2.   Key Normative Concepts

A major difficulty for societies emerging from a recent history of mass violence is how to contend with demands for justice, truth, and victim acknowledgment while recognizing the need for stability and peace. Although clearly important, a successful democratic transition requires more than strong leadership. Legacies of violence require that a society engage certain ethical questions in order to arrive at a just peace and meaningful reconciliation.

In the previous chapter I introduced the five normative concepts that underpin my theory of reconciliation as respect. In this chapter, I discuss the concepts at length. I begin by presenting the idea of mutual respect and then introduce the corollary concepts: truth telling, accountability, victim recognition, and rule of law. Although important, each on its own is insufficient for reconciliation. They help promote the development of the norm of mutual respect, which is the heart of reconciliation in badly fractured societies. As will become clear throughout the following discussion and in subsequent chapters, though the concepts themselves are at first framed in general terms, their actualization—that is, how they are manifested in any particular situation—varies significantly, and thus are not part of a universal template. The idea, in other words, is not to create a programmatic model of reconciliation that guarantees a particular outcome when certain steps are followed (which is not, in my opinion, possible). Rather, these normative concepts should be seen as crucial ethical challenges that continuously arise in transitional settings, and demand attention because they point to some of the very basic questions of how a society relates to itself, its past, and its future.

How should a society deal with a legacy of severe human rights abuses? What ethical and practical considerations should frame a response? Societies have turned to international and domestic criminal and civil trials to secure punishment (or amnesties to prevent further violence), truth commissions and other investigative institutions to investigate the past, public apologies and reparations for victims, and museums and monuments meant to preserve history. It is not uncommon to find a combination of measures, even though they may undermine one another. But without an understanding of what a society is trying to accomplish, and without some basic principles to guide its decisions, continued animosity and tensions are likely.

Mutual Respect

Reconciliation requires moving away from estrangement and distrust to a situation of respect and tolerance of others, including, crucially, former enemies. Reconciliation is the achievement of mutual respect across society. Such a statement immediately requires some elaboration.

To understand mutual respect, we need to define what it means to be a person, morally speaking. A healthy identity develops from intersubjective recognition among equals, which includes reciprocal recognition of claims to moral worth and dignity. Dignity is a fundamental property of what it means to be a person, as it points to the value of autonomy that is at the core of a healthy sense of self. Kant's second formulation of the categorical imperative, for example, argues that to treat another merely as a means, with no dignity, is a moral wrong (1998). While Kant did not espouse a theory of intersubjective morality, the notion that dignity is a fundamental value is at the basis of both deontological and intersubjective theories of morality. This is particularly important to victims and others who have suffered political abuse and stigmatization and remain mistreated and devalued. A society that seeks to be reconciled must create conditions for the recognition of all citizens as bearers of moral worth and dignity.

We recognize moral worth and dignity by showing respect for others. This is primarily a public relationship—or as Arendt (1989, 243) puts it, of “the domain of human affairs”—and differs from esteem, which refers to holding someone in special regard due to some particular trait of exceptional worthiness that we celebrate in her. Respect is not reserved for exceptional circumstances or for exceptional people;¹ rather, it follows from recognizing others as having inherent rather than instrumental moral worth by virtue of being persons. We show respect by behaving in ways that express that value. We do not, for instance, “respect” another when we measure one's value according to our own desires or goals, for such attitudes and behavior merely relegate the other's value to a measure of utility based on our own calculus. Respect requires that we recognize the other's claim on us to her moral worth and dignity, and we consequently have an obligation to treat her in a way that expresses this recognition. We are responsible for the ways in which we treat others.

Another way of expressing this is to say that as morally autonomous agents, we ought to treat others in such ways that recognize them also as morally autonomous. To the extent that we are autonomous, we can choose to act vis-à-vis others in certain ways that are constrained by the moral claims they make on us. Autonomy need not mean monadic individualism, that is, the notion that individuals are theoretically “prior” to socialization or broader cultural influences, but only that personhood includes the possibility of making moral decisions, of choosing. If there were no choice involved and our actions were biologically or socially over determined, it would make no sense to speak of moral autonomy or moral responsibility. Without autonomy, there would be no morality. The very idea of respect entails this conception of agency, since without it what appears to be recognition of the value of others is merely coincidental indifference bereft of any normative substance.

Respect is a reciprocal norm: It requires the mutual recognition of moral worth between subjects. Furthermore, it assumes that in engaging with others, we have an obligation to give them reasons for our actions and values that could affect them. We owe them, as moral beings whose dignity we recognize, an account of why we treat them the way we do. In this sense, reason is not private, but intersubjective. It is not morally sufficient for me to be satisfied with my own reasons for values or actions that may affect others; it is not sufficient, in other words, to assume that the values we hold are agent-neutral (in Christine Korsgaard's terms), and thus do not require that we justify them to others (1996). This would be a sign of disrespect, for it disregards the other's ability to judge my reasons and her right to respond to me. Rather, because I acknowledge others as moral, rational beings like me, it is incumbent upon me to give reasons they could recognize as valid. To respect another is to take seriously one's ability to comprehend and judge my reasons and respond to them, and also expect that person to do the same for me. She can not simply assume that the reasons for treating me in a particular way are normatively satisfactory or show respect for my dignity; she must offer reasons. This by no means requires that we ultimately reach a consensus on ends or policies, nor does it point to deep solidarity or social harmony, but it does mean that we are obligated to engage with one another as moral equals who deserve respect. Nor does it assume that we are strictly rational creatures bereft of emotion. We are also emotional beings who experience pain and joy and develop meaningful commitments and relationships with others, and part of giving reasons is a recognition of the importance of these broader, complex relations for a healthy identity. Giving reasons to others, and expecting the same from them, is an expression of this respect.

Understood in this way, respect refers to something greater than the liminal requirements demanded by political theorists like Chantal Mouffe, who sees political life as fundamentally antagonistic with little room for considering others as moral equals (1997). Respect implies the possibility of discussion and deliberation among former enemies, and reflects the ability to see another as a fellow human being. Nevertheless, it falls short of the deep fraternity that some thinkers place at the core of reconciliation. It certainly falls far short of what theologian and political theorist Miroslav Volf approvingly calls the “will to embrace”:

The will to give ourselves to others and welcome them, to readjust our identities to make space for them, is prior to any judgment about others, except that of identifying them in their humanity. The will to embrace precedes any “truth” about others and any reading of their action with respect to justice. This will is absolutely indiscriminate and strictly immutable; it transcends the moral mapping of the social world into “good” and “evil.” (2001, 42)

My understanding of respect does not transcend notions of good and evil, since it takes the problem of accountability quite seriously. Without just punishment and the rejection of impunity respect is nearly impossible, and we should not mistake it for absolving perpetrators of responsibility for their actions. Instead, respect underscores the importance of replacing social relations based on fear and domination with tolerance toward others in public and semipublic life. Tolerance is ancillary to respect; to the extent that we see others as deserving recognition as fellow citizens and moral equals, we tolerate their political claims, ways of life, beliefs, and so on, as long as these do not violate the fundamental dignity of others or require a return to violence.

The term tolerance is not without its ambiguities, and an enormous amount of literature has been generated to define its conceptual content and boundaries (Mendus 1998, 1999; Nehushtan 2007; Scanlon 2003). Michael Walzer has usefully identified a range of understandings, from “simply resigned acceptance of difference for the sake of peace” to an “enthusiastic endorsement of difference” (1997, 10–11). Here, I understand tolerance as more than resigned acceptance to include a basic recognition of others as moral beings. But this does not mean we should expect an “enthusiastic” endorsement of former enemies. Indeed, tolerance is for those with whom we strongly disagree. Tolerance includes several core elements: (1) it assumes the existence of continued, perhaps ineradicable disagreements with persons or beliefs, which are all more pronounced in the contexts under discussion here and (2) it requires that one's views are not forced on those with whom one disagrees, or that one not threaten violence or other sanctions if they do not change their ways (Locke1983; Newey 1999). Disagreements and even signs of disapproval toward others are fundamental aspects of social life, even more so following mass political violence, but their expression must remain within the bounds of minimally acceptable social behavior (i.e., fall short of outright violence or repression). And yet it implies that we continue to see ourselves as part of the same community as those with whom we disagree—in other words, we stay in a relationship with them and recognize that we have certain obligations toward them, and they toward us. Tolerance, as part of a larger understanding of reconciliation, is not simply disapproval and complete separation. Rather, tolerance requires that people stay in some nontrivial relationship with one another, and thus points toward a more significant level of reciprocal accommodation and recognition.

Tolerance, then, is the acknowledgment of difference and disagreement combined with a commitment to remain part of the same community. Beneath the disagreements and divisiveness there must be a basic, mutual recognition of moral worth and dignity—of respect—that serves to limit our reactions toward others. It is more than a unilateral act of one toward another, since it is premised on reciprocity. Tolerance and respect develop over time and require that the other normative concepts discussed below be adequately addressed. In this sense, respect develops partially in tandem with the actualization of these other norms, but only after they have been pursued successfully for some period. Over time, mistrust may recede as former antagonists learn to live with one another, although perhaps not embrace each other. It is only then, after erstwhile enemies come to see each other as citizens and moral equals (even if politically at odds), that respect can be said to take root. And it is then that we can talk about reconciliation, for at that point former enemies have made the difficult but important transformation of recognizing and accepting each other as members of the same polity and begin working together. To make sense of this, we now move to the corollary concepts: truth, accountability, recognition, and rule of law.

Truth

Public knowledge of past crimes is an important component of societal reconciliation, for without this perpetrators remain unchallenged, victims are further degraded, and acrimony and resentment between opposing groups remain, undermining future relations and destabilizing political and social life. An accurate rendering of the past can help mitigate these phenomena, first by undermining the self-serving justificatory narratives of perpetrators—and in the process holding them publicly accountable, even if only symbolically—and second, by offering extended recognition to victims and survivors. A generally accurate historical account can also frame future discussions about the past. Depending on the context, truth telling may largely confirm what many already believe happened, rather than uncover unknown crimes. In South Africa, for example, many blacks saw the truth commission's work as a vindication of the struggle against apartheid, as public truth telling provided a broader approval of the widely shared knowledge that the white regime had committed gross and systematic atrocities (though some whites expressed shock at the commission's findings). In other cases, however, truth telling essentially undermines denial. During periods of mass violence, the state may deny responsibility for abuses and employ disappearances and similar strategies that are explicitly designed to obscure the nature of its actions, as occurred in Argentina, Chile, and Guatemala (Garretón 1992). Uncovering the truth in these instances contains both an informative and critical dimension.

The aforementioned are fundamentally consequentialist arguments, for they identify the moral salience of truth telling in its ability to foster profound changes in societies. But public knowledge of the past alone is insufficient to secure reconciliation. New knowledge may only further antagonize opponents and rekindle conflict. But combined with other moral imperatives, truth telling can promote a new political order predicated on mutual respect and tolerance. Truth cannot provide all of this, nor does it serve as a curative elixir capable of magically erasing bitterness and distrust, but the alternative—continued denial, unchallenged dominance of perpetrator justifications, and further victim degradation—is certainly less likely to promote trust, and at any rate is morally unsatisfactory because of the contempt for fellow citizens that it implies. Admittedly, consequentialist arguments remain probabilistic at best, for it is impossible to prove axiomatically that truth telling results in a better society, especially if truth is divorced from other moral imperatives. However, as I argue below, when combined with the other normative concepts, truth telling can work to promote mutual respect and tolerance while remaining practically viable. Without a reasonable understanding of what happened, the other goals of reconciliation are unlikely to be met.

A second set of reasons, essentially non-consequentialist, relies on the moral right to knowledge by victims and society as a whole. Some scholars, such as David Crocker (2000) and Juan Méndez (1997), argue that there exists a deontological aspect to truth telling: Victims have a “right to know” about past abuses, regardless of the broader social consequences of publicizing such knowledge. International human rights law has also endorsed the right to knowledge. UN Special Rapporteur Louis Joinet has written that there exists a non-derogable “fundamental right to truth.” This right is understood as both an individual and collective right. “Its corollary is a ‘duty to remember,’ which the state must assume, in order to guard against the perversions of history that go under the names of revisionism or negationism” (1997, I, II). Theo Van Boven (1993) has also cited the right to truth as a basic right for victims of human rights violations.²

The deontological formulation requires further development, however: It should be conceptualized as both a right that victims have and a duty owed them through which truth telling becomes an expression of respect for their intrinsic moral worth—that is, an important element not captured in more consequentialist arguments. Truth telling acknowledges the dignity and moral value of victims, signaling that past abuses were morally wrong and require repudiation. At its core, a deontological formulation is certainly powerful and intuitively compelling. It justifies itself through the empathy we feel for victims and the claims they make on us to be recognized. By emphasizing a deontological dimension, we place victims at the center of reconciliation while encouraging society to reflect on its responsibility and obligations to those who were harmed.

Truth may be a fundamental aspect of reconciliation, but we are still in need of a satisfactory definition of the term. As I conceive it, reconciliatory truth contains three aspects: (1) an objective, forensic component, (2) a phenomenological, experiential component, and (3) a broadly narrative component that works to combine the first two aspects into an intelligible account that explains the past. I discuss each of these in turn, and then identify some specifically theoretical difficulties that can arise with truth telling.

Factual Truth

Since I accept that there is no socially relevant “fact” prior to its interpretation, I use the term factual truth simply to refer to those various types of events and phenomena that are employed in an interpretive process that invariably gives them meaning and places them in a longer chain of comprehension.³ Primarily, factual truth deals with identifying and investigating specific instances of political atrocity, such as massacres and extrajudicial killings, tortures, rapes, and other empirically identifiable violations of rights. It is concerned with empirically ascertainable events and actions, as well as the concomitant rules and procedures of verification that these require.

But cataloguing discrete violations is insufficient to arrive at a general understanding of political violence. Analysts must also investigate institutionalized patterns of repression, systematized policies of persecution, responsibility of actors and specific units, command structures, and legal codes that facilitated political violence. This is crucial because violence is often part of a broader state policy, and thus to focus solely on specific events and particular perpetrators ignores the larger administrative, legal, and political apparatus that facilitates such violations, as well as those at the top who give the orders for their execution. Mark Osiel has written perceptively on what he terms “administrative massacres,” as “the large-scale violation of basic human rights to life and liberty by the central state in a systematic and organized fashion, often against its own citizens, generally in a climate of war—real or imagined” (1997, 9). “Administrative massacres” captures the organized, large-scale and systematized nature of these violations, showing how disparate abuses are in fact often connected through state- centered policies.

For many societies reckoning with past atrocities, the population is often unaware of the extent of the violations. Certainly, there may be a broad understanding that abuses were committed, but their nature is not well-known. How systematic and organized were the abuses? Were they officially sanctioned? Were the targets legitimate political or military foes? How extensive were the violations? These and similar questions determine how justifiable the violations seem in the eyes of the population, whether they were inexcusable crimes or “unfortunate excesses” toward some defensible end. Investigation is crucial. Publicizing hidden violations, however, requires choices of selection and presentation. These choices then make, implicitly and explicitly, claims of moral responsibility, for certain actors are named as perpetrators, others as victims, and still others as morally responsible bystanders. In Althusserian terms, truth telling interpellates (Moon 2008). In an important sense, then, searching for the truth is both a historical and political enterprise; it creates winners and losers and reframes historical understanding. Factual truth may be concerned with investigating and classifying empirically identifiable phenomena, but the very process of selection and presentation points to its political nature.

Phenomenological Truth

I term a second form of truth as phenomenological. It deals with the physical and psychological legacies of violence on victims and their families, and the ways in which survivors make sense of these experiences. Fear is not easily framed in “objective” language. Nevertheless, personal stories offer insights that are missed in forensic reports. Consider Chanrithy Him's description of the aftermath of a Khmer Rouge (KR) attack on her town in 1975:

The destruction of something so familiar draws us closer. We dash toward the crumbled buildings, and the stench grows stronger. On the ground along the way, we see a soldier's camouflage hat and burnt pieces of wood from the classrooms. As we move even closer, the smell grows even stronger and buzzing flies swarm. Before our eyes lie piles of dead soldiers in destroyed bomb shelters that had been constructed in rectangular spaces where the flowerbeds used to bloom, between the steps to each classroom. Big flies with greenish heads and eyes swarm the gaping wounds in the soldiers’ decaying bodies. One blown-away leg lies beside the step to the first classroom, lonely and morbidly out of place. One soldier's crooked body lies on top of another soldier's, his mouth frozen open in excruciating pain.

I am nine years old.

Never have I seen so much death. For a moment I am hypnotized and spellbound by the ways these soldiers have been killed. […] My stomach begins to move. The breakfast I ate makes its way up my throat, followed by dizziness. Only then do I get hold of myself and feel the repeated tug, the persistent pull of Thavy's brother's hand. (2000, 60)⁴

The power of this account echoes Hans Erich Nossack's claim that “personal truth is the only reality,” and to speak it is “revolutionary” insofar as it forces us to confront the immediacy of political terror (2004, xii). But the importance of such stories goes beyond their particularity and immediacy. Victims bring order and meaning to their experiences through narration, simultaneously connecting experience to collective accounts by employing empirical examples. Telling personal stories can “break the silence” imposed through trauma, crucial to the development of a collective understanding of the past (Lira 1997, 5). It establishes a connection between the individual and collective memory, facilitating broader comprehension of what occurred. Thus, phenomenological truth has two dimensions: (1) as a reflection of the individual's personal experiences and (2) as a way of enriching and giving greater substance to social history. But while phenomenological truth plays a role in informing collective understandings of the past, its creation is a dialectical process: Its resonance emerges to the extent that it is both a response to and a further development of broader narratives—that is to say, it is not wholly the product of the speaker but also formed and shaped by other accounts of the past. As Claire Moon (2008) rightly notes, our understandings of violence shape and are shaped by already existing narratives. It is precisely this dialectical process that connects individual experiences to narratives about social conflict, providing a wider and empirically deeper understanding of the past. This is exactly what happens with phenomenological truths of the type presented in Chanrithy Him's account of the Khmer Rouge attack. Her story, and her memoir in general, provides us with a particularized account that both informs and deepens our understanding of the KR era, bringing greater immediacy to our understanding of the Cambodian genocide. And similar to factual truth presented above, phenomenological truth also includes a political aspect, insofar as certain facts and experiences are privileged in presentation; this in turn defines some people as violators, others as victims, and still others as bystanders. This rhetorical element makes these stories both powerful and also potentially problematic, as the combination of fact and empathy can provide a misleading account of events.

The difficulty, then, is how to understand the combination of factual and phenomenological truth as something that is a broadly accurate representation without falling into either a naïve correspondence theory or a radically fractured, deconstructionist notion that replaces the possibility of shared understanding with incommensurability.

Narrative Truth

In transitional settings, any historical inquiry is political charged. Making sense of what constitutes a “fair” appraisal of the past is problematic because of the political and ethical stakes involved and the difficulty of combining factual and phenomenological claims in a coherent, convincing manner. Any such enterprise is also interpretive: It requires making sense of each part, each “fact,” in relation to the whole story, which in turn affects the larger narrative picture. The goal of such a hermeneutic combination is to create larger narratives that make sense of events by placing them in culturally intelligible—and persuasive—language. In this book, narrative truth refers to these broader social understandings.

Keeping these narrative concerns in mind, it becomes clear that factual truth is not simply the cataloguing of crimes and creation of command-responsibility charts, but more importantly points to the selection (and implicit nonselection) of certain facts and people in an effort to make a larger, coherent story about political responsibility. Not all politically motivated tortures, murders, and rapes receive equal attention in the broader account. Some are used as paradigmatic examples of political atrocity, while others are downplayed because of questions about mens rea (intent), political salience, or other reasons. This is, of course, a political enterprise, insofar as decisions must be made about what events deserve special attention. Accordingly, deciding this is fraught with all sorts of difficulties concerning selection. Phenomenological truths also engage in a kind of hermeneutic of relevance. Individual stories are not totally agent driven, but rather are formed by and inform presiding social narratives, and in the process contribute to our historical understanding. A courtroom may provide a framework for victims to recount their experiences, while simultaneously restricting such presentations through the technical and procedural demands of the trial. Nationalist discourses about historical experiences (e.g., revolutionary struggles, past conflicts, moments of national glory) may be used to justify recent actions—often by all sides—while cultural discourses on responsibility, civility, and threats to the community can serve as the backdrop to debates in the public sphere. Transnational narrative discourses of human rights and popular sovereignty are important, as well as other political-ideological discourses (e.g., Marxism, Fascism, Maoism, anti-imperialism) within which phenomenological and factual truths are framed.

Nevertheless, if narrative truth as I have presented it here aspires to synthesizing truths into a larger overarching story about the past, it is not the case that these truths cohere unproblematically. Personal understandings of an event or series of events emphasize different experiences and downplay others, and can often contradict one another. There is, then, a certain openness that is constitutive of truth claims, insofar as they can never be complete (i.e., totally reflective of all of the events and experiences they represent) nor final. The hermeneutic synthesis required to make larger narratives intelligible is always open to revision, allowing for new interpretations of the past that emphasize some aspects and minimize others. However, this is not the same as wholesale relativism.

Rather, we need a critical history that presents the past as accurately as possible, while also ensuring that victims’ stories are not erased, nor abuses minimized through ideological manipulation of the past to lessen responsibility. A critical perspective eschews what Nietzsche called “monumental” histories, where the present is justified through the unreflective appeal to the past or any transcendental claim justifying “necessary” actions for its realization (1997 67–70). We need a perspective that interrogates given truths, and thus begins the admittedly difficult and politically delicate process of reconstructing a past that is not based on denials and convenient justifications. This is not an effort at radical deconstruction (if that means a relentless interrogation that risks resulting in incommensurable perspectives with little coherence), but rather a project that remains sensitive to its own assumptions.

Truth is fundamental for reconciliation. Societies need a basic understanding of past events to assign responsibility and resist impunity. The rule of law requires knowing which institutions are most responsible for abuse if they are to be reformed and made transparent and accountable. Punishment, too, requires knowledge of perpetrators, and without some semblance of the truth victims cannot secure the recognition they deserve as human beings; thus the norm of mutual respect stays weak. Certainly, truth alone will not reconcile former enemies. Victims who retell experiences publicly often desire more than merely being heard, even though this is important to them; they may seek reparations, justice for the guilty, and the state's guarantee that it will not violate their rights in the future. Admittedly, new revelations may re-antagonize former enemies and threaten stability, at least in the short term. If we seek to secure a basic condition of mutual respect, the commitment to truth must be matched by a commitment to accountability, recognition, and the rule of law. As such, the truth is necessary but certainly insufficient for securing justice and promoting respect.

Accountability

While truth is undoubtedly important for reconciliation, some form of accountability for wrongdoers is also necessary. Arguably, complete justice would require that all perpetrators be held responsible for past abuses and be appropriately punished—a kind of Kantian commitment to just deserts for all who deserve it. The immense practical obstacles to achieving this in the context of most transitions, as well as the risk of having it degenerate into a new “terror,” make it undesirable—however, at the very least, there must be a commitment to seeing that some “justice is done.” Before turning to the theoretical and practical limits of accountability, I discuss its normative content.

The Deontological Component of Accountability

Accountability requires publicly holding someone responsible and punishing him or her for an identifiable wrong or violation. While accountability can take many forms, it contains a constitutive element of sanction. This is essentially a form of retributive justice.

As John Borneman (1997) discusses, many (though certainly not all) legal theorists follow Aristotle in dividing justice into two spheres: distributive and corrective—two paradigms that miss the punitive dimension of retribution. Distributive justice, which is concerned with allocating resources according to some principle of fairness or equality, does not explain the element of punishment found in retribution. Nor does corrective justice. The latter implies returning to the status quo ante, so that the violator and the victim arrive at the relationship existing prior to the harm through the ruling of some impartial third party. For Aristotle, the goal of corrective justice is to compensate the victim for his or her “loss,” and in the process remove the perpetrator's “gain,” so that “when a measured value is assigned to the suffering, the terms gain and loss are appropriately used. Thus, the fair is the mean between the greater and the less…. That which would be correctively just, then, would be the mean between the loss and the gain” (1984, bk. V, ch. 7, lines 12–20). This assumes that we can return to some prior state, and furthermore that we should focus on the status of the victim, though not necessarily punish the violator. The perpetrator, for example, could be forced to return stolen property or compensate for its destruction, thus addressing the injustice caused, but not be punished for the theft in any meaningful sense. Retribution, however, includes a punitive element.⁵ First, it means punishing a perpetrator for his or her actions, and implies that victims are bearers of certain moral rights of dignity and worth (thus certain actions are seen as violations of these rights), and second, it means that perpetrators are also moral actors, to the extent that they are morally responsible for their actions and should thus be held accountable. Retribution means that we understand ourselves as agents with moral rights, and perpetrators as morally responsible actors (Murphy 1998).

Retribution, then, need not be pursued solely for utilitarian ends. While certainly retribution may have an instrumental bent—deterrence, enforcement of rules, expression of intolerance of impunity—this does not exhaust its power. Rather, it is also non-teleological: Violators are punished precisely because they are morally responsible for their actions, and not (solely) because punishing them would benefit the greater good. Kant underscores this when he writes, “Punishment by a court…can never be inflicted merely as a means to promote some other good for the criminal himself or for the good of society. It must always be inflicted upon him only because he has committed a crime” (1996, 105).

Retribution can also be understood from the perspective of the victim and her moral claims to respect. A victim expresses self-worth by showing resentment toward the violator and seeking to punish him (or have the violator punished). Unlike Nietzschean envy, resentment here is anchored in the anger that results after experiencing some moral wrong. Indeed, resentment is an expression of self-worth, for it shows the value one places on oneself and subsequent outrage at its violation. Without a sense of human dignity and value, there would be no resentment. It is hard to imagine a person who respects herself, much less others, who does not resent purposeful harms inflicted on her.

Though there can be teleological aims to retribution, and indeed below I indicate what some of these are, retribution is centered on the wrongness of a particular act and the violator's responsibility for the commission of this act. As such, retribution is also communicative, insofar as it communicates our moral condemnation of the violation to its author and to society, a point not lost on Kant (1996) when he argued that any punishment must be expressed publicly, even if the community were to be dissolved immediately afterward. However, retribution should not be confused with revenge, even though the two terms are often used interchangeably. While revenge also draws its force from the apparent justness of punishment, it risks collapsing into cycles of violence, bereft of any sense of proportionality and with little likelihood of ending. Retributive justice differs from revenge through its inclusion of procedural and substantive protections for the accused, and shifts from the victim's demand for immediate punishment to a measured response based on the rule of law. Unlike vengeance, it requires an evaluation of responsibility and an appropriate penalty.

I use retribution (and accountability) as a measured, legitimate response to harms against people. This formulation serves as an important heuristic device and moral ideal to orient our discussion of post- atrocity justice.

The Teleological Component to Accountability

To the extent that we can normatively separate retribution and vengeance, retribution can serve the broader moral enterprise of reconciliation. I indicated earlier that retribution should not be seen only in instrumentalist terms. Still, under the proper institutional constraints, it can further the project advocated in this book in at least two ways. First, a commitment to accountability sends a clear message that continued impunity will no longer be tolerated. By holding violators accountable for their behavior, the new state condemns past abuses and signals the importance of the rule of law. Second, and more broadly, such a commitment reaffirms (or perhaps affirms for the first time) social values of respect for human rights. These points are closely tied to one another, but I discuss them separately in order to tease out their importance.

Eliminating impunity is possibly the most defensible “forward looking,” or instrumentalist, use of retribution in transitional settings. Continued impunity is dangerous for a fragile democracy, weakening the rule of law and frequently jeopardizing the very existence of the new government. Particularly where former elites maintain some political, economic, or even military power, prosecutions may be used as an important way of dissolving this power. Along these lines, a typical defense of retribution rests on its promotion of the common good by sanctioning those who violate society's norms. This utilitarian argument, with roots as far back as Cesare Beccaria (1995) and reformulated by Jeremy Bentham (1995), normally assumes the existence of a relatively stable society with a functioning and impartial judiciary. Transitional societies require that we expand our focus to the role of the state as violator. In transitional situations, the emphasis shifts to a hypothetical: What would political and social life be like with the continued existence of significant authoritarian enclaves? Ruti Teitel has perceptively made this point: “Rather than an argument for punishment in the affirmative, the argument [in transitional situations] is generally made in a counterfactual way; What result if there is no punishment?” (2002, 28). The answer is the continued and possibly pervasive undermining of the new order by elites from the previous regime, who would feel unrestrained in their actions.

Accountability also expresses a repudiation of past practices and a commitment to change future relations based on principles of respect for human rights and the law (Kirchheimer 1961; Shklar 1964). Punishment signals the government's interest in establishing a new political and social order rooted in the rule of law by underscoring the importance of human dignity and subsuming state interest to human rights. It creates a wall between then and now, and publicly shows that state power must be limited. This is a pedagogical exercise, in many respects. By holding perpetrators accountable, the state signals to society that human rights and respect for individuals should become dominant social norms. Whether this is successful, of course, partly depends on the perceived fairness of the trials themselves—however, accountability reaffirms these norms by highlighting the wrongness of their transgression. In this sense, accountability not only means drawing a line between the past and the future but also signals the importance of endorsing new social values.

Challenges to Accountability

Although accountability is a crucial component for the establishment of a new regime, it nevertheless faces certain pitfalls. I have already mentioned the possibility of punishment degenerating into vengeance, which is a real threat when transitional regimes do not emphasize their commitment to the rule of law and instead replace one illiberal order with another. There are several other recurring problems with retribution.

First, formal retribution can be destabilizing. Trials and demands for punishment undoubtedly antagonize accused elites, who may threaten renewed violence. The threat of destabilization is a serious one, and entering into the delicate calculus of punishment versus stability poses significant dangers for new and fragile regimes. While there is no definitive theoretical solution to this problem, accountability—such as identifying perpetrators, writing new histories that detail complicity and responsibility, and encouraging massive social mobilization against perpetrator elites—should also be pursued in other (nonlegal) social spheres, such as civil society and elite discourse. Here, a commitment to unearthing the truth of violations is particularly important—though legal action may not always be possible, at least the truth can serve to condemn violators in the social imaginary.

Of course, excessive demands for retribution are also problematic. It is not uncommon to see the accused sacrificed for social cohesion, turning them into scapegoats for the sake of “moving forward.” Successor regimes are always confronted with the formidable task of establishing a new basis for solidarity, and certain strategies—such as high- profile prosecutions of a handful of prominent or particularly violent perpetrators—can function nicely as symbolic evils with which to suture open wounds. The symbolism of human rights trials, characterized as they are by powerful discourses of villains versus the innocent, or evil versus good, can have a far stronger impact than the bureaucratic reform of the security apparatus and judiciary. At risk, of course, are the due process protections of the accused, shed for the sake of social solidarity. Trials can be made to function as rituals to purge the sins of the past from the present (Walzer 1992). Though vengeance may be a part of this, it need not be; the motive may simply be to select some perpetrators as sacrificial lambs for social solidarity, a more instrumentalist (rather than vindictive) strategy, though a feeling of popular revenge certainly helps. Scapegoating not only affects trials but also broader, category-based forms of punishment such as lustration (purging), where an entire group of people such as party members are systematically denied basic rights, including voting, employment opportunities, government office, or something similar. In the first scenario, a small number of individuals are blamed for widespread violations, even though most mass atrocity involves broad institutional and organizational resources implicating a wide array of people in various ways. In the second scenario, guilt is extended to an entire class of people, who are summarily excluded from certain activities or positions, even though complicity can vary widely among individuals (Wilke 2007). In both situations the desire for social cohesion may undermine justice.

A third problem with accountability concerns the relationship between prosecutorial selectivity and historical representation. The successor regime's desire to show its commitment to accountability, often through a selective number of high-profile cases, can result in a misrepresentation of the past. Certainly, financial and resource constraints are a prime source of selectivity. Given that in most cases the scope of abuse was massive and the complicity widespread, a transitional society with limited resources can only afford to prosecute some violators. The question then becomes “Whom do we prosecute?” The choices rely on a number of factors about potential defendants, including their responsibility for particular abuses, public demands for their punishment, the likelihood of conviction, and what the new government can gain from a successful trial. Regardless of the choices made, greater public scrutiny will be drawn to a particular group of individuals while limiting broader historical understandings of the past. Indeed, the past actions of the defendants may become misleadingly representative of all of the regime's actions, thereby risking simplification of what are invariably complex events with many actors.

Closely related is a fourth problem. The repercussions of trials, with their symbolically powerful findings of criminal liability and victimhood, can continue to have an effect on social relations for a long time afterward. Future relations can be shaped for the worse by the public stigma that flows from prosecutions: Those who have even tenuous connections to the prior regime may continue to be disparaged or ignored, regardless of the substance of their political positions, while victims may enjoy continued moral capital well beyond what they should expect, allowing them to exercise political influence over issues that have little to do with their past experiences or victim status. The possibility of creating a shared future where former enemies can see one another as members of the same community can suffer when past culpability continues to delegitimize opposing viewpoints and vigorous public debate. One need only think of the way that under President Paul Kagame, even legitimate political opposition in Rwanda is often accused of treasonous aims.

A strictly liberal legal understanding of culpability requires that individuals be held responsible only for those actions over which they had some control (Fletcher 2001). While understandable, such an approach risks focusing on too small a group of perpetrators, and thereby missing the morally ambiguous category of bystanders. Bystanders, meaning persons who through their inaction and even tacit support facilitated the commission of wrongs though they may have been in a position to oppose or denounce them, carry a particular form of responsibility that lies outside judicial guilt.⁶ How are bystanders to be judged, if at all? In some respects, bystanders point to the limits of accountability, and bystander responsibility will be debated in the public sphere among political elites and civil society groups, as well as among individuals in private and semipublic places. However, their responsibility falls outside the purview of courts, and this combined with the troubling fact that substantial public support, or at least acquiescence, is often necessary for political violence means that an understanding of responsibility will likely be highly contentious and politicized. Truth commissions may be able to explore at least some aspects of this, but ultimately bystander accountability will largely be contested in political and civil society, and among individuals.

Although all of these issues point to the limits of accountability, its importance should not be dismissed. Retribution helps promote the rule of law and provides at least some recognition of victims, for it signals the importance of condemning those who harmed them while reinforcing values of human rights and dignity. Accountability is thus a crucial element in the larger project of reconciliation.

Recognition

A common legacy of political violence is the continued contempt and devaluation of victims. Often, victims are seen as having “deserved” their fate, particularly if they belong to groups that have been historically marginalized or disparaged (e.g., indigenous groups in Latin America). Such devaluation makes it difficult for reciprocal respect and tolerance—and hence, reconciliation—to develop, since these require that former enemies come to see one another as moral equals. In response, a society should seek to recognize the moral status of victims as equals, restoring their sense of dignity and establishing their legal rights as citizens. Such an idea of recognition is, certainly, tied to accountability, since both pivot on reinterpreting victims as bearers of moral value and violators as deserving punishment. Nevertheless, while accountability is a crucial element of reconciliation for the reasons discussed above, it does not, in and of itself, focus on victims. Prosecutions and other forms of retribution may provide victims a measure of justice, but these are not aimed primarily toward them. Indeed, following a successful trial, victims may still suffer disparagement and marginalization while continuing to live with the terrible consequences of what happened to them.

Additionally, fellow citizens are less likely to view victims as equals while the state continues to ignore them. Recognizing victims as moral agents may help undermine apologist perpetrator narratives by recasting the consequences of violence in terms of the victims’ experiences. This change in emphasis, from violators to those who suffered, can bring attention to the ways in which the endorsement of violent and exclusionary ideologies resulted in crimes with actual victims. This may erode the exclusionary political project of the perpetrators and, perhaps, create the possibility of achieving a more inclusive understanding of plurality and respect. In the following section I discuss the elements of a theory of recognition in transitional settings and then follow with a consideration of some theoretical obstacles and problems peculiar to this context.

Theorizing Victim Recognition

Survivors’ reactions to a legacy of violence are complex and varied. Many may call for prosecutions of perpetrators, others may demand programs to help them cope with trauma, and still others simply prefer burying the past (Chakravarti 2008; Verdeja 2000, 2007). The varied responses point to the difficulty of fashioning a general theory of victim recognition that holds across cases. Nevertheless, we can posit one objective: to restore victims’ dignity and provide adequate material support so that they may create meaningful lives, and do so in such a manner that neither patronizes them nor undermines their sense of moral agency. The goal should be to provide symbolic and material reparations to victims while simultaneously not degrading them as impotent, lacking in agency, or incapable of achieving self-respect and worth.⁷

The debate surrounding the value of recognition has a long history. Most contemporary theorizations begin from the Hegelian position of identity construction based on a dialogical model of interaction. For theorists in this tradition, recognition is a reciprocal relation whereby subjects see each other as equals entitled to respect. Recognition is thus a principle aspect of subject formation. Beings become full individuals through mutual recognition, underscoring the fundamentally intersubjective (i.e., social) nature of identity formation. This idea that recognition through social praxis is fundamental to stable and healthy identities has been developed by a number of authors (Benjamin 1988, 1995; Taylor 1994). Axel Honneth, for example, argues that a healthy notion of the self is a fundamental element of the good for individuals, and he elaborates the requirements for undistorted identity as consisting of three key components: (1) self-confidence, developed through affective relations between intimates and others who are emotionally proximate, (2) self-respect, accorded through the legal discourse of rights and implying the individual's capacity for autonomous moral action, and (3) self-esteem, developed through participation in communal activities and contributions to a meaningful, ethically substantive social life. These components are all developed through dialogical interactions with other, equal subjects. Moreover, they are crucial for a healthy subject. Without them, the individual risks degenerating into pathologies of self-hatred and denigration (1996).⁸

Some thinkers take these insights significantly further, arguing that the dynamics of individual recognition are mirrored at the macro level. Charles Taylor has discussed how patterns of individual misrecognition parallel those of groups: If the self can suffer mistreatment through devaluation, the same holds for entire groups that are consistently oppressed or suffer discrimination. They are unable to actualize themselves satisfactorily, and natural cultural expression and maturation are truncated or, even worse, fatally arrested. For Taylor, “A person or group of people can suffer real damage, real distortion, if the people or society around them mirror back to them a confining or demeaning or contemptible picture of themselves” (1994, 25). Consequently, certain groups require recognition of their uniqueness in some special, institutionalized manner, a claim that goes beyond the kind of recognition predicated on social equality—in other words, one that is difference-blind.

The criticisms of these approaches are well-known. Seyla Benhabib (2002) has persuasively argued that Taylor and others such as Will Kymlicka fall into traps of cultural essentialism, reifying group identities and privileging authenticity claims above basic justice concerns. Nancy Fraser (2003) has criticized these overly psychologized multicultural approaches for a variety of reasons, including their inability to define satisfactory criteria for distinguishing between just and unjust authenticity claims (and the implicit essentialism this springs from), their reductive assumptions about the primacy of recognition over injustices rooted in political-economic relations, and their inability to theorize from a more objective, sociological position that can distinguish between institutionalized/systematized patterns of subordination requiring justice and culturally salient differences that do not. Benhabib's and Fraser's observations are especially helpful in studying post-atrocity societies. Victim groups often make authenticity claims and special recognition demands in transitional settings. On the face of it, these claims appear quite legitimate, and indeed they often are; the individuals in question suffered devastating violations, offenses that may have been facilitated by historical patterns of categorical subordination and discrimination, but whose overwhelming and immediate barbarity carry a poignant demand that we acknowledge their experiences in some nontrivial fashion. The problem is in how we should acknowledge them, and what should be the criteria. Should we recognize victims as a way of enabling their ethical and moral self-realization, as Honneth would argue? While this approach may seem compelling, a theory of victim recognition based wholly on ethical self-realization runs into conceptual challenges because it is incapable of drawing the line between what constitutes satisfactory recognition and what exceeds it. Claims of self-realization, as Fraser points out, are “usually considered to be more restricted” than justice claims precisely because they are based on “historically specific horizons of value” (2003, 28). The difficulty here arises with the potential development of so-called “cultures of victimhood,” where similar experiences become a shared horizon of authenticity that demands categorical respect based not on the content of any particular claim but rather on the status of the speakers. In other words, in some scenarios victim group elites may transform their status as victims into a badge of irreproachable righteousness used to make morally suspect claims (such as a right to oppress internal members), or point to their status as a way to dismiss otherwise valid criticisms or challenges. What if the elites of a particular group that suffered massive human rights violations, say an indigenous group, argue that proper recognition of their identity requires that the state not interfere with the internal subordination of a particular subgroup, such as women (Okin 1999; Warnke 2000)? Should the state accede on the principle that this particular group was victimized and it now requires recognition? It is not uncommon for a victim group to claim special recognition rights in the aftermath of massive violations, in effect trading on its moral capital to gain further rights, resources, or autonomy.

Of course, victim demands are not always morally dubious. Far from it. In fact they can often be sound, but the principle of ethical self-realization does not give us the conceptual tools necessary to decide which claims are legitimate. For Taylor and Honneth, intersubjective recognition is a necessary condition for achieving undistorted, healthy identity. “The conception of ethical life,” writes Honneth, can articulate “the entirety of intersubjective conditions that can be shown to serve as necessary preconditions for individual self- realization” (1996, 173). However, to base a theory on claims of ethical self-realization leaves us incapable of discriminating between justice claims and (nonuniversalizable) authenticity claims. A primary concern should be to ensure that any model of victim recognition focus on questions of justice, or “right” (i.e., rather than the “good” entailed by an approach premised on ethical self-realization). Recognition of victims is crucial in transitional societies, but the aim should be for victims to reestablish their dignity and self-worth in such a way that permits them to be full participants in social and political life. This does not mean that all ethical claims are illegitimate. Rather, it means that these claims should be honored to the extent that they promote what Fraser (2003, 29) calls “reciprocal recognition and status equality,” a goal that is unachievable if victims continue to find themselves excluded, marginalized, devalued, and forgotten.

Additionally, some theories of recognition tend to reduce all forms of injustice to symbolic misrecognition, while saying little about material inequality, except to consider the latter as a predicate form of injustice. Honneth subsumes the latter under the former when he writes, “the conception of recognition, when properly understood, can accommodate, indeed even entails, a modified version of the Marxian paradigm of economic distribution…” (Fraser and Honneth 2003, 3). However, victims often receive symbolic acknowledgment from the state, including an official apology or monument in their in honor, but receive no material support. And yet, they are just as often left impoverished following mass violence, particularly where an entire ethnic group was targeted (e.g., the indigenous in Guatemala, blacks in South Africa). In this context, an apology is insufficient for social reintegration. Consequently, material inequality requires theoretical elaboration in conjunction with symbolic forms, rather than being subsumed into the latter. The goal should be to recognize their experiences as a step toward overturning patterns of discrimination and violence.

Any theory of victim recognition, then, should include both material and symbolic components, while avoiding claims of ethical self- realization to anchor it. As I have argued elsewhere, it does not follow that we must privilege liberal individualist rights and reject all collective claims (Verdeja 2008). That would miss the collective nature of violations, where groups were targeted (however defined by the perpetrator) and abuses carried out systematically. It is necessary, however, to distinguish between those policies that protect culturally essentialist claims and those that promote status parity among citizens. A theory that seeks status parity as a goal targets both symbolic misrecognition and material maldistribution. In terms of symbolic recognition, it should emphasize the elimination of cultural views that prevent individuals from recognizing each other as fellow citizens, to achieve what Fraser has termed the “intersubjective condition” of parity of participation (2003, 36). This requires the positive revaluation of “disrespected identities” and, more generally, cultural diversity, as well as the delegitimation of those social values that worked to justify violence and misrecognition (2003, 47). In terms of redistribution, it necessitates addressing economic marginalization that prevents individuals from participating as equal citizens, and secures the “objective condition” of participation parity. This may include a number of initiatives, such as monetary compensation for abuse and increased development programs in places targeted by the violence. The ultimate goal is to restore victims’ dignity and self-worth so that they may participate fully in social, economic, and political life, achieving “reciprocal recognition and status equality” with their peers (2003, 36). Without both material and symbolic strategies to correct past injustices, such a goal will remain unrealized. Achieving this goal requires both symbolic and material (i.e., redistributive) claims of justice. I use the terms victim recognition and victim acknowledgment in this broader sense of status parity (Fraser 2003) that includes both symbolic and material components.⁹

As I understand it, an account of victim recognition consists of four ideal-typical dimensions: “symbolic” and “material” along one axis (a typology of acknowledgment), and “collective” and “individual” along another (a categorization of recipients). These dimensions trace the scope and type of acknowledgment that should be accorded, and though different mechanisms are appropriate within each created space, they contribute to a coherent conceptualization of victim acknowledgment (Verdeja 2006). Below I introduce these four dimensions and follow with a discussion of the normative challenges that each faces.

In most cases of large-scale atrocity, crimes are directed at groups of some type, such as cultural, ethnic, religious, national, ideological, racial, or economic, and often, targeted groups cover different categories. In addition, targeted groups can overlap several different categories, and violations may also be gender-specific (De Vito 2008; Jones2006a, 2006b). Because of this broadly collective dimension, victim acknowledgment requires theorization of a collective symbolic element. Recognizing this means publicly highlighting that violations were part of an organized, coherent strategy against designated collective “enemies,” and not merely occasional “excesses” on the part of the perpetrators. Collective symbolic recognition requires recognizing both the way strategies of repression targeted victims as a group, and society and the state's moral obligation to recognize their experiences and treat them as equal citizens. This means fighting discourses that blame victims for what happened to them. Such symbolic recognition can be made in many ways, including through official apologies, public atonement, developing public spaces to honor victims, and establishing museums, monuments, and days of remembrance to promote and preserve collective memory.

Crimes, of course, are not merely collective. Individual symbolic acknowledgment consists of the need to recognize victims as individuals and not simply place them in a residual category, thus reducing them to an amorphous group of passive, voiceless survivors. This includes drawing attention to how violence and repression affected individuals qua individuals, and reminds us that “victim experiences” are always more than the aggregate of similar stories, pointing to the importance of recognizing actual persons who suffered in deeply personal ways. Such recognition can in practice be unattainable. Recognizing all victims in a meaningful sense is impossible when confronted with crimes of this magnitude, and not everyone can be given a space to speak publicly. Nevertheless, sensitivity to personal experiences is important because it underscores that victims are actual persons, not simply statistics. As the Argentine writer and torture survivor Jacobo Timerman (2002) reminds us, individual suffering is always more than a symbol of systematic crimes, and part of the process of rehumanizing victims requires attention to this fact.¹⁰ Indeed, such recognition contributes to reaffirming their status as citizens, for it reflects a sensitivity toward fellow humans that is a crucial element of any political order based upon democratic principles of equality and mutual respect. Certainly, symbolic individual recognition is not equivalent to the liberal democratic rights that accompany citizenship, but it is an important prerequisite. Without such recognition of individuals as individuals, and as equals who deserve respect, it is unlikely that victims will maintain their status as citizens.

Symbolic recognition furthers victims’ sense of dignity and self- worth, while reaffirming their place as fellow citizens—however, it does not address the material marginalization that is a common legacy of violence. Often, survivors are left impoverished by widespread or systematic violations, and special attention to their economic status is required if reparations are to be more than merely symbolic. Consequently, reparations should include some form of material support that gives victims the capacity to lead meaningful and productive lives. One form this can take is collective material reparations. These reparations provide resources to victimized groups as a way of obtaining the material basis and security required for them to participate fully in social, political, and economic life. They may include initiatives such as employment and housing assistance for groups whose economic situation was directly affected by the violence, physical and psychological support for trauma, and infrastructural investment in targeted communities (e.g., better roads, sanitation programs, rural education campaigns, credit allowances for economic development). Although the nature of the programs requires sensitivity to context and the particular needs of the victims, the programs share two characteristics: (1) they are for groups that were targets of violence (and are thus collective) and (2) they are dependent on the redistribution of economic resources, with the aim of enhancing victims’ lives so that they may realistically pursue their life plans. In Guatemala, El Salvador, and Peru indigenous groups were the primary victims of political violence, and truth commissions called for significant investments in public education, housing, employment, and economic development to offset the legacy of economic inequality inherited from the civil wars. These measures can help raise the standard of living of the most damaged communities and contribute to reintegrating marginalized groups into society.

Combining reparations programs with broader development programs can be challenging. However, some theorists have argued that combining the two can have positive results.¹¹ To be sure, society would most likely benefit from increased economic development in poor areas. Nevertheless, combining these two programs can undermine the normative aspect of reparations, since doing so may submerge the specifically moral dimension of reparative justice beneath broader state policies to combat poverty (Wilson 2001). For many victims, reparations are a moral acknowledgment of wrongful suffering, and subsuming them into development strategies obscures this. What the state may consider reparations may very well be part of the duties it has toward its people as citizens. While employing a discourse of reparations may result in greater political and moral capital for state elites, it confuses the normative specificity of reparations with broader obligations. Therefore, any reparations program should be sensitive to this risk, and should be crafted in such a way that maintains its distinctly normative dimension, for example, by explicitly invoking elements of symbolic recognition (e.g., apologies, days of remembrance), even if carried out simultaneously with general economic and infrastructural development plans.

Finally, there is an individual material component to theorizing victim recognition. This, too, is a form of distributive justice, insofar as it addresses the importance of redistributing resources to victims, but places greater emphasis on the autonomy of individuals than the collective dimension discussed above. Of course, no compensation can substitute for death or torture, and in this sense money—or any reparatory measure—is always inadequate. But compensation can have an impact for economically destitute victims and show that the state's recognition is not merely symbolic but also material and practical. Individualized reparation schemes normally include familial rehabilitation through access to medical, psychological, and legal services, compensation for losses that can be measured financially, economic redress for harms that are not easily quantifiable, and restitution of lost, stolen, or destroyed property. Individualized payment has the benefit that it maximizes autonomy by allowing victims to use funds as they see fit, and thus minimizes the paternalism inherent in collective material reparations (De Greiff 2006; Goodin 1989; Lomasky 1991).

Victim recognition is a crucial element of reconciliation. Whereas a commitment to truth seeking and to accountability has an impact on victims, a special focus needs to be given to individuals who suffered massive wrongs, not only as a way of reintegrating them into society as fellow citizens but also as a means of recognizing their worth and dignity as fellow humans—that is, as a way of according them moral respect.

Rule of Law

The final normative component of reconciliation as respect is the rule of law. Many commentators have identified the reestablishment of the rule of law—understood as cogent general rules that constrain the actions of the state and establish minimum legal protections for the citizenry—as an important element in democratic transitions (Hampton 1994; Kleinfeld 2005; Nino 1996; Plunkett 1998; Scheuerman 1994). Adopting the rule of law reflects the state's commitment to preventing the recurrence of such violations by reforming relevant state institutions, such as the judiciary and security apparatus. For transitional societies, however, it means more than this. It also means redrawing the boundaries of politics. Of particular importance is tracing some basic normative principles emphasizing that political differences will not be resolved violently—rather, political contestation remains bound within formal and informal spheres of deliberation and negotiation. Consequently, there are two dimensions to the rule of law: one institutional and the other normative. Although I focus on the latter in this book, both elements are necessary for reconciliation, and I briefly discuss the institutional component below before turning to the normative aspect.¹²

Institutional Dimension of the Rule of Law

Institutional reform of the rule of law in post-atrocity societies normally addresses the judiciary, police/state security apparatus, and prison system, though the specifics of this largely depend on the condition of the legal system at the point of the transition.¹³ These conditions can vary widely. Rama Mani (2003) has usefully outlined the relative functionality and legitimacy of different systems across numerous transitional cases. Mani notes that in some transitional cases, the legal system still operates, though its legitimacy has largely eroded after years of sanctioning clearly abusive policies by the state. Specifically, the judiciary continues to function and refers to existing law and statutes. Trials, however problematic, are still held for political detainees. Nevertheless, the rule of law is employed as an instrument of repression by the state, and is considered illegitimate by broad sections of the population. The state uses the law as a veil behind which it coerces the judiciary—and often the parliament, if one exists—for its own political ends. Chile, Argentina, and South Africa fall under this rubric. In these situations, legal positivism is insufficient to guarantee popular trust in the state's commitment to the rule of law because the space between written law and popular conceptions of justice remains so wide.

In other instances the rule of law may be significantly weaker. Here, a judiciary and legal order exists throughout the period of violence, but enjoys no autonomy or claims to impartiality. Most persons have no recourse to the legal system; the law becomes an instrument of power for corrupt elites. Public trust in the law disintegrates completely and the rule of law is progressively evacuated of any meaning, ultimately becoming a parody of itself. El Salvador, Guatemala, and Zimbabwe fit this category almost perfectly.

In yet other situations the entire legal system may collapse and no organized system of law, however corrupt and illegitimate, survives. Under situations of chronic war and massive population dislocation, a country can fall into chaos and power is exerted largely through direct violence and explicit coercion. In these conditions, formal legal systems are dismantled or dissolve on their own. Cambodia under the Khmer Rouge and Somalia exhibit this situation most accurately—in neither country were even traces of a legal order left after the violence.

This triptych is not meant to be exhaustive, and cases may show signs of all three stages at different historical periods. Identifying variations in the demise of the rule of law points to the multiplicity of obstacles that reform efforts face, and it follows that different cases exhibit differing needs based on their contexts. Nevertheless, what unites all of these societies is the need for a robust, fair, and transparent legal order as an institutional sine qua non of long-term reconciliation.

In all of these scenarios, building public trust in state institutions is an important aim of reform. As such, the commitment to the rule of law is not only a technical matter of policy implementation and efficiency but also a normative enterprise. The state fosters trust by showing it recognizes the rights of its citizens and places clear, consistent limits over its authority. This includes respecting standard liberal rights such as freedom of speech and association; the right to personal bodily integrity and property; protection against arbitrary detention, arrest, or exile; freedom of movement; and other rights that place restrictions on arbitrary state power. It should also include some basic rights that seek to maximize the individual's ability to participate in the political life of his or her country, such as the right to vote.¹⁴

Normative Dimension of the Rule of Law

To the extent that the rule of law is a normative concept, it also directly engages with the constitution of the political realm. How do we understand politics in a post-conflict scenario? What are the limits of political contestation, that is, the boundaries between acceptable politics and violent coercion? The reforms discussed above, while fundamental, do not answer these questions. The commitment to the rule of law also highlights the ability of citizens to discuss and debate politically relevant issues without turning to violence or threats of violence. In the aftermath of violence, it is not uncommon for groups to threaten to abandon the realm of debate if they feel that it is unlikely that they will achieve their aims. Even securing this liminal state can be exceedingly difficult, as recent experiences of violence and terror are likely to feed demands for vengeance and sow mistrust.

Admittedly, the idea of a liminal commitment to peace still falls short of the “rule of law” concept as normally understood in liberal democratic theory. This thin conceptualization has more in common with the agonistic notion of contestation found in the works of Schmittians like Chantal Mouffe.¹⁵ For Mouffe, politics is characterized by “the vibrant clash of political positions and an open conflict of political interests” that emerges from the basic friend/foe political category. Because democracy is never fully realizable in any substantive sense (say through the articulation of a Rousseauian “general will”), the best we can expect is a model that accepts the “impossibility of a world without antagonism” while demanding that “the opponent should not be considered an enemy to be destroyed but as an adversary whose existence is legitimate and must be tolerated” (1997, 4). That is, the most we can hope for is a political order that allows for as much contestation as possible without devolving into a violent politics of power. Mouffe's understanding of politics as agonistic is helpful because it draws attention to the incompatibility of essentialist political claims and democratic politics: The attempt to fuse the two, through authoritarian mass mobilizations and plebiscitary politics, is often the harbinger of future political violence. But her understanding of politics as barely containing conflict is normatively too thin, particularly in transitional situations where political discourse is already badly impoverished and there is little if any commitment to peaceful debate. Some deliberative democrats have attempted to move beyond this liminal understanding while still maintaining the complexity and irreducibility that “radical democrats” such as Mouffe emphasize.

David Crocker has been one of the most forceful defenders of deliberation in transitional settings. Drawing from the works of James Bohman and Amy Gutmann, Crocker argues that society “should aim to include public debate and deliberation in its goals and strategies for transitional justice” (2000, 108). Nevertheless, he is careful to note that the commitment to deliberation is not in itself a solution to pervasive inequalities or animosity. He acknowledges that it is unlikely that in “any given society, there will be full agreement about the aims and means for dealing with past abuses,” especially in a nation recently experiencing massive violence (2000, 109). But the goal of deliberation committed to the rule of law is to attenuate deep disagreements through public debate, allowing all affected sides to participate and seek “morally acceptable” compromises. In transitional situations, achieving this constitutes a significant political victory. A somewhat stronger understanding institutionalizes deliberation through proceduralist mechanisms of legitimacy formation. This, in turn, requires a deliberative process governed by principles of equality and reciprocity among participants and the right to initiate, debate, and question the content of discussion. The procedures themselves grant legitimacy to the outcome of debate. As Seyla Benhabib (1996, 72) notes, “Procedures can neither dictate outcomes nor define the quality of the reasons advanced in argumentation nor control the quality of the reasoning or rules of logic and inference used by participants. Procedural models of rationality are undetermined.”¹⁶ Thus, the substantive good of social life must remain open; the model remains silent on what constitutes ethically appropriate social relations. This is one crucial difference in the construction of the political sphere between a democratic and a totalitarian regime. In the latter instance, laws and procedures are subordinated to the creation of an ethically homogeneous society, with all of the violence and “purification” that this entails. In a democratic polity, the conflict of values and interests is taken for granted, and thus the procedural mechanisms in place to process and contain them offer the grounds for legitimacy. The law, understood as a codified series of procedural norms that both limit and permit the possibility of debate (i.e., limits debate insofar as it prohibits the recourse to violence, and permits debate by granting a space for deliberation free from coercion and threat of force), helps move a society from violence toward deliberation. The commitment to the rule of law, then, means agreeing to deliberate within a recognized system of rules rather than turning to violence to solve disagreements. As such, it is more robust than the liminal conception arrived at by Mouffe (1997), but remains wary of the integrative understanding of politics promoted by advocates of “forgiveness in politics,” discussed in the first chapter. This understanding of deliberation is essentially proceduralist: It focuses on the procedures developed through open and fair deliberation to give future decisions their legitimacy and binding power.

While this notion of rule of law may seem theoretically overburdened since it includes some of the normative criteria of deliberative democratic theory, I do not believe it is, at least not in the sense I use it here. The commitments to robust debate through meaningful political participation can be satisfied only if basic principles of equality before the law are adopted and violence is rejected. It is in this sense that I use the rule of law to address the importance of peaceful, impassioned debate to replace the politics of violence, and as a fundamental component of reconciliation. I do not mean it in a theoretically more substantive sense requiring a particular set of institutional mechanisms and theoretical links between power, money, and solidarity, as found in, for example, Habermas's Between Facts and Norms(1996). While these discussions are important, the theoretical postulates developed from analyzing consolidated, capitalist democracies are of a different order from the cases under study here. Nor do I follow Habermas in positing a strongly rationalist grounding to deliberation. As I discuss in Chapter 5 on civil society, any theory of deliberation for post-atrocity societies cannot assume that competing groups mutually espouse rational rules of discourse, a commitment to forging consensus, or even have shared background norms that allow for a common understanding of the rules of debate.

Alone, the normative dimension of the rule of law may seem silent on issues of social justice and the uneven political standing of different participants. In the aftermath of massive political violence, a model that emphasizes deliberation without reflecting on the material conditions necessary to ensure that participants actually have some equal standing in debate could be accused of naïveté, or worse, deliberately perpetuating modes of domination under the false pretense of liberal (or discursive) equality. But in keeping with the fundamentally integrative approach of my understanding of reconciliation, I emphasize that this goal must be understood in concert with the material component of victim recognition discussed earlier. As indicated there, victims cannot be expected to make meaningful, dignified lives and become full citizens without securing a substratum of material well-being. As such, the model proposed here is not silent on social justice but merely separates it from the theoretically distinct concerns of the rule of law and deliberation (though of course, if debates about distributive justice become salient in the public sphere, the distinction is minimized). Furthermore, a commitment to the rule of law requires a certain measure of accountability, for only if authoritarian enclaves are removed and citizens are bound by principles of open debate without the threat of violence, can the rule of law as normative principle be achieved.¹⁷ It is, thus, constitutively interrelated with other principles.

I have laid out the principle elements of my theory of reconciliation, which centers on moral respect with the corollary norms of truth, accountability, victim recognition, and the rule of law. One may understandably ask whether this conception of political reconciliation is overly Kantian and Western. Does the focus on respect allow it to speak to societies whose cultural values are more communitarian and less focused on the individual? What of those places where morality is understood in collective terms and the individual is not at the center of moral theorizing? These are important concerns, and I should stress that I am not arguing for a full-fledged theory of liberal reconciliation, premised on privileging the moral individual as conceptually prior to society. The concept of respect here emphasizes the intersubjectivity of moral identity formation, and while it admittedly has conceptual roots in philosophical language associated with Kantian individualism and rationality, it is not reducible to its origins. Indeed, the basic moral categories of dignity and self-worth emerge in victim narratives in myriad post-conflict settings across the world, even if they are not framed in this specific language (Beah 2008; Hatzfeld 2008; Lifton 1991; Szymusiak 1999). There are many moral vocabularies tied to particular religious, cultural, and philosophical traditions that speak to the importance of the individual's dignity, and Enlightenment thought is only one of these traditions. In any case, my discussion of respect is neither premised on monadic individualism, as should be evident from my emphasis on intersubjective moral recognition as one of its cornerstones, nor based on Kantian rationality devoid of emotional or expressive content, as I showed in discussing the importance of phenomenological truth and recognition. My account of reconciliation as respect, then, should not be taken as an endorsement of liberal individualism or rationality over social theories of ethical life. I believe that reconciliation as respect is sufficiently elastic to operate usefully in the types of cases under consideration here. More importantly, perhaps, is the simple observation that survivors in numerous cultural contexts—from Latin America to Southeast Asia to Africa to North America—often frame human rights violations not only as physical and emotional harms but also as moral wrongs. That is, they experience and retell these harms in moral language premised on expectations of moral respect, and their anger and resentment draw on the assumption that the perpetrators should have honored their claims to dignity and to moral personhood. It is this idea of respect as recognition of a person's inherent dignity that I am stressing here and that emerges across settings of mass violence.

Establishing whether such reconciliation has been adequately achieved is obviously quite difficult. Of course, sociological studies, anthropological ethnographies, and surveys can offer much-needed insights into overall patterns (Gibson 2004), but here I want to put forth a rough way to measure progress: When conflict-era forms of political identification are no longer the primary ways of determining political loyalty, some success has been achieved. Political violence and the rhetoric surrounding it depend on a strongly binary logic of identity. In-groups use language that constructs a tightly knit community while simultaneously disparaging and dehumanizing out-groups: This is political identity logic at its most theoretically elegant and empirically venomous. These distinctions can remain relatively stable, reinforcing systematic devaluation and social exclusion (Apter 1998; Levine and Campbell 1972; Staub 1989; Tajfel and Turner 1986; Zimbardo 2004). While it is impossible (and undesirable) to remove all distinctions and forms of differentiation from political life, since a constitutive element of politics is in fact differentiation, we can say that mutual respect and tolerance have been achieved to the extent that previous loyalties (i.e., from the period of violence) are mitigated through the development of alternate, overlapping political identities. Thus, the traditional in-group/out-group distinctions are no longer evident, or at least are no longer primary. This does not mean that political contestation is over, that economic stability is just over the horizon, or that democracy is finally and thankfully here. But it does mean that the most salient forms of political identity that characterized the era of violence can no longer mobilize the passion and viciousness they once did. This is a sign that former enemies are working together, even if only toward their new set of shared interests.

The conception of reconciliation as respect is intimately tied to the other normative concepts presented in this chapter. A culture that values tolerance and espouses deliberation rather than violence to resolve differences cannot emerge where the past is unexamined or where terrible crimes are justified. An honest and truthful understanding of the past is morally necessary, even though this can be painful for victims and unsettling for those who supported the perpetrators. Accountability, too, is necessary, for without some sign that impunity is inexcusable, victims continue to feel marginalized and citizens are not particularly moved to treat one another as equals. Impunity, and the language of superiority and contempt that often accompanies it, reflects a deep disdain for the rights of individuals. Closely tied to this is the importance of recognizing victims as moral equals and fellow citizens. Without meaningful efforts to recognize them, victims are likely to remain mistreated or ignored, receiving none of the respect and dignity they deserve. And the rule of law plays an important role in guaranteeing that personal rights will be respected, the state will remain bound by law, and political differences should be resolved peacefully, even if still contentiously. All of these normative concerns are important if reconciliation is to mean more than merely the temporary absence of violence where the powers of the past remain able to intimidate and coerce their opponents.

The key concepts of reconciliation are normatively dependent on one another, and thus the manifestation of one requires the development of the rest. Each normative concept is related to the others—they engage one another and presuppose each other, thus achievement of one is only partial if not accompanied by the others. The multivalent model is integrative, in the sense that it includes all of the concepts as crucial aspects for its sound actualization. They are necessary to promote a conception of reconciliation understood as reciprocal respect, which is the most we can expect from transitional societies. The following chapters are devoted to showing how the concepts interact across social space and how they may reinforce or undermine the project of reconciliation in different settings. We now turn to reconciliation at the level of political society.