Deconstruction's Legal Career
J. M. Balkin

Deconstruction began as a series of techniques invented by Jacques Derrida, Paul de Man, and others to interpret literary and philosophical texts. These techniques, in turn, were connected to larger philosophical claims about the nature of language and meaning. One such assertion is that the repetition of a text in a new context often subtly changes its meaning. There could be no better example of this principle than the career of deconstruction itself. To be adapted to the needs and concerns of the legal academy, deconstruction had to be translated and altered in significant ways, making it more flexible, practical, and attentive to questions of justice and injustice. This essay discusses some of the changes that deconstruction underwent as it moved from philosophy to literature and then to law. Its transformation eventually produced a deconstructive practice in law that emphasizes a sensitivity to changes in interpretive context, a pragmatic attitude towards conceptual distinctions, and a careful attention to the role of ideology and social construction in legal thought.

Deconstruction was first imported from Continental philosophy to American literature departments and later migrated to American law schools. Deconstruction became fashionable in America about the same time as reader response theory, which held that the meaning of a text is produced as the reader encounters it. As a result, deconstruction became wrongly associated with the improbable claim that texts mean whatever readers want them to mean. This notion is not only a misinterpretation of deconstruction, but also of reader response theory.

This idea would have seemed especially silly in Europe, where deconstruction arose out of an earlier philosophical movement called structuralism. Because deconstruction was understood as a reaction to structuralism, it is sometimes called a "post-structuralist" philosophy.

Both deconstruction and structuralism are antihumanist theories; that is, they tend to emphasize that people's thought is shaped and determined by structures of linguistic and cultural meaning. Both deconstruction and structuralism asserted that people are culturally and socially constructed, and that they internalize culture much in the same way that they internalize a natural language. The linguistic analogy is particularly appropriate: A speaker of English cannot make the words of that language mean whatever she likes; more importantly, she doesn't even want this to be the case, because part of being an English speaker is having internalized a sense about what is the proper way of talking and thinking. For the structuralist and the deconstructionist, language speaks us as much as we speak it.

As this example demonstrates, an antihumanist approach tends to deemphasize people's autonomy; it views people as the product of cultural forces largely beyond their control. In its most extreme forms, it tends to dissolve the individual (sometimes called "the subject") into structures of social meaning. Deconstructionists attacked the structuralist contention that there were universal and fixed structures of meaning that shaped all human thought. They argued that the structures of social meaning are always unstable, indeterminate, impermanent and historically situated, constantly changing over time and accumulating new connotations. But these criticisms didn't challenge the idea that individuals were constructed by larger social and linguistic forces, and deconstructionsts didn't dispute that individuals were not fully in control of the forces of social and linguistic meaning.

How this antihumanist philosophy got translated into the claim that people can make texts mean whatever they want is something of a mystery. The indecipherability of a lot of deconstructive writing probably contributed to the confusion. In addition, the point of much of contemporary literary criticism is to come up with new and original readings of older texts. This project doesn't sit well with a philosophy that asserts that the critic, like the author, doesn't have very much control over what she thinks.

When deconstruction moved from literature departments to the legal academy, it was modified further. Legal academics on the left, and particularly feminists and members of the Critical Legal Studies movement, latched on to deconstruction as a way of challenging legal orthodoxies. They assumed pretty much without question that deconstructive techniques could be adapted for the purpose of critiquing unjust legal doctrines and advocating more just arrangements. Once again, this assumption is rather puzzling. It is true that many literary deconstructionists identified with the political left. But they were using deconstruction to show the impenetrability, mutability, and conceptual incoherence of all texts, not simply the texts produced by political conservatives. In fact, in the literary world many people argued that deconstruction was a profoundly conservative movement: The recognition that any text could be deconstructed and that all meanings were unstable led inevitably to political quiescence. Nevertheless, these warnings did not raise much alarm in legal circles. Legal academics began deconstructing left and right, or, more correctly, almost exclusively to the left. No attempt was made, at least at first, to consider whether deconstruction had a necessarily or predominantly progressive slant, or whether it on the contrary it was particularly unsuited to political critique because it threatened to undermine any political program or conception of social justice.

Bringing deconstruction into law schools also required a new focus on the point of deconstruction. Legal theorists were primarily interested in using deconstruction for normative or critical purposes. They wanted to criticize some (but not other) doctrinal distinctions as incoherent, they wanted to show that some (but not other) parts of the law were unjust and needed reform, and they wanted to demonstrate that some (but not other) ways of thinking had undesirable ideological effects that concealed important features of social life and therefore promoted or sustained injustices. But each of these tasks faced two central problems.

The first difficulty was that literary deconstruction was not primarily devoted to these normative and critical purposes. Indeed, literary deconstruction spent much of its time showing the ambiguity, uncertainty, and impenetrability of all literary texts, the reversibility of all positions, and the instability of all theoretical conceptions. It did not focus specifically on pointing the way to a more just world.

To use deconstruction lawyers would have to modify it so that it could serve a critical and normative function. Deconstruction would have to mean something different for lawyers than it would for literary critics and philosophers. Deconstruction would become a series of rhetorical strategies for criticizing certain legal doctrines and legal arguments in order to show that they were unjust, ideologically biased, incomplete, or incoherent. It would necessarily discriminate between better and worse positions and interpretations, and it would state its conclusions in the language of normative prescription. Deconstructive legal argument, in short, would become a form of normative legal thought. Among legal scholars, Pierre Schlag has been virtually alone in recognizing this change and speaking out against it.(1) Schlag has thus taken the seemingly paradoxical position of being against the use of deconstruction for normative purposes.(2)

The second problem in bringing deconstruction into legal thought arose from the first. Implicit in the task of normative criticism was the assumption that other legal arguments and legal doctrines were not in need of such critique, even though they too could be deconstructed in theory. Thus, legal deconstruction had to distinguish what was deconstructible (everything) from what was unjust, practically incoherent, or practically unworkable. Legal deconstruction necessarily aimed only at certain targets, and not others. Left legal scholars wanted to deconstruct Justice Scalia's opinions on affirmative action, but not Justice Brennan's establishment clause opinions. In legal scholarship at least, deconstruction necessarily had a pragmatic orientation. If it wasn't broken, one wouldn't deconstruct it.

Of course, this selectivity was always present in literary and philosophical deconstruction too, although the reasons for selecting particular texts for deconstruction were somewhat different. Derrida himself was very particular about which texts he chose to deconstruct, because he thought that some (but not others) were rich sources of philosophical insight. Deconstruction's univeralistic pronouncements about the problematic nature of all texts, all language, and all readings tended to obscure the actual and selective use of deconstruction by literary critics and philosophers. The movement of deconstruction to law made this selectivity more perspicuous. It made clear that whatever deconstruction itself might be, deconstructive argument is, and always has been, a rhetorical technique used by scholars for pragmatic purposes.(3) Derrida, after all, did not go around deconstructing his own arguments to show that they were incoherent.

The movement to law also revealed more clearly that deconstructive analysis involves a series of repeatable rhetorical devices or tropes that can be adapted to many different problems and situations. Derrida and his followers have always insisted that deconstruction is not a method, and that it cannot be reduced to a set of techniques. But this assertion is undermined by their actual practices of reading and argument. After all, if deconstruction was to be perpetuated in the next generation of graduate students, these students had to learn how to do it, and this meant that there had to be some set of skills that could be repeated and transmitted from teacher to pupil. As a result, the actual practice of deconstruction has tended to revolve around a relatively small handful of rhetorical techniques. This claim is of course, as scandalous among deconstructionists as the claim that there are stock answers to many seemingly inscrutable Zen koans and the revelation that there is in fact no Santa Claus, but it is no less accurate for all of its unpleasantness.

The most basic forms of deconstructive argument involve the manipulation of conceptual oppositions.(4) One looks for the "privileging" of one pole of an opposition over the other. This privileging can occur in a text, an argument, or a social or historical tradition. It can be presupposed or stated overtly. Given a conceptual opposition between A and B, A is privileged over B if A is the general case and B the special case, if A is primary and B secondary, if A is normal and B is deviant, if A is of higher status and B is of lower status, or if A is central and B is marginal. A can also be privileged over B if A is considered more true, more relevant, more important, or more universal than B. Conceptual oppositions exist in larger networks of cultural meanings and associations. Hence A can have many opposites besides B, depending on the context in which it is considered. (Liberty, for example, can be opposed to slavery or to equality.) A concept can be privileged in many different ways in opposition to many different concepts and many different things can simultaneously be privileged over it.

Once identified, a conceptual hierarchy can be deconstructed in a number of ways. Normally the privileging of A over B is justified by reasons, either explicit or implicit. So the deconstructor can ask whether the reasons why A is privileged over B actually apply to B as well, or the reasons why B is thought subordinate to A are actually also true of A. Alternatively, one can try to show that A is a special case of B, or that A's existence or conceptual coherence depends on the thing it excludes or subordinates, namely, B. The point of these various techniques is to rethink the relationship between conceptual opposites and observe similarities and conceptual dependencies that were previously hidden or submerged. Of course, such a deconstructive analysis, once undertaken, can always be extended further. In theory, the possibilities of deconstruction are endless, but in practice the process of deconstruction must come-- at least temporarily-- to a halt. This fact further reveals the essentially pragmatic nature of deconstructive analysis in law that was always present but was not always so clearly recognized in literary and philosophical versions.

A good example of how one deconstructs a privileging is Derrida's well-known discussion of speech and writing.(5) According to Derrida, many philosophers have valued speech over writing because speech is thought to be closer to the essence or "presence" of thought. The speaking individual thinks as she speaks, and her words immediately convey what she wants to say. Writing, on the other, hand, is an inferior mode of expression because it can be separated from the presence of the writer's thoughts. It can be repeated in new contexts and misunderstood, and the author is not always present to correct any misunderstanding.

It is by no means clear that Western thought has featured the sort of pervasive prejudice against writing that Derrida describes. In any case, his deconstruction of the privileging runs as follows: If writing is less valuable because it consists of signs that can be separated from the author's thoughts, this is also true of speech. Speech is also just a set of repeatable signs that can be divorced from the author's thoughts and intentions; consider a recording of a person's speech played over and over out of context. In fact, speech is just a special case of those features of writing that philosophers considered inferior and untrustworthy. Speech is also representational and repeatable in new contexts that change its meaning and create the possibility of misunderstanding. In this sense, speech is just another kind of writing. Of course, when we say that speech is a kind of "writing," we are really using "writing" as a symbol that stands for its most salient characteristic-- a linguistic representation that can be separated from its author. The point of Derrida's deconstruction is to show that apparent opposites actually share important characteristics-- their semiotic nature as repeatable signs.

The basic technique of reversing conceptual privilegings has obvious applications to legal and policy argument. Law is full of conceptual oppositions because it is full of distinctions. A distinction creates a conceptual opposition because it separates things inside the category from things that fall outside it. Given a doctrinal or theoretical distinction between A and B, the legal scholar can locate the justifications for the exclusion of B from the class of A's and then proceed to deconstruct them. She can argue that the justifications for the distinction undermine themselves, that categorical boundaries are unclear or at odds with the proffered justifications, or that the boundaries shift radically as they are placed in new contexts of judgment.

Critical scholars in the feminist and critical legal studies movements made the most frequent and familiar use of deconstruction in law.(6) They employed deconstructive techniques to discover and critique ideological commitments they claimed underlay legal doctrine. Deconstruction has proved useful for ideological critique because ideologies often work through forms of privileging and suppression: Certain features of social life are privileged in thought and discourse, while others are marginalized or suppressed. Deconstructive arguments try to recover these subordinated or forgotten elements in legal thought and legal doctrine.

A good example is the familiar critique of contract law offered by CLS scholars.(7) It contrasts a dominant story about human relations in contract law with a marginalized story. It then imagines how the marginalized story might be expanded to rethink the law of contract. According to the CLS critique, The dominant story of human relations in contract law assumes autonomous individuals freely choosing the terms of their bargains, and accepting full responsibility if they choose badly. However, the Critics point out, much contract law (and much contracting) does not fit this model. Rather, it is best understood as invoking an alternative story about human relations. In this story, individuals cooperate with and rely on each other; they expect that they and others will not take advantage of each other even when this might be technically permissible. Individuals are interdependent and need to cooperate to survive. They often lack information, are emotionally vulnerable, and are often overborne by circumstances not of their own making. Institutions like the market can be unfairly coercive and oppressive even as they purport to be the home of freedom and self-realization. Deconstructive readings of contract law would explore how this countervision, if taken seriously, would change our understandings of appropriate contract doctrine.

A related approach studies the legal justifications offered within a body of doctrine. Often these are organized around a dominant principle or set of principles that are, in turn, connected to visions of human nature and social relations. The deconstructionist looks for a set of marginalized counterprinciples. These counterprinciples are either entirely suppressed or are conceded to have force only in exceptional or deviant situations. In this way one shows that the body of law features a hierarchy of privileged and subordinated elements and justifications. The deconstructionist then tries to show two things: First, these subordinated counterprinciples have a greater significance and importance to maintaining the intellectual coherence of the body of doctrine than most people had thought. Second, if one took these suppressed counterprinciples seriously and expanded their sphere of application and influence, they might radically change the content of legal doctrine.

Each of the above techniques are variations on the basic idea of reversing conceptual hierarchies. Yet another technique focuses on the rhetoric of legal argument. It studies the stylistics and word choices in a text. Often the rhetorical features of a text undermine or contradict the argument made by the text: What the text does is often in tension with what it says. Deconstructors can also look for unexpected relationships between seemingly unconnected parts of a text, or use the marginal elements of a text as an uncanny commentary on what appear to be its central elements.(8) Deconstructors also can play with the multiple meanings or the etymology of key words in the text to tease out possible conflicts or ambiguities. Puns and plays on words are often invoked to show interesting connections and unexpected tensions between different parts of the argument. Although these techniques focus on apparently peripheral elements of the text that seem unrelated to the logic of its argument, they often help display disturbances or problems in the logic of the text that might otherwise have gone unnoticed.

For example, Justice Scalia has recently argued that the human rights guarantees of the Constitution should be determined by looking to the longstanding traditions of the American people.(9) It turns out that "tradition" comes from the same word as "betrayal."(10) Both involve a handing over. Claiming to speak in the name of tradition can also be a kind of betrayal in several different ways. First, hewing to one's vision of tradition obliterates other interpretations of the past and other alternatives for the future. Thus a tradition can be a kind of extradition, banishing other perspectives and handing them over to their enemies, so to speak. To respect tradition is also to betray, submerge, and extinguish other existing and competing traditions. It can lead us to focus on a falsely unitary or unequivocal story about the meaning of the past when we should recognize the past as a complicated set of perspectives in tension with each other. Tradition never speaks with one voice, although, to be sure, persons of particular predilections may hear only one. Finally, to act in the name of a tradition is often to betray the tradition itself, by disregarding the living, changing features of a tradition and substituting a determinate and lifeless simulacrum.

These rhetorical techniques are actually another example of the general strategy of reversing privilegings and hierarchies. We ordinarily assume that the figural features of a text are supplementary and inessential to the coherence of its argument. But deconstructive techniques assume that there is no permanent and acontextual boundary between the figural and logical aspects of a text. More generally, deconstruction argues that rhetoric has an important though neglected function in supporting our conceptions of what is reasonable. Metaphor, metonymy, and other figures play unacknowledged roles in the construction and judgment of legal and political argument. Sometimes figural elements provide essential support for the reasoning we find in legal and political texts; just as often, they uncannily comment on and undermine this reasoning.(11)

Deconstructive readings are not premised on the notion that texts have no meaning or that their meanings are undecipherable. Rather, deconstruction argues that texts are always overflowing with complicated and often contradictory meanings. The predicament that deconstruction finds in texts is not the lack of meaning but a surplus of it. Similarly, the point of deconstructing conceptual oppositions is not to show that concepts have no boundaries, but rather than their boundaries are fluid and appear differently as the opposition is placed into new interpretive contexts. Deconstruction is not a mechanical demonstration of total indeterminacy. Deriving interesting results from deconstructive techniques is a skill that requires sensitivity to changes in interpretive context. Moreover, deconstructing a legal distinction does not necessarily show that it is incoherent. That is a pragmatic judgement to be made by the interpreter. All legal distinctions are in principle deconstructible, but not all are unworkable; their usefulness depends on the context in which they are employed.

To take a well-worn example, critical scholars have often deconstructed the familiar distinction between public and private choice and responsibility. But the reason to deconstruct this distinction is not because these concepts have no value or use. The problem is rather that they are often used in a conclusory fashion that hides the important policy choices at stake.

The deconstruction of the public/private distinction softens the boundaries between these two categories, but does not abolish them. For if we try to abolish the distinction between public and private power-- by arguing that everything is an effect of public power or a choice for which government is responsible-- we will find the concept of the private returning in a new guise. Public and private form exist in a relationship of différance-- a technical term that means that the two concepts are mutually dependent as well as mutually differentiated. Because they are mutually dependent, one cannot get rid of one without destroying the other. Moreover, to abolish the distinction between them privileges their differentiation over their mutual dependence, suppressing or hiding the latter. This new privileging is itself subject to further deconstruction.

One might deconstruct the distinction between public and private to argue that the state has a duty to prohibit private acts of racial and religious discrimination. It is no argument that these harms arise from private preferences. After all, the state protects its citizens from many other economic and social harms produced by private choice: minimum wage and maximum hour laws are two examples. The government's failure to protect its citizens from the harms of racial and religious discrimination is a delegation to private parties to inflict harms that it does not permit in other contexts. It is a regulatory choice for which the state is ultimately responsible.

But our deconstruction cannot wholly rid us of the public/private distinction, for it will always reappear in a new guise. For example, does it follow from the previous argument that the state has a duty to provide legal relief to parties who were spurned by prospective spouses or lovers because of racial or religious differences? There are good reasons to think that the state should not be held responsible for these discriminatory acts. After all, the selection of one's spouse or lover is a matter of individual choice and the government should not interfere with it. But to accept this argument we must reinscribe the notion of a realm of private power, choice, and responsibility that is distinct from the public. Our deconstruction of the public/private distinction has itself been deconstructed.

To be sure, one could deconstruct this deconstruction. The claim that the government should not interfere with one's choice of spouse or lover is problematic; the government always shapes one's choices through its selection of legal regulations. After all, the government's failure to proscribe spousal discrimination when it actively proscribes workplace discrimination can hardly be viewed as a neutral stance or a position of nonintervention into private decisionmaking. Its selectivity surely affects the conditions under which people choose their spouses and lovers. Moreover, people's preferences in spouses and lovers are also influenced by their education and upbringing. The state chooses to instill many different values through its educational system --including racial and religious tolerance-- but not the importance of nondiscrimination in choice of spouses. Its particularity on this point belies its claim that it has no responsibility for producing a society in which people are unwilling to intermarry. Of course, these responses are hardly irrefutable, but to refute them one has to reinscribe the notion of private choice, power and responsibility in a new form. Thus, one might object that regardless of the state's role in shaping choices through regulation and education, the state should not be permitted to brainwash its citizens to do its bidding even if citizens make choices that the state does not approve of. The distinction between public and private constantly reappears, no matter how often we attempt to cast it out.

This example leads to a larger point. When we deconstruct conceptual oppositions, we are not necessarily trying to show that they form a false dichotomy. We are trying to show that they form a nested opposition.(12) A nested opposition is a conceptual opposition in which the two terms "contain" each other. The metaphor of "containing" one's opposite stands for a number of related concepts--similarity to the opposite, overlap with the opposite, being a special case of the opposite, conceptual or historical dependence upon the opposite, and reproduction of the opposite or transformation into the opposite over time. These forms of containment share a sort of Wittgensteinian family resemblance--they all bear similarities to each other, although one cannot point to a single property that all have in common.

The most general way of stating the relationship between the terms of a nested opposition is that they bear a relationship of mutual dependence and differentiation. The point of deconstructive analysis is to show how this similarity or this difference has been suppressed or overlooked. It tries to find difference and antinomy in purported similarity and similarity and mutual dependence in purported differentiation. As a result, deconstructive arguments usually emphasize the contextual nature of judgements, because the relationship between conceptual opposites and the relative meanings of the opposed concepts change as they are inserted into new interpretive contexts.

Like the opposition between public and private power, Derrida's famous example of the hierarchy of speech and writing is a nested opposition. Derrida deconstructs the distinction between them by showing that both are special cases of the same semiotic phenomenon. But a nested opposition is not the same thing as a false opposition. Even after Derrida deconstructs the hierarchy of speech over writing, he has not abolished the distinction between speech and writing, for speech is still spoken and writing is still written. Derrida could only claim to have effaced the distinction between them if he maintained that the differences between them are unimportant to their true nature. An essential property of speech and writing is their ability to function as repeatable signs; an accidental or inessential property is the fact that they are produced by a voice or with a pencil. However, this commits Derrida to a distinction between essential and inessential properties, a distinction he would be unlikely to accept. It would be especially ironic if he did: The deconstructor can only justify the effacement of distinctions by asserting the essential character of other properties and distinguishing them from accidental properties. This effaces one distinction only to assert another, necessary one.

The concept of nested opposition avoids these difficulties; the reliability of a conceptual opposition depends on the context in which it is understood and the purposes to which it is put. In some interpretive contexts, the opposition is useful and makes sense; while in others it does not. The recurrent error that deconstructive arguments identity is the overextension of conceptual distinctions to new contexts and new purposes where they may be inoperable, obstruct understanding, and promote injustice. But deconstruction does not argue for the destruction of all conceptual oppositions. The latter view is not only incorrect, but incoherent, for deconstructive arguments implicitly rely on conceptual oppositions to do their work.

The notion of a nested opposition is related to another important deconstructive concept: iterability. Iterability is the ability of a sign to be repeated again in a new context. When a sign is repeated in a new context, it takes on a new set of cultural meanings and associations, which are both similar to and different from the previous incarnations. Thus, the iteration of the same sign in a new context creates a nested opposition between the two significations. In repetition the same text occurs twice, but its meanings are partially different. Thus, repetition always creates the possibility of a divergence or opposition within a unity of meaning, because new contexts reveal new and different cultural associations. Derrida sums up this phenomenon in his aphorism that "iterability alters."(13)

One application of the principle of iterability is the phenomenon of "ideological drift." Politicians and theorists alike continually make arguments that appeal to abstract policies and principles. But when a policy or principle is repeated over and over again in new cultural surroundings, its meanings and cultural associations begin to change, and hence its political valence begins to shift.(14) A good example is the trope of colorblindness in equal protection law. In 1896 (the year of Justice Harlan's famous dissent in Plessy v. Ferguson(15)) colorblindness was a radical assertion of racial equality; by 1996 it had become the rallying cry of racial conservatives. The social context in which the trope of colorblindness appeared had changed in the intervening years. As a result, liberals who supported affirmative action and conservatives who opposed it could both claim to be heirs of Martin Luther King and his dream of an America where children were judged not by the color of their skin but the content of their character. For liberals, King's appeal to colorblindness in the 1960's was a call for an end to racial subordination in all of its forms, while for conservatives the trope of colorblindness applied to the advantage of whites as well as blacks. Liberals thought that the trope had outlived its usefulness when it was no longer connected to the project of ending racial subordination, while conservatives claimed that it continued to embody a fundamental principle of political fairness, as valid now as a hundred years ago.

The phenomenon of ideological drift is related to the multivalent meanings of a tradition, and to the important connections between tradition and betrayal. The phenomenon of ideological drift guarantees that the concrete exemplars and symbols of a tradition will take on multiple and conflicting meanings over time. As a result, different groups can claim to be faithful adherents of the tradition and yet wish to continue it in radically different ways. Traditions are thus the result of interpretive struggles between adherents all of whom claim to be faithful to the tradition. However, each group wishes to consolidate and continue it in ways that seem like a betrayal from the perspective of the other camps. We might say then, that the seeming unity of any tradition is actually, on closer inspection, a complicated set of nested oppositions, whose conflicts appear only with the passage of time and the arrival of new circumstances. Traditions often try to submerge and suppress their multiple meanings, enshrining some interpretations as orthodoxy and banishing others as heresy. Yet the multiplicity of meanings, and the instability of interpretations continue to emerge incessantly as the tradition travels through history.

1. Pierre Schlag, "`Le Hors' de Text, C'est Moi:' The Politics of Form and the Domestication of Deconstruction, 11 Cardozo L. Rev. 1631 (1990).

2. Pierre Schlag, Normativity and the Politics of Form, 139 U. Pa. L. Rev. 801 (1991); Pierre Schlag, Normative and Nowhere to Go, 43 Stan. L. Rev. 167 (1990).

3. J.M. Balkin, Tradition, Betrayal, and the Politics of Deconstruction, 11 Cardozo L. Rev. 1113 (1990).

4. J. M. Balkin, Deconstructive Practice and Legal Theory, 96 Yale L.J. 743 (1987).

5. Jacques Derrida, Of Grammatology (Baltimore: Johns Hopkins Univ. Press 1976).

6. See, e.g., Gary Peller, The Metaphysics of American Law, 73 Calif. L. Rev. 1151 (1985); Clare Dalton, An Essay in the Deconstruction of Contract Law, 94 Yale L.J. 997 (1985); Gerald Frug, The Ideology of Bureaucracy in American Law, 97 Harv. L. Rev. 1276 (1984).

7. Robert Gordon, Unfreezing Legal Reality: Critical Approaches to Law, 15 Fla. St. Law Review 196 (1987); Roberto M. Unger, The Critical Legal Studies Movement (Cambridge: Harvard University Press 1986); Dalton, Deconstruction of Contract Law, supra.

8. See, e.g., J.M. Balkin, The Domestication of Law and Literature, 14 Law and Social Inquiry 787 (1987), which focused on the dust jacket and the dedication page to Richard Posner's Law and Literature: A Misunderstood Relation.

9. Michael M. v. Gerald D., 109 S. Ct. 2333, 2344 n.6. (1989).

10. Balkin, Tradition, Betrayal and the Politics of Deconstruction, supra.

11. See, e.g., J.M. Balkin, The Footnote, 83 Nw. U. L. Rev 275 (1989), a deconstructive reading of United States v. Carolene Products, 304 U.S. 144 (1938), and its famous footnote four, 304 U.S. at 152 n.4.

12. The theory of nested oppositions is developed more fully in J.M. Balkin, Nested Oppositions, 99 Yale L. J. 1669 (1990).

13. Derrida, Jacques. Ltd., Inc. abc...., Glyph 2:167ff, (1977).

14. J.M. Balkin, Ideological Drift and the Struggle over Meaning, 25 Conn. L. Rev. 865 (1993).

15. 163 U.S. 537 (1896).

Although deconstruction is naturally associated with literary interpretation, its applications to legal interpretation are far less clear. Deconstructing legal texts is not primarily designed to show the impenetrability or undecideability of these texts. The connection between deconstructive theory and the practice of legal interpretation lies elsewhere. Deconstruction is helpful in showing the sensitivity of legal meaning to changes in interpretive context, and uncovering the competing policies and potentialities buried in the words and expressions of legal texts. Often key words in a statute or doctrine disguise a nested opposition of possible interpretations. Deconstruction can uncover these warring conceptions and bring them to the surface.

Finally, deconstructive theory emphasizes that no theory of legal interpretation can be foundational in the sense of offering a primary or central method. None of the familiar principles of text, intention, structure, precedent, consequences, or natural justice can stand as a self sufficient justification for legal interpretation. Nor can any one be elevated above the others as a general rule. Rather, deconstruction argues that interpretation is a pragmatic enterprise drawing on each of these modes of argument in a creative tension. The art of legal interpretation is the art of using the multiple tools of interpretation without being able to rely on any single tool as foundational.(16)

One might think that the text should be the foundation of legal interpretation, with the other modes of argument as mere supplements brought in only in cases of interpretive difficulty. Where meaning is plain and clear, resort to these supplements is unnecessary. Yet even the most obvious examples of seemingly unambiguous texts produce interpretive problems when they appear in new and unexpected contexts. Take, for example, the U.S. Constitution's requirements that each state shall be represented by two senators and that "no State, without its consent, shall be deprived of its equal Suffrage in the Senate."(17) During Reconstruction, the Radical Republican Congress refused to seat delegations from the former Confederacy until they agreed to ratify the Fourteenth Amendment.

The seemingly unproblematic character of the equal suffrage clause becomes complicated once we recognize that a plain meaning approach would tend to undermine the legality of the Fourteenth Amendment.(18) The purpose of this example is not to show that meaning is radically indeterminate, but that its determinacy depends on close attention to context and careful judgment using a number of competing factors, none of which can be considered foundational.Above all, one should not confuse this example with the claim that text and intentions are unimportant to interpretation, and that consequences are all. The deification of consequences is simply foundationalism in another form. Just as text and intention cannot be taken as the touchstone of all legal interpretation, neither can consequences. Implicit in the textuality of a statute is the possibility that its best interpretation can fail to be the best social policy. If consequences were the only determinant of interpretation, there would be no reason to have a written text at all. One could simply decide the best course to take in each case.

Although deconstructive arguments may be found in critical race theory, feminist, and postmodern legal scholarship, deconstruction first emerged most clearly in the work of the Critical Legal Studies Movement. CLS scholars were attracted to deconstruction for three reasons. First, CLS scholars emphasized the instability and indeterminacy of legal doctrines and the political ideologies that lay behind legal reasoning. Deconstruction's discovery of mutability in meanings and conceptual boundaries seemed to support these views. If deconstruction showed that all legal decisionmaking and all legal categories were flexible and mutable, this might seem to buttress the claim that something other than legal reasoning-- like political judgment-- lay behind legal decisionmaking.(19)

Second the force of the deconstructive critique applied beyond legal meanings to all social meanings. Deconstruction seemed to suggest that social structures themselves were also unstable and indeterminate. This meshed well with CLS claims that legal consciousness was based on the "false necessity" of social and legal structures that seemed reasonable in theory but were oppressive in practice.(20)

Finally, deconstruction seemed attractive to CLS scholars because it held that all texts undermined their own logic and had multiple and conflicting meanings. CLS scholars could use deconstructive techniques to "trash" traditional legal arguments and legal distinctions by showing that they were fundamentally incoherent.

Nevertheless, CLS's appropriation of deconstruction was problematic. To begin with, deconstruction's claims of indeterminacy proved too much. If legal argument was indeterminate, so too was political argument, and so CLS scholars could not assert that it was really politics and not legal reasoning that decided cases. Moreover, the idea that judges could manipulate indeterminate legal language seemed to assume an autonomous legal subject who was in control of her beliefs and meanings. But deconstruction argued that individuals were socially constructed and social meanings were largely beyond their deliberate control. Hence it was difficult to claim that legal reasoning was a disguise for political reasoning. Rather, the consciousness of legal scholars and judges was constructed by and their thought shaped by the language of law as much as by the categories of politics.

In fact, deconstruction pointed to a quite different theory of judicial decisionmaking. Instead of arguing that legal doctrine was indeterminate, one could argue instead that social construction placed constraints on legal decisionmaking and produced the felt sense that some arguments were better than others. Legal consciousness as a subset of political consciousness created the sort of legal subjects who naturally understood that certain kinds of claims were "on-the-wall" and others "off-the-wall." Ideology, then, was not a source of indeterminacy but of constraint.(21)

Second, the antihumanist assumptions of deconstruction tended to undermine the notion of "false necessity" implicit in CLS scholarship. Social structures and legal doctrines might be "contingent" in the sense that they did not have to take any particular form, but once they were in place they would not melt away simply by an act of will. Moreover, changes would have to be built using the social meanings already in place. Individuals constructed to see social structures and legal categories as normal and natural would not easily be able to transcend the limits of their perspectives, because they were not in full control of how these perspectives were formed in them in the first place. Deconstruction did suggest that legal and social structures had unstable and flexible boundaries, but it did not imply that these structures could easily be altered through individual thought or effort. Even if legal concepts had multiple meanings, it did not follow that individuals would be able to manipulate and change the shared social meanings of these concepts in anyway they liked; moreover, their social construction suggested that they would not even desire to change shared meanings in some of these ways.(22)

Third, using deconstruction to demonstrate the incoherence of the categories and distinctions of legal orthodoxy proved entirely too much. If the concepts and categories used by the status quo were surely deconstructible, so too would be any concepts and categories offered by critical legal scholars. If deconstructibility meant incoherence, then it also meant the incoherence of any positive progressive program for critical legal studies and of any radical alternatives to liberal legal thought.

British critical thinkers tried a somewhat different strategy -- they argued that mainstream legal thought used rhetorical figures that helped disguise and sustain serious injustices in the legal system.(23) But this approach also tended to prove too much. One could not condemn these moves simply because they were rhetorical. After all, deconstructive arguments suggested that rhetorical and figural language would always play some role in legal and political argument; the only question was whether this influence would be acknowledged or suppressed. Exposing the rhetoricity of contemporary legal argument and legal doctrine could not without more delegitimate it, because the rhetorical nature of practical argument and practical judgment is inescapable.

We can draw two important lessons from CLS's encounter with deconstruction. First, CLS scholars were most successful using deconstruction pragmatically rather than programmatically, as a technique for the analysis of specific problems in specific areas rather than a means of showing global incoherence and indeterminacy in liberal legal thought. CLS scholars used deconstructive methods most fruitfully to tease out suppressed principles in doctrine, unearthing the clash of principle and counterprinciple in seemingly unitary and coherent bodies of law. Deconstructive arguments also came in handy when critical scholars showed how justifications for particular legal doctrines and distinctions undermined themselves, and how the ideologies underlying these doctrines marginalized or suppressed important features of human life. These targeted and focused uses of deconstruction were much more successful precisely because deconstruction works in the interstices of specific texts and specific problems, inviting contextual judgements rather than grand generalizations. Indeed, such generalizations are a prime target of deconstructive scrutiny.

Second, critical scholars employed deconstruction most successfully when they came to terms with deconstruction's antihumanist heritage, which reoriented the so-called indeterminacy critique often associated with CLS. The point was not to show how legal doctrine could be manipulated and used to justify a particular judge's politics. Rather, it was to show how well meaning legal decisionmakers could see the world in such a way that they really felt that they had no choice in how to apply the law.

Traditionally debates over indeterminacy had been organized around the problem of the "rogue judge:" the ideologue whose political predilections led him or her to bend the law to his or her own will. Because of these dangers, the rogue judge had to be constrained by rules and doctrines. That is why the indeterminacy critique seemed so threatening; critical scholars claimed that many legal rules and doctrines could not do the work demanded of them. Critical scholars infuriated defenders of legal liberalism because they asserted that the shackles that could bind the rogue judge were often illusory.

But deconstruction transformed this classic jurisprudential problem. The difficulty was not that ideology made law indeterminate but that it produced a brittle, oppressive determinacy. The problem was not the rogue judge but the sincere judge. This judge was always bound, not simply by doctrine but by the limits of his or her imagination. Social construction caused individuals to understand the world in ways that made it difficult for them to envision alternative ways of ordering law and society.(24)

I noted earlier that many left legal academics had assumed without much debate that deconstruction could be harnessed to promoting a progressive political agenda. In 1987, these assumptions were given a rude shock when scholars learned of Paul de Man's wartime journalism for a pro-Nazi Belgian newspaper. De Man, of course, was one of the major figures in the development of literary deconstruction. In the past literary critics had occasionally charged that deconstructive claims that all social meanings were unstable, all interpretations provisional and all truths uncertain suggested that no definitive moral or political consequences could be assigned to any event or any text. Hence deconstruction was politically conservative or at the very least led to political passivity. Now critics of deconstruction began to insinuate that de Man embraced interpretive theories that revelled in ambiguity and indeterminacy in order to assuage his guilty conscience over collaborating with the Nazis. The ensuing controversy surely affected de Man's friend Jacques Derrida, a Jew who was a teenager during World War II. Whether directly or indirectly, the de Man affair led many thinkers, including Derrida, to consider the question of deconstruction's relationship to justice much more closely.(25)

In response to his critics, Derrida insisted that deconstruction had always been focused on normative questions, and particularly questions of justice. He even offered the provocative claim that "Deconstruction is justice."(26) (At the same time Derrida insisted that justice is impossible, which, one presumes, implies that deconstruction is also impossible.) Drucilla Cornell's "philosophy of the limit" builds on Derrida's work, combining it with feminist legal theory and the work of Emmanuel Levinas.(27) In her view, deconstruction has an important relationship to ethics for two reasons. First, deconstruction calls attention to our ineradicable difference from other persons. Hence, Cornell argues, deconstruction emphasizes our responsibility to respond to and recognize the other as an other. This responsibility in turn demands that we make ourselves open to the views and concerns of other persons. Second, Cornell argues that all systems of positive law have gaps and paradoxes that cannot be overcome. Deconstruction is justice because justice resides in these gaps and paradoxes. In Derrida's terms, justice is an "impossible demand." Deconstruction helps us to see this impossibility in all systems of positive law.

Despite Cornell's efforts on Derrida's behalf, Derrida's own remarks on the relationship of deconstruction and justice seem muddled. Whatever the philosophical underpinnings of deconstruction might be, deconstructive techniques (for example the inversion of conceptual hierarchies and the use of iterability) have no necessary ethical stance. It is easy enough to produce deconstructive arguments for both sides of any policy question, especially since all texts are capable of deconstruction. Derrida's suggestion that deconstruction is always on the side of the angels leads to the implausible position that if a purportedly deconstructive argument were used for a base or immoral purpose, it was not really deconstruction after all.(28)

In equating deconstruction and justice, Derrida has attempted to take the moral high ground in the face of the assaults leveled at deconstruction (and at himself) following the de Man affair. But what Derrida and his followers have not fully faced, I think, is that in practice deconstructive argument is a species of rhetoric, and, like all rhetoric, it can be used for good or for ill depending on how it is wielded. Deconstruction is no more and no less noble than the forms of rhetoric that deconstructionists repeatedly discover in philosophical and literary texts. Deconstruction cannot flee from its own rhetoricity, or the normative consequences of that fact. The deconstructive claim that "iterability alters"-- that texts take on new and conflicting meanings when they are inserted into new contexts-- surely applies as much to deconstruction itself as to any of its objects. Thus, it is not surprising that deconstructive arguments can be invoked by the political right as well as the political left, and that they can serve many different and conflicting positions.

Is there any relationship between deconstruction and justice, then? I believe that the strategy of trying to show that deconstruction is a necessarily good thing is the wrong way of approaching this question. Rather, we should ask ourselves what kind of deconstruction would assist us in the discussion and analysis of what is just and unjust. The question is not whether deconstruction is justice but what kind of deconstruction could assist us in the pursuit of the just. Such a deconstruction always has a critical or ameliorative purpose; we use it to figure out what would be more and less just, even though people may disagree about these questions, and even though the deconstructor may ultimately be mistaken about them.(29)

It is important to note that not all deconstruction has this critical or ameliorative purpose. After all, we might simply deconstruct texts in order to discover their multiple meanings and internal tensions. In order to distinguish these other possible uses of deconstruction I call give this normative and critical form of deconstruction a special name. For reasons to be described presently, I call it transcendental deconstruction.

If we deconstruct law for a critical purpose it must be because we believe that there is some gap or divergence between the law and what justice requires. In other words, the critical use of deconstruction presupposes a conceptual opposition between law and justice. However, deconstructive theory also tells us that every conceptual opposition can be reinterpreted as a nested opposition. So there is a complex relationship of mutual dependence and differentiation between these two concepts. What is this relationship?

Laws apportion responsibility, create rights and duties, and provide rules for conduct and social ordering. But law can never achieve perfect justice. Law is always to some extent unjust. At the same time, our notions of justice can only be articulated and enforced through human laws and conventions. Although we may have an ideal of a justice that always escapes human law and human convention, the only tools we have to express and enforce our ideal are laws and conventions. Our conception of justice relies for its articulation and enforcement on the imperfect laws, conventions and cultural norms from which it must always be distinguished.

This, then, is the nested opposition of law and justice: Human law, culture, and convention are never perfectly just, but justice needs human law, culture, and convention to be articulated and enforced. A fundamental inadequation always exists between our sense of justice and the products of culture, but we can only express this inadequation through the cultural means at our disposal.

Yet if human legal creations are always to some degree unjust, justice cannot be fully and finally determined by any positive norms of human law, culture or convention; for these positive norms must also fall short of our value of justice. Thus, we must think of justice as an insatiable demand that can never be fulfilled by human law. We must postulate a value of justice that transcends each and every example of justice in human law, culture and convention. In this way deconstructive argument brings us to recognize a transcendent value of justice. Hence, the critical use of deconstruction becomes "transcendental" deconstruction, because it must presuppose the existence of transcendent human values articulated in culture but never adequately captured by culture.(30)

The debates about deconstruction's relationship to justice show how far deconstruction has mutated since its introduction into the legal academy. This is indeed a case where "iterability alters." While legal deconstruction remains similar to its literary and philosophical forbearers, it has adapted itself to the peculiar concerns of legal academics. On the one hand, legal deconstruction has emphasized the rhetorical nature of deconstruction, the selectivity of deconstructive arguments and their implicit appeal to transcendent values. On the other hand, each of these features was already present in literary and philosophical deconstruction, although perhaps not noticed or emphasized. Legal deconstruction has, in this sense, deconstructed the practice of deconstruction itself, revealing more clearly its conflicting commitments and assumptions, both that which it privileges and that which it suppresses. Yet before one rushes to label the development of legal deconstruction a betrayal of the deconstructive tradition, one might well recall the admonition-- made by deconstruction itself-- that the relationship between tradition and betrayal is always more complicated than it first appears.

Footnotes

16. For two different versions of this thesis, see Phillip Bobbitt, Constitutional Interpretation (Oxford: Basil Blackwell 1993) and William N. Eskridge, Dynamic Statutory Interpretation (Cambridge: Harvard University Press 1993).

17. U.S. Const., Art. I, sec. 3; Art. V.

18. See Bruce Ackerman, We The People: Volume Two: Transformations (forthcoming Harvard University Press 1997).

19. Frug, Ideology of American Bureaucracy, supra; Dalton, Deconstruction of Contract Law, supra.

20. Peller, The Metaphysics of American Law, supra.

21. See J.M. Balkin, Ideology as Constraint, 43 Stan. L. Rev. 1133 (1990).

22. Stanley Fish, Doing What Comes Naturally: Change, Rhetoric, and the Practice of Theory in Literary and Legal Studies (Durham, N.C.: Duke Univ. Press l989).

23. Costas Douzinas, Ronnie Warrington, and Shaun McVeigh, Postmodern Jurisprudence: The Law of Text in the Texts of Law (London: Routledge 1991); Peter Goodrich, Languages of Law: From Logics of Memory to Nomadic Masks (London: Weidenfeld and Nicholson 1990); Peter Goodrich, Legal Discourse: Studies in Linguistics, Rhetoric and Legal Analysis (London: Macmillan Press 1987).

24. See Balkin, Ideology as Constraint, supra.

25. See, e.g., Jacques Derrida, Like the Sound of the Sea Deep Within a Shell: Paul de Man's War, 14 Critical Inquiry 590 (1988) (Peggy Kamuf trans.)

26. Jacques Derrida, Force of Law: `The Mystical Foundation of Authority', 11 Cardozo L. Rev. 919 (1990).

27. Drucilla Cornell, The Philosophy of the Limit (New York: Routledge 1992).

28. See, e.g., Jacques Derrida, Biodegradables: Seven Diary Fragments, 15 Critical Inquiry 812, 827 (1989)(rejecting the claim of a critic that deconstruction can be used for either "fascist" or "liberal" purposes)(Peggy Kamuf trans.).

29. J.M. Balkin, Being Just With Deconstruction, 3 Social and Legal Studies 393 (1994).

30. J.M. Balkin, Transcendental Deconstruction, Transcendent Justice, 94 Mich. L. Rev. 1133 (1994). The argument for a transcendent value of justice is "transcendental" because it claims this value is a necessary presupposition to deconstructive arguments about justice.