics. There is not a trace of realism in Quetelet's account of the average man. The average man is at once the entity that permits scientific judgment of man and the necessary correlate of that judgment. Once human nature loses its metaphysical status, individuals can be judged only with reference to the social and, more precisely, with reference to the average man.

The theory of the average man, then, is simply a new—and altogether modern—means of individualizing the members of a population. This is what we mean today when we make reference to norms and the normal. The notion of the average man corresponds to a new way of judging individuals—the only way that is scientifically possible, in fact: Quetelet writes,

I have always felt it to be impossible to estimate the degree of courage, or what we must regard as such, contained in the acts of an isolated individual, for what standard of measurement could we possibly adopt for such a quantity? Would we observe this individual for a long enough time and in sufficient depth to take into account all of his actions and estimate the relative value of these acts of courage? What tribunal could possibly pass judgment on these actions, and would there be a large enough number of them for us to reach a satisfactory conclusion? Who could guarantee that in the course of these observations, the individual in question did not undergo some major change? When we work with a large number of individuals, these problems disappear almost entirely, particularly if we mean only to understand something about the relations between them and nothing of a more absolute nature.^[12]

With his theory of the average man, Quetelet proposes a means of specifying individuals with reference to their position within a group, rather than by paying close attention to their essence, their nature, or their ideal state of being. The theory of the average man, then, is an instrument that makes it possible to understand a population with respect only to itself, and without recourse to some external defining factor.

The insurer's "risk" (an objective principle based on calculation and distributions) corresponds directly with the notion of the average man outlined in Quetelet's social physics. The concept of risk makes no reference to nature (as in a metaphysics) or to morality (according to some ideal notion of what man should do or be). Instead it allows the group to make social judgments with respect to itself in a way that always reflects the current state of society and is based on normative, rather than prescriptive, evaluation.

Risk is at once calculable and collective, and these two characteristics are dependent upon one another. Accidents and misfortune are individual occurrences, but risk is a profoundly social phenomenon. Moreover, risk can only be calculated for an entire population. The task of the insurer is to constitute this population through a process of selection and division of risks. Insurance socializes risk, transforming each individual into a part of a whole. The function of insurance is to constitute mutuality, consciously in the case of mutual societies and less consciously in the case of anonymous companies with premiums.

Insurance is therefore more than a scheme that allows individuals to protect themselves against loss for a small fee because of the benefits of mutuality. Reducing it thusly makes it indistinguishable from more primitive forms of mutual aid and solidarity such as the confraternity or the corporation. The central characteristic of insurance is not that it spreads out the cost of individual damages over a large group, but that it provides a justification for this kind of division that has no basis in charity or fellow feeling and is based on a rule of justice, a rule of law. According to Eugène Reboul,

Insurance is the application to human affairs of the rule of possibilities that determines the fate of individuals apart from society before chance has made its own division among them and disposed of the common fund of property according to its own logic. So that equity is preserved, each person must take upon himself a proportional part of the risk that may bring him good fortune or mishap.^[13]

This "proportional part" defines risk for the insurer. The abstract principle behind this reasoning is that the natural distribution of luck and misfortune, i.e., chance, is fundamentally just. Chance must be allowed to play out its whims, and it is up to individuals to protect themselves as they are able to.

Legal judgments were traditionally based on an attempt to discover the cause of damages—it was essential to find out whether damages were the result of an unpredictable natural event or whether they could be attributed to a particular person or institution who would then be required to bear responsibility for the damages. The insurance system, by contrast, proposes an entirely different idea of justice: causality is superceded by the notion of a distribution of a collective burden according to a fixed rule that determines the contribution of each individual. Insurance then offers a new rule of justice that refers no longer back to nature but rather to the existence of the group, a social rule of justice that the group is free to determine for itself, and on its own terms.

At the close of the nineteenth century in industrial Europe, the technology of risk and the institution of social insurance form the basis for a new way of thinking about politics. Insurance becomes social, not so much because new kinds of risk have come into being but because society has come to understand itself and its problems in terms of the principles of the technology of risk. At the end of the nineteenth century, the term insurance designates both a set of institutions and the structure that orders the regulation and functioning of modern society.

This account presupposes the establishment of new relationships between the notions of insurance and the state. Insurance is not imagined simply as an institution or system within the state for which the state must provide an order or organizing principle; rather, the state can now be conceived of in terms of the actuarial view of society. Insurance is no longer a simple subordinate function of the state but an essential part of the state's organization that affects its very nature—the state itself becomes a vast system of social insurance.

The institution of the prison as the universal penalty for crime at the beginning of the nineteenth century marks, in Foucauldian terms, the birth of a "disciplinary society," one whose organization obeys the logic of the norm at the level of micro-power. Likewise, the growth of the insurance industry during the nineteenth century, along with the beginnings of social insurance and the development of large-scale social welfare systems, marks the birth of the "insurance society," in which the norm functions similarly.

In its technical sense, the term normalization does not refer to the production of objects that all conform to a type. Rather, it involves "providing reference documents for the resolution of standard technical and commercial problems that recur in the course of interchange between economic, technical, scientific and social partners."^[14] Normalization, then, is less a question of making products conform to a standard model than it is of reaching an understanding with regard to the choice of a model. The Encyclopaedia Britannica stipulates in its article on standardization that "a standard is that which has been selected as a model to which objects or actions may be compared. In every case a standard provides a criterion for judgment." Normalization is thus the production of norms, standards for measurement and comparison, and rules of judgment. Norman F. Harriman writes, "A standard may be concisely defined as a criterion, measure or example, of procedure, process, dimension, extent, quantity, quality, or time, which is established by an authority, custom, or general consent, as a definite basis of reference or comparison."^[15] Implicit within the concept of normalization is the notion of a principle for measurement that would serve as a common standard, a basic principle of comparison. Normalization produces not objects but procedures that will lead to some general consensus regarding the choice of norms and standards. This definition of normalization lends a certain paradoxical allure to the history of the term. Normalization is a practice that only became aware of itself as such at the start of the twentieth century; the term itself dates from this period (1928), as do the first national and international organizations concerned with the establishment of norms.^[16]

To those who first worked on normalization, the concept appeared as a sort of universal ordering principle. All the institutions that make society possible, no matter how primitive, such as language, writing, money, instruments of measurement, habits and customs, all suddenly seemed to derive, at least in retrospect, from practices of normalization.^[17] The normalizing process had accompanied humanity in every stage of its development. It served a primary social function by regularizing human conduct and by facilitating both technical progress and communication. No social object could escape normalization, and society would be inconceivable without it, for norms and standardization had always played an essential role in social development. Above all, though, normalization played an

essential part in the constitution of systems of communication—the norm is what transformed linguistic signs into a common language. There was thus a certain reciprocity between society and the norm, and between the norm and language. Thus, the modern exponents of normalization viewed normalization as a basic principle of socialization. There can be no society, they argued, without norms, codes, common standards of measurement, and basic principles of communication. Technical normalization was simply a question of constituting a society of producers and consumers and providing it with a common language and common institutions.

One basic difference between technical normalization and earlier forms of socialization lay in the fact that modern normalization was self-consciously and actively willed rather than simply tolerated or accepted, as in earlier periods. Where the population had once been passively subjected to the norm, now certain elements within it were actively seeking to direct and manage the process of normalization.^[18] This was a global development that concerned not only individual producers or sectors of production but the activity of production itself and with it the activity of consumption.

The systematic character of technical normalization differentiates it from earlier processes to which it has some resemblance—for example, the Venetian ship-building industry, which from the sixteenth century onward was organized according to the principle of the division of labor.^[19] Technical normalization has another genealogy entirely. Normalization is the language of the engineer, and its successful integration as a part of modern institutions marks the moment when this technical language could attain to the status of a common language. The institutions of normalization all grew out of associations of mechanical and electrical engineers that were founded during the second half of the nineteenth century in every industrialized nation.^[20] One line of ancestry for the idea of normalization therefore lies in the scientific and technological transformations that accompanied industrialization. Normalization took on a real institutional existence with the creation of the first official bureaus of norms and standards during the First World War.

The demands of wartime production were a second point of origin for normalization movements, since coordinated production, interchangeability of parts, and compatibility of products are all dependent on the establishment of norms.^[21] In the period immediately following the war, normalization appeared to be an inevitable requirement of production in the modern world, and seemed to imply a general peacetime mobilization of the population. Future industrial and economic productivity all seemed to depend on normalization, and industrial leaders saw it as an inescapable necessity.

Thus normalization became something more than a technique to be adopted or neglected at will. Instead, it began to appear as the essential structure that would provide the framework for production and exchange everywhere in the

world. Herein lies the third line of ancestry for normalization: it is both economic and sociological. The demand for normalization in industry indicates an increasing awareness among industrialists that they form a society, and that this society requires its own language, its own codes, and its own specific forms of regulation. The technical normalization movement in industry signals the moment when industrialists and government officials begin to recognize that industrial growth, the appearance of new needs, and mass consumption all contribute to the creation of a new productive system that is distinct from its predecessors not only in terms of its techniques and productive capacities but also in terms of its instruments of communication and its rules.

It is essential to distinguish between the functions and objectives of normalization and its techniques. The functions of normalization are well known: simplification, unification, and specification.^[22] Simplification involves reducing the number of models for objects, choosing between products that resemble each other too closely, and eliminating any superfluous models. Unification means establishing fixed characteristics of objects so that objects are compatible and interchangeable. Specification is a process of reaching a precise understanding about standards for the quality of manufactured products. All of these functions are part of a larger program for rationalizing production by reducing waste, regularizing production so as to minimize the effects of economic fluctuation, and adapting production to demand as efficiently as possible.^[23] In other words, one of the aims of technical normalization is to gain a certain measure of control over time.

Industrial normalization cannot be reduced to the pursuit of these objectives alone, however, for the essential thing here is the technique that makes it possible. Despite the tendency of contemporary authors to insist that there are many different kinds of norms—terminological norms, norms for spatial measurement, and qualitative norms—any one of these varieties of norm would be inconceivable without the others.^[24] This mutual interdependence of norms can be explained by the fact that what is really being normalized is language. Normalization begins with vocabulary:

The first thing to develop when one begins to examine a particular problem is a means of specifying terms precisely: there must be a single term for each thing and a single meaning for each term. . . . Equally, since the elaboration of norms clearly determines technical terminology, product nomenclature, and the forms of symbolic representation in diagrams, establishing a set of standard terminological points of reference is the first task in the process of normalization.^[25]

This normalization of vocabulary extends even to systems for notation and writing: the signs and locutions that characterize common usage are less than ideal for the purpose of precise technical expression. Words are soon joined by numbers and drawings that are themselves normalized.

Linguistic normalization also encompasses syntax. The language of normalization has its own grammar and logic. This normalized language serves a specific mode of thought that it must both call forth and translate: it comprises a style of analysis, and a way of categorizing and breaking down objects, tasks, and needs, at once segmenting them and individualizing them. At the same time that it fabricates a language, normalization serves as a principle of objectification and a producer of objectivity.^[26] This artificial language serves to prevent ambiguities. It is a language of precision and certainty, a language without puns, stylistic figures, or interference—the language of perfect communication. However, the language of technical normalization also implies the institution of a new relationship between words and things.

Fundamentally, normalization is the process of turning this language, with its vocabulary and syntax, into a common language, a general principle of communication that functions in much the same way as the system of thought that it expresses. This language must function not only within the limited context of a single industry but within the sphere of relations between various industries, and in the field of relations between producers and consumers. Within this language, the demands of buyers, sellers, producers, and consumers must all be expressed, refigured, and readjusted with respect to one another. Normalization establishes the language that allows these different groups to understand one another and to form a society.^[27] Its central projects are to institute new common standards of measurement while searching for appropriate rules of analysis and expression, and to teach this language to all those who are involved in one way or another in the system of economic exchange.

Normalization is thus the institution of the perfect common language of pure communication required by industrial society. But what will this language say, and what will be its content? What makes these new norms anything more than a precise and perfected version of communicational tools that have already been in existence for some time? How does one evaluate the requirements of this language, and its performance in communication—in other words, what is the norm for industrial normalization? One of the specific characteristics of technical normalization is that requirements and performance are defined according to principles of relativity and solidarity. In terms of industrial normalization, the measure of a productive norm is a norm for consumption and vice versa.^[28] Normalization forces each individual to imagine the ordering principle behind his activity not only with respect to some ideal of perfection that he might attain in isolation (such an ideal isolation has no meaning in a normative system), but with respect to a determined need that must be satisfied. Normalization is a means of organizing that solidarity which makes each individual the mirror and measure of his fellow.

For Harriman, "The idea of perfection is not involved in standardization."^[29] In place of perfection is the "one best," or the relative best, with reference to

industrial capabilities and economic capacities, specific uses and requirements. Normalization is a means of assigning value that renders absolute standards of perfection meaningless.^[30] The good is figured in terms of adequacy—the good product is adequate to the purpose it was meant to serve. Within the normative system, values are not defined a priori but instead through an endless process of comparison that is made possible by normalization. The norm is the best relational principle. It expresses a compromise: the compromise among norms in accord with the general principle of normative solidarity, the compromise between technical capability and industrial capacity, or the compromise between production and need. Technical normalization, or standardization, provides an example of valorization that makes no reference to universals, where equilibrium has replaced the absolute as the value of values. A standard may become stable or regular, but it is only temporarily so. The standard is a form of compromise, the common denominator, a point of reference that is destined to disappear—a measurement that expresses the relation of a group to itself, even that of a group as large as the entire population of the globe.

These different properties become apparent once again in the procedures for establishing standards. Standardization is not a form of legislation, nor is it a process that can be carried out by decree. In other words, standardization is not a state function. Rather, it presupposes the creation of associations where all interested participants—producers, consumers, engineers, scientists—can negotiate the common standard according to their respective requirements. There is a kind of democracy specific to the standardization process.^[31] In general, this democracy functions on two levels: on the first are the organizations that represent the various kinds of productive activity and are competent to decide upon standards; on the second are the organizations of standardizing associations that verify the compatibility of various norms among themselves according to the principle that "standards must form a perfectly unified whole."^[32] This principle makes standardization into an infinite task.

Discipline, insurance, and standardization ought not to be conflated. However, all three practices can be subsumed under the term norm . How can we think about the relationships between them, then? How might comparison of them clarify our conception of the norm? It is worth noting at the outset that each of these three sets of practices is marked by a tendency to relentless proliferation: discipline becomes normative as it becomes generalized and as it shifts from negative to positive functioning, and the logic of the norm is what makes this generalization possible.

Ever since its legally problematic debut at the start of the nineteenth century, the insurance industry and its spheres of influence have expanded almost incessantly. Today there is scarcely a social problem that is not dealt with in terms of

risk: public hygiene, health issues, pollution, social maladjustment, and delinquency have all come to make sense in insurance terms. Social security has also helped to make insurance the essential form of social relations. Technical normalization, too, seems to require extensive development: normalizing a product means normalizing both production techniques and the needs to be satisfied by production.

These normative procedures are implicated in a process of expansion that only stops when each has exhausted the possibilities for further extending its jurisdiction. They are also related to one another, however, in such a way that each pulls the others into a kind of normative spiral. Modern techniques for managing accidents in the workplace provide a good example of this: the insurance industry's way of coping with accidents resulted in the birth of a science of workplace safety, ergonomics, which is clearly related to the development of scientific organization and management principles. The demands of social hygiene certainly benefited the industrialization of construction, which was itself taken over and encouraged by the development of construction insurance. We might also chart the structure of normative networks, and in so doing we would gradually come to see how a norm on one level is related to a norm on another, a safety norm to a normative level of performance, for example, or a disciplinary norm to a productive norm, or a productive norm to a norm of population. "Norms," explains Canguilhem, "are relative to each other in a system, at least potentially. Their correlativity within a social system tends to make this system into an organization, that is, a unity in itself, if not by itself and for itself."^[33]

Just as norms can only exist socially, there can be no such thing as a norm that exists in isolation, for a norm never refers to anything but other norms on which it depends. Norms communicate among themselves, shifting from one level or field of their existence to another according to a kind of modular logic. The norm finds meaning only in relation to other norms: only a norm can provide a normative value for another norm. The paradox of the norm is that before one can exist, there must already be another. If a norm exists, the entire space in which it appears becomes a normative space. Thus it would be an error to say that The History of Sexuality , which extended the scope of the normative to the state and the populations within its jurisdiction, continues or completes Discipline and Punish , which merely situated the normative at the level of discipline. This displacement is part of the logic of the norm. When the norm appears, it establishes itself necessarily as an order: the normative order that characterizes modern societies.

This correlative quality of norms provides us with a methodological insight: it is essential to distinguish between the norm itself and the apparatus, institution, or technique of power that brings it into action and functions according to its principles. The norm (or the normative) is no more specific to discipline than it is to insurance or standardization. The norm in particular cannot be characterized as the exertion of a punctilious or minute form of power or imagined in

terms of the microphysics of its engagements. Norms are linked neither by scale (macro or micro) nor by the characteristics of their objects, whether they are bodies, populations, or things. Hence the ubiquity of the normative, which can no longer be confused with the exercise of power that it informs. If power is exerted according to a set of physical constraints, the norm falls within the province of a metaphysics of power.

What, then, is a norm? It is a way for a group to provide itself with a common denominator in accordance with a rigorous principle of self-referentiality, with no recourse to any kind of external reference point, either in the form of an idea or an object. The normative process can obey a variety of different logics: the panoptical logic of discipline, the probabilistic schema of insurance, or the communicative logic of the technical norm. These three logics have the same form: in each case, the rule which serves as a norm, by virtue of which everyone can measure, evaluate, and identify himself or herself, will be derived from those for whom it will serve as a standard. A strange logic, this, which forces the group to turn back in upon itself and which, from the moment it establishes itself, will let no one escape its purview.

The norm implies a rule of judgment, as well as a means of producing that rule. It is a principle of communication, a highly specific means of resolving the problem of intersubjectivity. The norm is equalizing; it makes each individual comparable to all others; it provides the standard of measurement. Essentially, we are all alike and, if not altogether interchangeable, at least similar, never different enough from one another to imagine ourselves as entirely apart from the rest. If the establishment of norms implies classification, this is primarily because the norm creates classes of equivalency.

But the norm can also work to create inequalities. This is, in fact, the only objectivity that it provides: the norm invites each one of us to imagine ourselves as different from the others, forcing the individual to turn back upon his or her own particular case, his or her individuality and irreducible particularity. More precisely, the norm affirms the equality of individuals just as surely as it makes apparent the infinite differences among them. The reality of normative equality is that we are all comparable; the norm is most effective in its affirmation of differences, discrepancies, and disparities. The norm is not totalitarian but individualizing; it allows individuals to make claims on the basis of their individuality and permits them to lead their own particular lives. However, despite the strength of various individual claims, no one of them can escape the common standard. The norm is not the totality of a group forcing constraints on individuals; rather, it is a unit of measurement, a pure relationship without any other supports.

Normative practices, based on the notions of equality and the common standard, are compatible with the existence of a certain kind of law. The normative allows us to understand how communication remains possible even within a historical moment characterized by the end of universal values. The norm is a means

of producing social law, a law constituted with reference to the particular society it claims to regulate and not with respect to a set of universal principles. More precisely, when the normative order comes to constitute the modernity of societies, law can be nothing other than social.

This kind of law possesses two remarkable qualities: first, it is no longer based on a model in which the law emanates from a sovereign will. In a normative order, there is no room for the sovereign. No one can pretend to be the subject that establishes the norm; norms are created by the collectivity without being willed by anyone in particular. The norm is the group's observation of itself; no one has the power to declare it or establish it. Undoubtedly, the norm gives the group a certain sovereignty over itself, but that sovereignty does not derive from a contract. Although it presents itself as an expression of the general will, legislative sovereignty within the normative order is mere appearance, a form or fiction necessary to ensure the community's respect for the common standard.

Secondly, although there can be a parliament within a normative order—and practically speaking, there are usually many, since they have a tendency to multiply—there is no legislator. This position is empty, and definitively so. The law is no longer valid as an expression of the general will or the common interest. Rather, it is valid by virtue of its normative quality. Parliament no longer establishes the fundamental principles of law; it can only set forth regulations. A normative economy of obligations allows us to imagine a law without obligations or sanctions. The supporters of technical normalization have made it amply clear that whenever a regulation is propounded, a norm has been negotiated. The validity of a norm derives from the fact that it is not imposed from outside but that it observes itself without requiring obedience. Within a normative space, constraint is more of an obstacle than an aid. At the United Nations, for example, arguments have been made for "resolutions" and "recommendations" that do not have the binding force of treaties and serve instead as points of reference for evaluating the conduct of states. Of course, they are most effective when they express a consensus. More precisely, these resolutions and recommendations are the expression of norms. The norm eliminates within law the play of vertical relations of sovereignty in favor of the more horizontal relations of social welfare and social security.

The norm, then, is a means of producing the common standard, a rule for common judgement that makes law possible in modern societies. It functions within the bounds of three defining conditions.

The first involves the constitution of a homogeneous field of positive values. The norm makes visible and records only the sheer phenomenality of phenomena. The normative gaze does not seek to penetrate to the inner substance of things. Instead, it remains on the level of pure facticity, never going beyond

this to attain a deeper appreciation of its objects. Facts are sufficient in themselves; they simply exist, neither as appearance nor as essence. Processing them is not a question of unmasking them or interpreting them because for the normative way of seeing, a fact refers to other facts and not to an original cause. Thus, in the normative order one simply moves from one visible surface to another, indefinitely.

This positivism based on pure facts is fundamental to the normative order. It allows the norm to appear as both a principle of objectivity and as a common language. If there were a norm for the norm, it would be this rigorous positivism of exteriority. The normative institution presupposes a dual decision that functions in both a negative and a positive sense. In its negative formulation, the language of the norm assumes that it is always possible to distinguish facts from their interpretations. From the normative point of view, all interpretation is subjective; all explanation, opinion, and theory are simply forms of metaphysics. In its positive formulation, the normative allows for communication that is independent of all philosophical or religious conviction. Within the realm of the norm, faith and knowledge are definitively separated. The category of the normative itself presupposes the creation of a purely descriptive language in which syntax and vocabulary would always succeed in containing the slippage of meaning that occurs in metaphor. If the possibility of secular politics is founded on the constitution of a sphere of objective interpretation, then the norm is eminently secular.

As I have suggested earlier in this essay, norms are anything but natural; facts are never simply given. It is essential to distinguish between facts and their interpretations, and statistical probability has played a major role in establishing this distinction. In this respect, the science of statistics resembles the language of the norm both in its vocabulary and its syntax. It also functions as a common language because it produces objectivity independent of any doctrine. Of course, there can be no objectivity without objectification. Statistics and probability are techniques of objectification that produce facts whose objectivity, liberated from all metaphysics, can function as a common language. This is not to say that either one is neutral, or that there are such things as pure facts, but simply that they create the possibility of objectivity.

The second characteristic of the norm is its relativity . This directly contradicts the idea that the norm represents some kind of absolute . A norm is a self-referential standard of measurement for a given group; it can make no pretense to bind anyone for an indefinite period, as a law can. This is not to say that norms are ephemeral, for they are enormously durable. But they are also inconstant, almost by definition. In the eyes of the business community, this capacity for adaptation and flexible response to changing conditions makes normalization superior to laws or regulations as a management technique. Part of the norm's value derives from the fact that it is so completely time-bound.

Similarly, the norm can never be universal. Ever since the work of Emile

Durkheim, both sociology and cultural ethnology have repeatedly returned to the idea that the validity of a norm can never extend beyond the bounds of the group which that norm describes. As Durkheim notes, "A fact can be termed pathological only in relation to a given species. . . . It can only be termed normal in relation to a particular phase, likewise determinate of its development."^[34] The same holds true for biological norms: standards of health are not the same for everyone, and those suitable for adults would be less so for small children or elderly people. Because norms are relative, it makes no sense to apply a particular set of norms established for one group to other unrelated groups or cases.

The relativity of norms has often been interpreted to their detriment, as though the fact that normative rules bear the marks of their historical context were enough to make them invalid. Relativity does not necessarily imply relativism. If a norm's sphere of validity cannot extend beyond the bounds of the group that establishes it in the first place, this is precisely because norms are neither equivalent nor interchangeable. In short, there is system of valorization specific to norms that is altogether unrelated to the Kantian criteria for value.

Finally, norms involve polarity. Canguilhem has commented that the relationship between the normal and the abnormal is not "a relationship of contradiction and externality, but one of inversion and polarity."^[35] As we have already observed, the abnormal is not outside the realm of the normal; the division between the normal and the abnormal occurs on the basis of inclusion rather than exclusion. However, if the normal and the abnormal can only be distinguished along a continuous spectrum of possibilities, are real distinctions even possible?

This is the biological problem of deciding on the status of the anomalous. Recalling Canguilhem once again, we are reminded that anomalies are part of the normal in much the same way that mutation is an essential part of biological life.^[36] Just as in statistics, there are never any real constants—only differences of various sorts. But if the norm is based on variation, how can we describe one particular sort of variation as abnormal? Biological anomalies can be considered abnormal less because they diverge from an a priori model of their type than because the anomalous individual will experience his or her difference as a handicap or obstacle in the business of life. If all possible forms are not normal, it is not because some forms are naturally impossible but because the various possible forms of existence are not all equivalent for those who must exist in them.^[37] The separation between the normal and the abnormal occurs at the point in the relationship between a living entity and its environment where equilibrium is completely disrupted, and the distance between environmental requirements and individual performance becomes too great. If environmental requirements change, performance does too, and along with them the location of the boundary between the normal and the abnormal.

We can also explain the normative assignment of value in politics on the basis of these ideas. Social groups impose demands of various sorts (e.g., industrial or

educational demands) that serve as the standard by which individual performance is measured and that allow individuals to be classified and placed in a hierarchy. Abnormality is defined as a handicap or inability. It no longer refers to a natural quality or property of being; instead, it signals some aspect of the group's relation to itself. The relationship between the normal and the abnormal thus becomes an unstable threshold. At the same time, the political stakes in the fixation of this boundary become increasingly apparent. Opposition to a particular technique and the demands associated with it implies a will to modify the threshold for exclusion. Inversely, debate over the frontiers of the normal and the abnormal is meaningless in the absence of some effort to alter the social conditions that have produced the boundary.

The normative society is a strange one: like all other societies, it excludes various individuals and groups, but its tactics for exclusion in no way imply any kind of natural prejudice. It has its own demands, which are never natural and always social. In normative societies, political life is always highly polemical and concerned primarily with the establishment of a balance between the various claims of individuals and groups, a stable social state. The achievement of specific ends is less important than the maintenance and negotiation of this state, since in the normative society social good and stability are one and the same.

To the extent that norms are unstable, one might object that they cannot function either as a common standard or as a precondition for law. After all, how can a rule serve as a common reference if it is constantly changing and can offer no security to those who will have to make decisions based on it? Doesn't a rule have to be fixed, unchanging, and outside the influence of those who are going to use it? Can a law whose rules are constantly changing still be considered a law?

The norm is that which, as a rule, is least arbitrary. The evidence provided by averages and statistical regularities suggests that normative objectivity does exist, at least for particular moments and particular situations. Certain social facts do recur reliably in obedience to a sociological law of inertia which can be read as proof that the life world has found its equilibrium in a specific normative identity. A priori, a normative order may seem to be constructed as if anything were possible. However, even if we do believe that everything is possible in law, in practical terms the possibilities are predetermined and relatively limited. The rationality of the norm has introduced us to a kind of positivistic pragmatism that cannot be grasped in absolute terms. What we must understand is that there is no need to impose a law on the living in order to ensure regularity in their behavior.

I have attempted to elucidate Foucault's rather enigmatic claim at the end of The History of Sexuality that, as a result of the rise of bio-power, "we have entered a phase of juridical regression in comparison with the pre-seventeenth-century societies we are acquainted with."^[38] This remark might well be construed to mean

that there has been some kind of decline of law and legality in modern society.

However, Foucault's project was neither to announce the imminent disappearance of law, nor to criticize bio-power in the name of law. Foucault was concerned less with the place of law in the exercise of power (that is, whether bio-power is compatible or not with the exercise of law) than with the use of law as a "model" for analysis, a principle that makes power intelligible.^[39] Foucault was interested in the relationship between the juridical and the political, and between power and law, as a means of determining the conditions necessary for an analysis of the mechanisms of power. Power does not necessarily function through law. Instead, law serves to camouflage the machinations of power. An adequate description of monarchical power would have to include law, for law is the language the monarchy provides itself with in order to legitimate its own exercise of power. However, in the case of bio-power, any reference to the juridical is illusory, since the language of bio-power is purely technical and has almost nothing to do with the law as such. Foucault's analysis leaves open two questions: first, if the juridical is an inappropriate category to use in interpreting bio-power, how do we make sense of all those "instruments of the law" (codes, constitutions, laws, regulations) that have developed and expanded during the era of bio-power? Second, if the action of norms replaces the juridical system of law as the code and language of power, what role remains for law?^[40]

I have tried to argue that the contemporary legal apparatus is not coterminous with the juridical, as Foucault describes it, and that the normative and the juridical are essentially opposed. Further, I have attempted to delineate the structure of the normative on the basis of two examples: insurance and industrial standardization. I have broadened the definition of the norm to include that form of the common standard produced through the group's reference to itself and demonstrated, finally, that law cannot be understood simply in terms of its formal expressions (constitutions, codes, laws). These must all refer back to what functions in society as a common standard, a normative and objective basis for judgment. Thus we can now imagine a history of the law in which law is no longer conceived in essentialist terms, and an account of the inevitable decline of essentialist law becomes unnecessary.

In both the insurance system and industrial standardization, the norm appears as a technique for the production of a common standard of measurement. No society can exist without something akin to this common standard, a common language that binds individuals together, making exchange and communication possible. The norm is one part of a long history of the common standard, a lesser instance of a larger category. This articulation of the norm and the common standard opens up a variety of research perspectives, allowing us to explore modernity in terms of measurement techniques and standards. Societies

become modern at least partly by virtue of transformations in their instruments of technical, political, and social measurement. What did the French Revolution bring about, after all, if not an enormous transformation in systems of measurement? The introduction of the metric system, the institution of a truly national language, calendar reform, and the creation of the Civil Code are all examples of this. Similarly, the institution of constitutional democracy was a means of producing a common political standard. One might also read the history of the social sciences in the nineteenth century as the formation of so many instruments intended to furnish modern societies with social and political measurements. Thus we might well assess social modernity in terms of the transformation a given society may have experienced in its techniques of measurement.

—Translated and adapted by Marjorie Beale

Law, Boundaries, and the Bounded Self

Jennifer Nedelsky

In the American tradition of constitutionalism, property has served as a powerful symbol of rights as limits to government. This notion of "rights" functioning as "limits" to "government" involves a complex set of abstractions and metaphoric links that nevertheless is taken as common sense by most Americans. Perhaps the clearest form of that common sense is "Government can't take what's mine" or the more elegant "A man's home is his castle." These phrases convey an image of property as a source of security whose sacredness acts as a barrier even to the power of the state. The enduring power of this image reflects (among other things) the original importance of property in shaping the American conception of rights as limits to the legitimate scope of the state. This conception is, in turn, a part of a deeper phenomenon: the focus on boundaries as the means of comprehending and securing the basic values of freedom or autonomy. The importance of property in American constitutionalism both reflects and exacerbates the problems of boundary as a central metaphor in the legal rhetoric of freedom.

I have argued in Private Property and the Limits of American Constitutionalism that in the early development of the American Constitution property served as the defining instance of the notion of rights as limits, and that this notion is an attempt to address the inevitable tension between the individual and the collective.^[1] I concluded that the attempt is deeply flawed and that the problems go beyond the particular limitations of the focus on property. In the end, the new vision of constitutionalism we need requires something more than the replacement of property with other rights that can serve as boundaries. We need a new conception of the tension between the collective and the individual, for which boundary is not an apt metaphor. My purpose here is to explore that rejection of boundary.

Drawing on this earlier work, I start by discussing the American Constitution as an instance of the ways boundary has been central to our conceptual and institutional framework and give an account of how the Constitution came to have the boundarylike structure that it does and of what role property played in the emergence of that structure. I offer a critical appraisal of the lasting consequences of the original focus on property that is within the conventions of constitutional discourse. My argument leads, however, to the claim that the particular limita-

tions of the focus on property are themselves reflective of the deeper problems that made property an attractive focus in the first place: a picture of human beings that envisions their freedom and security in terms of bounded spheres. I then begin to explore why boundary is destructive as a central metaphor for addressing the real problems of human autonomy.

I turn next to what I came to see as problematic about this view of boundary as destructive: the boundary metaphor is also used in other discourses, ones that are far removed from the traditions of legalism and, more important, ones I trusted and respected. I look at examples of these other invocations of boundary and the possibility of envisioning boundaries not as the rigid walls of property but as fluid, shifting, and permeable. But I conclude that the boundary metaphor still provides the wrong focus for exploring the best possibilities of human autonomy. We need a language of law whose metaphoric structure highlights rather than hides the patterns of relationship its constructs foster and reflect. Finally, I consider the emergence of new mythic structures, new visions and metaphors that may provide this necessary alternative.

I

One of the ways of understanding both the power and the perversity of American constitutionalism is to examine its origins. In my earlier work I discussed the importance of property in the design of the Constitution of 1787 and the subsequent rise of judicial review. Here I will just give the core of my argument, particularly my claims about the distortions that the focus on property both expressed and gave rise to.

During the 1780s the newly independent confederation of states seemed to pose problems for which the revolutionary conceptions of politics were inadequate. The new governments had solved the problem of "no taxation without representation," but many men of property felt that their rights were still not secure. During the Revolution they had proclaimed that "a man is a slave if his property can be taken without his consent." By the 1780s that consent had been ensured through representation, but the new representative governments were themselves threatening property. The depreciating currency and debtor relief laws promulgated by virtually every state were widely seen as violations of property rights—and thus illegitimate even though passed by duly elected representatives. For many of the great statesmen of the time, the threats these laws posed came to be understood as part of a more general problem. Their important (and now trite) insight was that democracy solves some problems of tyranny but brings its own: the tyranny of the majority.^[2]

The forms of representation that the colonists had demanded to secure the rights of Englishmen turned out to threaten those very rights. Property in par-

ticular turned out to be just as vulnerable to debtor relief measures in the republic as it had been to unjust taxation under the monarchy. Indeed, many thought his would always be so: property would be inherently vulnerable in a republic because the many would always be poor and the few rich; what would prevent the many from using their numerical power in the legislature to take the property from the few? The problem of designing a republican government that could provide security for property was a central one for the Federalists, whose views prevailed at the Constitutional Convention of 1787. It was an immediate problem that had to be solved if the republic was to survive, and, for the most thoughtful Framers, it became the defining instance of the larger problem of securing rights against the threat of majority oppression.^[3] Ironically, it is because this original preoccupation with property was not limited to a crass concern with protecting the interests of the rich that it has had such a lasting and destructive legacy. Originally invoked as the defining instance of the larger problem of securing justice and liberty in a republic, property indeed came to define the terms of that problem for at least one hundred fifty years.^[4]

There were many complex consequences to this original focus on property. To begin with, the problem of property arose for the Framers because their conception of it was inseparably tied to inequality. The link to inequality was liberty. Property was important for the exercise of liberty, and liberty required the free exercise of property rights; this free exercise would inevitably lead in turn to an unequal distribution of property. Property thus posed a problem for popular government because this inequality required protection; those with property had to be protected from those who had less or none. Without security, property lost its value. And the threat to security was inevitable, for (the Framers presumed) it was in the very nature of a productive system of private property that many, perhaps most, would have none.^[5] It was this inequality, which the Framers both feared and accepted as natural, that skewed their conception of republican government.

The republic the Framers envisioned required the security of a right to which the majority posed a constant and inevitable threat. Defense against such a threat was a problem quite different from that arising from the general insight that in a republic the majority may oppress the minority. It is one thing to say that everyone's rights are vulnerable to the possibility of majority oppression. It is another to say that an essential ingredient of the republic is the protection of rights that the majority will never fully enjoy, will always want more of, and will therefore always want to encroach upon. The vulnerability of property bred a fear and, perhaps, contempt of the propertyless, who were to be the vast bulk of the people. With this fear came an urgent sense of the need to contain the people's threatening power.

Given the historical context, we can understand both the focus on property and the insistence that it stood for broader issues and deeper values. The Framers

who focused on property were not crass materialists of either the self-interested or philosophical variety; they were not devoted to property for its own sake. Their preoccupation with property had its origins in the connections they presumed between property and other basic human goods, in particular liberty and security. Property came to be not merely a link to those basic values but the symbolic focal point for the effort to make republican government compatible with both liberty and security. Property was a powerful symbol of these goals because it could both crystallize complex aspirations and problems and provide a practical focus for dealing with them. Property was an effective symbol in part because it was not merely a symbol but a concrete means of having control over one's life, of expressing oneself, and of protecting oneself from the power of others, individual or collective. In addition, the need for security in order for property to serve those purposes expressed the important link between security and liberty: although the two values are not the same, and can be in tension with one another, some level of security is necessary for liberty to have meaning.

Property effectively captures this link between liberty and security in that it literally loses its meaning without security. We mean by property that which is recognized to be ours and cannot be easily taken from us—hence the connection between property and what are seen as the sources of its security (and thus, in part, its meaning): law and government. Property is a right that requires collective recognition and enforcement. In ways quite different from freedom of conscience, it requires the involvement of the collective for definition and defense and thus is peculiarly vulnerable to collective power—at the same time that one of the basic purposes of property is to provide a shield for the individual against the intrusions of the collective. Property defines what the society, or its representative the state, cannot touch (in the ordinary course of things). It defines a sphere in which we can act largely unconstrained by collective demands and prohibitions. But the definition and protection of that sphere must reside with the collective itself. Property thus captures the essence of the problem of self-limiting government. But a focus on property also provides a distorted image of the problem.

Property (at least as the Framers understood it) must distort because it makes inequality rather than liberty, or individual autonomy, the central problem of government. (Note that what follows is not an argument that there must be equality of property. It is an argument about what happens when property, conceived of as inherently unequal, becomes the central symbol for the protection of individual rights.) The Federalists presumed that the threat inequality posed to property captured the inevitable threat democracy posed to individual freedom and security. But the Federalists were wrong: not all rights, all components of liberty, must be enjoyed unequally. Perhaps even property need not be enjoyed unequally, but the Framers' conception of property had inequality—and thus fear, anger, and resentment—built into it. The centrality of inequality skewed the Federalists' comprehension of the basic problem of republican government. They

failed to see that protecting the rights of the propertied few against the demands of the many is not the same as protecting individuals from the ever-present possibility of collective oppression. The cost of the Framers' insight is that they cast the general problem in the terms of the particular—and we have continued to think in those terms.

The tension between the individual and the collective does not inevitably concern inequality, power, and domination, but the problem of protection of unequal property does. In accepting vast economic inequality as a given, and the contours of property rights as obvious, the Framers were in fact focusing on protecting the few from the many, not the individual from the collective of which he or she is a part. The Framers' focus on property turned attention away from the real problem: the need for an ongoing collective formulation of individual rights in a political culture that respects both democratic decision making and individual freedom, and that recognizes the need to sustain the inevitable tension between them. The Framers recognized this tension, but they were preoccupied with one dimension of it: insulating property from democratic decision making. Rights became things to be protected, not values to be collectively determined. The most thoughtful Framers transformed a widespread fear about threats to property into a sophisticated analysis of the inherent problem of majority oppression. But in doing so they also transformed this general problem into a question of how to contain the power of the people.

The 1787 Constitution did indeed "insulate" property rather than set up rigid boundaries to protect it.^[6] Its solution was a carefully structured system of institutions that would minimize the threat of the future propertyless majority in large part by minimizing their political efficacy. This undermining of democracy was justified in part by an articulation of competing categories of civil and political rights. Civil rights, with property as the leading example, were the true ends of government; political rights were merely means. It was therefore no sacrifice of basic principle to tailor the means to the end. But this conceptual hierarchy remained institutionally fluid in the Constitution of 1787; there was no clear answer to how the priority of civil over political rights was to be enforced. There were few formal declarations of rights as limits to the power of the federal government (or as we would put it today, few rights specified as limits to the legitimate scope of democratic decision making).^[7] Rather, the Constitution the Framers wrote protected civil rights in more subtle ways by channeling the power of the people in order to minimize their threat to property and civil rights generally.

By 1800 this channeling seemed insufficient to contain the threats posed by Jeffersonian democracy. With the rise of judicial review the conceptual hierarchy of civil over political rights hardened into an institutional one: the judiciary wielded the power to strike down the outcomes of the democratic process in the name of constitutionally protected rights. Those rights were proclaimed as clear

boundaries to the legitimate power of the state, boundaries to be defined and defended by the judiciary.^[8]

We are now coming to my central point. The vision of constitutionalism sketched above—the tension between democracy and individual rights, the hierarchical distinction between civil and political rights, the notion of rights as boundaries—not only has the dark underpinning of inequality but rests on a flawed conception of the individual, a conception captured, amplified, and entrenched by its association with property. The boundaries central to American constitutionalism are those necessary to protect a bounded or "separative" self: the boundaries around selves form the boundaries to state power.^[9] Now, the boundedness of selves may seem to be a self-evident truth, but I think it is a wrong-headed and destructive way of conceiving of the human creatures for whom law and government are created.

II

Much of our constitutional protection can be understood as a (misguided) attempt to protect individual autonomy. The primary content of this underlying conception of autonomy is protection from the intrusion of the collective. The autonomy the American system is designed to protect can be achieved by erecting a wall of rights around the individual. Property provided an ideal symbol for this vision of autonomy, for it could both literally and figuratively provide the necessary walls. The perverse quality of this conception is clearest when taken to its extreme: the most perfectly autonomous man is the most perfectly isolated.

Everyone is familiar with, or at least would immediately recognize as intelligible, the image of rights as boundaries defining the sphere within which human autonomy (or freedom or privacy) resides.^[10] Certainly within Anglo-American legal theory that image is routine. These images abound, for example, in Charles Reich's famous "new property" article in which he tries to expand the traditional meaning of property to cover government largess:

The institution called property guards the troubled boundary between individual man and the state. . . . Property draws a circle around the activities of each private individual or organization. Within that circle, the owner has a greater degree of freedom than without. Outside, he must justify or explain his actions, and show his authority. Within, he is master, and the state must explain and justify any interference.^[11]

Reich wants to redefine property so that it can continue to define protective spheres, but ones that are appropriate for a regulatory state.

Boundary images have a powerful hold even over those who have some sense of their limitations. Laurence Tribe, for example, tells us that we cannot rely

solely on spherelike conceptions of human autonomy: "Meaningful freedom cannot be protected simply by placing identified realms of thought or spheres of action beyond the reach of government. . . . Ultimately, the affirmative duties of government cannot be severed from its obligations to refrain from certain forms of control." But then Tribe reaches immediately for the sphere metaphor even as he is trying to suggest the difficulties of defining it: "The tight interdependence of persons in advanced industrial and post-industrial societies precludes any attempt to define the protected sphere in terms of conduct that is wholly self-regarding or needs that are wholly personal."^[12] And he soon moves to repeated references to the "invasion" of personality.^[13]

Not surprisingly, the image of bounded space is invoked particularly often in discussions of privacy. For example, Robert Post writes that the tort of intrusion "guards against the penetration of private space." This image explicitly draws on boundary metaphors found in nonlegal discourse, for example Georg Simmel's image of "the ideal sphere [that] lies around every human being" and that "cannot be penetrated, unless the personality value of the individual is thereby destroyed." Post then makes use of Erving Goffman's metaphor of the "territories of the self": "We indicate respect for a person by acknowledging his territory; conversely, we invite intimacy by waiving our claims to a territory and allowing others to draw close."^[14]

The examples from Post suggest far more than the success of American constitutional images; they suggest that the image of protective boundaries as essential to the integrity and autonomy of the self is deep and pervasive in our culture. Must my challenge to the rhetoric of boundaries then extend beyond the peculiarities of the property-based boundary forms of American constitutionalism? Yes—at the least because, as I noted earlier, I do not accept the position that we should simply remove the perversions of the original focus on property by replacing it with other rights to serve as boundaries, and more broadly still because the boundaries of American legal discourse effectively capture a wider (and deeper) phenomenon. So I must pose my question more broadly: What is wrong with boundary imagery?

Let us start by focusing on autonomy again, for that is the aspect of the individual self for which both law and other discourses routinely invoke boundary. As I see it, individual autonomy is a capacity, not a static human characteristic to be posited as a presupposition of legal or political theory.^[15] This capacity must be developed; it can flourish or become moribund. What is essential to the development of autonomy is not protection against intrusion but constructive relationship. The central question for inquiries into autonomy (legal or otherwise) is then how to structure relationships so that they foster rather than undermine autonomy. The boundary metaphor does not direct our attention to this question. Instead it invites us to imagine that the self to be protected is in some crucial sense insular, and that what is most important to the preservation of such a self is

drawing boundaries around it that will protect it from invasion (or at least that that is the most crucial thing the law can do).

I should make clear before proceeding further that all metaphors inevitably distort or shape our perceptions in the sense that they hide some dimensions of the phenomenon they refer to and highlight others. And it is, further, not an option simply to avoid metaphors. I accept George Lakoff and Mark Johnson's view that all our thinking is metaphoric in structure.^[16] There is in that sense nothing peculiar or perverse about the recourse to metaphor to explore the issue of autonomy. The problem arises exactly because metaphors are so important in how we experience the world. (For example, Lakoff and Johnson open their book with a compelling picture of how the dominant metaphor of war structures our experience of what argument is and how we ought to engage in it.) It is therefore crucial that the reigning metaphor for autonomy and other basic aspects of selfhood be one that directs our attention where it belongs.

What actually makes human autonomy possible is not isolation but relationship, first with parents, then teachers, friends, and, potentially, agents of the state (although, in fact, many relationships with the state, such as the receipt of welfare, are structured so that they undermine rather than foster autonomy). Practices of childrearing may therefore provide fruitful images and insights into the nature of the relationships that foster autonomy. The image of the child developing autonomy teaches us not to mistake interdependence and even dependence as antithetical to autonomy. That image is, of course, very different from and far more helpful than that of a man securely ensconced on his property, whose boundaries are well guarded against intrusion.

There are practical virtues to better imagery. If we understand autonomy as made possible by relationship rather than by exclusion, we can better understand the genuine problem of autonomy in the modern state. Our central problem today is not maintaining a sphere into which the state cannot penetrate but fostering autonomy where people are already within the sphere of state control or responsibility. The problem is best addressed by focusing on how the interactions between state and citizen are structured. And with that question in mind we can look, for example, to the practices of administrative law and bureaucratic decision making for both guidance and warning. More broadly, we change our whole conception of the relation between the individual and the collective when we see that the collective is a source of autonomy as well as a threat to it. We would no longer take "a boundary guarded by property" or "invasion" as suitable metaphors to guide us in designing the legal structures to optimize that relationship.

I want to offer another, very different perspective that reveals the perversity of conceiving of selfhood and autonomy in boundary terms. Andrea Dworkin has a powerful and disturbing indictment of the meaning of sexual intercourse in our society, which, as she interprets it, is inseparable from domination. What is striking, and important for our purposes here, is that her analysis assumes the

understanding of humanness as boundedness. Since that boundedness is either nonexistent or routinely violated in women, they are less than human. Here are examples of her vision:

In the experience of intercourse, she loses the capacity for integrity because her body—the basis of privacy and freedom in the material world for all human beings—is entered and occupied; the boundaries of her physical body are—neutrally speaking—violated.

A human being has a body that is inviolate; and when it is violated, it is abused. A woman has a body that is penetrated in intercourse: permeable, its corporeal solidness a lie. The discourse of male truth—literature, science, philosophy, pornography—calls that penetration violation. . . . She, a human being, is supposed to have a privacy that is absolute; except that she, a woman, has a hole between her legs that men can, must, do enter. . . . That slit which means entry into her—intercourse—appears to be the key to women's lower human status. By definition, as the God who does not exist made her, she is intended to have a lesser privacy, a lesser integrity of the body, a lesser sense of self, since her body can be physically occupied and in the occupation taken over.^[17]

Dworkin vacillates in her argument between suggesting that intercourse must have this meaning because of its essential nature as invasion and suggesting that in our society this is the inevitable social construction of the meaning of intercourse. I think the social construction version is compelling (although I do not accept the claim of inevitability). It also shows us the deep perversity of making boundedness the central metaphor for humanness. When boundary is the routine metaphor for integrity, privacy, or the conditions of autonomy, it is indeed not surprising that intercourse "has in it, as part of it, violation of boundaries, taking over, occupation, destruction of privacy."^[18] It is not enough to meet the force of Dworkin's images with the conventional argument that it is only nonconsensual boundary crossing that constitutes violation. Of course, she has reasons why we should doubt that intercourse is every fully consensual. But her point goes deeper than that. Consensual invasion is an odd concept, and if boundary is central to our conception of self, it will not be surprising that intercourse comes to be seen and experienced as invasion or violation, even if consented to. When boundary is central and intercourse is violation, women come to be seen and experienced as something less than fully human. (And of course women's boundaries seem indistinct in other ways: they blur with nature, with their children, their families, their lovers. Their contours do not seem hard, clearly defined or well protected.)

Our project should not be to try to shore up women's boundaries (as opposed to a genuine sense of self, which does need strengthening), nor should it be, at least in the long run, to find ways to draw circles of protection around women that are the same as men's. (It is sometimes revealing, however, to see what odd results obtain from such efforts, particularly with respect to women and their fetuses, nursing infants, and children generally.) And my point is not that the boundary metaphor is apt for men but destructive for women. Rather, it is mis-

leading about the nature of human selfhood and thus about the institutions and concepts that will foster the flourishing of that selfhood.^[19]

I think the most likely objection to my rejection of boundary (at least from those sympathetic to the social dimensions of selfhood) is that the function of boundaries is to structure relationships, indeed to structure them in ways that "enable an autonomous self to emerge."^[20] It is, of course, true that our conceptions of boundaries, including legal ones, do structure our relationships. But do they do a good job of it? And does the focus on boundary in fact help us understand what those relationships are or should be? Relationships structured around boundaries have not been optimal, and, equally important, the boundary metaphor consistently inhibits our capacity to focus on the relationships it is in fact structuring. I will turn later to an inquiry into privacy and property to explore the distortions boundary language brings. But first I want to examine the sources of my own doubts about rejecting boundary.

III

As I was finishing my book on property, I began to notice the frequency of boundary metaphors. In particular, I realized that two quite different sorts of literature I was engaged with gave great weight to boundaries. The first was child-development literature. The form in which that use of the metaphor is familiar to everyone is "Children need limits" or "It is important to set clear boundaries for children." In fact the metaphor recurs repeatedly in related forms such as this: "My [sexual feelings for a parent] are all right. There is nothing wrong with me for having them. My parents will help me keep them within bounds."^[21] These metaphors sound so familiar that they seem to be obviously "right." They are related to another spacial metaphor: the idea of creating a "safe space" in which the child can develop. This is, I think, closely related to the idea underlying the bounded sphere of rights for the adult. If, therefore, the boundary metaphors ring true in the context of children, should that at least relax my suspicion of the metaphors' capacity to capture the conditions for adult selfhood?

Although child-development literature might at first seem to be well removed from legal discourse (and thus a sort of "independent" validation of boundary metaphors), it not surprisingly reflects the dominant modes of thought in our society. Much of this literature is aimed at helping parents raise their child so that he or she fits in well, an enterprise that requires being attuned to (even if not theoretically conscious of) prevailing values and modes of thinking. For example, Dorothy Corkville Briggs's book Your Child's Self-Esteem neatly presents in the space of six pages the connection between private property, the notion of autonomy, and the self conceived of as essentially separate.^[22] First, the author explains

that "only by practicing separateness can the child capture the feeling of autonomy"; indeed, the "Terrible Twos" would be better labeled "The Age of Separateness."^[23] She then proceeds to offer some helpful suggestions about how to aid this emerging selfhood by respecting the child's efforts at autonomy. But among these suggestions are those based on her view that "separateness means possession":

Possession is one device the young child uses to hammer out autonomy. Consequently, ownership takes on special meaning to the toddler set. To them, separateness means the right to possess.

Just as babbling comes before talking, so owning comes before sharing. To fully share, a person must FIRST fully possess. None of us can share what we don't have. And the little child needs time to get the feel of ownership thoroughly worked into his experience before he can let go.^[24]

Now I trust that the culturally specific, not to say ideological, quality of this picture is clear. The author's concept of sharing is largess. It is the situation in which one who has control of an object chooses to let another interact with it. It is of course tautologically true that one cannot learn this form of sharing unless one first learns about ownership. And there is indeed a very complex set of ideas the child needs to incorporate—first, that some things are "his" and that that means he can do virtually whatever he wants with them, regardless of other people's feelings, needs, or preferences; and then, that he may wish to respond to their feelings by "sharing," and indeed it would be "good" if he did, but it is up to him. Exercizing that power of control and choice not only over objects but over other people is treated as an important part of autonomy. These implicit lessons about property, power, and relationship are no doubt important in our society, and they are closely related to a notion of selfhood as separateness. But surely this is also an odd (though by no means idiosyncratic) notion of "sharing": we can only "share" by largess those things that are our "own," i.e., not truly shared. Suppose, by contrast, that the basic task were to communicate to a child that much of her environment is such that several people (and other creatures) have need of it and claims on it—that is, it is "shared." Here her selfhood would not be hammered out in possession, but developed in the context of the rules of reciprocal connection. We would end up with a different picture of sharing and of the self that shares.

It is possible that early ownership and the power to dispense largess are essential to the separateness so widely identified with autonomy. But this striking link alerts me to the presence of the same ideological framework and metaphoric structure in the apparently different contexts of law and political theory. The boundary metaphors are not independent similarities but the result of the same conception of selfhood, autonomy, and property.

What I am still left with is the powerful intuitive appeal of these metaphors in the context of children. Perhaps their persuasiveness there should make us more open to the underlying ideology even after we have recognized it as such. (Of course this is actually a common form of argument: that children are inher-

ently possessive shows that a regime of private property is natural and thus desirable.) To proceed with this approach, we need to go "behind" the boundary metaphor, to unpack what it stands for. We can then ask whether the notion of boundary focuses our attention on what we really think matters. But I will defer this exploration until we consider another example of the use of boundary.

If some of the child-development literature seems to share an ideological framework with legal discourse, my next example is a self-conscious and radical rejection of that framework. Starhawk is a feminist author who uses the tradition of witchcraft to explore issues of "magic, sex, and politics," as one of her books is subtitled.^[25] She offers an inspiring alternative framework for thinking about political action and personal psychological transformation. But even here, at this safe remove from conventional thought, I find the invocation of boundaries.^[26] The metaphor arises (as before) in an effort to describe "safe space." Starhawk offers ideas for what makes groups sufficiently safe either for planning radical political action or for exploring the powers of magic or for personal transformation. Of course the space here is around not individuals but groups, and it is in large part for the purpose of bonding together in solidarity. But boundary remains an important metaphor:

Somewhere a line of demarcation must exist, dividing the group from what is not the group. In Witchcraft, one of our magical tools, the athame , the knife, is used to make divisions. With it we draw the magic circle, which divides ritual space and time from ordinary space and time. A group, too, needs a circle around it to define it.

The athame is a powerful tool: it is double-edged to remind us that any separation cuts both ways. Boundaries contain our power so that it can deepen and intensify, and they may keep out what could threaten or disrupt our group. But boundaries may also exclude those who might benefit us or bring us power. A boundary is always, in essence, somewhat arbitrary and false: an island of separation carved out of the rippling whole.

Nevertheless, we cannot trust unless we feel safe from intrusion. Each member of a group needs to have some control over who comes in. Each group needs some sense of focus and purpose, which necessarily excludes other possibilities. A group needs a shape and an edge, a skin. Like skin, a boundary both separates and interacts with the world, keeping some things out, letting others in. Boundaries can be thick or thin, solid or permeable, fixed or elastic. But a group with no boundaries is not a safe place to be.^[27]

Starhawk clearly distinguishes her vision of selfhood from the "set-apartness" she sees as a widespread and dangerous delusion. And she distinguishes the safety in solidarity she advocates from the perverse and doomed effort to defend ourselves "behind the barriers we erect, like the walls of ancient citadels, to guard us from the world."^[28] Yet she also draws parallels between her invocation of boundary and the more conventional usages:

Every member of a household needs areas of physical and emotional privacy that are respected by others. . . . We all also need some physical place that is ours alone to control; if not a room, then a corner, a desk, a special altar spot. We can give a young child a special box, saying, "This is yours, to keep whatever you want in it. I will never open it or look in it unless you tell me I can."^[29]

Starhawk is of course right that the concept of boundary conveys a clear image. And it is crucial that we have compelling images for our most important values and concepts. So we must ask again whether the boundary metaphor is helpful when Starhawk uses it, and, if so, why would we exclude its power from the important and difficult problem of mediating the relationship between the individual and the collective?

(It is worth noting that Starhawk herself seems willing to use boundary for this purpose: "In reality, what laws and rules do best is to impose limits on hierarchical power. . . . The Bill of Rights, for example, is an instrument designed not to control the people, but to control the government. It outlines what laws Congress shall not make."^[30] Starhawk is literally right about the history of the Bill of Rights, but not about the general purpose of judicial review nor the uses to which the Bill of Rights has been put. The people often want to use the government to effect some action they think is important, and the Bill of Rights serves as the limit to what they can do. Starhawk is missing the basic insight behind the Constitution: in a democracy we cannot simply think of the government in opposition to the people. In some ways the people are the government, and it is the people and their propensity for injustice that the Framers of 1787 wanted to limit. The boundary problems here are as complex as in the personal and group relationships Starhawk understands so well.)

One of the general problems with the boundary metaphor, like all metaphors that are so deeply established that they appear natural and obvious, is that it obscures the questions it was intended to answer, it closes down rather than invites inquiry. But in Starhawk's case the necessary unpacking is easy because she is (in keeping with her philosophy) forthright about the objectives and values she has in mind. What then are boundaries supposed to accomplish? We can combine this inquiry with the one I postponed earlier: What is really at issue in the search for "safe space" for a group to work or a child to develop in? The answers are similar. Both require a sense of trust and security; an ability to count on protection; some kind of comprehension of expectations, consequences, practices, or of the rules governing situations. It is said that a clear sense of boundaries aids all of these things. But what are these "things"? They are feelings the subjects in question must be able to have, and they point to the sorts of relationships necessary for these feelings to flourish. Let us look a little more closely at one issue of boundary language in the context of children to see whether the notion of boundary helps us focus on those feelings and relationships.

Parents are supposed to provide boundaries so a child knows that if she does something really bad or dangerous, her parents will stop her. She can use her energies freely and expansively, knowing that her parents will step in if necessary. And it is very plausible that if the parents fail to provide this sense of boundary, the child will feel unsafe and may set up her own internal, often rigid and unnecessarily confining boundaries to protect her from the risk of doing something "really" bad.

Now I am very sympathetic with this notion of the need for a child to have confidence in her parents' attention, reliability, good judgment, concern, and capacity to act—in short all those attributes and affects necessary for the parents to stop the child from doing anything really dangerous or harmful. I also think that it is very important that the parents be able to communicate effectively about what sorts of behavior they consider unacceptable and to be clear and consistent about what they treat as falling into that category (what is known as setting clear boundaries). But I doubt whether the admonition to set clear boundaries attunes the parents to the feelings the child needs to have or to the quality of the relationship that will foster those feelings. Surely the injunction will often be heard to be essentially a requirement for rules, rather than an invitation to think about what the child (or parent) is feeling and why.

Of course part of the function of effective metaphors is that they provide a useful shorthand for complicated patterns of relationship. By invoking the metaphor we do not have to think everything through from scratch. There may be some circumstances where the shorthand of boundary, even if translated as "make clear, consistent rules," is as much as one can hope to get a parent to listen to. Nevertheless, the images the concept of boundary invites do not seem to me optimal, even as a shorthand. They focus the mind on barriers, rules, and separateness, perhaps even oppositional separateness. They do not direct attention to the nature of the relationship between the parent and the child, and that, not rules or their clarity, is the essence of what provides the child with "safe space," the sense of confidence and security to explore and develop.^[31]

In addition, boundary imagery teaches both parents and children that security lies in walls. The image of bounded space as essential to autonomy reinforces the image of bounded selves. And of course the boundary rules of possession send far less subtle (if no less complex) messages about power, about who is entitled to it (those with property), what its entitlements are (you may ignore the wishes of others), and that its essence is to be wielded by some over others. The child learns that his world is to be "mastered," and that this mastery includes learning to wield or bow to the power of property in appropriate ways. Boundaries structure relationships, but they do not help us to understand or evaluate those structures, and often the structures are undesirable.

Is that true even for Starhawk's uses of boundary? Of course, she avoids the use of the word property, but her example of "safe space" is part of her discussion of establishing boundaries, and the sort of control she describes is what some people mean by property and everybody agrees is a component of property. The concept of space as Starhawk uses it is a way of organizing patterns of respect and expressing the capacity for commitment. When given the special box, the child learns that her parent respects her need to be able to hide things, to control who sees them and when. And she learns about one of the great features of human relationship, the capacity to make and receive promises. The box is a good vehicle for these forms of communication, particularly if promise and respect rather

than ownership are the central messages about the box. The box seems to work as a focal point for defining rather than denying relationship. It takes its basic meaning from the promise of the parent to the child, which is quite different from the message "This is yours," meaning the child's will takes priority, or "This is your little sphere in which you may exercise power over others."

In the group dynamic the same basic issues arise, with one exception. Here it is important not simply that everyone knows what the expectations are but that they know they share in determining them. Part of Starhawk's use of boundaries is to insist that everyone should know where the power lies in the group, who counts as a member (who is "in"), and who determines membership. For her, exploring the issue of boundary brings these questions into focus.^[32] This is the opposite of asserting the presence of a boundary, whose quality of givenness closes down inquiry (like "That's mine!").^[33] But the most important dimension of establishing group boundaries seems to be mutual commitment, the capacity to bind oneself and others in joint action and restrictions on action. Here we find a close parallel with the issue of constitutional limits on democratic decision making. The essential problem is the same: How can the group meaningfully bind itself with respect to its future actions? And, in a sense, the puzzle is the same: How is it that one decision can control future decisions, what sets the initial "boundary-setting" decision apart, and how do we know which things belong in which decision-making category? In constitutional politics we can recognize this as the basic problem of self-limiting government.

The invocation of boundary language here does not help but hinder, although it has been the very essence of American constitutionalism. At least in the constitutional context, treating rights as boundaries has given them a reified quality that does not direct our attention to the ongoing process of determining those boundaries. We see the rights, the boundaries, as things, not as a particular set of decisions that stands in a special relation to other decisions, thus inviting ongoing inquiry into the nature of that decision making.

Starhawk tries to avoid this reification by acknowledging the artificial, even "false" quality of boundaries and by insisting that boundaries need not have the wall-like quality that we (with our property-oriented vision) ordinarily assume. She insists that boundaries can be "thick or thin, solid or permeable, fixed or elastic." And most important, she reminds us that boundaries are not just barriers that separate but points of connection and contact. The human skin is perhaps the most compelling alternative to the wall image of boundaries: it is permeable, slowly and constantly changing while keeping its basic contours, and a source of sensitive connection to the rest of the world.

Will such new images of boundary help? Yes, to the extent that they focus the mind on connection and patterns of relationship. But in the traditional legal spheres we are still better off without them. The imagery associated with boundary is too well established, too wall-like, too closely tied to a separative self. We have thought of the problems of the self and the collective in boundarylike

terms for so long that they invite no new modes of inquiry; they shield our understanding of reality. Starhawk says that "conflicts in groups often develop about boundaries, and generally masquerade as something else."^[34] In constitutional discourse troubling problems of the meaning of democracy and the process of defining basic values masquerade as boundary issues. The concept of boundary has become more of a mask than a lens.

One of the things the language of boundaries masks is power. We can see that most clearly by returning to our starting point of property. Although property rights are no longer important constitutional boundaries, property is itself conceived in boundary terms and continues to be seen as an important source of autonomy. Everyone is familiar with the notions that property gives us control, property allows us to express ourselves by shaping the space around us, property provides privacy and security, owning a house gives a sense and impression of permanence. And in almost every case it seems as though the physical contours of the property are essential to these desirable experiences. But, of course, property really is a set of legal rules and norms that structure power and relationships. The boundaries the law defines and enforces are a means of wielding power, of shielding power and of shielding from power. The rules of property tell us who has to ask whom for what, and how much power or powerlessness they will have in their request. As Robert L. Hale explained a long time ago, the power to exclude that our legal structure of property gives us is the starting point of all contracting, all negotiation over use of, access to, and exchange of property and labor.^[35]

The focus on the naturally bounded, thinglike quality of "property" obscures the fact that the power we derive from it in no way inheres in the object but is allocated to us by the state. The power to exclude exists because it is backed by the power of the state, and all the other dimensions of property and its relation to autonomy and security flow from that power. The boundary metaphor permits us to indulge in focusing on the experiences we can have in, on, and with our property (whose value I do not deny) and ignore the patterns of relationship shaped by the power to exclude. Private property permits us to flaunt power at the same time that we deny its state-created nature. And that denial has of course sustained the distorted quality of the prevailing discussions of the "free market" versus state "interference" via regulation. A discourse of property that always kept at its center the relationships structured by property backed by the power of the state at its center could not sustain the myth of the free market. Our focus on boundary turns our attention away from relationship and thus away from the true sources and consequences of the patterns of power that property constitutes.

Privacy raises the same basic issues. It entails relationships of respect. Of course, where we treat bounded spheres as indexes of personhood, respecting those boundaries constitutes respecting persons. The question here, as always, is whether the focus on boundaries is the best way for us to understand the sorts of relationships we want to foster. We associate privacy so closely with boundary

imagery that I expect a suggestion to abandon boundary will be heard as rejection of the concept of privacy itself. That is not my intention. The concept of privacy captures, illusively, important values such as people's capacity to decide for themselves some of the ways they will or will not enter into relationship with others. But without trying to unravel it here, I think we can best foster those values by focusing directly on the expressions and patterns of relationship that foster and express respect for people's needs, rather than on "respecting boundaries," which draws our minds away from relationship. As with "boundaries" for children, the metaphor may be said to be a shorthand for the relational values at issue. But I doubt that it helps us think about them creatively and constructively. It presumes answers to questions that we should not take as given.

Boundaries structure relationships. But they structure them badly, in part because boundary imagery masks the existence of relationships and their centrality to concepts like property and privacy. When the dominant metaphors turn our attention away from relationships, we cannot give either the relationships or the legal concepts that mask them the critical scrutiny they require.

IV

Having pointed to the problems the boundary metaphor gives rise to, I must come finally to the most basic question: Can we do without it? Is there not something essentially bounded about us that makes us routinely invoke boundary metaphors to describe those things we experience as basic to our selfhood—like property, privacy, and protection against the power of others, individually and collectively? I think not. What is true is that the dominant notion of selfhood is that of the "separative self," and we take that notion so much for granted that our boundedness seems natural and essential.

For example, Lakoff and Johnson make a startling "mistake" in their discussion of the body as a source of imagery and metaphor. Despite their recognition of cultural diversity, of the way in which some metaphors "are so natural and so pervasive in our thought that they are usually taken as self-evident, direct descriptions of mental phenomena," they treat the experience of the body as a "container" to be exactly such a self-evident, direct description of "our" experience of the body. They assert that "we are physical beings, bounded and set off from the rest of the world by the surface of our skins."^[36] But Catherine Keller has a very different experience: "Our skin does not separate—it connects us to the world through a wondrous network of sensory awareness. . . . Through my senses I go into the world, and the world comes into me. It is precisely in embodiment that the many are becoming one and the outer becoming inner."^[37] Lakoff and Johnson tell us that "we experience the rest of the world as outside of us."^[38] But Susan Griffin reminds us, "For the part of the mind that is dark to us in this culture,

that is sleeping in us, that we name 'unconscious,' is the knowledge that we are inseparable from all other beings in the universe. Intimations of this have reached us."^[39] Lakoff and Johnson say, "Each of us is a container, with a bounding surface and an in-out orientation."^[40] But for Alfred North Whitehead, "Each actual entity is a locus for the universe."^[41]

As Lakoff makes clear in his later work, it is important for Johnson's theory that this experience of the body as container is unmediated by concepts and then forms the basis for the metaphoric structure of concepts.^[42] Johnson's basic idea of looking at the connections between our concepts and our experience of the body is surely an excellent one. But there is a great deal in feminist literature that should make us wary of any assertion that our experiences of our bodies are unmediated through culture and language and thus concept and metaphor.^[43] "The body as container" seems a perfect example of the metaphoric structuring of our experience of the body, as in Andrea Dworkin's reflections on the bounded body. (It seems worth noting as well that many of our boundary metaphors do not, as I noted above, have a very "skinlike" quality to them. On the contrary it often seems that we conceive the boundaries to our selves to be like walls: non-porous, hard, clear, and defendable against invasion.) Boundary metaphors are indeed ubiquitous and have a deep coherence with the basic values of our culture. But there are better ways of understanding this phenomenon than positing "the body as container" as a primal, universal, unmediated experience.^[44]

Catherine Keller, for example, offers a sweeping, brilliant exposition of Western culture and the centrality in it of sexism and the "separative self." The conception of the self that she sets out to question has remarkable resonances with both Johnson's purportedly immediate experience of the body and the model of child development I discussed earlier:

To be a self, must I be something separate and apart? How else could I be myself: Myth and religion, philosophy and psychology center our civilization on the assumption that an individual is a discrete being: I am cleanly divided from the surrounding world of persons and places; I remain essentially the same self from moment to moment. Common sense identifies separateness with the freedom we cherish in the name of "independence" and "autonomy." The assumption that selfhood requires separation is even rooted in language. The Latin for "self," se , meaning "on one's own," yields with parare ("to prepare") the verb "to separate." For our culture it is separation which prepares the way for selfhood.^[45]

In Keller's view we cannot fully understand the genesis or power of this vision of selfhood unless we see its connection with sexism: "Separation and sexism have functioned together as the most fundamental self-shaping assumptions of our culture." And until we recognize their interdependence, "the old world view will retain the momentum of unconsciousness."^[46] I find this complex interdependence particularly resistant to condensation, so I can offer only glimpses of her basic insights. She says that, "Fear of merger and self-dispersion motivates all insistence on separate selfhood. But let me suggest that in such fear of self-loss

lurks a profound fear of women." She opens her introductory chapter with a revealing quote from C.S. Lewis that suggests a further link to the issues this essay began with, the need to protect the individual from the threat of the collective: "In the hive and the anthill we see fully realized the two things that some of us most dread for our own species—the dominance of the female and the dominance of the collective."^[47] Over the course of several chapters, she draws a connection between the fear of women, the fear of the collective, and the fear of the "oceanic feeling," of "an indissoluble bond, of being one with the external world," which is associated with the mother.^[48] In Freudian psychology, she tells us, "this indissoluble bondedness would then belong to the vestiges of earliest childhood, lingering only as an abnormal and regressive infantilism."^[49] But it is this very bondedness that Susan Griffin suggests is the fundamental truth of the world, even though it is "dark to us in this culture." And despite the fear and condemnation, "intimations of this [truth] have reached us."^[50]

Keller merges mythology and psychology with philosophy—drawing on Whitehead in particular—to provide an alternative image of selfhood, true to these "intimations." She tries both to explain the fierce resistance to such an alternative—"Creative connectivity then appears as chaos and confinement, as an undifferentiated heap, a constrictive—maternally monstrous—mix of matters"^[51] —and to take seriously what "differentiation in relation" or integrity without oneness might mean.^[52] She draws on Williams James's "streams of self" and Whitehead's immanence to envision a self that is "an event, a process, and no fixed substance, no substantive."^[53]

In the course of constructing an alternative, Keller helps us understand the relation between the separative self and the problems of power and domination—and the ways those problems are in turn linked to the boundaries of constitutionalism, property, and privacy. Here her views link well with those of Dorothy Dinnerstein and Jessica Benjamin: the separative self is on an endless and doomed search for security, a security that seems possible only in power and domination.^[54] Thus the sought-after walls of protection (like property) are those that entail domination. Keller focuses on control, a classic virtue associated with property:

The separative ego as we have characterized it—as self-objectifying and other-exclusive—cannot separate its strategies of self-perpetuation from its drive to control. In its emphasis upon self-control, being "on top of things," it is simultaneously keeping the influent others under control as well. Domination is its best defense, and retreat its familiar back-up plan. And these defensive strategies inadvertently confirm the truth of internal relations: that the world gets inside us, gets under our skin, does not keep a respectful distance.

Control is the age-old alternative to connection. The denial of internal relations issues in external manipulation.^[55]

These links between the separative self, control, and domination offer a new perspective on the distortions of inequality that the focus on property brought to American constitutionalism. The way the Framers' sense of the need to control

the threat of the people distorted the democratic institutions they designed no longer looks like an anomaly born of a particular historical preoccupation with property. Rather, property appears as a logical preoccupation for separative selves, providing the sought-after illusion of security through power, domination, and isolation. And property can crystallize the fears of loss of control, of intrusion, and of the threat of impending chaos if things are not under control. An emerging democracy stirs up (not entirely unwarranted) fears of the collective (and, in 1787, repeated references to the need for a strong, manly republic untainted by feminine vices). Property focuses these fears in ways that are paradigmatic of the efforts of separative selves to protect themselves through boundaries: the protection of those boundaries is inevitably tied to fear and domination and to the inequalities of power necessary for security through domination. The protections the Framers sought from property (and for property) were inseparable from domination. But from this perspective, the inequality and fear of the people that shaped American constitutionalism is not anomalous but characteristic of the protection in boundaries sought by separative selves.

American constitutionalism has the power it does in part because its metaphoric structure resonates so powerfully with the fears and (elusive because illusory) goals of the separative self. And the fact that we see the boundary metaphor and its distortions not only in the discourse of law but everywhere reflects both the metaphoric coherence of our culture and the centrality of the separative self to the culture and the coherence.

Keller's exploration of the power and pervasiveness of the separative self helps us understand our culture, including constitutionalism (which she does not address). But her purpose is not acceptance or accommodation but transformation. If we accept Keller's (and others) accounts of the depth of the separative self (and its fears and illusions) in our mythology as well as our philosophy, our psychology (both as discipline and as self-consciousness) as well as law, we know we need more than a good theory to change things. In this case I am not thinking of political action (which is of course required) but of new symbols, myths, and metaphors to replace the old.

In contemplating this overwhelming task, I was tremendously encouraged by Estella Lauter's Women as Mythmakers: Poetry and Visual Art by Twentieth-Century Women .^[56] Lauter persuades me that she sees in this art the makings of a new mythology. She recognizes, of course, that the potential for myth that she sees in the work of these individual women will only become genuine myth if it continues to resonate and be echoed again and again. But she thinks the germs are there, the flashes of individual genius that can capture an emerging spirit and give it focus and power. We will need such images not only to sustain and inspire us but to make a credible case that alternatives are really possible. We will need a new vocabulary, new metaphors to invoke if we are not to be sucked back into the forms we are resisting even as we argue against them.

Recall the claim that started my inquiry here: we need new ways of conceiving of the tension between the individual and the collective for which boundary is not an apt metaphor. Let me give you just a sample of what Lauter offers:

Over and over again this "chordal" analysis of common images and themes leads us back to a view of relationships with the world in which the customary boundaries are not preserved. Inside is barely distinguished from outside, and the distinction matters only momentarily. What matters is the flow of energy from one realm to another, so that life is sustained. . . . The images (or bundles including images, stories, themes, ideas) that are repeated in the separate lines of myth I treat here include antinomies rather than mediating them.^[57]

Boundary mediates. Our new conception will focus on the complexities of the interpenetration of individual and collective. Boundaries and mediation imply a separation and opposition that does not capture the complex, fertile, and tension-laden interconnection between self and others that a transformed constitutionalism must respond to. Lauter helps us to see that the necessary forms of reconception are possible by providing examples of boundary-dissolving metaphors.^[58] And, equally important, she tries to meet the fears that Keller has warned us will be stirred up by any effort to transcend the boundaries crucial to the separative self. Writing on Susan Griffin's poem "Matter," she finds it to be

a ritual gesture, an incantation, above all an image of an idea of self that is incandescent or evanescent. "I have no boundary," the speaker says, and with these words she undoes centuries of mythology. . . . The "center," the self that is defined by its categorical differences from others, does not hold, as Yeats predicted it would not. But "things" do not fall apart, nor does the human being relinquish consciousness. We merely participate in reality differently—as one among equals.^[59]

We can approach the daunting task of reimagining the rhetoric of law and freedom with this inspired reassurance in mind.

V

There is something profoundly and I think irreducibly mysterious about the combination of individuality and "enmeshedness," integrity and integration that constitutes the human being. We have constructed a multitude of myths and metaphors to capture the extraordinary capacity for human action, for initiation, for the creation of something genuinely new out of all the multiplicity that has shaped and constituted the actor up to that moment of action. The notions "human will," "agency," "autonomy," "freedom" each insist that not even the fullest account of that multiplicity can ever account for or determine the moment of action. And the structures of philosophy and law built around these notions insist further that this extraordinary capacity is to be cherished and protected.^[60] American mythology (the pioneer, the yeoman farmer, the rugged indi-

vidual, the self-made man) and the institutions and rhetoric of American constitutional law (which also have a kind of mythic status) have done an impressive job of capturing a particular vision of this dimension of humanness. But the very separateness thought essential to this dimension distorts our understanding by splitting it off from, and setting it up in opposition to, the integration, interpenetration, and unity that are also part of our humanness and without which the capacity for action would not exist.

We have seen this oppositional splitting off take the form of the fear of the collective. In the American context the fear was originally focused around the protection of property, the symbol for separative freedom, power, and control, ever vulnerable to invasion by the many. The Constitution was designed to ward off that threat, and although the protection of minority rights has sometimes taken more benign forms, it usually carries a tone of contempt and fear of the many. One can be concerned, for example, about the tone in which Northern liberals speak about Southern bigots even if one has nothing but revulsion for the inherent violence of racism. We need to become attuned to the distortion of the values we care most deeply about. For example, we need to see how our separative vision of freedom of speech has blinded us to the ways in which the speech of some silences the voices of others by virtue of the patterns of power and dominance entailed in the protected speech.^[61] An exclusive focus on containing the threat of the collective cannot capture the genuine values of "freedom of speech."

The boundaries so central to American law are the boundaries that feel desperately necessary to the separative self to keep the threatening others at bay—a task whose impossibility only fuels the desperation. When I say the task is impossible, I do not mean to imply that the boundaries do nothing. They do protect us from certain kinds of threats. But equally (or more) important, our boundary-setting rights protect us from the seemingly overwhelming responsibility that would flow from a recognition of unity. This is, I think, a frightening form of the "oceanic feeling," intimations of which have reached us. We fear being "invaded," "taken over," not just by threats but by demands—the overpowering demands of those in pain and hunger all around us. We wall ourselves off from their cries—genuinely do not hear them most of the time, even though we "know" they are there—by telling ourselves that we are "within our rights," that rights define our obligations as well as our entitlements, and that as long as we have violated no one's rights, we are doing nothing wrong in our daily nonresponsiveness.

That particular form of freedom would, I think, be radically transformed if we were to come to see ourselves as "inseparable from all other beings in the universe." The fear of such a transformation is old, deep, and not crazy. I think we see it lurking under the Framers' assertions of the illegitimacy of redistributing property as well as in modern attacks on the welfare state. Throughout American history anxieties about redistributive legislation have been accompanied by cries of "Where will it all end?" If such a redistribution (in the form of paper money,

or a progressive income tax, or workers' compensation, or socialized medicine) is permitted, what boundaries can we count on? Won't the demands of the collective become insatiable, finally devouring us and the orderly society that keeps chaos and the collective at bay? Lauter acknowledges some grounds for these fears: "The female envisioned (or hoped-for) is one of vastly increased responsibility."^[62] And that responsibility will not be fully compatible with the scope of privileged freedom that our rights have carved out for us.

As we develop the new myths and metaphors that truly combine rather than mediate antinomies, we will have to find new ways of comprehending and responding to our responsibilities.^[63] And of course we cannot do so by accepting law as the realm of boundaries, leaving the rest (in good Kantian fashion) to morality or compassion. That division makes sense only for the separative self. We will need new ways of capturing the mystery of human creativity, the spontaneous combustion of human action. But the puzzle of a nonsubmerged selfhood amidst connective responsibility cannot be solved by walls of rights. We need to take our traditional concepts like property and ask what patterns of relationship among people and the material world we want, what patterns seem true to both integrity and integration. Those questions do not necessarily preclude a concept of property, but they imply a focus not on limits but on forms of interaction and responsibility for their consequences.

In rejecting the categories of the past, we should never underestimate the task. The American Constitution is only a particularly vivid modern form of an age-old effort to use the concept of boundary to mediate, to "grasp," to "bring under control" the illusive mysteries of human freedom and connection. Without the boundary metaphor, the structure of legal conceptions of freedom disintegrates. But some of that disintegration is already underway. At least in the academic legal community there is no consensus on the basic meaning of law or of the values it is supposed to protect. New metaphors are a genuine option because they are in fact emerging. We are in a period of flux where our presuppositions are in doubt. It is therefore possible to exercise some deliberate choice about the frame of reference through which we see the world. We can try to transform our own language, push it in the direction of the barely articulated "intimations" that have reached us. Disintegration entails promise. If we can let go of our walls of rights, the reintegration is likely to be far fuller and more promising.

Notes

Introduction: The Relatively Autonomous Discourse of Law

Generalization in Interpretive Theory

Theories of Constitutional Interpretation

Facing Facts in Legal Interpretation

The Placement of Politics in Roberto Unger's Politics

Enlightenment Epistemology and the Laws of Authorship in Revolutionary France, 1777–1793

Norms, Discipline, and the Law

Law, Boundaries, and the Bounded Self

Contributors

FRANÇOIS EWALD is a research associate at the Centre national de recherche scientifique in Paris. He has participated in the French government as part of the Ministry of Humanitarian Action, and recently published L'Etat providence (1986) as well as a collection of excerpts from the French civil code.

CARLA HESSE is Assistant Professor of History at the University of California, Berkeley. Her book Publishing and Cultural Politics in Revolutionary Paris, 1789–1810 is forthcoming from the University of California Press.

JENNIFER NEDELSKY is Associate Professor of Law and Political Science at the University of Toronto. She is the author of Private Property and the Limits of American Constitutionalism: The Madisonian Framework and Its Consequences .

ROBERT POST teaches constitutional law at the School of Law (Boalt Hall), University of California, Berkeley. He writes on the legal regulation of communication, and he is presently working on a history of the United States Supreme Court from 1921 to 1930.

KIM LANE SCHEPPELE teaches political science, law, and public policy at the University of Michigan. Legal Secrets: Equality and Efficiency in the Common Law was published in 1988 by the University of Chicago Press. The present essay is part of a book-length work, Nothing But the Truth , forthcoming.

MARTIN STONE is Assistant Professor of Law and Philosophy at Duke University and a doctoral candidate in philosophy at Harvard University.

JOSEPH VINING is Hutchins Professor of Law at the University of Michigan and author most recently of The Authoritative and the Authoritarian (Chicago, 1988). His current work is on legal conceptions of mind.

Index

A

Abnormality, 140 , 157 -58

Adams, John, 37 n12

Adorno, Theodor W., 41 n96

Alienation, 36 , 89 , 95

Amendments, to U.S. Constitution:

First Amendment, 23 , 24 , 29 ;

Fifth Amendment, 67 n2;

Sixth Amendment, 67 n2;

Seventh Amendment, 67 n2;

Fourteenth Amendment, 30 , 36 n4

Antifederalism, 67 n2, 185 n7

Appeals courts, 2 -3;

and law/fact distinction, 42 -44, 48 -49, 61 -62, 68 n3

Aristotle, 96

Artificial intelligence, 8 -9, 10 n1, 12 n12

Attainder, 50 , 70 n37, 71 n40

Authority:

and canon formation, 3 ;

of Congress, 21 -22;

of constitutional law, 19 -20, 24 , 26 -28;

of democratic consent, 21 -23, 26 , 28 , 35 , 39 n72, 40 n79;

and historical interpretation, 19 -23;

and institutional division of labor, 61 ;

intellectual, 82 ;

judicial, 15 , 26 , 34 , 65 ;

of legislatures, 6 -7;

and legitimation of discourse, viii ;

and narrative construction, 65 ;

of national ethos, 23 , 26 , 28 , 29 , 30 , 32 -36, 39 n72;

and speaking for law, ix , 6 -7;

and stare decisis , 19 -20, 26 -28, 33 , 39 n72, 40 n74

Authorship:

accountability of, 114 , 115 , 118 -20;

and contracts, 4 , 112 ;

and copyright law, 110 -11, 132 nn2,6, 136 n68;

economic determination of, 110 , 114 ;

and epistemology, 109 , 114 -17, 128 -30;

and individualism, 109 -10, 113 , 114 -17;

and intent, 4 , 5 , 7 -10, 21 , 104 n20;

legal determination of, 109 , 110 , 111 -12, 113 , 117 ;

and literary privilege, 111 -30;

and literary style, 116 ;

political determination of, 113 , 114 , 129 , 131 ;

and poststructuralism, 109 ;

and public service, 126 , 129 -30;

and speaking for law, 6 -7.

See also Literary property; Publishing trade; Texts; Writing

Autonomy:

and boundary metaphor, 162 -63, 167 -72;

and individuality, 162 , 165 , 167 -68, 182 , 186 n14;

of legal discourse, viii , ix , xiii ;

political, 95 , 99 , 103 n14, 108 n64;

of self, x , xiv -xv, 95 , 99 , 168 -72, 186 n14

Averages, law of, 144 -46

Avery v. Everett , 52 -54, 59 -60, 75 n79

B

Banishment, 72 n47

Barfield, Owen, 12 n18

Baudin, P.C.L., 136 n83

Beaumarchais, Pierre-Augustin Caron de, 125 , 127

Benjamin, Jessica, 180

Bentham, Jeremy, 76 n88

Berkeley, George, 94

Bill of Rights, 18 , 22 , 174 , 185 n7

Bio-power, xi -xii, 138 -39, 158 -59

Body:

discipline of, xi , 141 ;

experience of, 170 , 178 -79, 188 n43

Bork, Robert, 33 -34, 41 n95

Boundary metaphor:

and autonomy, 162 -63, 167 -72;

and body, 170 , 178 -79;

and child development, 169 , 171 -76, 178 , 179 ;

and constitutional limits on democracy, 167 , 174 , 176 , 177 , 181 ;

experience structured by, 169 ;

and fluidity, 163 , 173 , 176 ;

and group dynamics, 173 -74, 176 ;

and legal discourse, viii , x , 171 , 173 , 181 , 184 ;

and privacy, 167 , 168 , 170 , 177 -78;

and property rights, 162 -63, 167 -68, 171 -72, 177 ;

relationships structured by, 171 , 175 , 178 ;

and selfhood, 167 -72, 176 , 178 -80, 182 ;

and sexuality, 169 -70, 188 n58;

transformation of, 181 -84

Bourdieu, Pierre, xiii , xvi n7

The Bramble Bush (Llewellyn), 68 n8

Brennan, William J., 16 -18, 22 -25, 28 , 32 , 37 n31

Briggs, Dorothy Corkville, 171 -72

Brown v. Board of Education , 30 , 34

Bureaucracy, 36

Burger, Warren, 17 , 21 , 25 , 28 , 29

Burke, Edmund, 103 n14

Burnet v. Coronado Oil and Gas Co. , 40 n75

C

Calvin, Jean, 139

Canguilhem, Georges, 139 , 153 , 157

Canon formation, 3

Cardozo, Benjamin, 47 -48

Catachresis, 74 n72

Causality, 143 -44, 147

Censorship, 111 , 115 , 117 , 120

Chénier, Marie-Joseph, 127 -28, 137 n91

Child development, 169 , 171 -76, 178 , 179

Cicero, 139

Civil death:

and civil rights, 51 , 72 n44;

and common law, 50 , 53 , 70 n37, 71 n41;

and forfeiture, 50 -54, 70 n37, 73 n56, 75 n79;

and inheritance, 50 -59, 66 , 71 nn38-40;

and law/fact distinction, 60 -61;

legal consequences of, 50 -52, 64 , 72 n44;

as legal fiction, 50 , 54 -59, 64 , 74 n73;

and marriage, 51 , 72 n42;

and statutory interpretation, 59 -61

Civil Disobedience (Thoreau), 82

Code de la librairie , 111

Colbert, Jean-Baptiste, 112

Collective, tension between individual and, 162 , 166 , 183

Comédie française, 122 -23, 124 , 125 -26

Common law:

and attainder, 50 , 70 n37;

and civil death, 50 , 53 , 71 n41;

and corruption of blood, 50 , 70 n37;

in England, 50 , 53 , 67 n1, 70 n37, 72 n45;

and forfeiture, 50 , 70 n37;

and lawsuits, 50 , 71 n41;

and property rights, 185 n8;

and wealth maximization, 105 n44

Communitarianism, 32 -33, 87 , 103 n11, 189 n63

Computers, 8 -9, 10 n1, 12 n12

Comte, Auguste, 143

Condorcet, Marie Jean Antoine Caritat, marquis de, 115 -24, 126 -30

Constitution, U.S.:

attainder in, 50 , 71 n40;

Bill of Rights in, 18 , 22 , 174 , 185 n7;

democratic decision-making limited by, 163 -67, 174 , 176 , 177 , 181 ;

Due Process Clause in, 25 ;

Equal Protection Clause in, 30 ;

Establishment Clause in, 13 -14, 15 -18, 20 , 24 ;

and property rights, 162 -67, 174 , 183 ;

provisions for juries in, 67 n2;

ratification of, 21 -23, 29 , 40 n77, 67 n2.

See also Amendments

Constitutional interpretation:

and authority of consent, 21 -23, 26 , 28 , 35 , 39 n72, 40 n79;

and authority of ethos, 23 , 26 , 28 , 29 , 30 , 32 -36, 39 n72;

and authority of law, 19 -20, 26 -28;

and counter-majoritarian difficulty, 15 , 25 , 34 , 41 n98;

doctrinal, 16 -17, 19 , 20 , 26 -28, 30 , 31 -32, 33 , 35 , 40 nn79,84;

and freedom of speech, 25 ;

and hermeneutics, 25 -26;

historical, 17 -23, 25 , 26 , 28 -30, 32 -35, 40 nn77,79;

and individualism, 25 , 26 , 33 ;

and judicial review, 14 -15, 20 -21, 163 , 166 , 174 ;

and liberalism, 21 , 25 ;

and metaphors, 32 -33;

methodology of, ix -x;

and noninterpretivism, 31 ;

and original intent, 17 -18, 21 -23, 25 ;

and property rights, 162 -67, 181 ;

and racial equality, 30 ;

responsive, 18 , 19 , 23 -26, 28 , 29 -35;

and right to privacy, 25 ;

and stare decisis , 16 -17, 19 , 26 -28, 33 , 39 n72, 40 nn74-75;

and superliberalism, 89 -90, 93 ;

and textualism, 14 , 36 n9

Constructive discharge, 59 , 75 n75

Context:

constraint of, 81 , 83 -86, 94 , 105 n47;

and meaning of language, 69 n13;

self-revision of, 90 , 93 , 94 -100

Contracts:

and authorship, 4 , 112 ;

and fraud, 49 ;

implied, 59 , 75 n77;

in superliberalism, 91 , 93

Conventionalism, 43 -44

Cook, Walter Wheeler, 68 n8

Cooper, Edward H., 67 n3

Copyright, xii , xiii , 110 -11, 132 nn2,6, 136 n68. See also Literary property

Corneille, Pierre, 128

Cornell, Drucilla, 80

Corruption of blood, 50 -51, 54 , 70 n37, 71 nn39-40

Counterfeit money, 72 n42

Counter-majoritarian difficulty, 15 , 25 , 34 , 41 n98

Courts:

appeals, 2 -3, 14 , 42 , 44 -49, 61 , 68 n3, 76 n82;

division of labor in, 42 , 61 , 67 n1;

institutional design of, 42 , 44 , 61 , 67 n1, 76 nn83,85;

law/fact distinction in, 42 , 44 , 61 -62, 67 nn1-3, 68 n7;

legal fictions in, 58 , 63 ;

trial, 14 , 42 , 61 , 67 n3

Critical Legal Studies, xiv -xv, 78 -79, 83 , 86 , 99 , 101

Critical Legal Studies Movement (Unger), 83 , 88 -97, 99 -101

Critique of Pure Reason (Kant), 81

Culture:

and autonomous self, x , 168 ;

and bodily experience, 179 ;

and constitutional interpretation, 30 ;

and individualism, 109 -10;

law as mediation of, vii -viii, ix , xii , 131 ;

and legal narrative, 63 , 65 ;

and political subjectivity, xii -xiv;

in public domain, 129 , 131 ;

role of authors in, 109 -11, 131

D

Daly, Mary, 188 n51

Darnton, Robert, 132 n7

Dascal, Marcelo, 37 n10

Death, in juridical systems, xi , 138

Declaration of the Rights of Man and the Citizen , 121

Deconstruction, xiv

De Man, Paul, 74 n72

Democracy:

and authority of consent, 21 -23, 26 , 28 , 35 , 39 n72, 40 n79;

and bourgeois revolution, 130 ;

constitutional limits on, 163 -67, 174 , 176 , 177 , 181 ;

and counter-majoritarian difficulty, 15 , 34 , 41 n98;

and judicial decision-making, 15 , 25 , 34 , 36 , 89 -90;

legitimized by debate, 13 ;

and public domain, 129 ;

and standardization, 152 , 160 ;

and superliberalism, 79 -80, 89 -90;

as threat to individuality, 166 -67;

as threat to property rights, 163 -67, 174 , 183 .

See also Social democracy

Democracy and Distrust (Ely), 89

Derrida, Jacques, 107 n63

Desbois, Henri, 110

Descartes, René, 81 , 95

Desertion, 64 -65, 76 n87

Diderot, Denis, 114 -17, 122 , 123 , 124 , 128 -30

Dinnerstein, Dorothy, 180

Disciplinary order, 140 -41, 152 -53

Discipline and Punish (Foucault), 140 -41, 153

Discourse:

and boundary metaphor, 163 ;

and contextual constraint, 84 -86;

and historicism, 85 ;

and possessive individualism, 109 ;

and social will, 100 .

See also Discourse, legal; Language; Metaphors

Discourse, legal:

autonomy of, viii , ix , xiii ;

and bio-power, xii ;

and boundary metaphor, viii , x , 171 , 173 , 181 , 184 ;

cultural mediation of, vii -viii, ix , xiii , 131 ;

and jurisgenesis, vii ;

and law/fact distinction, 55 -59;

and legal fictions, 55 -59, 63 -65;

legitimation of, viii , xii ;

and narrative construction, 62 -63;

ordinary description in, 55 , 73 n61;

and political action, viii , xiii ;

and social change, xi , 184

Divorce, 51 , 64 -65

Dred Scott decision, 15

Due Process Clause, in U.S. Constitution, 25

Durkheim, Emile, 157

Dworkin, Andrea, 169 -70, 179

Dworkin, Ronald, 47 -48, 68 n10, 69 n23, 107 n62

E

Ellerson v. Westcott , 66

Ely, John, 89 -90

Emeri, Jacques-André d', 118

Empiricism, 94

Encyclopédie , 114 , 116

England:

attainder in, 50 , 70 n37;

civil death statute in, 50 -53, 70 n37, 75 n79;

common law in, 50 , 53 , 67 n1, 70 n37, 72 n45;

forfeiture in, 50 -53, 70 n37, 71 nn38-40, 73 n56, 75 n79;

publishing trade in, 110 , 134 n32

Enlightenment, 114 -17, 128 , 130

Epistemology, and literary property, 109 , 114 -17, 128 -30

Equal Protection Clause, in U.S. Constitution, 30

Equal Rights Amendment, 28 -29

Essay on Human Understanding (Locke), 117

Essentialism, 159 , 188 n43

Establishment Clause, in U.S. Constitution, 13 -14, 15 -18, 20 , 24

Ethos, authority of, 23 , 26 , 28 , 29 , 30 , 32 -36, 39 n72

Ewald, François, viii , xi -xii, xiii

F

Facts:

and applicability of texts, 2 -3;

and cases of first impression, 53 , 73 n55;

and law/fact distinction, x , 42 -44, 54 -62, 67 n1;

and legal fictions, 54 -59;

in normative system, 155 -56;

statistical, 143 -44, 156 ;

textualization of, 44 ;

and value/fact distinction, 79 , 80 , 98 , 107 n62

False Necessity (Unger), 79 , 88 -94, 96 , 99

The Federalist Papers , 22

Felonies, and civil death, 50 , 54 , 70 nn37-38, 71 nn39,41

Feminism, 173 , 179 , 186 n22, 188 n43, 189 n63

Fichte, Johann Gottlieb, xiv

Fictions, legal:

Bentham's critique of, 76 n88;

and civil death, 50 , 54 -59, 64 , 74 n73;

and law/fact distinction, 54 -59, 65 ;

and legal discourse, 55 -59, 63 -65;

as metaphors, 56 , 61 , 63 -65;

and the self, 94 -95

Finnis, John, 105 n45

Fish, Stanley, 47 , 48 , 69 n13, 85 , 105 n47

Forfeiture, 50 -54, 70 n37, 71 nn39-40, 73 n56, 75 n79

Formalism, xv , xvi n7, 83 , 100 , 104 n18

Foucault, Michel, ix , xiii -xiv, 81 , 84 , 153 ;

and authorship, xii , 109 -11, 113 , 115 , 130 ;

and bio-power, xi -xii, 138 -39, 158 -59;

and disciplinary order, 140 -41;

and juridical system, xi , 138 -39, 158 -59

Foundationalism, 83 , 94

Fragments sur la liberté de la presse (Condorcet), 115 , 119

France:

absolutism in, 111 -14, 117 ;

Burkean political critique of, 103 n14;

Enlightenment in, 114 -17, 130 ;

and French Revolution, xii , 110 -11, 117 -31, 160 ;

publishing trade in, 110 -31

Frank, Jerome, 68 n8

Freedom:

and boundary metaphor, 162 -63, 167 -68;

and integrated individuality, 182 ;

and power, xiv ;

of the press, 117 -21, 124 , 134 n25;

of speech, 25 , 183

Freud, Sigmund, 105 n46, 180

Freund, Paul, 21 -22

G

Gadamer, Hans-Georg, ix , 25

Galileo, 142

Geertz, Clifford, 69 n14

Gelbart, Nina Ratner, 134 n25

Germany:

publishing trade in, 110 , 134 nn31-32

Ginzburg, Jane C., 132 n6

Goffman, Erving, 168

Green, Leon, 67 n1

Greenblatt, Stephen, xiii

Griffin, Susan, 178 -79, 180 , 182

Gyn/Ecology (Daly), 188 n51

H

Habermas, Jürgen, 36

Hale, Robert L., 177

Harriman, Norman F., 148 , 151

Harris, Roy, 12 n12

Hegel, Georg Wilhelm Friedrich, 78 , 83 , 94 , 95 , 99 , 104 n18, 106 n59

Hegemony, and constitutional interpretation, 30 , 40 n85

Heidegger, Martin, 80 , 82 , 102 n3, 107 n63, 108 nn63,67

Hein, Hilda, 188 n58

Hell, François, 123 , 125

Hemery, Joseph d', 133 n18

Héricourt, Louis d', 112 , 114

Hermeneutics, and constitutional interpretation, 25 -26

Herzog, Don, 40 n78

Hesse, Carla, viii , xi , xii , xiii

Hexter, J. H., 74 n69

Histoire de l'édition française , 110

Historicism, 81 , 85

History, and constitutional interpretation, 17 -23, 25 , 26 , 28 -29, 32 , 33 , 34 , 35 , 40 n77

History of Sexuality (Foucault), 138 , 153 , 158

Hobbes, Thomas, 33 -34

Homes, Oliver Wendell, 23 , 32

Horkheimer, Max, 41 n96

Hoy, David, 25

Hume, David, 87 , 94 , 106 n59

I

Individualism:

and authorship, 109 -10, 113 , 114 -17;

and constitutional interpretation, 25 , 26 , 33 .

See also Individuality; Self; Subjectivity

Individuality:

and autonomy, 162 , 165 , 167 -68, 182 , 186 n14;

and boundary metaphor, 167 -68;

and constructive relationships, 168 -69;

and disciplinary order, 141 ;

integrated, 182 ;

and normative system, 141 , 154 ;

and privacy, 186 n14;

tension between collective and, 162 , 166 , 183 ;

threatened by democracy, 166 -67.

See also Individualism; Self; Subjectivity

Inheritance:

and civil death, 50 -59, 66 , 71 nn38-40;

legal interpretation of, 45 -48, 59 -60;

and life insurance, 49 , 70 n36;

and marriage, 51 ;

and murder, 45 -46, 49 , 52 -54, 59 , 66

Institutions:

and bio-power, 138 ;

and cultural mediation, vii ;

design of legal, x , 42 , 44 , 61 , 67 n1, 76 nn83,85;

and normative system, 138 , 141 ;

and power relations, xvi n17;

and social insurance, 147 -48;

and social plasticity, 93

Instrumentalism:

legal autonomy denied by, xvi n7;

and property rights, 121 , 130 ;

and rationality, 36 ;

and self-expression, 95 -96;

and social plasticity, 91 -93

Insurance:

and causality, 143 -44, 147 ;

and inheritance law, 49 , 70 n36;

as normative system, 141 -42, 152 -53, 159 ;

and risk, 141 -42, 146 -47;

and state organization, 147 -48, 153 ;

and statistical probability, 142 -44

Intent:

and authorship, 4 , 5 , 7 -10, 21 , 104 n20;

and constitutional interpretation, 17 -18, 21 -23, 25 ;

and law/fact distinction, 43 ;

legislative, 45 -46, 59 -60, 69 n18

Intentionalism, 44

Interpretation:

and appellate opinion, 2 -3;

and contextual constraint, 84 , 88 ;

of contracts, 4 -5;

and conventionalism, 43 -44;

and conversation between reader and text, 25 ;

of delimited situations, 2 , 3 ;

and identification of texts, 1 -3;

and intentionalism, 44 ;

and interpretivism, 44 ;

and law/fact distinction, 42 -44, 48 -49, 60 -62, 67 n2, 68 nn8-10, 69 n14;

and literalism, 44 ;

and narrative construction, 65 ;

and parasitism, 81 ;

and phenomenology, 14 , 36 n7;

statutory, 5 , 45 -48, 59 -61, 67 n2, 69 n23;

and structuralism, 44 ;

superseded by normative system, 156 ;

and textualism, 14 , 36 n9;

of torts, 4 ;

of wills, 45 -46.

See also Constitutional interpretation

Intersubjectivity, 154

Isaacs, Nathan, 43

J

James, William, 180

Jefferson, Thomas, 37 n12

Johnson, Mark, 169 , 178 -79

Judiciary:

anti-democratic decisionmaking in, 15 , 26 , 34 , 89 -90;

in appeals courts, 42 , 44 , 76 n82;

authority of, 15 , 26 , 34 , 65 ;

and cases of first impression, 53 , 73 n55;

and constitutional interpretation, 14 -15, 20 -21, 163 ;

and counter-majoritarian difficulty, 15 ;

and judge-centeredness, 5 ;

and judicial review, 14 -15, 20 -21, 163 , 166 , 174 ;

juries as constraint on, 42 ;

and law/fact distinction, x , 42 -44, 60 -63, 67 nn1-2, 68 nn7-8,10;

legal fictions used by, 59 -60, 63 , 65 ;

and narrative construction, 63 , 65 ;

and politics of context, 85 ;

and statutory

Judiciary

interpretation, 44 -47, 59 -60, 67 n2, 69 nn18,23;

in trial courts, 42

Juridical system, xi , 138 -39, 141 , 158 -59

Juries:

and antifederalism, 67 n2;

as constraint on judiciary, 42 ;

cultural background of, 62 -63;

and law/fact distinction, x , 42 , 61 -63, 67 nn1-2, 68 n7, 76 n82;

and narrative construction, 62 -63, 75 n82, 76 n83

Jurisgenesis, vii

K

Kant, Immanuel, 24 , 81 , 95 , 99 , 103 n11, 107 n63, 142 , 143 , 157 , 184

Keats, John, 91

Keller, Catherine, 178 -82

Kennedy, Duncan, 86

Knapp, Steven, 1 , 104 n20

Knowledge and Politics (Unger), 78 -83, 87 -88, 90 , 100

Kripke, Saul, 103 n11

L

Lafayette, Marie-Joseph de, 118

LaFontaine, Jean de, 128

LaHarpe, Jean François de, 122 , 126

Lakanal, Joseph, 137 n91

Lakoff, George, 169 , 178 -79

Lameth, Charles de, 118 , 120

Language:

and authorship, 4 ;

catachresis in, 74 n72;

and computers, 8 -9;

constraint of, 86 ;

context of, 69 n13, 86 ;

and legal fictions, 55 -59;

as normative system, 149 -52, 156 ;

and semantics, 8 , 55 -59;

and social will, 100 ;

and syntax, 8 , 151 , 156 .

See also Discourse; Metaphors

Lanthenas, François, 135 n47

Lauter, Estella, 181 -82, 184

LeBreton, André-François, 114

Le Chapelier, Isaac-René-Guy, 126 , 137 n92

Lefort, Claude, 13 , 36

Legality:

political, 81 -82, 83 , 86 -88;

of syntax, 8

Legislatures:

authority of, 6 -7;

and constitutional amendments, 29 ;

and politics of context, 85 ;

religion in, 13 -14;

and statutory interpretation, 47 , 69 n18

Lemon v. Kurtzman , 16 -17, 20

Lettre historique et politique sur le commerce de la librairie (Diderot), 114 -15

Lewis, C. S., 180

Libel, 115 , 117 , 120 -21, 125 , 128

Liberalism:

and bourgeois democracy, 130 ;

and constitutional interpretation, 21 , 25 ;

and fact/value distinction, 107 n62;

and self as property, 89 .

See also Superliberalism

Libertarianism, 106 n58, 124

Linguet, Simon-Nicolas-Henri, 115

Literalism, 44

Literary property:

and absolutism, 111 -14, 130 ;

and accountability, 114 , 115 , 118 -20;

and copyright, xii , xiii , 110 -11, 132 nn2,6, 136 n68;

and economic determination of authorship, 110 , 114 ;

and epistemology, 109 , 114 -17, 128 -30;

free distribution of, 117 , 120 , 121 -22, 124 , 129 ;

and individualism, 109 -10, 113 , 114 -17;

and legal determination of authorship, 109 , 110 , 111 -12, 113 , 117 ;

limited ownership of, 119 -25, 128 -29;

and literary privilege, 111 -29;

monopolization of, 112 -13, 115 , 117 -18, 122 , 124 , 125 ;

and natural rights, 110 , 119 , 121 , 122 , 130 ;

pirating of, 118 , 127 ;

and political determination of authorship, 113 , 114 , 129 , 131 ;

private ownership of, 109 , 114 -17, 123 , 130 ;

and public domain, 113 , 126 , 128 -29, 130 -31;

social ownership of, 116 -17, 119 , 124 , 126 , 128 -29;

theatrical works as, 122 , 125 -27

Llewellyn, Karl, 24 , 40 n77, 49 , 68 n8

Locke, John, 23 , 110 , 112 , 117

Louis XIV, 112

Loustallot, Elysée, 121

Louvet, Jean-Baptiste, 127

M

MacCormick, Neil, 68 n9

Mackie, John, 96 -97, 99

MacKinnon, Catherine, 189 n61

Malesherbes, Chrétien-Guillaume Lamoignon de, 133 n18

Marbury v. Madison , 15 , 19

Marin, François, 133 n18

Marriage:

and civil death, 50 -51, 72 n42;

and inheritance, 51

Marshall, John, 15 , 19 , 37 n12

Marsh v. Chambers , 13 -14, 15 -18, 19 -25, 32

Marx, Karl, xiv , 78 , 81 , 87 , 95

Meese, Edwin, 21

Mémoire sur les propriétés et privilèges exclusifs de la librairie (Linguet), 115

Mercure de France (periodical), 123

Metaphors:

bodily, 178 -79;

and constitutional interpretation, 32 -33;

experience structured by, 169 ;

and legal fictions, 56 , 61 , 63 -65, 75 n73;

and politics of legality, 81 -82;

superseded by normative system, 156 .

See also Boundary metaphor

Metaphysics, 80 , 81 , 83 , 99 , 102 n3

Michaels, Walter Benn, 1 , 104 n20

Mirabeau, Honoré-Gabriel de, 126

Modernism:

and normative system, 139 , 141 , 160 ;

and superliberalism, 79 -81, 93 , 99 , 100

Molière, 120

Moniteur (periodical), 127

Murder:

and inheritance law, 45 -46, 49 , 52 -54, 59 , 66 ;

and legal discourse, 56 -57, 73 n62, 74 n68;

and right of dower, 70 n31;

second-degree, 73 n49

N

Nagel, Thomas, 107 n62

Narrative:

constructed by juries, 62 -63, 75 n82, 76 n83;

and culture, 63 , 65 ;

and legal authority, 65

Nedelsky, Jennifer, viii , x -xi, xii , xv

Negative capability, 91 , 101

Nemours, Dupont de, 118

Neurath, Otto, 82

New York Mutual Life Insurance Co. v. Armstrong , 70 n36

Nietzsche, Friedrich, 23 , 79 , 87 , 94 , 95 , 97 , 99 , 107 n63, 108 n67

Nihilism, 79 , 83 , 89 , 96 , 97

Nominalism, 7 , 144

Nonet, Philippe, 24

Noninterpretivism, 31

Normative system:

abnormality in, 140 , 157 -58;

and bio-power, xi -xii, 138 -39, 158 -59;

and disciplinary order, xi -xii, 140 -41, 152 -53;

and exercise of power, 153 -55, 159 ;

facticity in, 155 -56;

individuals in, 141 , 154 ;

insurance as, 141 -42, 152 -53, 159 ;

interpretation superseded by, 156 ;

language as, 149 -52, 156 ;

and measurement techniques, 151 , 152 , 159 -60;

and modernism, 139 , 141 , 160 ;

and normalization process, 148 -52;

opposed to juridical system, xi , 138 -39, 141 , 158 -59;

polarity of, xii , 157 ;

politics as, 157 -58, 160 ;

and positivism, xii , 156 , 158 ;

proliferation of legislation in, 138 ;

relativity of, xii , 156 -57;

and standardization techniques, xii , 148 -50, 152 , 159 ;

and state organization, 147 -48, 153 , 155 , 158 ;

and statistical probability, 142 -44, 156 , 158 ;

values superseded by, 154 -56, 157

Nozick, Robert, 106 n58

O

Objectivism, 116 -17

Of Woman Born (Rich), 186 n14

Owens v. Owens , 49 , 70 n31

P

Panckouke, Charles-Joseph, 123 -24, 130

Passion (Unger), 79 , 83 , 85 , 86 , 89

Phenomenology, 14 , 36 n7

Philosophical Investigations (Wittgenstein), 86

Philosophy of Right (Hegel), 83 , 106 n59

Pitkin, Hanna, 23

Plasticity, social, 80 , 91 -93

Plasticity into Power (Unger), 91

Plato, 90 , 105 n42

Plessy v. Ferguson , 30

Politics:

and agonism of power, xiv -xv;

and authors' rights, 114 , 117 -31;

and autonomy, 95 , 99 , 103 n14, 108 n64;

and bio-power, 139 ;

and civil death, 51 ;

and constitutional interpretation, 19 , 33 , 163 -67;

and contextual constraint, 84 -86, 88 ;

and determination of authorship, 114 , 129 , 131 ;

in French Revolution, xii , 118 -20, 135 n45;

and legal discourse, viii , xiii ;

and legality, 81 -82, 83 , 86 -88;

and nihilism, 79 ;

as normative system,

Politics

157-58, 160 ;

and power balance between judges and juries, 42 ;

and rhetoric, 100 -101, 107 n63;

and self-limiting government, 176 ;

and self-revising context, 95 -100;

and social goals, 95 -99;

and standardization, 152 , 160 ;

and subjectivity, xiii .

See also Democracy; Liberalism; Social democracy; Superliberalism

Positivism, 5 , 156 , 158

Posner, Richard, 105 n44

Post, Robert, 168 , 186 n14

Postmodernism, xiv -xv, 33 , 80

Poststructuralism, 109

Pound, Roscoe, 46 -47, 48 , 69 n18

Powell, Jeff, 106 n49

Power:

and bio-power, 138 -39, 158 -59;

and boundary metaphor, 177 ;

and contextual constraint, 84 ;

and freedom, xiv ;

and institutions, xvi n17;

and juridical systems, 138 -39, 158 -59;

monarchical, 138 -39, 159 ;

Nietzschean will to, 108 n63;

and normative system, 138 -39, 153 -54, 158 -59;

and separative self, 180 -81;

Pragmatism, 37 n10, 158

Prayer, legislative, 13 -14, 15 -18, 19 , 36

Precedents, and law/fact distinction, 42 -43

Privacy, 167 , 168 , 170 , 177 -78, 186 n14

Private Property and the Limits of American Constitutionalism (Nedelsky), 162

Probability, statistical, 142 -44, 156

Proceduralism, 79 , 89 , 91 , 99 , 108 n64

Property, private:

and autonomous self, 162 -63, 171 -72;

and boundary metaphor, 162 -63, 167 -68, 171 -72, 177 ;

and common law, 185 n8;

and epistemology, 109 , 114 -17;

forfeiture of, 50 -59, 66 , 71 nn38-40, 73 n56, 75 n79;

and legal discourse, 56 -57;

and liberal self, 89 ;

and natural rights, 110 , 119 , 121 , 122 , 130 ;

quasiderelict, 59 , 75 n78;

and separative self, 167 , 180 -81;

threatened by democracy, 163 -67, 174 , 183 ;

and U.S. Constitution, 162 -67, 181 , 183 .

See also Inheritance; Literary property

Public domain, 113 , 126 , 128 -29, 130 -31

Publishing trade:

and absolutism, 111 -14;

and accountability, 118 -20;

and censorship, 111 , 115 , 117 , 120 ;

and copyright, xii , xiii , 110 -11;

and Enlightenment, 114 -17, 128 ;

and freedom of the press, 117 -21, 124 , 134 n25;

and French Revolution, 110 -11, 117 -31;

and libel, 115 , 117 -18, 120 -21, 125 , 128 ;

and literary privilege, 111 -29;

monopolization of, 112 -13, 115 , 117 -18, 122 , 124 , 125 ;

pirating in, 118 , 127 ;

and sedition, 115 , 117 -18, 120 -21, 125 , 128 ;

and theatrical works, 122 , 125 -27.

See also Literary property

Q

Quetelet, Alphonse, 144 -46

R

Racine, Jean, 120 , 128 , 137 n83

Rationality:

instrumental, 36 ;

normative, 158 ;

probabilistic, 142

Rawls, John, 79 , 88 , 105 n38, 185 n10

Realism, 43 , 67 n1, 68 n9, 89 , 146

Reboul, Eugène, 147

Reich, Charles, 167

Religion:

and constitutional interpretation, 13 -14, 15 -18, 19 , 36 ;

and normative system, 139 ;

and superliberalism, 79 , 82 , 99

Rich, Adrienne, 186 n14

Riggs v. Palmer , 44 -66

Rights:

of authors, 114 , 117 -31;

and boundary metaphor, 162 , 167 ;

civil, 51 , 72 n44, 166 -67;

natural, 110 , 119 , 121 , 122 , 130 ;

political, 51 , 72 n44, 166 -67;

property, 121 , 130 , 162 -63, 184

Ripstein, Arthur, 187 n36

Risk, in insurance system, 141 -42, 146 -47

Ritter, Joachim, 139

Roberts, Owen, 14

Robespierre, Maximilien, 99

Roe v. Wade , 30 , 40 n84

Roman law, 69 n19

Romme, Gilbert, 127

Rorty, Richard, 83

Rose, Mark, xii , 110 , 131 , 134 nn21, 32

Rousseau, Jean-Jacques, 120 , 128 , 130 -31

Royer, Jean-François, 118

Rule of law, vii , 9 , 16 , 19 , 27 -28, 35 , 39 n72, 185 n8

S

Sandel, Michael, 185 n10

Sartine, Antoine de, 114 , 133 n18

Say, Jean-Baptiste, 143

Schelling, Friedrich Wilhelm Joseph von, xiv

Scheppele, Kim Lane, viii , xii

Schiller, Johann Christoph Friedrich von, xiv

Scientific revolution, 142

Searle, John, 61 n13

Sedaine, Michel Jean, 137 n83

Sedition, 115 , 117 , 120 -21, 125 , 128

Seidman, Peter, 74 n73

Self:

autonomy of, x , xiv -xv, 95 , 99 , 168 -72, 186 n14;

and boundary metaphor, 167 -72, 176 ;

and constructive relationships, 168 -69;

contextual revision by, 93 , 94 -100;

integrated, 184 ;

and liberalism, 89 ;

and political power, xiii -xiv;

and self-expression, 94 -96, 98 ;

separative, xi , 167 , 171 -72, 178 -80, 183 -84;

and social goals, 95 -98;

and social solidarity, 86 .

See also Individualism; Individuality; Subjectivity

Selznick, Philip, 24 , 32 -33

Semantics, 8 , 55 -59

Sexism, 179

Sexual harassment, 75 n75

Sexuality, and boundary metaphor, 169 -70, 188 n58

Shils, Edward, 186 n14

Sieyès, Emmanuel, 117 -28, 130

Simmel, Georg, 168

Simpson, A. W.B., 43 , 74 n66

Situation:

legal interpretation of, 2 , 3 ;

pragmatic meaning of, 37 n10

Skepticism, 79 , 81 , 89 -90, 93 , 96 , 97 , 103 n11

Social democracy, 78 -79

Social structure, 80 -81, 88 , 91 , 97 , 186 n14

Social Theory (Unger), 82 -85, 89 -92, 94 , 97

Solidarity, social, 83 , 86 -88, 91 -92

Standardization, xii , 148 -49, 152 , 159

Stare decisis , 16 -17, 19 , 26 -28, 33 , 39 n72, 40 nn74-75

Starhawk, 173 -77

State:

and individual autonomy, 167 , 169 ;

as normative system, 153 , 155 , 158 ;

and property rights, 162 -67, 177 ;

and social insurance, 147 -48, 153 ;

and subjectivity, 78

Statistics, 142 -44, 156 , 158

Statutes:

constitutionality of, 14 -15;

interpretation of, 5 , 45 -48, 59 -60

Stith, Kate, 75 n82

Stoicism, 139

Stone, Martin, viii , xv

Structuralism, 44

Subjectivity:

and epistemology, 117 ;

and political power, xiii -xiv, 99 ;

and self-expression, 94 -95;

and the state, 78 ;

and transcendence, 108 n67.

See also Individualism; Individuality; Self

Summer, L. W., 106 n59

Superliberalism:

and constitutional interpretation, 89 -90;

and contextual constraint, 81 , 83 -86, 88 , 94 ;

and contextual revision, 90 , 93 , 94 -100;

as critique of liberalism, 78 -81, 89 ;

and formalism, 83 , 100 , 104 n18;

instrumentalism in, 91 -93, 95 -96;

legality in, 81 -82, 83 , 86 -88;

metaphysics in, 80 , 81 , 83 , 99 , 102 n3;

and nihilism, 79 , 83 , 89 , 96 , 97 ;

political program of, 78 , 80 , 83 , 89 -90, 98 -100;

proceduralism in, 79 , 89 , 91 , 99 ;

rhetoric of, 100 -101, 107 n63;

role of God in, 79 , 82 , 99 , 100 ;

and self-expression, 94 -97;

and skepticism, 79 , 81 , 89 -90, 93 , 96 , 97 ;

and social context, 80 -81, 83 -86, 90 , 93 ;

and social goals, 95 -99;

and social plasticity, 80 , 91 -93;

and social totality, 80 -82;

and solidarity, 83 , 86 -88, 91 -92

Supreme Court, U.S.:

and antidemocratic decision making, 15 , 26 , 34 , 89 ;

and constitutional constraint, 31 ;

and legislative prayer, 14 , 15 -18, 19 ;

and racial equality, 30

T

Texts:

application of, 2 -3, 25 -26;

authorship of, 4 , 6 , 104 n20;

and canon formation, 3 ;

and constitutional law, 21 ;

context of, 69 n13, 84 , 88 ;

conversation with, 25 ;

cultural mediation of, vii ;

identification of, 1 -3, 104 n20;

and intertextuality, ix ;

and law/fact distinction, 43 -44, 60 ;

and meaning as reception, 131 ;

phenomenological reading of, 36 n7;

and textualism, 14 , 36 n9

Thayer, James Bradley, 43

Theology:

and subjectivity, 108 n67;

and superliberalism, 79 , 82 , 99 , 100

A Theory of Justice (Rawls), 79 , 88

Thoreau, Henry David, 82

Thus Spake Zarathustra (Nietzsche), 94 , 107 n63

Tolstoy, Leo, 3

Torts, 4 , 68 n7, 168 , 186 n14

Totality, social, 80 -82

Tractatus Logico-Philosophicus (Wittgenstein), 103 n11

Treason, 50 , 71 n39

Trespass, 75 n74

Trial courts:

division of labor in, 42 , 61 ;

and law/fact distinction, 42 , 61 -62, 67 n3;

and legislative prayer, 14 ;

role of judiciary in, 42

Tribe, Laurence, 167 -68, 185 n12

Trusts, 66

Turgot, Anne-Robert-Jacques, 115

Tushnet, Mark, 70 n38

Tussman, Joseph, 23

U

Ulpian, 69 n19

Unger, Roberto:

and Critical Legal Studies, xiv -xv, 78 -79, 83 , 86 , 99 , 101 ;

and critique of liberalism, 78 -81, 89 ;

metaphysics of, 80 , 81 , 83 , 99 , 102 n3;

and nihilism, 79 , 83 , 89 , 96 , 97 ;

political program of, 78 , 80 , 83 , 89 -90, 98 -100;

political rhetoric of, 100 -101, 107 n63;

theology of, xiv , 79 , 82 , 99 , 100 ;

works of:

Critical Legal Studies Movement , 83 , 88 -97, 99 -101;

False Necessity , 79 , 88 -94, 96 , 99 ;

Knowledge and Politics , 78 -83, 87 -88, 90 , 100 ;

Passion , 79 , 83 , 85 , 86 , 89 ;

Plasticity into Power , 91 ;

Social Theory , 82 -85, 89 -92, 94 , 97

Uniform Probate Code, 66

V

Values:

cultural mediation of, vii ;

and fact/value distinction, 79 , 80 , 98 , 107 n62;

fundamental, 89 ;

and legal discourse, vii -viii;

and nihilism, 79 , 89 , 96 ;

and self-revising context, 95 -97;

superseded by normative system, 154 -56, 157

Viala, Alain, 132 n7

Vining, Joseph, vii , viii , ix

Vitruvius, 139

Voltaire, 120 , 128 , 130 -31

W

Walden (Thoreau), 82

Walzer, Michael, 81

Weber, Max, 5

West, Robin, 186 n19

Whitehead, Alfred North, 179 , 180

Whitman, Walt, 83

Williams, Raymond, vii

Wills:

and civil death, 51 -53, 66 ;

interpretation of, 45 -48, 59 , 66 ;

and legal discourse, 56 -57;

and life insurance, 49 ;

and right of dower, 70 n31

Wittgenstein, Ludwig, 74 n70, 86 , 103 n11, 107 n63

Women:

and abortion rights, 39 n74;

and privacy, 186 n14;

and sexual boundaries, 169 -70, 186 n19

Women as Mythmakers (Lauter), 181 -82

Woodmansee, Martha, xii , 110 , 131 , 134 nn21, 31 , 32

Woolf, Virginia, 8 -9

Wright v. Wright , 51

Writing:

distinguished from reading, 5 -6;

expressive, 7 -8;

manipulative, 7

Wroblewski, Jerzy, 37 n10

Y

Your Child's Self-Esteem (Briggs), 171 -72

Preferred Citation: Post, Robert, editor. Law and the Order of Culture. Berkeley:  University of California Press,  c1991 1991. http://ark.cdlib.org/ark:/13030/ft9q2nb693/