Transcendental Deconstruction, Transcendent Justice-- Part II

Originally Published in 92 Mich. L. Rev. 1131 (1994).
Copyright 1999 Jack M. Balkin. All Rights Reserved.

Go to Part I
Go to Part II
Go to Part III
Return to Writings Online



III. Speaking in the Language of the Other

Derrida's third formulation of the relation between deconstruction and justice notes the etymological connections between justice and answering. To be just is to have responsibility, which is to respond to or to answer for something. Thus, Derrida speaks of justice as an "infinite demand." (67) However, not any answer will do. Justice, Derrida insists, requires one to address oneself to the Other in the language of the Other: It requires us to forswear our own way of thinking, talking, and looking at things in order to understand the Other in all of her singularity and uniqueness. (68) This requirement is ultimately impossible to attain, and hence the infinite demand of justice can never fully be satisfied.

Because justice demands that we address ourselves in the language of the Other, the law can never be fully just. The problem for law, Derrida argues, is threefold. First, law must speak in general terms, and therefore it must simplify and falsify the situation at hand. (69) Legal understanding never allows us to understand situations or the persons affected in all of their uniqueness. We must understand them instead filtered through a set of legal categories, or classes of situations, that lump them together with many other equally heterogenous and unique circumstances. The enforcement of the law according to these categories is a form of simplification and falsification, and this simplification and falsification are sources of injustice. (70)

Second, as James Boyd White has recently noted, the problem of justice is inherently a problem of translation. (71) For judges or other parties to speak in the language of another, they must translate the Other's language into their own. But translations are always imperfect. They never fully convey the sense of the original. Hence the very necessity of translation renders it impossible fully to speak in the language of the Other. (72)

Third, the requirement that law be impartial demands that we not speak in the language of a particular party, but in a language that is neutral and fair. (73) To speak in the language of only one of the parties risks the danger of undue partiality toward that person, for the situation will be completely described in terms of her experience and her concerns. This result is unfair because it may give short shrift to the experience and concerns of other parties. Hence, law, which requires fairness to all parties, must proceed in the language of neither one party nor the other, but in a third language that attempts - even if it does not always succeed - to be fair to both sides. Legal justice strives for an impartiality that is also impersonal. Yet this solution creates its own set of problems, for the neutral language of a third party fails to speak in the language of either party, and hence it doubly falsifies the situation by denying or obscuring the uniqueness and singularity of each side. (74)

Derrida's ethics of Otherness contains two separate imperatives. The first demands that we see a situation in all of its singularity. The second demands that we attempt to see things from the Other's point of view, using her vocabulary and her way of understanding the world. To deal justly with each of these two points, we must not conflate them, but rather deal with each separately - that is, respecting the singularity and difference of each.


A. Justice as the Recognition of Singularity

Derrida's demand that we see each situation in all its singularity is ambiguous. We could interpret it either as a claim of absolute difference among situations or only as one of relative difference. A claim of absolute difference means that we must see each situation as completely different from every other. A claim of relative difference means that we must see each situation as different from any other in some respects but not in others. Each situation is both different from and similar to every other situation; its uniqueness consists in the fact that this combination of similarity and difference manifests itself in different ways for each situation to which it is compared. Thus, A and B are both similar to and different from C; but A and B are unique because they are similar to and different from C in different ways.

We may state this distinction in another way. Consider three situations A, B, and C. They are all different. But are two of them more alike than the third? There are two positions we can take. One argues that all of them are absolutely different; consequently, no situation is any more like another than any other situation. The alternative position would insist that we cannot answer this question until we know what context the questioner has in mind. Given a particular context of decision, it will often be possible to say that two situations are more like each other than either is to a third; but this judgment may shift radically if the context of judgment is sufficiently altered. If we are concerned only with the question of weight, an elephant and a truck are more alike than either is to an amoeba. Yet if the context of judgment is shifted to the question of animate versus inanimate, the elephant and the amoeba are more alike than either is to the truck. This alternative position asserts the relative similarity and difference of all situations.

Does Derrida mean to suggest a theory of absolute difference or a theory of relative difference? If justice is an "infinite demand," perhaps we must keep trying to view a situation as different from any other in every respect. That would presume a theory of infinite difference. Yet, if Derrida means that justice requires us to assert the absolute difference of every situation, his claim is incoherent. It will be impossible to decide any case, because no case can be compared to any other. Because each case is completely different from all others, no case is a better point of comparison than any other. We cannot apply any consistent principle to different cases; hence, our judgment is merely one of fiat, for no decision is any more principled or unprincipled than any other. Conversely, we might also say that all decisions are equally principled. Because there are no relative degrees of comparison, any judgment is as good an exemplar of our principles as any other.

If Derrida's claim is based on a notion of relative difference, however, it accurately describes the predicament of just decisionmaking. (75) Each case is both similar to and different from every other, depending on how we look at it. The difficulty of just decisionmaking lies precisely in deciding what is the appropriate context of judgment. The question of principled consistency is the question of which cases our case is most like and which cases it is least like, given the appropriate context of judgment.

Consider the recent example of a seventeen-year-old high school student who was sentenced by a judge in Thomaston, Georgia to three years in prison for stealing an ice cream bar from the school cafeteria. (76) The judge defended his decision on the grounds that the case was a burglary, and the penalty for burglary was three years. He argued that the appropriate context of judgment involved the definition of burglary, the legislature's decision to fix the age of majority at seventeen, and his county's practice of uniform punishments for all violators of the same crime. His judgment was criticized on two grounds, each of which offered a competing context of judgment. First, what the student did was more like a schoolboy prank than a professional breaking and entering. Second, on the same day the student was sentenced to three years, the judge gave suspended sentences and fines to several people convicted of drug possession and drunk driving. (77)

We can only criticize the judge's decision if we assume the relative difference of situations - that is, only if we argue that this situation is both different from and similar to others. In order to differentiate this case from an "ordinary" case of burglary, we must be able to say that this defendant was a student like other students, that his action was a prank like other schoolyard pranks. In order to argue that it is unfair that drug users and drunk drivers should receive a lesser penalty, we must be able to assess comparative degrees of responsibility and harm between situations. Yet this means that we must already be able to see these situations as similar in some respects; this similarity is necessary for them to be comparable or commensurable according to some common metric. We cannot compare these situations if we assume that each is unique in the sense of absolutely different. We can only make such a judgment if we see each situation as relatively different. Thus, justice may require that we understand each situation in its uniqueness, but, ironically, this requires that we treat it like the situations that are most similar to it in the appropriate context of judgment. To recognize its uniqueness, we must also recognize its similarity to other situations.

The same criterion of relative difference applies when we seek a just understanding of persons who are different from us. It is important to try to understand and respect people who are different from us. To understand and respect their difference, however, we must first understand their similarity to us. We must try to see how their concerns and values are really similar to our concerns and our values, and thus, how the situation they find themselves in and their reactions to that situation make sense. At the same time, to grasp this similarity, to put ourselves in other people's shoes, we must recognize how our lives and theirs are different. That is why every attempt at understanding is a simultaneous assertion of commonality with and difference from the Other. If we unthinkingly assume that the Other is too much like us, we will never understand her actions when they diverge from our own; if we insist on our absolute difference from her, she will never be able to understand us.

The competing interpretations of absolute and relative difference offer two different accounts of the predicament of judging. The theory of absolute difference suggests that just judging is impossible because no situation is really like any other. All principled decisionmaking is completely indeterminate because we have no way of comparing situations when each is absolutely different. On the other hand, the theory of relative difference argues that doing justice is difficult because there are so many ways to see situations as similar as well as different. The problem is not that no two situations are ever similar; it is that there are too many ways in which situations are similar to each other, and we must try to parse out the right ways to assess this similarity. In other words, the secret of judging lies in determining the appropriate context of judgment. However, we can never fully determine the present context, and we can never fully know of the presence or absence of other events that might significantly alter the context of our judgment when we decide a case. Therefore we are always uncertain - at least to some degree - about the justice of our decision.

Note that the dependence of justice on context is much like the dependence of meaning on context. The indeterminacy of meaning and the uncertainty of judgment are both based on the indefiniteness of context. This view is consistent with the transcendental approach to deconstruction. In contrast, an approach that asserts the infinite difference of each situation is just the flip side of an approach that asserts that meaning is infinite. The former asserts the absolute difference of all situations and all people, while the latter asserts the absolute identity of all meanings. Both approaches lead to normative nihilism and a failure of understanding. As before, Derrida's arguments only make sense if his is a transcendental account of deconstruction.


B. Justice as Understanding the Other's Point of View

Derrida's ethics of Otherness has a second component: It employs a different sense of individuality and uniqueness. Under this view, justice requires one to speak in the language of the Other by trying to see things from the Other's point of view. (78) This conception of justice seems most attractive when we are the injurer or the stronger party in a relationship, or when we are in the position of a judge who is attempting to arbitrate between competing claims. For example, suppose that we are the State, the stronger party, the oppressor, or the injurer, or suppose that we are contemplating an action that might put us in such a position. It seems only just that we should try to understand how we have injured or oppressed the Other (or might be in a position to injure or oppress). We can only do this if we try to see the problem from the Other's perspective and understand her pain and her predicament in all of its uniqueness. The duty we owe to the Other is the duty to see how our actions may affect or have affected the Other; to fulfill this duty we must put away our own preconceptions and vocabulary and try to see things from her point of view. Similarly, if we are a judge in a case attempting to arbitrate between the parties, the ethics of Otherness demands that we try to understand how our decision will affect the two parties, and this will require us to see the matter from their perspective. Suppose, however, that we are not the injurer, but the victim; not the State, but the individual; not the strong, but the weak; not the oppressor, but the oppressed. Does justice require that we speak in the language of the person we believe is injuring or oppressing us? Must a rape victim attempt to understand her violation from the rapist's point of view? Does justice demand that she attempt to speak to the rapist in his own language - one which has treated her as less than human? Must a concentration camp survivor address her former captor in the language of his worldview of Aryan supremacy? We might wonder whether this is what justice really requires, especially if the injustice we complain of is precisely that the Other failed to recognize us as a person, refused to speak in our language, and declined to consider our uniqueness and authenticity.

When we move from Derrida's grand pronouncements on the ethics of the Other to the place "where he lives," to his writings on Paul de Man, we see this pleasant formula properly deconstructed. Derrida both adopts and rejects the formula in different contexts. For example, Derrida applies the formula when he demands that his critics read both him and de Man fairly. His first rule of just interpretation is "respect for the other, that is, for his right to difference, in his relation to others but also in his relation to himself." (79) To judge de Man fairly, one must speak to de Man in his own language and read him in the proper context.

Thus, Derrida insists that we evaluate de Man through the ethics of the Other. But what is the proper attitude to take toward de Man's critics? Derrida argues that he faces a serious problem: These critics cannot or will not read what he and de Man say; he describes them through the neologism ne pveulent pas lire - they can't/won't read. (80) These critics have failed the call of justice; they have not been open to the Other - de Man and Derrida. (81)

What does justice demand of Derrida in defense of de Man? Here Derrida's practice deconstructs his theory: The essay Biodegradables cannot be described as anything other than a polemic. Derrida is perfectly aware of this but considers it unavoidable:

Justice, it seems, does not always demand that one speak in the language of the Other, especially when the Other is not playing by the same rules. (83) Jesus might have advised his followers to turn the other cheek and to love their enemies, but this approach is not necessarily what justice requires or a particularly good strategy for achieving a just result. Not all encounters with an Other will involve willing participants in an open dialogue. Nevertheless, justice demands that the oppressor answer for her wrong, whether or not she admits her fault. Even if she makes no answer to her victim, she must answer for her crime. Justice demands satisfaction, even (and especially) if the miscreant is unwilling to provide satisfaction. The answer that an oppressive Other must provide to an oppressed Other - for example, a prison sentence or a money judgment - is not necessarily addressed to the victim in her language. It is not necessarily even understood by the injurer as an answer to the victim, or as an attempt to understand the victim in all of her singularity. Sometimes justice makes its demand precisely when people will not understand each other, when they will not treat others as equals.

Thus, perhaps unsurprisingly, justice demands that we deal with people and things not in a self-abnegating manner, but in the manner appropriate to the situation before us. Centuries ago, Confucius argued that the virtuous man is one who knows both how to love people to the extent they are deserving of love, and to hate people to the extent that they are deserving of hate. (84) Justice demands that we speak in the language of the Other to the extent that it is appropriate to do so because this would further justice, but it equally demands that we not do so when it would increase injustice.

We may connect this point to our earlier criticism of the notion of a "responsibility without limits." Derrida has argued that the ethics of Otherness imposes upon us a responsibility to speak in the language of the Other. However, because justice is a responsibility without limits, we might ask as before whether this responsibility to the Other is an infinite responsibility or merely an indefinite one.

Thus, there are two different interpretations of the ethics of Otherness. The first imposes an infinite duty; the second imposes only an indefinite duty. The first corresponds to a nihilistic conception of deconstruction; the second to the transcendental conception. The requirement of an infinite duty means that we must in every case use all the available means at our disposal to speak in the language of the Other; the requirement of an indefinite duty means we must make some attempt to speak in the language of the Other, and that the boundaries of our duty are uncertain and contextually driven. In the first case, the demand of justice is never satisfied because this demand is infinite; in the second case, we can never be certain that the demand of justice is satisfied because the duty it imposes is indefinite.

All of the difficulties with the ethics of Otherness arise from the assumption that our responsibility to speak in the language of the Other is infinite. We can restate the difficulty by relating it to a similar problem in understanding the views of another. This is the problem of hermeneutic charity. When we try to understand what another person means, we usually do so by trying to envision how what they are saying makes sense. As Hans-Georg Gadamer has argued, we must make an "anticipation of completion" that what another is saying is coherent and has a claim to truth. (85)

A stance of openness and interpretive charity is actually essential to the process of understanding. If we do not take this stance, we cannot be sure that our discovery of incoherence or falsity in another's position is due to a defect in their argument or our inability to understand it fully. (86) In other words, when attempting to understand another person, especially a person with a different world view from our own, we must be open to the possibility that the truth is more on their side than ours, that what they are saying is really true and valid - indeed, more true and valid than our own beliefs. Thus, we must be open to the possibility that our encounter with the Other will change our own views about what is true and good. (87) Understanding, then, is a kind of vulnerability or openness to the truth that the Other may have to express. (88) It always requires the possibility that our beliefs will be changed through our encounter with the Other. Indeed, Gadamer argues, if we do not come to our encounter with this hermeneutic openness, we can never achieve real understanding; at most we achieve a halfway measure, in which we withhold ourselves from true understanding because we withhold ourselves from the necessary hermeneutic vulnerability. (89)

The duty of hermeneutic charity in Gadamer's theory of interpretation strongly resembles Derrida's version of the ethics of Otherness. Given the common influence of Heidegger on both thinkers, this commonality should not be surprising. Yet Gadamer's duty of interpretive charity and openness to the object of interpretation raises a serious difficulty. If this duty is endless, it is hard to distinguish this duty from a duty to reach an agreement with the person we are trying to understand. We do not know, in other words, at what point we should cease our efforts to see the truth in the views of the other party and simply recognize that they are wrong or that their argument is incoherent. If we go too far, we risk the danger of what I call hermeneutic cooptation. (90) By repeatedly blaming the incoherence or wrongness of the argument of another solely on our insufficient failure to understand it, we place all the responsibility for intellectual change upon ourselves. There is the danger that our drive to understand the truth in the other person's views will lead us to be coopted by those views and brought into agreement with things we should not agree with, because they are false, misleading, or unjust. (91)

There is a further difficulty. If the Other's views treat us as objects or as persons who deserve no respect, the requirement of continual hermeneutic charity will require us to adopt ways of thinking and talking that are unjust to ourselves. That is why we hesitate to think that justice requires that a political prisoner strive to speak in the language of her former torturers, or that a Holocaust survivor attempt to understand her own situation in the language of Aryan supremacy. To require this sort of understanding is to require these people to injure themselves psychically through the duty of understanding. For want of a better name, let us call this the requirement of hermeneutic masochism. True justice toward another should never involve hermeneutic cooptation, and it should never require hermeneutic masochism.

An infinite responsibility to speak in the language of the Other creates the perpetual danger of hermeneutic cooptation and hermeneutic masochism. It leads to the ridiculous spectacle of the rape victim being asked to understand that, in the eyes of her attacker, she was really just an object for subordination and conquest. It leads to the conclusion that blacks should be more understanding when white police officers automatically assume that they are likely to be criminals. (92) An infinite responsibility to speak in the language of the Other can easily lead to perpetual justification of the Other, no matter how unreasonable their position. This is not what justice requires.

The postulation of an infinite duty is untenable. Yet we might still make sense of the ethics of Otherness by viewing the duty to understand as indefinite rather than infinite. We have some duty to speak in the language of the Other, but our duty is not infinite. Instead, justice demands that we make just the right amount of effort to understand the Other. Beyond that point, it is not only appropriate but necessary for us to recognize that the Other's views are incoherent or unjustified, and that our own position is more reasonable. We have a duty to be open to and absorb that part of the Other's point of view which furthers justice while disagreeing with the rest.

But if we have this responsibility, how will we know when to cease our efforts at understanding? How will we know when we have done all that justice requires? We cannot know the full contours of our responsibility in advance of our encounter with the Other. Each situation will be different, and our responsibility in each situation will depend heavily on the context of the encounter. Hence our responsibility to the Other, while not infinite, is nevertheless indefinite.

There is a further reason why our duty to the Other must be indefinite. It has to do with the symmetrical nature of Otherness. (93) We are always an Other to the person who is an Other to us. The ethics of Otherness seems most appealing when we sympathize with the Other because the Other is the oppressed, the victim, or a potential object of injustice. However, we cannot state as a general rule that only the oppressor needs to speak in the language of the oppressed, or the injurer in the language of the victim. An additional problem of indefiniteness arises because these categories are not always clearly defined. In different contexts, and from different perspectives, different people appear to be strong or weak, injurer or victim, oppressor or oppressed, judge or judged.

First, the roles of the two parties may shift radically depending on how one describes the situation. Consider two neighbors who make conflicting uses of their property. Suppose that Neighbor One needs to operate machinery to run her business while Neighbor Two needs peace and quiet for her health. The second neighbor regards the first neighbor as an injurer; Neighbor One should try to understand how important it is for Neighbor Two to have peace and quiet. But Neighbor One has a symmetrical complaint. If she has to stop operating her machine, she will go out of business. Neighbor Two's demand is unreasonable under the circumstances, and justice requires that Neighbor Two understand the significant costs her request will impose on Neighbor One. (94) In this situation each party is an Other to the other, and each is an Other to the judge who must decide the case. Thus there are at least four different potential duties to respond. Both neighbors must make some effort to understand the position of the Other, but the roles of oppressor and oppressed, victim and injurer, are not determined in advance. Similarly, the judge must attempt to understand the situation of each neighbor from each neighbor's perspective, but the extent of this duty depends on which neighbor is acting unjustly toward the other, and this is the very question that the judge must decide.

Thus, the scope of the duty owed to speak in the language of the Other depends on our definition of the roles of the parties - as victim or injurer, strong or weak - but this definition will in turn be affected by the scope of the duty to speak in the language of the Other. For example, the more we try to see things from Neighbor One's perspective, the clearer it may become that Neighbor Two is being unreasonable, oppressive, and playing the role of the potential injurer. Yet the opposite conclusion might follow if we attempt to see the situation from Neighbor Two's perspective. The scope of our duty to speak in the language of the Other does not exist before we decide what their respective roles are, but the roles each plays cannot fully be determined before we fix the scope of the duty; each feature of the situation provides the proper context in which the other feature is to be judged. Because of the mutual dependence and differentiation of these contexts, the scope of the duty toward the Other is indefinite. It is neither infinite nor nonexistent, but dependent on facts and circumstances that are never fully clear, and whose precise contours we cannot fully determine in advance. Thus, the duty to speak in the language of the Other becomes a duty without limits, but it is by no means an infinite duty as Derrida suggests.

The difficulty that produces indefiniteness may be subdivided into two problems: Let us call them the problem of certainty and the problem of proportion or degree. The problem of certainty arises when we cannot be sure whether a person is in fact an oppressor or an oppressed person. Justice may require us to decide this question, rather than use it as a basis for determining just action. Suppose a criminal defendant is accused of rape. If he really raped the victim, he is an oppressor. Justice demands that he recognize how he has harmed his victim and answer for his crime. If he is innocent, however, he will be oppressed if the State fails to do justice to him. The State must therefore ensure that it hears his story and understands the situation from his point of view; it must make sure that it does not convict him based on false evidence or unreasoning stereotypes. At the same time, it must not unthinkingly accept every piece of exculpatory evidence and every exculpatory account the defendant might offer, for that might create an equal and opposite injustice. In a criminal case we cannot determine in advance whether a person is guilty or innocent; that is precisely the purpose of a criminal trial. Given this situation, does justice demand that we speak to the defendant in his language or that he speak to us in our own? Our uncertainty in this regard leads to the indefiniteness of the duty.

The problem of degree arises because a person may at one and the same time be an injurer or oppressor to some degree and a victim or an oppressed person to another degree. Suppose that we are quite sure that a criminal defendant is guilty of rape. He nevertheless deserves to be treated with some respect; for example, the State should not be permitted to torture him to extract a confession, and it must give him an opportunity to defend himself in court. His crime may be less bad than other crimes of the same sort; if so, justice requires that he be given a chance to produce exculpatory evidence, and the State has a duty to consider it and lessen his sentence to the appropriate extent. If the State fails to protect the defendant's rights, he may be in the position of an oppressed party, despite his horrible crime. Thus, the State has a duty to speak in his language with respect to some features of the situation, but not with respect to others. These complications also produce an indefinite duty.

The case of Paul de Man is a perfect example of the indefiniteness of our duty toward the Other. The entire debate surrounding de Man concerns what role he should be assigned. Is de Man a victim of unjust accusation or a person who unjustly accused others (Jews)? Was he a collaborator, an ambitious man without a moral compass who sought to forget his sordid past, or was he an immature youth who made an early mistake yet matured into a respected scholar? The question of his status is inextricable from the question of what it means to be fair to him. Moreover, even if de Man willingly wrote antisemitic literature, justice requires that he be condemned only to the extent that he is responsible, but not to the extent that he is not responsible. In order to be fair to de Man, we must try to understand what he did and wrote, and this endeavor may require us to understand how he understood his actions and his writings. Nevertheless, we do not have to accept everything he said or did at face value or interpret every one of his writings or actions so as to exculpate him, just as we should not deliberately interpret his life in its worst possible light. Our duty of fairness is indefinite, but not infinite in either direction - toward exculpation or inculpation. We can be unjust either by refusing to speak in the language of the Other or by exclusively adopting that tongue.

Note that we cannot avoid these difficulties by insisting that all parties, whether victim or injurer, oppressor or oppressed, have an equal and infinite duty to speak in the language of the Other. An equal and infinite duty on all sides leads to either incoherence or injustice. It means that the rapist and the rape victim have equal duties to understand each other in terms of each other's conception of each other, and this requirement permits neither a determinable decision nor a just one. Justice demands that each "speak in the language of the Other" to the proper amount, to the proper degree, and in the proper circumstances. This duty is without limits not because it is infinite but because it is indefinite - because the question of duty to the Other is bound up with the very description of the situation which the duty concerns.

An indefinite duty, like an indefinite responsibility, is "without limits" because its contours are context bound and because this context cannot be fully determined in advance. What would an indefinite duty toward the Other mean? It would be a duty that can never fully be satisfied. However, it cannot be satisfied because we are uncertain about its scope, not because the demand is infinite. (95) Previously we identified the concept of an infinite responsibility with normative nihilism. The notion of an infinite duty to speak in the language of the Other - which is a total abandonment of ourselves to the language of the Other - also leads to a kind of nihilism. Hoping to efface the distinction between self and Other, we succeed only in effacing the self and its language, just as the effacement of all distinctions leads to the destruction of meaning. This endeavor leads to a "nihilism of the self." The preservation of the self is the preservation of its Otherness from the Other, which is also the preservation of its partial similarity to the Other, and a source of the indefiniteness of its duty toward the Other. When we understand deconstruction to require an indefinite obligation, we preserve the self and make sense of the demand of justice; but when we understand deconstruction to require an infinite demand, it must lead to effacement or nihilism.


IV. Deconstruction as an "Antitotalitarian"

Form of Analysis


Derrida's final suggestion connects deconstruction to an opposition to totalitarianism in thought. Derrida insists that what he has always practiced under the name of deconstruction "has always seemed to me favorable, indeed destined (it is no doubt my principal motivation) to the analysis of ... totalitarianism in all its forms, which cannot always be reduced to names of regimes." (96) He finds examples of totalitarian thinking in criticisms of Paul de Man; since these critics are so concerned with denouncing political totalitarianism, he argues, they should avoid reproducing the logic of totalitarianism in their judgments and readings. (97) In fact, deconstructive analysis, which they attack, is the best way to avoid totalitarian logic: "Deconstructions have always represented ... the at least necessary condition for identifying and combating the totalitarian risk" in discourse. (98)


What is a "totalitarian" logic of discourse or a "totalitarian" gesture in discourse? Derrida seems to identify it with various forms of oversimplification, falsification, or rushing to judgment. His examples describe features that he finds objectionable in various critical readings of de Man's work:


All of these sins of reading and understanding share an inattention - whether willful or innocent - to problems of context. They oversimplify by failing to spend the time or effort to see the multifaceted and complicated textures of meaning that attend any text or any event. (100) The totalitarian gesture, then, is oversimplification and inattention to complexity and context; the antitotalitarian gesture, which is just, is a corresponding attention to these features of texts.

Of course, a judgment whether one is being insufficiently sensitive to context is itself a contextual judgment. For example, in some contexts, dismissing another's arguments, making categorical distinctions, and even rushing to judgment may not be an entirely bad practice. (101) Derrida denies that "we have access to a complete formalization of this totalitarian logic" or can be absolutely exterior to it. (102)

Indeed, there is no "systematic coherence proper to each of them, since one must not confuse them too quickly with each other." (104) Conversely, there is no discursive act which is completely antitotalitarian, for there is no "property so closed and so pure that one may not find any element of these systems in discourses that are commonly opposed to them." (105)

Derrida's identification of deconstructive argument with antitotalitarianism is really a special case of his critique of logocentrism. The problem of logocentrism is the problem of categorical judgment. Categorical judgments are judgments that employ categories; whenever we predicate a property of an object, we place it in a category. We say that it is this way rather than that, that it goes into this box rather than another. In fact, categorical judgments are necessary to our thought; they are the basis of all judgments of similarity and difference. Of course the term categorical has another meaning as well - insensitivity to context. Categorical obligations are unconditional; to state something categorically is to assert it without regard to (at least some types of) context. Thus, categorical judgments, because they are categorical, are to some degree acontextual. They must lump some things together as similar and exclude others as different, without attending to the similarities across, or the differences within, the boundaries that they establish. Hence every categorical judgment is a sort of falsification or oversimplification of the situation. Of course that is precisely why categories are useful. In a world of infinite diversity, change, and differentiation, categories gather things together and treat them as similar so that the human mind can understand the world before it. One might think of categories, then, as heuristics that aid understanding through partial simplification. (106)


What is most remarkable about categorical judgment, then, is that it is simultaneously useful, adequate, and empowering in some contexts and deficient, inadequate, and misleading in others. We cannot do without categorical judgments of some type, yet if we do not pay sufficient attention to the context in which we make them, they may lead us away from what is true and what is just. Deconstruction helps us to recognize the discrepancy between the categorical judgments we make and the context our judgment overlooks. A perfectly just treatment of a situation would require us to understand the situation in all of its contextual richness. (107) Nevertheless, we must make categorical judgments of some type to articulate the very context we seek to uncover. Context itself must be describable in terms that are unavoidably categorical. Hence the process of deconstructive analysis, while "urgent," is also "interminable." (108) Deconstruction becomes "the tireless analysis (both theoretical and practical) of ... adherences" (109) to the totalitarian discourses that remain lodged even in one's own ways of thinking and which one is trying to combat.

Indeed, in articulating his point, Derrida falls prey to the very danger he warns against. Derrida labels all the various forms of oversimplification and acontextual judgment as "totalitarian." (110) But this comparison is itself a gross overstatement and highly misleading. To be sure, totalitarian regimes may make use of simplistic slogans and lump various persons together into categories of undesirables, but it hardly follows that every oversimplification deserves the name "totalitarian," especially given the powerful connotations that usually accompany this word. In Derrida's terms this comparison is in itself an oversimplification, a "totalitarian" move that provides Derrida's attack on de Man's critics with much more rhetorical force than it really deserves. It would be more appropriate - and more just - to argue that these critics do not read de Man in the proper context and with the proper degree of charity; but this accusation, even if true, in no way justifies the claim that, in misreading de Man, they are reproducing the logic of totalitarian discourse.

In fact, when we strip away its more obfuscating elements, Derrida's identification of deconstruction with antitotalitarianism is really better expressed in terms of deconstruction's continual allegiance to transcendent human values, which law, language, and convention never fully serve and always partially obscure. We deconstruct categorical judgments because they take us further away from truth; we deconstruct legal categories because they deviate from what is just. Nevertheless, a deconstruction in the service of justice is always premised on the possibility of reconstruction - that is, on the hope of some categorical scheme that would better articulate the appropriate context of judgment. If we do not believe that there is a better description, there is no point in deconstructing in the first place. We are simply substituting one description for another, without any assertion that one is better than another.

Does this practice in fact presuppose a transcendent norm of justice? We might deconstruct only in order to show that a categorical judgment fails to live up to the norms of our particular culture or legal system. In that case the ultimate ground of our deconstruction is to obtain increasing fidelity to positive norms. However, such a practice places the positive norms beyond criticism because they are the basis of deconstructive critique. Yet one can easily imagine cases in which this very refusal to critique positive norms would itself be totalitarian.

However, to say that positive norms are inadequate - and hence in order to deconstruct them - we must refer to values that lie beyond the norms we are critiquing and that serve as the source of our criticism, even if we believe that the values we wish to uphold are to some extent realized in our culture. Suppose that we denied that we need concern ourselves with transcendent values: Suppose we assert that we are only interested in engaging in an "immanent" critique. In other words, we say that we are using one aspect of our cultural norms to critique other aspects, and therefore we need make no reference to anything beyond the positive norms of our culture. For example, we might use the commitment to equality expressed in the Civil Rights Act of 1964 (111) to criticize the lack of civil rights for homosexuals in the United States. The question remains, however, why we saw a particular aspect of our cultural practices as a worthy basis for our critique and another aspect as unworthy. Since both are equally aspects of our culture, culture by itself cannot serve as a norm to decide between them.

We might say that one is a more central feature of our cultural norms than the other, but this leads to two different interpretations. If something is central because it is more prevalent, we refer only to a positive norm. Yet mere prevalence does not guarantee the worth of a cultural practice, unless our only goal is to reinforce positive norms for their own sake. For example, racial inequality may be central to a regime of Jim Crow or apartheid, but this fact does not make it a worthy basis for a critique of egalitarian norms that might exist elsewhere in the culture. Indeed, by reinforcing the most prevalent practices of a culture, we may reinforce its most deeply unfair elements. On the other hand, by "central" we could mean "more valuable" or "more just." In that case our judgment must also refer to a transcendent conception of value or justice that informs our notion of centrality. Thus, whenever we speak of the proper continuation of positive norms or about deciding between alternative interpretations of positive norms, we must eventually make use of transcendent norms of justice. Although we may find these norms partially realized in portions of our own culture, these inadequate articulations do not exhaust their meaning for us.




67. Derrida, supra note 4, at 955. In discussing Derrida's arguments

concerning justice I shall follow his practice of speaking about the "demand" of justice, or about what justice "demands," to describe what is just or unjust. Nevertheless, I should note at the outset that this familiar locution has the twin rhetorical effects of anthropomorphizing justice and downplaying human subjectivity, as I describe more fully below. See infra text accompanying notes 136-37.


68. See Derrida, supra note 4, at 949.


69. See id.


70. See id.


71. See James B. White, Justice as Translation 257-69 (1990).


72. See Derrida, supra note 4, at 949. This idea is related to a theme that Derrida borrows from Heidegger and Levinas - the Other, because it is an Other, always remains ultimately unreachable and unfathomable.


73. See id.


74. See id.; White, supra note 71, at 262-63.


75. This interpretation seems most consistent with his criticisms of

Levinas. See, e.g., Drucilla Cornell, The Philosophy of the Limit 53-55, 68-72, 83-85 (1992); Jacques Derrida, Writing and Difference 126-28 (Alan Bass trans., Univ. of Chicago Press 1978) (1967); Guyora Binder, Representing Nazism:

Advocacy and Identity at the Trial of Klaus Barbie, 98 Yale L.J. 1321, 1376 (1989).


76. Christopher Sullivan, Small Town Ponders Prison for Snickers Theft, Austin Am. Statesman, Sept. 19, 1993, at A10.


77. Id.


78. See Derrida, supra note 4, at 949.


79. Derrida, Paul de Man's War, supra note 8, at 644.


80. Derrida, Biodegradables, supra note 8, at 823. Derrida combines the "p" in pouvoir (can) and the "v" in vouloir (want) to create the expression "je ne pveux pas lire (I can/will not read)." Id. at 828; see also id. at 827, 843.


81. Indeed, so egregious is their practice of reading that at one point Derrida wonders whether they can even be considered to be morally responsible for their actions. Id. at 823. Because the category of ne pveut pas lire

"displaces the category of responsibility," Derrida insists that he is not passing judgment on his critics: "Moreover, I bear these five no ill will; I have nothing against them; I would even like (if only in order to avoid this spectacle) to help them free themselves from this frightened, painful, and truly excessive hatred." Id. However, given the tone of the rest of the article, one might be pardoned if one were tempted to take these protestations of good will at less than face value.


82. Id. at 872.


83. Indeed, Derrida insists, one does not even have to respond at all to accusations that are fundamentally unfair or disrespectful:


Is it necessary to respond to every interpellation, to everyone no matter who, to every question, and especially to every public attack? The answer is "yes," it seems, when time and energy permit, to the extent to which the

response keeps open, in spite of everything, a space of discussion. Without such a space no democracy and no community deserving of the name would survive. But the answer is "no" if the said interpellations fail to respect certain

elementary rules, if they so lack decency or interest that the response risks shoring them up with a guarantee, confirming in some way a perversion of the said democratic discussion. Yet, in that case, it would be necessary that the nonresponse be appropriately interpreted as a sign of respect for certain principles and not as contempt for the questioner.

Id. at 837.


84. Confucius, The Analects 4:3 (D.C. Lau trans., Chinese Univ. Press 1983) (1979).


85. Hans-Georg Gadamer, Truth and Method 261-62 (Garrett Barden & John Cumming eds. & Sheed & Ward Ltd. trans., Seabury Press 1975) (1960).


86. See id. at 263.


87. Cf. id. at 262.


88. See id. at 262.


89. Id. at 270.


90. See Balkin, supra note 27, at 163.


91. Id. at 160-61, 167-69.


92. See Patricia J. Williams, The Alchemy of Race and Rights 46, 58-79 (1991).


93. See Cornell, supra note 75, at 54; Derrida, supra note 75, at 128.


94. This example draws on the familiar Coasean point that both parties are causal factors in their conflicting use. R.H. Coase, The Problem of Social Cost, 3 J.L. & Econ. 1, 2 (1960).


95. We should not confuse the question whether our responsibility to the Other is infinite or indefinite with still another reason why the demand of justice may remain unsatisfied in a particular case: the impossibility of a commensurable remedy. Often subsequent reparations for an injury never seem adequate to compensate for a previous loss. But this impossibility is not because either the loss or the responsibility for the loss is infinite, in the sense of having a limitless magnitude. Rather, the problem is one of

incommensurability between the loss and any remedy we could offer. This


incommensurability arises from the fact that our lives change in response to the events that happen to us. Our situation after an injury cannot be made fully commensurable with the situation before because both we and the world around us have changed as a result and can never be the same again. The past, precisely because it is past, can never fully be redeemed. The nature and direction of our lives have been irrevocably altered by previous actions and events; we become different people because of what has happened to us.


Thus, if the defendant accidentally breaks the plaintiff's leg, she may be deeply sorry for what she has done. Yet her action has affected the lives of others in a way that cannot fully be repaired, no matter what good deeds she later performs, and no matter how much assistance she offers to the victim and the victim's family. In such circumstances, the reason why the injurer's

responsibility can never be fully satisfied is not because it is infinite in magnitude. The problem is that any remedy we could offer will be of the wrong kind, because we cannot relive the past. Thus, the fact that we live our lives in unidirectional time by itself can make subsequent remedies for finite harms incommensurable and hence essentially and perpetually inadequate.


96. Derrida, Paul de Man's War, supra note 8, at 648.


97. Id. at 645.


98. Id. at 647.


99. Id. at 646.


100. Id. at 645.


101. For example, Derrida himself thinks it important to condemn Nazism and political totalitarianism in all of their forms. See id.


102. Id.


103. Id.


104. Id.


105. Id.


106. Although this sort of argument is generally associated with conceptual relativism, we should note that it is perfectly consistent with a realist ontology. One might believe that linguistic categories lump together objects that are really similar in some respects, although different in others, and that the grounds of this similarity and difference are not simply a matter of human convention.


107. Thus, Derrida insists, "one must analyze as far as possible this process of formalization and its program so as to uncover the statements, the philosophical, ideological, or political behaviors that derive from it and wherever they may be found." Derrida, Paul de Man's War, supra note 8, at 646.


108. Id.


109. Id. at 648.


110. See id. at 645 ("Such a formalizing, saturating totalization seems to me to be precisely the essential character of this logic whose project, at least, and whose ethico-political consequence can be terrifying.").


111. Pub. L. No. 88-352, 78 Stat. 241 (codified as amended principally at 42 U.S.C. 2000a to 2000h-6 (1988)).


Go to Part I
Go to Part II
Go to Part III
Return to Writings Online