War crimes proceedings-investigations, indictments, trials, exoneration of the innocent, and punishment of the guilty respond to the question of how to deal with atrocities committed during armed conflict by treating the acts committed as criminal violations of law and their perpetrators as felons under the law. At first glance, applying this criminal justice model to wartime atrocities seems a straightforward and obvious step. Yet historically, international law has not functioned in the same way as domestic law, and to carry over into the international sphere procedures developed for domestic criminal justice required first that international law be understood not only as consensually binding, but as institutional in a way it has not been until comparatively recently.

Nonetheless, building on the long history of prohibited behavior in the traditions underlying international law, the concept of war crimes is now part of positive international law, and the examples not only of Nuremberg and Tokyo but also of the former Yugoslavia and Rwanda trials argue strongly for the acceptance of this concept as part of customary international law, the actual behavior of states. In order to address the argument for war crimes proceedings as a proper response to atrocities in contemporary armed conflicts, let us look more closely at the idea of war crimes as it has taken shape in international laws?

The concept of war crimes in international law is defined on the basis of two principal sources. The first of these, the law of armed conflict, incorporates the Geneva and Hague conventions and other relevant international declarations and treaties. Among other provisions, this body of law prohibits direct, intentional harm to noncombatants and protected persons and rules out the use of certain weapons deemed to be disproportionate and/or indiscriminate in their effects. Behind the positive law is a stream of normative thought going back through theorists like Grotius to just war concepts of jus in bello to ancient sources, as well as a tradition of statecraft that sought to limit the destruction of value brought about by wars. Ranging forward from the positive law are additional protocols and refinements seeking to amplify the limitations imposed and, since the early t98os, to interpret what had formerly been termed the law of war as applying to all armed conflicts, those interior to a state as well as international.

Since 1945, under the aegis of the United Nations, further development of both the law of armed conflict and human rights law and their cross-fertilization have led to their being joined together in the concept of international humanitarian law. The idea of war crimes today thus includes both violations of the law of armed conflict and crimes against humanity. This is reflected in the definition of the jurisdiction of the international tribunals established for the conflicts in the former Yugoslavia and Rwanda, which the Security Council defined to be over "persons responsible for serious violations of international humanitarian law" (former Yugoslavia; Security Council Resolution 827 [1993)) and "genocide and other serious violations of international humanitarian law" (Rwanda; Security Council Resolution 955 [19941).

The concept of war crimes in international law aims both to prevent atrocities during armed conflict and to provide for punishment of persons guilty of crimes of war when they have already taken place. These aims have different force at each of three stages relative to an armed conflict.

1. Before an armed conflict breaks out, the concept of war crimes carries the assumption that the values it seeks to protect will be recognized by potential parties to armed conflict, so that these parties will plan potential military actions accordingly and train the members of their fighting units to act so as to observe the laws of war and to respect human rights. Criminal behavior attaching to this stage, it follows, would be a decision to plan the armed struggle otherwise (in the case of political and/or military leaders) or participation in that decision (in the case of persons at lower levels, on down to individual members of military units), with the emphasis being on the responsibility of persons exercising higher levels of leadership.

3. After the conclusion of an armed conflict the concept of war crimes assumes a continuing effort, at both the domestic and the international levels, to identify persons suspected of war crimes and bring them to justice. War crimes trials set up at this stage, as in the cases of the Nuremberg and Tokyo trials and the tribunal for the Rwandan conflict (whose charge is limited to violations that occurred in 1994), aim at identifying, trying, and punishing persons guilty of violations of international humanitarian law that took place during the course of the conflict in question. An international tribunal is not the only means to the end, since individual countries have the right to arrest, prosecute, and punish persons for violations of human rights up to and including genocide; Israel's trials of Nazi war criminals provide a prominent example of this. Yet, addressing crimes of war by means of an international tribunal serves the important end of underscoring the conception of such crimes as violations of rules put in place by the international community as a whole, violations that strike at the very framework of international order.

Application of the war crimes model to contemporary armed conflicts such as we have been discussing raises certain questions. First, there is the question as to whether the precedent established by the Nuremberg and Tokyo trials is appropriate for dealing with atrocities in such conflicts, which are typically not international wars but civil conflicts in which the adversaries are divided by major cultural differences. While this is a serious issue, the question is now moot; its answer has already been provided by the creation of international tribunals for the conflicts in former Yugoslavia and Rwanda. Themselves forming a precedent, these tribunals make plain that violations of international humanitarian law are crimes of war not only in international conflicts but in conflicts between elements of the same society, and that the values embedded in such law are to be honored by parties to armed conflict whatever their particular cultural frame.

The answer to a second question remains very much open: what are the implications of war crimes proceedings for the process of bringing peace to the parties in conflict? There may in fact be no simple answer, since this process is a complex one, running from the achievement of a cease-fire through reconciliation of former enemies and the rebuilding of civil society. At the beginning of this process, where the concrete problem is how to bring the leaders of the opposing parties to a cease-fire agreement and how to maintain the cease-fire once established, a serious question is whether, as much policy argument would have it, a robust effort to investigate war crimes, identify persons responsible, and bring them before a court of justice is likely to interfere with achieving a cessation to the fighting. The reason, as already observed, is that some of the very persons whose agreement is necessary for a cease-fire may have been involved in the activities that war crimes proceedings would investigate. For such persons, avoiding the initiation of formal war crimes proceedings is a priority, and their position of authority makes it possible for them to seek to bargain away such proceedings in return for a cease-fire agreement. Correspondingly, they may reason that their interest is better served by continuing the fighting, if war crimes proceedings are set in motion. The result is that war crimes proceedings and negotiated settlements to conflicts are set against each other as mutually exclusive alternatives.

The problem is complicated by the fact that in the international arena support for each of these approaches to dealing with conflict and involvement in their implementation come from two significantly different constituencies: for war crimes proceedings, international lawyers and criminal justice specialists and institutions involved in monitoring and enforcing compliance with international humanitarian law; for negotiated settlements focused on ending the fighting, members of the diplomatic community, conflict resolution professionals, and institutions engaged in providing emergency relief and other services. Moral opinion splits between the two.

The rationale for a no-fault position reflects several concerns beyond that of how to end ongoing armed conflict as quickly as possible. One concern reflects the lack of congruence between the context of the war crimes trials after World War Il and that of contemporary armed conflicts in divided societies. The former followed an international war in which the Allied powers totally defeated the Axis and imposed terms of unconditional surrender, whereas negotiated settlements have become the preferred solution to internecine conflicts. The former thus reflected the character of war as a whole as a final arbiter of dispute; the war crimes trials effectively imposed the values of the victorious powers on the defeated powers in the context of their having submitted unconditionally. But in contemporary conflicts within divided societies, negotiated settlements produce a different result: the value systems of the parties to the conflict remain intact, and the adversaries agree in principle to some form of compromise in which both remain valid. War crimes proceedings in such cases, the argument goes, can only upset the balance and make the necessary compromise impossible to achieve.

A second concern is that if war crimes investigations lead to indictment and trial of civilian and/or military leaders in either or both parties to a conflict, no leadership may be left to carry out the terms of a negotiated settlement. Another possible outcome of such indictments is renewal of the armed conflict, insofar as such persons remain powerful in their own communities and may decide to make use of the armed force of those communities to protect themselves.

On this view, the alternative to a negotiated end to an armed conflict is to allow the fighting to continue either until one side wins, which may or may not serve the cause of justice, or until both sides are utterly exhausted, in which case the resulting stalemate will perpetuate the fighting, though at a lower level. The former possibility raises a problem of its own for war crimes trials to address atrocities during the armed conflict: that those committed by the winning side are, in practical terms, are likely never to be addressed.

As one whose professional career has focused on the development of restraints on the initiation of war and the conduct of war, and one who, moreover, holds that the deepest moral justification for use of armed force is the service of justice, I clearly belong among the constituency supporting war crimes proceedings. From my perspective, shaped by thinking of morality and war in just war terms, egregious violations of the rights of people caught up in a conflict constitute an injustice that it is immoral not to seek to remedy. Indeed, as I have argued above, such violations justify the interventionary use of military force to stop them and to seek to remedy them; war crimes proceedings serve the same ends. In particular cases, the best way to respond to a pattern of war by atrocity may be to use both measures: I have supported both, for example, in the case of the Bosnian conflict. This is not to argue for a knee-jerk response by these means to every instance of atrocities during a conflict; there must be a pattern of atrocious conduct, and even then, there may be better ways of dealing with it. Within the framework of just war reasoning, the test of last resort needs to be passed before resort to force is finally warranted in moral terms, and it may also be well to think of the institution of war crimes proceedings in this way. Legal sanctions and punishments are tools that serve higher ends; they are not ends in themselves. But in my judgment, demonstrated patterns of atrocious conduct by one or both sides in an armed conflict, as in many contemporary local wars, unambiguously satisfy this concern for caution.

Before the institution of war crimes proceedings, both moral thought within just war tradition and positive international law depended on whether retaliation is justified as a means of responding to atrocities and deterring future ones. The means of retaliation, while they should be robust enough to achieve their ends, were still, preferably, to respect discrimination; that is, they should be attacks against military targets, though noncombatants might be harmed collaterally. The threat of retaliation may be enough to prevent atrocities in the first place; successful retaliation during an armed conflict may prevent a pattern of warfare by atrocity from developing. The danger is that if retaliation fails to prevent recurrence of enemy atrocities, then a pattern of tit for tat may develop, and the conflict may escalate, perhaps with the side originally victimized turning to atrocious means itself. This is the dilemma posed on a much larger scale by nuclear deterrence, a strategy that depends on the threat of retaliation; if deterrence fails, then the resulting nuclear war will bring devastation to both sides.

Turning the matter around, a conscious decision not to pursue war crimes proceedings under international auspices when there is a pattern of egregious misconduct during an armed conflict undercuts the restraints on warfare set out in the law of armed conflict and human rights law, implying that future warfare by atrocity will also go unpunished. This, to my mind, is the most fundamental problem with the no-fault approach to conflict resolution: it weakens the restraints that help to keep wars from being fought according to the rule that anything goes, and it undermines the protection of human rights by bracketing out of consideration the idea of crimes against humanity. It would be a tragedy for all humankind if the limits on war and the protections against abuses of human rights now in effect in international law were to be diluted in this way.

Finally, a further argument in support of the establishment of international war crimes proceedings in the case of patterns of atrocity during an armed civil conflict is that doing so assists in the setting up of the rule of law, an essential step in the reconstitution of a civil society torn by conflict. Thus war crimes proceedings form part of an overall synthesis of responses to the problem of how to deal with armed conflicts in divided societies, a synthesis that looks beyond the problem of achieving a cease-fire to the long-term, but more fundamental, problem of how the society is to be reconstituted so as to be at peace with itself. Without justice, there can be no peace; in this, the medieval political philosophy based on Augustine remains a source of profound insight into human community.

Morally, then, the case for war crimes proceedings against persons responsible for atrocities during a conflict outweighs the argument for ending the armed stage of the conflict on a no-fault basis. Indeed, the moral case for war crimes proceedings rejects the idea that these are somehow opposed to the achievement of a workable cease-fire; rather, the effort to bring justice vis4-vis atrocities committed both helps to protect potential victims in the conflict so long as fighting lasts and serves to facilitate a more genuine peace when the fighting ends.