I. Introduction

In the 1980s and 1990s there has been an unprecedented degree of international attention to the application of the laws of war to contemporary conflicts. This body of humanitarian rules has been a major consideration in much international diplomacy, and has had considerable impact in some wars and military occupations. However, as in wars earlier in this century, implementation of the laws of war has been uneven. Basic norms have been violated in wars both international and internal. Horrific events in the former Yugoslavia since 1991, and in Rwanda in 1994, have impelled the United Nations Security Council to establish international tribunals in the hope of thereby restoring the effectiveness of rules after they have been flouted. The many efforts by the UN and other bodies to act against violations have involved a daunting array of problems, many of which had been only dimly foreseen.

Questions about implementation addressed in this study include: What are the formal provisions and mechanisms by which the laws of war are supposed to be implemented, and why have they been relatively little used? In practice, what other mechanisms of implementation have been used? How effective has been the increased involvement of the United Nations Security Council since the early 1980s in addressing violations of the laws of war? What have been the particular problems of implementation in some of the major armed conflicts around the world in the 1980s and 1990s? What has been the role of the International Court of Justice as regards implementation? Why have diplomatic efforts calling for observance of existing rules so often been unsuccessful? Do belligerent reprisals have a continuing place in ensuring observance of rules? Are other uses of force, not specifically designated as reprisals, important as a means of countering violations? What is the place of individual criminal prosecutions, and of demands for reparations from states responsible for major violations? How can the international community respond to the demands for amnesty which inevitably accompany peace negotiations? What conclusions can be drawn about how the question of implementation could be pursued?

States, by becoming parties to the 1949 Geneva Conventions, have undertaken "to respect and to ensure respect " for the main conventions on the laws of war. What exactly does this obligation imply, and how is it to be interpreted in today's circumstances? The difficulty that members of the international community face in attempting to ensure that rules are implemented, and to restore their effectiveness after they have been violated, should not be underestimated. The international system differs from domestic politics precisely in the fact that there is no strong central authority capable of enforcing the full range of rules that states and nonstate bodies are obliged to follow. On the international level, despite the increased role of international organizations including the UN, authority is still decentralized. It follows that international structures, organizations and rules have to be based in large measure on consensus, and rules have to be implemented largely through the states that are the main members of international society. When states (or, as in some recent cases, forces trying to establish a state) do not observe the rules, it is hard to make them change course. This problem is particularly severe so far as implementation of the laws of war is concerned.

A complicating factor in the application of the laws of war is that the majority of wars in the post-1945 world have failed to fall neatly into the category of "international armed conflict" - the only category of war to which the main body of the laws of war is formally and indisputably applicable. Most conflicts have been civil wars, or at least have contained a major element of such war. The application of the laws of war to civil wars raises both a legal and a practical problem. The legal problem is that governments have usually been reluctant to create or sign up to a body of law which would bind their freedom of action so far as dealing with armed rebellion is concerned: thus the treaty-based rules formally applicable in such conflicts have been inadequate, though there is now a tendency at the UN and elsewhere to view a wide range of humanitarian rules as applicable, as the establishment in 1994 of the International Tribunal for Rwanda confirmed. The practical problem is that civil wars are notoriously bitter, for several reasons: each side is likely to deny the legitimacy of the other, training in the laws of war may be limited, the neat distinction between soldier and civilian frequently breaks down and the scope for a compromise settlement of the war is usually slight. Trying to secure even a minimal level of observance of rules is peculiarly difficult in such circumstances.

When a civil war is internationalized, in the sense of involving foreign troops on one or both sides, there is a much stronger argument that the whole body of the laws of war is formally in force, especially as concerns the conduct of the outside forces involved. Even then, however, application may involve practical difficulties arising from the special characteristics of civil wars.

A related problem is how to ensure that the rules themselves are sufficiently realistic that they are capable of being applied by belligerents in the peculiarly difficult circumstances of war, that they reflect a real consensus, and that there is a serious intention to observe them. The progressive development and increasing complexity of the rules, especially in the past two decades, may have gone so far that the laws of war have lost meaningful contact with the thinking and actions of at least some states, armed forces, and non-state entities. Further, the manner in which the law is discussed and advocated may have caused problems. Rules which are not properly incorporated in the training and mentality of fighting forces obviously will not work.

Implementation can assume a variety of forms, of which war crimes trials are only one. The term "implementation" is used here to refer to the many ways in which states generally (including belligerents in an armed conflict) apply, and sometimes fail to apply, the international rules applicable in armed conflict. They can be grouped under three headings:

1. In peacetime: training, education and planning, including within armed forces.

2. In time of occupations and armed conflicts: the codes of conduct, rules, commands and actions of the governments and forces involved; and what one belligerent does, during or after a conflict, in response to another's alleged violations of the rules.

3. The actions of third-party states (i.e. those not directly involved in a conflict), non-governmental bodies, and international organizations, aimed at trying to secure compliance by belligerents, including when belligerents have committed major violations of the rules.

In this study I have generally used the term "laws of war". This refers to those streams of international law (especially the various Hague and Geneva Conventions) specifically intended to apply in armed conflicts. In some eyes the term "laws of war" is old-fashioned. However, it has merits. It accurately reflects the well-established Latin term for the subject of this enquiry, ius in bello; and it is brief and easily understood. It has two modern equivalents, both longer. The term "law applicable in armed conflicts" is unexceptionable, but adds little. "International humanitarian law" (IHL), often with the suffix "applicable in armed conflicts", has become the accepted term in most diplomatic and United Nations frameworks. However, it has the defect that it seems to suggest that humanitarianism (rather than interest or professional standards) is the main foundation on which the law is built, and thus invites a degree of criticism from academics, warriors and others who subscribe to a "realist" view of international relations. However "IHL" also has the merit that it is widely seen as encompassing relevant parts of the international law of human rights. Preference for the term " laws of war " does not imply a down-playing of the significance of human rights provisions, whose merging with the laws of war is one of the most significant developments in this field in recent decades.

In the conflicts of the 1980s and 1990s, the actual practice of belligerents has often fallen far short of what might have been expected granted the important developments in treaty law, the high level of formal adherence to treaties, and the unprecedented involvement of the United Nations (especially the Security Council) in matters relating to implementation of the laws of war. Gross violations of well established rules have occurred in the Iran-Iraq War, in the Iraqi occupation of Kuwait in 1990-91, in the former Yugoslavia since 1991, and in many other conflicts both international and internal. In most cases these violations attracted much diplomatic attention, but no effective response. Those engaging in illegal and inhumane practices could and did exhibit contempt not merely for the laws of war but also for those who sought so ineffectually to uphold them.

The problem of implementation has increasingly come to be seen as the central problem of the laws of war. Sir Frank Berman, Legal Adviser to the UK Foreign and Commonwealth Office, has said:

"It seems to many that the problem is not to discover what the law is, or how to apply it to the particular case, or even whether the existing rule is 'satisfactory' or not, but rather how to secure or compel compliance with the law at all".

A natural response to a pattern of violations of the laws of war is to call for new systems of implementation and enforcement. However, since at least 1949, treaties on the laws of war have included a range of formal provisions on implementation generally, including penal sanctions of various kinds; and there have been numerous efforts, including within a UN framework and also by governmental and non-governmental bodies, to reinforce these mechanisms of enforcement and to develop new ones. Why such provisions and efforts have had relatively little impact is the underlying question which needs to be explored.

If a high level of formal adherence by states to the major treaties were a guarantee of enforcement, there should be little problem. At mid 1995 the four 1949 Geneva Conventions had 185 parties - the same overall total as that of UN members, though the two lists were not quite identical. The two 1977 Additional Protocols had gained a respectable number of parties: Protocol I had 138 states parties, and Protocol II had 128.

Unfortunately, the question of how the laws of war are, or are not, implemented has not been the subject of a vigorous tradition of thought. Lawyers tend to think in terms of enforcement through legal processes after a violation, when implementation may take many other forms. Indeed, its most important aspect is implementation through education and training in well-organized armed forces. Much writing on implementation, including that published in law journals, has been narrowly legal or prescriptive in character. There has too often been a formalistic assumption that the main modes of implementation are, or ought to be, those laid down in the conventions. The problems faced by soldiers and decision-makers in armed conflicts have not been explored in depth. It has been assumed - understandably, but perhaps too easily that the main form of analysis involved is judging the behaviour of belligerents by a legal yardstick, when there is also a case for judging the laws of war by the harsh test of how they operate, or fail to operate, in the circumstances for which they were designed.

Analysis of the question of implementation can benefit from a more descriptive approach, looking systematically at the many difficulties, and opportunities, that have been encountered in applying the laws of war. Such an approach employs the methodologies not only of law but also of history, politics, international relations, and strategic studies. The major single-author work along such lines, Geoffrey Best's examination of whether international humanitarian law has worked well or not since the Second World War, reaches pessimistic conclusions. He draws a picture of a body of law with an impressive and admirable superstructure built on insecure foundations, of which perhaps the shakiest is the central, critical distinction between the soldier and the civilian. The law's impact has been much less than had been hoped. Sometimes, indeed, it has been little more than an instrument of propaganda warfare.

This study is based on six underlying propositions:

1. Difficulties in securing compliance are not unique to the laws of war, but arise also in many aspects both of international law more generally, and of the domestic law of states.

2. Implementation of rules of conduct in war is usually best achieved when parties to a conflict have a political and military culture, and a perception of their own interests, which is broadly favourable to observance. Implementation is therefore largely a matter of proper preparation in peacetime. States, their armed forces, their governments, and their legislatures, are among the most important entities for implementing the laws of war. Securing compliance by states (or non-state entities) after there have been violations of the rules is just one small part of the much broader process of implementation.

3. Punishing transgression of norms is particularly complex in cases in which offences are committed in what is perceived as a public cause, in which large numbers of people are implicated in different ways in the commission of the offences, and in which the state or non-state entity in whose name the offences were committed continues to exist and to protect its own citizens.

4. States which are neutral in a particular conflict, or which do not wish to extend or deepen a limited belligerent role, may be particularly reluctant to take measures to punish war crimes by a belligerent.

5. The actual processes by which compliance with law is induced are by no means the same thing as the formal treaty provisions to that end. Institutions whose role is outlined in the conventions, such as Protecting Powers and the International Fact-finding Commission, have hardly had any role in implementation. In practice, when there are legal procedures, they often assume a different character from that envisaged in the conventions: for example, commissions of inquiry rather than trials. In some extreme cases, it may be only possible to induce compliance with the law by making credible threats of reprisals against an adversary, or by major states showing a willingness to intervene militarily to uphold international standards - courses which manifestly involve numerous risks.

6. The public demand in many countries for more effective implementation of international norms regarding warfare will not go away.

(introduction...)

This section looks at the formal provisions regarding implementation as set out in the conventions from 1899 to 1981, and glances briefly at aspects of the wider range of pressures and mechanisms which may in fact be involved in processes of implementation. When in an actual armed conflict the laws of war do play a part in shaping the decisions made by belligerents, they may do so for complex reasons. Action in consonance with the laws of war may owe much to a wide range of political, military, diplomatic and ethical factors: these may include a fear of military reprisals, and an anxiety to project a reasonable image with the domestic public and with actual or potential allies.

The 1992 German tri-service military manual lists thirteen factors, mainly treaty-based, that "can induce the parties to a conflict to counteract disobedience of the law applicable in armed conflicts and thus to enforce observance of international humanitarian law ":

· consideration for public opinion;
· reciprocal interests of the parties to the conflict;
· maintenance of discipline;
· fear of reprisals;
· penal and disciplinary measures;
· fear of payment of compensation;
· activities of protecting powers;
· international fact-finding;
· the activities of the International Committee of the Red Cross (ICRC);
· diplomatic activities;
· national implementing measures;
· dissemination of humanitarian law; and
· the personal conviction and responsibility of the individual.

As has been recognized in many treaties and manuals on the subject, the laws of war are implemented largely through the medium of individual countries. It is usually through their government decisions, laws, courts and courts-martial, commissions of inquiry, military manuals, rules of engagement, and training and educational systems, that the provisions of international law have a bearing on the conduct of armed forces and individuals. The overwhelming majority of legal cases in connection with the laws of war have been in national, not international, courts.

Even where the problem is one of international enforcement following a violation - to get a foreign state or armed force to comply with the rules - the actions of individual governments have often been important. For example, neutral states may influence the conduct of belligerents, through private or public diplomatic pressure, economic inducements, embargoes, and even threats of military action. On the other hand, they are sometimes hesitant to do so, and when they do act their intervention is often rebuffed by belligerents.

One means of enforcing the law is reprisals. A reprisal may be defined as a retaliatory measure, normally contrary to international law, taken by one party to a conflict with the specific purpose of making an opponent desist from particular actions violating international law. It may be intended, for example, to make the adversary abandon an unlawful practice of warfare. The use of reprisals is controversial. They can on occasion be little more than a fig-leaf thinly disguising the resort to unrestrained warfare; and certain types of reprisal are now prohibited in 1977 Geneva Protocol I. At ratification, a number of states made declarations which, in interpreting some of the Protocol's provisions, appeared to keep open the possibility of reprisals. Italy's long statement of interpretation included the following: "Italy will react to serious and systematic violations by an enemy of the obligations imposed by Additional Protocol I and in particular its Articles 51 and 52 with all means admissible under international law in order to prevent any further violation". On occasion the threat or actuality of reprisals can be an important means of inducing restraint.

One other powerful instrument of enforcement is neglected in most discussions of the subject. Sometimes illegal conduct by a belligerent, including the commission of atrocities, may contribute to the formation of an international military coalition against the offending state; and may influence the coalition's willingness to use force. Such conduct has been a significant element in the building of many coalitions, including the anti-Axis alliance in the Second World War, the international coalition against Iraq in 1990-91, the intervention in Somalia in December 1992, and the decision by NATO and the UN to initiate "Operation Deliberate Force" in Bosnia-Herzegovina on 30 August 1995. Even the possibility of such a process is almost entirely neglected in the legal literature. It constitutes a little-recognized but important link between ius in bello (the law applicable in armed conflicts) and ius ad bellum (the law governing resort to armed conflict). It is discussed further at several points in this study.

The treaty provisions regarding implementation, discussed in subsequent parts of this section, are of many kinds. They include stipulations about the trial and punishment of individual offenders, and about reparations by states. They touch on reciprocity as a basis for observing the laws of war, and on the controversial issue of threats of reprisals as a means of enforcing the law. They contain a variety of arrangements, including monitoring, negotiating, and fact-finding, in order to secure implementation of the conventions. Despite this wealth of provisions, concentration on the treaty arrangements for securing compliance can easily mislead. Many of the formal written provisions have proved less important in practice than was hoped. Arrangements and forms of pressure that were not envisaged in the laws of war treaties have sometimes had more impact.

1. From 1899 to the Second World War

The 1899 and 1907 Hague Conventions on Land War, and the Regulations annexed to them, are vague on the matter of ensuring compliance. Article 1 of the 1899 and 1907 Hague Conventions requires the powers to issue instructions to their land forces in conformity with the Regulations. Article 3 of the 1907 Convention says that a belligerent party violating the Regulations "shall, if the case demands, be liable to pay compensation". In addition, Article 56 of the 1899 and 1907 Hague Regulations makes a vague reference to legal proceedings in the event of violation of its rules about certain types of public property. Nothing more is said about how these or other provisions are to be enforced. The many striking omissions regarding enforcement exposed the Hague system to the accusation that it was based on unduly optimistic assumptions.

However, the relative paucity of formal provisions in the Hague Conventions and Regulations did not mean that there was no implementation system at all. The central assumption, of which the above-summarized provisions are a mere reflection, was of a responsibility on states to ensure that the rules were observed and offenders brought to justice. This assumption has many weaknesses, of which the most obvious, easy to identify but hard to remedy, is that most governments have been, quite understandably, reluctant to prosecute their own servants in cases where their violations of the laws of war were carried out while pursuing government policy. It is this problem above all which has sustained an unbroken series of calls for some diminution of national sovereignty so far as the punishment of war crimes and crimes against humanity is concerned. In twentieth century practice, the Hague Regulations have provided the basis for numerous trials, appeals, arbitrations, and inquiries - not only by national, but also by international, bodies. Their modes of implementation have been richer than those specified in the original texts.

After the First World War a number of international agreements included reference to the punishment of war crimes. The 1919 Treaty of Versailles, Articles 227 to 230, required Germany to surrender for trial members of its armed forces charged with violations of the laws of war. However, by subsequent arrangement with Allied governments, Germany itself tried German offenders, very few of whom were convicted The Versailles Treaty also imposed heavy reparations on Germany on account of its perceived responsibility for the outbreak and therefore the costs of the war - a matter of ius ad bellum rather that ius in bello. The 1929 Geneva Convention on Wounded and Sick, Article 29, provided for punishment of violations on the basis of national penal legislation. The 1929 Geneva Convention on Prisoners of War, Article 86, provided for Protecting Powers to monitor observance of the Convention. both these 1929 Conventions are now defunct, having been superseded by the 1949 Conventions.)

By contrast, a major laws of war agreement of the inter-war years, the 1925 Geneva Protocol of Gas and Bacteriological Warfare, said nothing about implementation other than in the noble but vague formulation "that this prohibition shall be universally accepted as a part of International Law, binding alike the conscience and the practice of nations" On ratification, France and many other states remedied the gap by making it clear that they would cease to be bound by the Protocol if their enemies used the prohibited weapons. Here was a clear threat of reprisals. Fear of reprisals almost certainly played some part in subsequent cases of non-use of such weapons, including during the Second World War and the 1991 Gulf War.

2. The Post-Second World War Trials

There were very many war crimes trials at the end of the Second World War, mainly in national courts of the victorious powers and of the countries they had liberated. The best-known were the international military tribunals at Nuremberg and Tokyo which tried major Axis war criminals. They constitute the major precedent for implementation through international trials. Today, when new international tribunals in respect of war crimes have been established, the post-1945 international tribunals merit re-examination.

The bare facts of the two international tribunals are easily summarized. They originated in Allied declarations in London (13 January 1942) and Moscow (1 November 1943) which declared the prosecution of war crimes to be one of the Allies' principal war aims. On the basis of these statements much work was done to prepare for the numerous national trials. This left the special problem of what to do about major figures whose crimes were not tied to any particular location. On 8 August 1945 the London Agreement for the Prosecution and Punishment of the Major War Criminals of the European Axis was concluded. Article 6 specified three types of crimes:

· 6(a) crimes against peace (relating to ius ad bellum);
· 6(b) war crimes (violations of the laws or customs of war); and
· 6(c) crimes against humanity (discussed further below).

On this basis, the International Military Tribunal at Nuremberg sat from 20 November 1945 to 1 October 1946. Of the twenty-two defendants, all but three were found guilty on some charges. Twelve were sentenced to death. Seven received prison sentences ranging from ten years to life. The International Military Tribunal for the Far East, based on the same principles as at Nuremberg, was held in Tokyo from 3 May 1946 to 12 November 1948. There were twenty-eight defendants, of whom two died during the trial, and a third was found mentally incompetent. All twenty-five remaining were found guilty on some charges. Seven were sentenced to death. Sixteen were given life imprisonment, one twenty years, and one seven years. The Tokyo Tribunal has been even more heavily criticized than its counterpart at Nuremberg, not least because it placed comparatively less emphasis on the well-established category of war crimes against protected persons such as prisoners and inhabitants of occupied territories.

The significance of the Nuremberg and Tokyo tribunals is notoriously a subject of controversy. Some aspects of the controversy are not relevant to the present inquiry, such as certain procedural deficiencies and the question of whether the charge of waging aggressive war was well founded in existing international law, and fairly applied, especially at Tokyo. Three other grounds of criticism are closely related to present-day problems of applying the law. These are: (1) that the tribunals applied a body of law some aspects of which had not before 1945 been clearly enunciated in treaty form, or were in treaties which were not fully applicable to the events under scrutiny; (2) that the tribunals were one-sided, as possible war crimes committed by the Allies were neither fully considered there nor dealt with elsewhere; and (3) that large numbers of guilty individuals were either not prosecuted at all, or were treated too leniently. The first and second of these grounds feed into the commonly-uttered view that these trials constituted victor's justice.

The point about whether new law was being developed deserves attention today because it sheds light on the issue of whether there are some standards of behaviour so basic that they must apply to all states and their citizens, in peacetime and in wartime, irrespective of whether or not those standards are clearly spelt out in treaties to which the state in question has subscribed. Several elements are involved.

The first element is the notion of "crimes against humanity". To some extent these were simply war crimes writ large, but in addition they could encompass actions before as well as during the war, and could apply to a government's offences against its own citizens. However, the wording of Article 6(c) of the 1945 Charter of the Nuremberg Tribunal suggested that they had to be committed "in execution of or in connection with any crime within the jurisdiction of the Tribunal" - a curious proviso which blunted the impact of this innovative category. Like the simultaneous development of human rights law in the Charter of the United Nations, the introduction of the idea of "crimes against humanity" can be seen as creative law-making. It did have some basis, albeit shadowy, in existing law. For example, the 1899 and 1907 Hague Conventions had stated in the Martens Clause that in cases not included in the Regulations, "the laws of humanity" remain a source of law of key importance. Whatever the validity of the basic concept of "crimes against humanity", it had even less concreteness and independent existence in the Nuremberg judgment than it had had in the London Charter. Its role was yet further attenuated in the Tokyo trial. Subsequent developments, however, including in the Barbie case, and in the statutes of the international tribunals for the former Yugoslavia and Rwanda, suggest that this legal category is growing in importance and is reinforcing the idea that a wide range of crimes and factual situations is subject to international legal rules.

A second aspect of new law-making arose from the way in which the judgment of the International Military Tribunal at Nuremberg asserted the universal applicability of a key part of existing law, the 1907 Hague Convention IV on land war. The Hague Convention's Article 2, its "general participation clause", had stated that its provisions were applicable only if all the belligerents in a conflict were parties to the Convention. The defence at Nuremberg contended that because some belligerents, Albania being one example, were not parties to the Hague conventions, therefore Germany had not been bound by these conventions, at any rate vis-is non-parties. The Tribunal's judgment said that during the Second World War this treaty applied to all countries, because it had by then become customary law. With this statement, the judges from the victor powers were saying that irrespective of the adherence or otherwise of states to particular accords, irrespective of the legal niceties and the small print, there had to be some minimal universal standards.

All this was part of that aspect of Nuremberg which significantly tempered the central role of governments by injecting the idea of personal responsibility in respect of universal standards. The doctrine that superior orders are not a defence against a charge of war criminality was another manifestation of this approach. In short, the old idea of a society of states was having to yield, however slowly, to a society where governments were subject to certain over-arching principles, and could not order around their citizens, even their officers, just as they wished. Modern ideas of legal positivism, based on treaty law alone, were subtly yielding to older ideas of natural law.

As to the accusation that the Nuremberg and Tokyo trials were onesided, it is easy - and it was well done by the defence lawyers at the trials- to point out that there had been terrible deeds on both sides in the war. On some matters, such as submarine warfare and city-bombing, the Allies, just as much as the Axis powers, had ignored existing treaties and legal principles. The Soviet record regarding treatment of prisoners had been appalling. However, there was one major category of activity on which the law was clear, and in respect of which there was little comparison between Axis powers and at least most of the Allies. This category was in the end a main basis for the conviction of most of the major war criminals. It concerned, not combat itself, but treatment of those more or less hors de combat The Axis atrocities against many of those who were directly under their control - whether Jews, prisoners of war or inhabitants of occupied territories - formed the strongest ground for conviction. It cannot have been wrong to punish these acts. They were clear violations of the most elementary principles of decency, as well as being contrary to the 1907 Hague Land War Convention and the 1929 Geneva Prisoners of War Convention.

The accusation that the many trials at the end of the Second World War did not go far enough is serious. The administration of justice was, as perhaps it had to be, extremely selective. Many German and other individuals who had been accomplices in mass murder escaped the Allied net, or else were considered too useful to the Allies to be prosecuted The attempt at Nuremberg to establish "organizational guilt" through membership of criminal organizations such as the SS (Schutzstaffeln) was not successful. As time wore on, the Allies rapidly lost enthusiasm for criminal prosecutions, and thousands of cases were not pursued. The "denazification programme" was a preferred if still flawed substitute In the end, an implicit principle of Nuremberg and Tokyo was to hold highly-publicized trials of a few leaders primarily responsible for a process of criminality in which hundreds of thousands had in fact been culpable in one way or another.

While these and other criticisms of the Nuremberg and Tokyo trials are well founded, they do not for the most part suggest that the trials were not worthwhile. The trials responded to wholly exceptional circumstances. They greatly strengthened the idea that the conduct of states and armies was subject to some over-arching legal standards however imperfectly enforced in this instance.

The post-Second World War trials may have been "victor's justice", but in retrospect that is not only their greatest weakness but also their greatest strength. The fact that they were victors enabled them to gather the necessary evidence, arrest a high proportion of the main suspects, and hold trials. There were few worries about whether the trial process might actually worsen the conflict with the Axis powers, since they were already defeated. The costs of the process were underwritten The contrast with the International Criminal Tribunal for the Former Yugoslavia, which has experienced difficulties in such matters as getting hold of suspects and securing adequate resources, is striking.

A questionable part of the legacy of Nuremberg is the creation of expectations that, in general, trials are an appropriate way to handle war crimes issues. As Telford Taylor has put it:

"Also, in terms of enforcement, whether the charge is war crimes or crimes against humanity, I think it is a mistake to expect that the device of a criminal trial is the major way in which the enforcement of those limitations and obligations is going to be achieved. As one vho has taught criminal law for several years, I always try to instill in my students a basic appreciation that most law enforcement is voluntary. Therefore, in the international field as well, the idea that trials alone (or statutes and treaties) can bring about the reforms and remedies that we hope for is misplaced reliance .

Since 1945, many countries have conducted national trials for war crimes committed in the Second World War. Often there has been special legal provision to permit such prosecutions long after the offences. Thus the UK government, while failing to take action regarding Iraqi war crimes in the 1990-91 conflict over Kuwait, and acting over the objections of the House of Lords, passed the War Crimes Act 1991, enabling UK courts to try offences arising from the Second World War. The UK government appeared to be pressing ahead towards possible trials of octogenarians who had held middle- or low-ranking positions in the Second World War. An argument for the UK government's approach was that these individuals were under UK jurisdiction, and hence it was actually possible to take action. To date, extensive investigations of possible suspects have only led to one indictment.

3. The Post-1945 Conventions: General

After two world wars the need for some penal and other mechanisms for securing compliance with the laws of war was self-evident. The major treaties in the field since 1945, particularly the four 1949 Geneva Conventions, contain an unprecedented range of provisions about dissemination, instruction of armed forces, humanitarian and monitoring tasks during armed conflicts, and repressing breaches These provisions, discussed below, did not follow a single standard form, and contained many innovations.

A general trend in the post-1945 conventions has been to try to get beyond the previous unsatisfactory situation in which it was assumed that implementation was fundamentally an internal matter for states. The Nuremberg and Tokyo tribunals had self-evidently taken a bold step beyond the idea that states could be relied on to punish their own nationals. In several ways, itemized in the following sections, the Geneva Conventions and other laws of war agreements advanced the concept of international involvement in implementation - especially through proposals regarding fact-finding and universal jurisdiction for war crimes. Three agreements since 1945 have specifically provided for a United Nations role in securing implementation of their terms and in dealing with violations.

Common Article 1 of the four 1949 Geneva Conventions is often, and increasingly, viewed as providing a basis for states to involve themselves in ensuring implementation of the conventions not merely within their own territories and areas of operations, but also more generally. It says: "The High Contracting Parties undertake to respect and ensure respect for the present Convention in all circumstances". The words "and ensure respect", which had not appeared in previous conventions, have been widely seen as signifying that states, whether or not involved in a particular conflict, have a responsibility to help ensure implementation of the conventions wherever and whenever they are being violated. Such an interpretation of common Article 1 was already evident in Jean commentary, in which he said, in expansive terms going beyond the narrow confines of legal analysis: "It follows [..] that in the event of a Power failing to fulfil its obligations, the other Contracting Parties (neutral, allied or enemy) may, and should, endeavour to bring it back to an attitude of respect for the Convention". This view has prevailed in much subsequent analysis and advocacy, including at the ICRC.

This interpretation of Article 1 does not appear to have a basis in the negotiating history of the 1949 Geneva Conventions. In the various meetings at Stockholm and Geneva, the words "to ensure respect" had another meaning: to ensure that the whole population of a country which was party to the conventions would respect the law in all circumstances, even perhaps in the case of civil war. There appears to be little or nothing in the records of the 1949 Diplomatic Conference to suggest an awareness on the part of government delegates, or indeed ICRC participants, that the phrase "to ensure respect" implied anything beyond this.

Whatever the original intention behind it, the interpretation of common Article 1 as implying a duty to promote implementation generally has helped to bring the question of implementation of the laws of war more centrally into the discourse of states and the activities of international organizations. It played some part in the EC Dublin summit Declaration on the Middle East of 26 June 1990. States are indeed at liberty to interpret, or reinterpret, their obligations under Article 1 in this way; and it may be especially appropriate to do so in view of developments since 1949, including the world-wide concern over laws of war issues. Yet states need to be aware that in so doing they are entering difficult territory, probably not envisaged by the negotiators of the 1949 conventions, and containing pitfalls for the unwary.

4. The Post-1945 Conventions: Humanitarian, Monitoring and Fact-Pinding Tasks

The 1949 Geneva Conventions put considerable, perhaps too much, emphasis on that long-established diplomatic institution, the Protecting Power, as a means of ensuring implementation of their terms during armed conflicts. Pictet defines a Protecting Power thus:

"A Protecting Power is, of course, a State instructed by another State (known as the Power of Origin) to safeguard its interests and those of its nationals in relation to a third State (known as the State of Residence). It will be seen at once that the activities of a Protecting Power are dependent on two agreements: the first between the Power of Origin and the Protecting Power and the second between the Protecting Power and the State of Residence".

The 1977 Geneva Protocol I, Article 2(c), offers the following definition for the purposes of the Protocol:

"'Protecting Power' means a neutral or other State not a Party to the conflict which has been designated by a Party to the conflict and accepted by the adverse Party and has agreed to carry out the functions assigned to a Protecting Power under the Conventions and this Protocol".

The 1949 Geneva Convention IV makes extensive provision for Protecting Powers to look after the interests of civilians, including in occupied territory. Protecting Powers have a general duty to protect the interests of the parties to the conflict; they are to lend their good offices in cases of disagreement either about the interpretation of the Convention or about its application; they can facilitate the establishment of hospital and safety zones and localities; they are to be informed of any transfers or evacuations in or from occupied territory; they can verify the state of food and medical supplies in occupied territories; they must be informed of all charges instituted by the occupant against protected persons involving the death penalty or sentences of two years or more, and they have various other rights and duties as regards legal proceedings; they can supervise the distribution of collective relief shipments to internees; and they can go to all places where protected persons are, particularly to places of internment, detention and work. In addition, Protecting Powers have responsibilities regarding the observance of the 1954 Cultural Property Convention; and are the subject of numerous detailed provisions in the 1977 Geneva Protocol I.

The 1949 Geneva Conventions do allow for the possibility that the Protecting Power arrangements might fail to come into operation. A common Article in all four conventions provides that certain organizations can assume all the functions of the Protecting Powers. Such a substitute may be, by agreement between the parties, "an organization which offers all guarantees of impartiality and efficacy"; failing this, in exceptional circumstances it may be a neutral state appointed by the Detaining Power; failing this, a humanitarian organization such as the ICRC may assume, if not all the traditional functions of the Protecting Power, at least the humanitarian functions performed by Protecting Powers under the 1949 Conventions.

Quite separately from the provisions regarding Protecting Powers, the 1949 Geneva Convention IV leaves much scope for activities by impartial humanitarian organizations such as the ICRC, which is mentioned in no less than 22 of its 159 articles.

In practice, the various formal provisions for the role of Protecting Powers have been of little use, because states in conflict with each other have almost always been unwilling or unable to agree on the appointment of such Powers. Many writers have pointed to the weaknesses of the Protecting Powers arrangements, depending as they do on a tripartite consensual basis. Yet there have been some cases of Protecting Powers having a role during armed conflicts, albeit in a manner different from what the conventions envisaged. They were used, for example, in the 1971 India-Pakistan War, in which one Protecting Power acted on behalf of both sides in a conflict, representing each to the other; and in the 1982 Falklands War, in which the duties of the Protecting Powers were mainly general diplomatic representation, rather than the exercise of humanitarian functions under the Geneva Conventions.

In most conflicts, it has been the ICRC which has borne the heat and burden of the day. ICRC representatives have actually carried out most of the various humanitarian and monitoring tasks provided for in the Geneva Conventions; this has been so even in cases where Protecting Powers have been appointed to look after certain interests of the belligerents.

A further mechanism aimed at securing implementation of the Geneva Conventions is the International Humanitarian Fact-Finding Commission. 1977 Geneva Protocol I, Article 90, provides for the establishment on a permanent basis, with periodic elections, of an International Fact-Finding Commission to:

"(i) enquire into any facts alleged to be a grave breach as defined in the Conventions and this Protocol or other serious violation of the Conventions or of this Protocol;

(ii) facilitate, through its good offices, the restoration of an attitude of respect for the Conventions and this Protocol ".

The International Humanitarian Fact-Finding Commission has in fact been set up. Under the terms of Art. 90(1)(b), which stipulated that it could only be established when twenty states agreed to accept its competence, the Commission, which has fifteen members, was duly elected at a meeting in Bern in June 1991. At its second meeting, in July 1992, the Commission unanimously adopted the final draft of its rules of procedure, and became operational. Since then it has done some preparatory work, not least, in the delicate words of its President, trying "to draw the international community's attention to its avaiblity.

An inquiry can be set up in two situations, both of which depend on states to take an initiating role. First, a party to a conflict can request an inquiry, which can only be established with the consent of the other party or parties concerned; or secondly, a state having accepted the automatic competence of the Commission makes an allegation against another state which has likewise previously accepted the Commission's competence. So far, there is no evidence that states will use this new mechanism. Franse Hampson has written:

"As experience with the Geneva Conventions has shown, the mere existence of a fact-finding mechanism does not mean that it will be used. The challenge is there. It remains to be seen whether States will rise to meet it".

Why, in its three years of existence, has use not been made of the International Humanitarian Fact-Finding Commission? One part of the problem is the continuing reluctance of many states to make a declaration accepting its competence. As of 30 June 1995, of the 138 states parties to 1977 Protocol I, 46 had made a declaration accepting the competence of the Commission. A more fundamental problem may be the reluctance of states, including those which have accepted the Commission's competence, to invoke its services in view of the fact it may have to deal with issues as sensitive as the applicability of the Conventions, and the characterization of particular acts as "grave breaches " or "serious violations".

The relevance of the Fact-Finding Commission is above all called into question by the fact that in the years since it was established the UN Security Council has established ad hoc mechanisms for investigating and taking action regarding violations, most notably in connection with the wars in the former Yugoslavia and Rwanda. Indeed, Erich Kussbach has gone so far as to say: "After a long period of inability to act, the Security Council has seemingly become the master of collective security and apparently is about to take over, step by step, the responsibility for the administration of humanitarian law. The future will show how the Security Council will be able to cope with its heavy responsibilities". Although his optimism about a UN system of collective security is open to serious criticism, the ad hoc arrangements under UN Security Council auspices do have many advantages over the Commission: it is not necessary for individual states to initiate the process; states or other entities can be investigated irrespective of whether they have accepted the competence of the Commission; the relevant body of law to be applied can be identified separately in each instance, and can thus be appropriate to the particular conflict and the facts alleged; the range of problems and situations which can be investigated is therefore greater, since it is not limited to clear cases of international armed conflict; there are fewer obstacles to publication of the outcome of an investigation; and the fact-finding process can be linked to action in the form of prosecutions. Once again, implementation in practice has assumed forms significantly different from what the conventions provide.

5. The Post-1945 Conventions: Punishment and Compensation

The various conventions concluded since 1949 contain far more on the punishment of violations than did their predecessors.

The 1948 Genocide Convention, Article VI, contains provisions for the trial, whether by national or international tribunals, of persons charged with offences under the Convention. However, these provisions are puny and inadequate when the treaty's full title (Convention on the Prevention and Punishment of the Crime of Genocide), and the scope and seriousness of the problem which it purports to tackle, are borne in mind. Article VIII breaks new ground in specifying that any contracting state "may call upon the competent organs of the United Nations to take such action under the Charter of the United Nations as they consider appropriate for the prevention and suppression of acts of genocide

The 1949 Geneva Conventions introduce a special category of "grave breaches", and outline a system of penal sanctions for persons committing them. Convention IV defines grave breaches thus:

"Grave breaches to which the preceding Article relates shall be those involving any of the following acts, if committed against persons or property protected by the present Convention: wilful killing, torture or inhuman treatment, including biological experiments, wilfully causing great suffering or serious injury to body or health, unlawful deportation or transfer or unlawful confinement of a protected person, compelling a protected person to serve in the forces of a hostile Power, or wilfully depriving a protected person of the rights of fair and regular trial prescribed in the present Convention, taking of hostages and extensive destruction and appropriation of property, not justified by military necessity and carried out unlawfully and wantonly".

The immediately preceding article, which is common to all four 1949 Geneva Conventions, requires states parties to enact any necessary legislation to punish grave breaches of the conventions; and to search for people who have committed, or ordered, such breaches. States may also hand such suspected persons over for trial in another state, provided a prima facie case has been made out. States are also to take measures necessary for the suppression of breaches other than grave breaches. This article follows the well-trodden path of relying on states for enforcement, but it does contain significant new provisions in that regard. National courts, which are to have penal jurisdiction to implement the Conventions, are to have jurisdiction over all individuals, regardless of nationality, alleged to have committed grave breaches. This system has obvious, but inevitable, weaknesses. In practice, states not involved in a conflict have proved reluctant to fulfil the obligation to try suspected offenders. A related difficulty concerns extradition. If a state is unwilling to punish a war criminal residing in its territory, it may prove difficult for another state to secure extradition, as the provision in the common article of the conventions is purely permissive. The easy way out for any war criminal is still to go to a country which does not have the political desire to punish him or her, and does not have extradition agreements with those who do.

Overall, the punishment provisions of the 1949 Geneva Conventions have proved vulnerable to the familiar objection that governments are generally reluctant to prosecute their own servants for violations of the laws of war. Further, courts may be reluctant to act independently of the political stance of their respective governments. The punishment provisions of the 1949 Geneva Conventions are also open to the possible objection that they allow for trials to be held in wartime, when passions may be high and the evidence incomplete. Finally, some problems may arise from the fact that the penal provisions, by providing for action by the state, do not envisage legal actions by injured parties. This is in contrast with the rather different procedures in human rights law.

The payment of compensation, or reparations, has consistently been provided for in the laws of war as one means of making amends, and has often been practiced. This whole approach is very different from provisions for trials, as it involves for the most part taking action against the state as a whole, rather than against individuals; and it is often decided by political-diplomatic rather than judicial bodies. Compensation was mentioned, as noted above, in the 1907 Hague Convention IV, Article 3, the words of which were adapted only slightly in the 1977 Geneva Protocol I, which says: "A Party to the conflict which violates the provisions of the Conventions or of this Protocol shall, if the case demands, be liable to pay compensation. It shall be responsible for all acts committed by persons forming part of its armed forces".

As far as non-international armed conflicts are concerned, the post 1945 international conventions that touch on this subject make hardly any provision for punishment or compensation. Common Article 3 of the 1949 Geneva Conventions, and the 1977 Geneva Protocol II, do establish some minimal rules regarding such conflicts, and states do have the right to punish breaches of those rules. Despite this, Denise Plattner was basically right to conclude in 1990: "The rules establishing international responsibility for violations of IHL applicable in non-international armed conflicts are yet to be made". Since then, the adoption by the UN Security Council of the Statute for the International Tribunal for Rwanda (which, in contrast to the tribunal on Yugoslavia, is predicated on the assumption that the conflict in Rwanda is non-international) provides some legal reinforcement to the claim that failure to observe certain basic humanitarian rules is an international offence even in civil wars.

6. Other Mechanisms of Implementation

Implementation of the laws of war often involves use of mechanisms which were not specifically envisaged in the conventions. These mechanisms are of many kinds. In their mutual relations, states commonly use the whole range of methods open to them, from diplomatic notes to economic sanctions and threats of war, as means of trying to enforce compliance; while within states a huge variety of factors may affect compliance, including the political and ethical culture of the state and its citizens. What follows are simply a few illustrations of the range of mechanisms which may in fact operate.

Fact-finding, especially public exposure of violations of the laws of war, has often been done through the media. It was through the press, for example, that details of the 1968 My Lai massacre in Vietnam were exposed. Similarly, it was though the press and TV that the inhumane regime in some prison camps in the former Yugoslavia was publicized. Sometimes the presence of UN peace-keeping forces in an area can help to make war crimes known.

Within states, a wide range of administrative measures have had an important part in bringing the practice of armed forces and the principles of the laws of war into some kind of relation with each other. For example, official enquiries and reports are often a means of establishing facts, and of re-examining and bringing about changes in government policy on particular issues.

Such reports can be a basis for applying international rules to internal situations. For example, throughout the troubles in Northern Ireland from 1968 to 1994 the UK government was consistent in viewing the situation there as essentially internal and low-level in character. At least until 1972 it did not go much beyond the position that the principles in common Article 3 of the 1949 Geneva Conventions were germane whether or not they were formally applicable. However, in 1972 Lord Gardiner's minority report, which was accepted by the Government, was an interesting example of asserting the wider relevance, even in an internal conflict, of certain international legal standards, including some from the main body of the four 1949 Geneva Conventions.

When, following the Israeli intervention in Lebanon in 1982, there were massacres of Palestinians at Sabra and Shatilla camps in Beirut in September 1982, it was an Israeli official report which helped establish the facts surrounding these events, and reminded Israel that certain well established standards had to apply not only to the actions of the Israel Defence Forces but also to those paramilitary forces operating in conjunction with them.

Unofficial commissions and inquiries, set up by non-governmental organizations, can also play an part in fact-finding, and in expounding the rules applicable to a particular situation. One such commission was established following the Israeli invasion of Lebanon in 1982.

Regional organizations can also have a role in reminding parties to conflicts of the relevance of international standards and in applying pressure for their implementation. They may do this through their representative, executive or judicial bodies. There have been many such cases in the history of the European Union and its predecessors. The role of the European Court of Human Rights in dealing with numerous cases from Northern Ireland is an example. The Court's decision of 27 September 1995 about the Special Air Service killing of three Irish suspects in Gibraltar in 1988 exposed considerable British government sensitivity about UK military actions being subject to European court decisions.

The European Commission of Human Rights played a significant role in issues arising from armed conflict when it heard the Cyprus v. Turkey cases. The Government of Cyprus vigorously asserted that violations of human rights by Turkey in the Turkish-occupied areas (including the detention or murder of some 2,000 missing Greek Cypriots and the refusal to allow more than 170,000 Greek Cypriot refugees to return to their homes) were contrary to the European Convention on Human Rights, and were matters of legitimate international concern. The applications of the Government of Cyprus were ruled admissible by the European Commission of Human Rights on 26 May 1975 and 10 July 1978 - a significant recognition in principle of the applicability of international human rights law to occupied territories.

At a more political level, from the late 1980s onwards member states of the European Community made protests to Israel regarding its policies in the occupied territories, and suspended or delayed ratification of trade agreements.

What is striking about the some of the implementation mechanisms briefly outlined here is their largely political character. They tend to involve attempts to change policies perceived as illegal or inhuman through the application of pressures of various kinds. Even when it is asserted clearly that violations of the laws of war have occurred, such attempts are not necessarily linked to demands for trials of individuals. Sometimes such attempts are one-sided, showing only limited understanding of the complexities of a conflict and the different viewpoints of belligerents.

Where court cases have followed from violations of the laws of war, they have often assumed a very different form from the state trials of war criminals envisaged in the conventions. They have often been civil rather than criminal in character. One unusual and controversial example is the Israeli innovation of establishing a right to petition the Supreme Court of Israel against arbitrary or illegal acts by the occupant. Another interesting example is the current attempt of former prisoners of war under the Japanese, and Asian women forced into prostitution by occupying Japanese armies in the Second World War, to gain compensation for illegal Japanese conduct through Japanese courts.

7. The Involvement of the United Nations

Since about 1980, crises over implementation have focused to an unprecedented extent on the United Nations, and more particularly on the Security Council. The UN's involvement in issues relating to the international law of armed conflict goes back a long way. The wartime allies established the United Nations War Crimes Commission on 20 October 1943, at the same time as they were working towards the creation of the United Nations Organization. The 1948 Genocide Convention, Article VIII of which made significant reference to the UN, was negotiated at the UN General Assembly. A further major landmark was General Assembly Resolution 2444 of 19 December 1968 on "Respect for Human Rights in Armed Conflicts". Literally hundreds of UN General Assembly resolutions have used the laws of war as a basis for criticizing the actions of particular states; Israel's conduct in the occupied territories has been condemned with particular frequency.

In 1977, two treaties made explicit provision for a major UN role in implementing the laws of war. The Convention on the Prohibition of Military or Any Other Hostile Use of Environmental Modification Techniques (the ENMOD Convention) envisaged that the UN, especially the Security Council, would handle investigations of suspected violations. The Geneva Protocol I foreshadowed the further involvement of the UN in matters relating to the laws of war when it stated: "In situations of serious violations of the Conventions or of this Protocol, the High Contracting Parties undertake to act, jointly or individually, in co-operation with the United Nations and in conformity with the United Nations Charter".

Since at least the mid-1980s, the UN's involvements in matters relating to the laws of war have focused more on the Security Council than the General Assembly, and more on implementation than on development of the law. The Security Council's involvement in this area illustrates the wide range of issues in which the Council can become engaged once it is seen as capable of reaching agreement. It also illustrates the difficulties of attempts to ensure implementation.

If the assumption is that the UN is becoming the centre of a system of collective security, then its concurrent rule as a supervisor and arbiter of the implementation of the laws of war may well seem a logical corollary. In the post-Cold War era, many have hoped, and some still hope, that the UN has the possibility of establishing some kind of general system of collective security. However, for those who are sceptical as to whether the traditional difficulties of proposals for collective security have even been addressed, let alone overcome, the Security Council's role in laws of war issues looks especially problematic. If the Security Council is not capable of tackling effectively even a modest proportion of threats to the peace, will it be any more effective in the lesser task of securing implementation of the laws of war?

One obvious problem with the UN's role in respect of the laws of war is that the Security Council is necessarily selective as to which issues it tackles. Due principally to the existence of the veto, the Security Council did nothing about alleged violations of the laws of war during the Vietnam war in the 1960s and 1970s, nor during the Afghan war in the 1980s. It was the General Assembly, not the Security Council, that passed most of the resolutions critical of Israeli conduct in the occupied territories.

In some conflicts since the mid-1980s, as briefly summarized in the following sections of this study, the UN Security Council has dealt with laws of war issues. It has addressed two fundamentally distinct aspects: first, the investigation and punishment of major violations by belligerents; and second, the management of UN-authorized forces, whether in peace-keeping or enforcement actions, in a manner consistent with the laws of war. An opinion on the effectiveness of these roles will be offered in the conclusions of this study.

8. The International Court of Justice

The International Court of Justice (ICJ) at The Hague has long had certain limited roles in respect of implementation of the laws of war. There are specific references to the ICJ in the 1948 Genocide Convention; and the 1954 Hague Cultural Property Convention. However, the Court's Statute, with its built-in limitations on what type of cases may be brought to it and by whom, is likely to mean that it will only have to look at a minority of issues concerning the laws of war.

Many cases brought before it have involved key laws of war matters: for example, the Corfu Channel case in 1949; and Nicaragua v. USA in 1986.79 Both these cases involved the principle that a state laying mines at sea is obliged to give notification of their location in order to protect the security of peaceful shipping. The Court now has the politically more sensitive and intellectually more complex issue of the legality of nuclear weapons to consider in the case brought by the World Health Organization and the UN General Assembly.

Many cases have involved issues analogous to, and potentially relevant to, laws of war problems. The United States Diplomatic and Consular Staff in Teheran case concerned the treatment of individuals under the protection of international law in an emergency situation. The Frontier Dispute (Burkina Faso/Mali) case raised the question of interim measures of protection. The 1971 Advisory Opinion on Namibia involved several germane matters, including the use of a sanction: termination of a League/UN mandate as a response to failures to observe certain rules of restraint.

In cases concerning the former Yugoslavia, the ICJ has been asked to answer very complex political questions touching on the laws of war. This is most notably so with the case brought by Bosnia and Herzegovina against the Federal Republic of Yugoslavia, Case Concerning the Application of the Convention on the Prevention and Punishment of the Crime of Genocide. This case is currently proceeding slowly.

In many of these cases on which it has reached decisions, the ICJ has performed a useful service by clarifying the content of the laws of war and their application to particular and often complex circumstances; and by publicizing fundamental principles which should inform the policymaking of states in matters relating to the use of force. However, there are limits to what the ICJ can achieve. Many states are reluctant to let cases concerning their own survival be settled by a distant conclave in The Hague. They may also worry about the slowness of some (but certainly not all) of its proceedings. When it is asked to comment in a general way on complex issues which are bones of contention among statesmen and lawyers - as in the nuclear weapons case currently before it - the Court's decision, whatever it is, may not be found universally persuasive, let alone decisive. The ICJ may look weak if it tries to avoid certain difficult issues because they are not justiciable, or because they are not quite the types of matter with which the Court is charged to deal. However, it may look even weaker if it reaches a decision which is then not fully implemented. In some cases it could be very difficult to secure implementation of the Court's decisions.

(introduction...)

In some of the international wars of the past two decades, belligerents have displayed both an acceptance of the applicability of the laws of war and a willingness to apply them in practice. Despite some questionable acts on both sides, this was largely the case in the 1982 Falklands War between the UK and Argentina - perhaps because this was a war in which civilians were only marginally involved, and in which neither of the belligerent countries faced a threat to its entire existence. In some other conflicts, including the 1990-91 Gulf Conflict, the laws of war have been applied more by one side than by the other. In yet other conflicts, especially those involving more extreme threats to the existence of communities and countries, implementation of the laws of war has been patchy or non-existent. Important issues have emerged from these varied experiences, including several very different responses by the UN Security Council to violations of the laws of war. Four conflicts are considered in detail in this section: the 1980-88 Iran-Iraq War, the 1990-91 Gulf conflict, the wars in the former Yugoslavia since 1991, and the internal conflict in Rwanda in 1994. There is also a brief look at the International Conference on the Protection of War Victims, held in Geneva in 1993, which attempted to grapple with the question of implementation.

1. Iran-Iraq War 1980-88

The long war between Iran and Iraq was in many respects a traditional inter-state war. The main treaties of the laws of war were incontestably applicable. There were violations of fundamental rules in such matters as the treatment of prisoners, the use of gas, and attacks on neutral shipping. The UN had two major involvements in respect of the laws of war: first in connection with treatment of prisoners, then in connection with use of gas.

In January 1985, acting on his own behalf, the UN Secretary-General dispatched a mission to Iran and Iraq to investigate conditions under which POWs were being held; this included investigating an incident which had occurred at a POW camp in Iran in October 1984, in which a number of Iraqi POWs were killed or injured. This led to a very thorough report; whether it greatly benefited POWs is doubtful.

Iraq's use of gas during the war gave rise to one of the first major uses of the UN Security Council as a "monitor" of the laws of war - a role which proved, and continues to prove, problematic. Both Iran and Iraq were parties to the 1925 Geneva Protocol prohibiting the use of gas and bacteriological war. The Secretary-General commissioned a number of investigations into reports of the use of gas. On 21 March 1986 a UN Security Council statement for the first time criticized Iraq by name over the use of gas. On 26 August 1988 the Security Council unanimously adopted a resolution condemning "the use of chemical weapons in the conflict between Iran and Iraq" 87 There was also considerable evidence of Iraqi use of chemical weapons against its Kurdish minority.

After the end of the war in 1988, various diplomatic efforts were made to reinforce the Protocol's provisions. In January 1989, at a specially convened Conference in Paris on the Prohibition of Chemical Weapons, attended by representatives of 149 states, a final declaration was adopted which solemnly condemned use of such weapons and reaffirmed the prohibition in the 1925 Protocol. Iraq agreed to this declaration, claiming that the threat it had faced from Iran had been terrible and exceptional. The conference's final declaration gave strong support for the role of the UN Secretary-General as an investigator of violations. However, as a special correspondent of the Financial Times wrote, "Past experience has made the UN Secretariat sceptical about how free a hand it will get".

In January 1993, five years after the end of the Iran-Iraq War, the Chemical Weapons Convention was signed in Paris; this is first and foremost a prohibition of manufacture and possession of such weapons, not just of use, and thus belongs more in the category of arms control than laws of war. It is not yet formally in force. This treaty has been seen as overcoming a perceived weakness of the 1925 Geneva Protocol, namely that it prohibited use but not possession. However, there is some risk that the Chemical Weapons Convention, when it eventually enters into force, could actually weaken the prohibition on the use of chemical weapons. This is because it leaves some uncertainty about the sanction that would be employed in the event of violations. The threat of retaliation in kind, which had sometimes buttressed the old 1925 Geneva Protocol regime, would be absent in future. It had only been where that threat was absent, because the victim state lacked any capacity to threaten retaliation in kind, that chemical weapons had been employed. Instead, Article XII of the 1993 treaty provides for the application of collective measures by States Parties, including, in cases of particular gravity, bringing the issue to the attention of the UN General Assembly and Security Council. Whether this will prove effective in practice remains to be seen.

The events surrounding the Iraqi uses of chemical weapons in the 1980-88 war illustrate several inter-linked themes which have since recurred in other conflicts:

1. there is clear violation of a well established rule of the laws of war;

2. following an investigation under UN auspices, the Security Council condemns violations;

3. an international diplomatic conference in Europe solemnly condemns violations and calls for action; and

4. nothing further happens regarding the particular violations that have occurred.

This episode may have reinforced a lesson which the Iraqi leadership had no doubt also learned from the supine response of the powers and of the UN to the original Iraqi attack on Iran in 1980: that Iraq could ignore the solemn pronouncements and spasmodic condemnations issued by the international community.

2. The 1990-91 Gulf Conflict

The laws of war issues in the 1990-91 Gulf conflict - that is, the August 1990 Iraqi occupation of Kuwait, and the war of January-February 1991 which brought it to an end - have been considered in several official US publications. There has also been much analysis, official and unofficial, in a wide range of other publications, and from other national perspectives.

During the occupation and war there was extensive public reference to well-estabhshed international norms about particular aspects of the conduct of occupations and armed conflicts, not least in the matter of seizure and treatment of hostages, treatment of prisoners of war, attacks on civilians, non-use of chemical weapons, and wanton destruction including damage to the environment. Further, this war threw light on some general questions. First, did Iraq's violations of the Geneva Conventions and other rules of war, and the strong political reaction that these violations caused, contribute to the hardening of opinion against it, and thus to the formation of the multi-national coalition? Second, can conducting operations on a basis of acceptance of the laws of war assist in maintaining the unity and coherence of a coalition action with United Nations authorization? Third, can the laws of war contribute significantly to the maintenance of restraints in war? Fourth, what can be done about the post-war punishment of violations?

The 1990-91 Gulf Conflict is an interesting case of a war in which one side, by and large, took laws of war restraints seriously on a wide range of matters, dealing both with combat and with the treatment of prisoners and others under the control of a belligerent. The other side, while not in principle rejecting all idea of the laws of war, did ignore them on a range of issues. The case therefore confirms that in practice implementation does not always have to be a matter of reciprocity between the parties; there are other reasons - legal, political and practical - for one side to implement them even if its adversary does not

The twenty-eight member military coalition probably gained from its attention to the laws of war in a number of ways. First, in the months after the Iraqi occupation of Kuwait on 2 August 1990, the widespread public attention to illegal Iraqi policies, including seizure of hostages, helped to reinforce support for the coalition cause. It did so both domestically within the coalition countries, and internationally in helping to mobilize support for their position. Second, once Desert Storm began in January 1991, the emphasis on restraint and accuracy in targeting had similar benefits so far as the international consensus was concerned. Third, the promise of good treatment of Iraqis who left their vehicles and/or surrendered, a promise backed up by a major leaflet campaign, may have encouraged many Iraqi soldiers to end their participation in hostilities.

The first two of these points suggest the particular salience of the laws of war to coalition warfare under UN auspices. They also suggest that there are in fact some complex and subtle connections between ius in bello and ius ad bellum. Iraq's violations of ius in bello helped to weld the coalition together, reinforcing its sense of the legitimacy of its cause. The laws of war provided one means for harmonizing the different practices of states and justifying them to anxious publics. Curiously, there was no serious suggestion in the 1991 Gulf War that forces acting in the name of the UN ought to be in some way privileged, and to have greater rights in some matters than their adversaries. That issue would come up in Somalia.

This is not to assert that, in laws of war terms, coalition conduct was perfect. There were many defects. As in the 1982 Falklands War, some battle incidents raised the question of what is proper evidence that troops wish to surrender. For example, on the second day of the war US forces attacked an oil platform on which Iraqi forces were allegedly trying to surrender. A US Navy board of investigation concluded that the US commanding officer's actions did not violate the law of armed conflict, but his failures to investigate, evaluate and report on the Iraqi white flags represented a serious lapse of judgement. This US Navy investigation was an instance of implementation of aspects of the laws of war being seen as a national responsibility, and being handled in the context of maintaining professional military standards.

Perhaps the most questionable acts by the coalition were those resulting in civilian deaths. In the Amariya bunker attack on 13 February 1991 there were approximately 300 civilian casualties. It can be claimed that this was not clearly, perhaps not at all, a violation of the laws of war, as it had not been the intention of the coalition to bring about such a terrible result. This disaster appears to have been due to reliance on faulty intelligence.

Many other events in the war similarly suggest the inherent difficulty, perhaps impossibility, of maintaining a clear line between attacks on the military and attacks on civilians. The coalition's use of accurate air-delivered weapons undoubtedly did something to increase the chances of air warfare conforming to a greater extent than hitherto with basic principles of the laws of war, especially as regards discrimination in attacks. On the other hand, in this war there were many coalition attacks on military targets such as bridges and power stations which also served civilian functions. There was also much collateral damage, faulty intelligence, and so on. Suggestions that we might be entering a new era of clinical warfare are probably misplaced.

Iraq violated the laws of war in countless ways: looting, taking of hostages, treatment of prisoners, Scud attacks on cities, and attacks on oil installations resulting in damage to the environment. Most of these acts would have been violations in any circumstances, but were particularly egregious cases on account of the absence of a serious strategic rationale. In the event, they achieved few military results. The judgement that Iraq derived no serious military advantage from its violations is persuasive. Many Iraqi war crimes involved the coalition in considerable effort, for example in dealing with the Scud threat during the war and in restoring the oilfields after the war, but they were never likely to be militarily decisive.

After the suspension of coalition military activities on 28 February 1991, the coalition governments suddenly became quiet on the subject of the responsibility of Saddam Hussein and colleagues for major war crimes. This contrasted with the earlier emphasis on war crimes in Security Council Resolution 674 of 29 October 1990. After the cessation of hostilities, the Security Council passed some long and very detailed resolutions on the cease-fire, reparations, dismantling of Iraq's capability for chemical warfare, and so on. One of these, Security Council Resolution 687 of 3 April 1991, is the longest ever passed by the Security Council. Yet. nothing was said on the subject of personal responsibility for war crimes. Similarly, in less than three months after the cessation of hostilities some 64,000 Iraqi prisoners of war were repatriated without any attempt to sift out those suspected of war crimes - a process which might have delayed repatriation by years.

There were genuine difficulties in pursuing the war crimes issue First and foremost, Saddam Hussein would have been difficult to arrest even had the coalition military action had more offensive goals. After the end of hostilities, it would have been awkward to call for his arrest as a war criminal at the same time as negotiating cease-fire terms with his government. Further, outside powers were reluctant to press for trials if local powers would not join them in doing this. There were hazards in limiting trials to the conflict of 1990-91, as the Iraqi regime had engaged in criminal activities externally and internally both before and after that episode.

However, the failure to take any action against the Iraqi leaders exposed a serious problem regarding the laws of war, namely, the difficulty of securing enforcement even after clear evidence of violations The Pentagon ended its Final Report pointedly: "A strategy should be developed to respond to Iraqi violations of the law of war, to make clear that a price will be paid for such violations, and to deter future violators". At a minimum, it would have been possible to make a statement to the effect that major war crimes occurred, involving grave breaches of the Geneva Conventions, that there is personal responsibility for these crimes, and that under the Geneva Conventions any state is entitled to prosecute. Such a statement could have been made by the coalition powers, the UN General Assembly, or the Security Council. The United States did eventually, in a little-noted war crimes report prepared in 1992 and issued by the UN in March 1993, say all of these things, but did not put great political emphasis on the matter. The absence of a coherent coalition policy on major war crimes was highlighted by the fact that some junior figures - Iraqi officers who happened to have been caught in Kuwait at the time of the cease-fire - had been put on trial in Kuwait for lesser offences.

The failure to hold a major war crimes trial after the 1991 Gulf War exposed a central problem of the laws of war. When a state is not willing to prosecute its own government leaders and officers, there is frequently no other practical mechanism for bringing alleged offenders to justice. It is not scarcely responsible to think in terms of some form of supranational justice being applied, when the power, the mechanisms and/or the will are lacking. Even when an opportunity for a trial presented itself, there was no rush to take action. When in August 1995 Lt.-Gen. Hussein Kamel, a senior Iraqi leader implicated as a war criminal, defected to Amman, there were not many calls for his trial. Governments were evidently more concerned to encourage defections from the Iraqi regime than to punish past misdeeds.

When a war ends, especially if it is seen as having resulted from an illegal and aggressive act by one side, or was characterized by widespread damage or looting, there is often a demand - which has a basis in the conventions - for reparations, compensation, and the return of stolen goods. Following this tradition, the UN Security Council, in the second paragraph of its Resolution 686 of 2 March 1991, demanded that Iraq:

"b) Accept in principle its liability under international law for any loss, damage or injury arising in regard to Kuwait and third States, and their nationals and corporations, as a result of the invasion and illegal occupation of Kuwait by Iraq;

[...]

d) Immediately begin to return all Kuwaiti property seized by Iraq, to be completed in the shortest possible period ".

The wording of this resolution implied that compensation was demanded principally on account of considerations of ius ad bellum - i.e. Iraq's initiation of the conflict by its invasion of Kuwait, and all that flowed from that. Iraq was considered to have state responsibility. This approach meant that it would not be necessary to show that all damage was directly committed by Iraqis as distinct from the coalition forces repelling them; nor that any Iraqi acts complained of were necessarily violations of the laws of war (ius in bello).

The idea of compensation was given flesh by the subsequent establishment of the United Nations Compensation Commission under the terms of Security Council Resolution 692 of 20 May 1991 This is based on the principle that Iraq is internationally responsible for its unlawful acts. By April 1995 2.6 million claims had been filed for a total of approximately US$ 174 billion. Since Iraq has so far been unwilling to sell oil, the only funds available to the Commission have come from the partial liquidation of Iraq's assets abroad.

Neither in the Iraqi case, nor more generally, are reparations and compensation necessarily an alternative to war crimes trials. Both paths can be pursued simultaneously, as they were, disastrously, in the 1919 Treaty of Versailles at the end of the First World War.

Some saw the post-war demands for reparations, and the sanctions on Iraq, as part of a great opportunity to shape a fair method of compensation, and a bold application of UN Charter procedures. The approach of seeking financial redress on the basis of state responsibility has some obvious advantages: matters can be settled in negotiations between states, sometimes relatively quickly. However, the fact of a country as a whole being held liable for the entire costs of a war is intensely problematic. It is likely to involve a whole population in paying for offences committed by a minority among them. The process of payment may drag out for decades and cause dangerous political resentments against those imposing the penalties. If submitting a few individuals to trial and punishment seems dangerously selective when a larger number may be responsible, punishing the whole population over a long period is open to the accusation of being indiscriminate.

The 1990-91 Gulf Conflict also raised important questions about neutrality. In different ways, the idea of neutrality has had a significant place in provisions regarding the implementation of the laws of war, including the provisions for Protecting Powers, and for action by the ICRC. Further, the idea that states can be neutral vis-is particular conflicts is fundamental to the laws of war; and 1907 Hague Convention V on Neutrality in Land War firmly associated neutrality with impartiality.

The whole idea of neutrality is undergoing significant change as a result of a number of factors, including the collective character of international action, especially sanctions, under the UN Charter. Following the Iraqi invasion of Kuwait in August 1990, various Security Council resolutions imposed a range of obligations on all states to take part in sanctions and even to provide a measure of assistance to the coalition military actions to enforce the blockade. Security Council Resolution 678 of 29 November 1990, which authorized the use of force, did not call on, still less require, all UN member states to take part in military action, but it did request them to provide appropriate support. Subsequently, some traditionally neutral states, including Switzerland and Austria, went so far as to permit overflights by US military aircraft. In summary, UN resolutions and the practice of states in the crisis suggested that there was still some space for neutrality in the sense of non-participation in hostilities, but less space for neutrality in the sense of complete impartiality.

The policy of the Swiss government regarding the Iraq-Kuwaconflict had an influence on debates within the International Committee of the Red Cross (ICRC) about the ICRC's status. The ICRC is of course one of the principal bodies concerned with monitoring implementation of the Geneva Conventions. Both Switzerland and the ICRC (which had always been a quintessentially Swiss body, and remains based in Switzerland) are committed to policies of neutrality, but the policies are not identical. In the wake of UN decisions, Switzerland - autonomously, as it is not a UN member - decided to apply wide-ranging economic and financial sanctions against Iraq. This may have fed Iraqi suspicions about the impartiality of the ICRC's operations. These events connected with the 1991 Gulf War form part of the background to a key change in the ICRC's status. On 19 March 1993, the ICRC and the Swiss government signed an agreement on the ICRC's legal status, in which the Swiss government recognized the ICRC as an international, rather than Swiss, organization. The ICRC has simultaneously become involved in various UN activities, including participation in the UN's Department of Humanitarian Affairs, established in March 1992. In this connection, the ICRC has had to specifically reiterate its independence and impartiality.

The developments described do not invalidate conceptions of neutrality, whether of states or of the ICRC, nor do they suddenly destroy the roles of neutral entities in application of the laws of war. What they demonstrate clearly is that notions of neutrality are changing. In a wide range of matters - including implementation of sanctions, conduct of military operations, and organization of humanitarian relief the UN framework has become more important than before. The UN may well not be neutral, or not be perceived as neutral, in a particular conflict. To some extent at least, the idea, enshrined in treaties, of implementation by neutrals may be becoming overlaid by the idea of implementation by UN-related bodies.

3. The Wars in the Former Yugoslavia since 1991

Ever since the Croatian part of the Yugoslav war began in dune 1991, the war in the former Yugoslavia has been characterized by extensive atrocities. Many acts - including the terrorization, killings and expulsions of civilian inhabitants in so-called "ethnic cleansing", and also the cases of rape - were violations of existing rules, whether those applicable in international or in internal conflicts, or those from the human rights stream of law. This grim fact can be explained in many ways, but high among them is that this is a war about state creation, in which a principal purpose of certain belligerents is to achieve an object which itself involves violations of the laws of war: expulsions of populations, and their replacement by other populations.

During the 1990-91 Gulf Crisis, some people had criticized the UN Security Council for not paying enough attention to laws of war issues m its various resolutions. In the Yugoslav crisis, Security Council resolutions from mid-1992 onwards drew attention to violations of the laws of war and demanded that they cease. For example, Security Council Resolution 764 of 13 July 1992 reaffirmed that all parties to the conflict are bound to comply with their obligations under international humanitarian law, and that persons who commit or order the commission of grave breaches are individually responsible. Two months later, Resolution 771 of 13 August 1992 called on states to collate substantiated information on violations of humanitarian law, and also said that if the parties failed to comply, then the Council would take 'further measures".

The London Conference on the former Yugoslavia of 26 and 27 August 1992 - a joint EC and UN initiative - echoed such appeals, decided to "take all possible legal action to bring to account those responsible for committing or ordering grave breaches of international humanitarian law", and announced that the Co-Chairmen "have undertaken to carry forward a study of the creation of an international criminal court ".109

When the belligerents paid no attention to these various statements, the Security Council in Resolution 780 of 6 October 1992 asked the Secretary-General to establish an impartial Commission of Experts to examine evidence of grave breaches of international humanitarian law in the former Yugoslavia. The five-member commission chaired by Prof. Frits Kalshoven of the Netherlands was created during October, and produced an interim report that was issued on 10 February 1993.

The decision to set up an international tribunal was influenced by the political and moral pressure, which was strong in many countries, to do something about Yugoslavia. Since the international community was unable to agree on any major intervention or other decisive action, the Tribunal was one of the few options left. Thus, on 10 February 1993 at a time of great demand for action over Yugoslavia, and on the same day that the Kalshoven Commission's interim report was issued, US Secretary of State Warren Christopher announced "a series of new steps that President Clinton has decided to take with regard to the former Yugoslavia''. In the course of announcing these he said: "The President is seeking the urgent creation of a war crimes tribunal at the United Nations to bring justice and deter further atrocities". Less than a fortnight later, in Res. 808 of 22 February 1993, the Security Council agreed in principle that "an international tribunal shall be established for ¿he prosecution of persons responsible for serious violations of international humanitarian law committed in the territory of the former Yugoslavia since 1991".

This led to the Report of the Secretary-General of 3 May 1993, which contains the Statute of what was to become the International Criminal Tribunal for the Former Yugoslavia, and also explains the reasoning behind its establishment. The report's rationale for establishing the Tribunal has several notable features. First, on a procedural issue, the report is defensive as to why the Security Council is involving itself in the setting up of the Tribunal, when the normal course would be "the conclusion of a treaty by which the States parties would establish a tribunal and approve its statute". It says, in a statement whose realism and sense of urgency cannot be faulted, that one disadvantage of the treaty approach is that "there could be no guarantee that ratifications will be received from those States which should be parties to the treaty if it is to be truly effective". This raises the question whether, if certain states would not ratify the treaty, they are likely to co-operate fully with the Tribunal as set up by the Security Council.

On the substantive issue of what the International Criminal Tribunal for the Former Yugoslavia might be expected to achieve, the Secretary-General's report of 3 May 1993, echoing Res. 808, seeks to justify the Tribunal largely in terms of its practical effect in relation to the ongoing conflict. In so doing it sounds naively optimistic:

"the establishment of an international tribunal would bring about the achievement of the aim of putting an end to such crimes and of taking effective measures to bring to justice the persons responsible for them, and would contribute to the restoration and maintenance of peace".

The "International Tribunal for the Prosecution of Persons Responsible for Serious Violations of International Humanitarian Law Committed in the Territory of the Former Yugoslavia Since 1991 " was finally established by a Security Council decision of May 1993, and was inaugurated in The Hague on 17 November 1993. With Judge Antonio Cassese as President, it has made progress in establishing itself generally, in drawing up Rules of Procedure and Evidence, and in preparing cases. Indictments were issued against Dusan Tadic on 13 February 1995, and against Milan Martic, Radovan Karadzic, Ratko Mladic and 21 others on 21 and 25 July 1995. By September 1995 a total of 43 people had been indicted, but it was very uncertain whether more than a very few would ever appear before the tribunal.

Despite the evidence of progress, it remains to be seen whether the International Criminal Tribunal for the Former Yugoslavia can fulfil the purposes indicated in the UN Secretary-General's report of 3 May 1993. In particular, it may be doubted whether there will be anything like a major trial of major leaders. Instead, there are likely to be trials of lesser figures, for example those who happen to have travelled to a country which was willing and able to arrest them and extradite them to The Hague. Alternatively, there may simply be indictments and arrest warrants which effectively confine those under suspicion to stay in states which do not co-operate with the Tribunal. The whole process will take years or decades, not months - a further reason why it is not necessarily wise to suggest that such a process is a preliminary to restoration of peace.

It was obvious from the start that achievement of the goals for which the Tribunal was established might be blocked by the following factors:

1. the probable need, in efforts to end the war, to negotiate with the very people who are wanted for war crimes, and to agree some kind of amnesty;

2. the problem of getting evidence which proves the guilt of specific named individuals - a far more difficult matter than proving in a general way that war crimes occurred;

3. the difficulty of getting suspects arrested and brought to The Hague - the statute having, probably rightly, ruled out trials in absentia;

4. the difficulty of getting witnesses to come to The Hague to give evidence, and of protecting them thereafter; and

5. the difficulty of getting adequate and reliable financial resources for what must be a very extensive process of investigation and trial, especially as the UN General Assembly has ultimate control over funding, and is anxious about the gravitation of powers within the UN toward the Security Council.

In respect of war crimes in the former Yugoslavia, there remains a role for national courts. There have been a few national trials already in connection with the Yugoslav events, including at least one such trial in Bosnia, and doubtless there will be more. The establishment of the International Tribunal does not do away with the requirement in the 1949 Geneva Conventions for all states to see to the punishment of grave breaches, wherever they occurred. However, Article 9 of the Statute of the International Tribunal confirms that "it shall have primacy over national courts".

The creation of the International Criminal Tribunal for the Former Yugoslavia has naturally given rise to pressures to treat with equal seriousness other offences in other places. One consequence has been a revival of long-standing proposals for the establishment of an international criminal court. As Theodor Meron put it in 1993:

"The establishment on an ad hoc tribunal should not stand alone, however, as a sole or adequate solution. The world has failed to prosecute those responsible for egregious violations of international humanitarian law and human rights in Uganda, Iraq and Cambodia. To avoid charges of Eurocentrism this ad hoc tribunal for the former Yugoslavia should be a step toward the creation of a permanent criminal tribunal with general jurisdiction. The drafting of a treaty on a permanent tribunal, on which work has begun by the UN International Law Commission, should be expedited, providing an opportunity to supplement the substantive development of international law by an institutional process''.

Although there are grounds for scepticism regarding the International Criminal Tribunal for the Former Yugoslavia, and its counterpart for Rwanda discussed below, its failure could be a great disaster. Governments, having willed it into existence, should be encouraged to co-operate fully with the Tribunal, not least in assisting it in gathering evidence, and in providing urgently needed material and human resources. In the event that there are peace agreements in former Yugoslavia involving some element of amnesty, it is not self-evident that outside powers in general, or the UN Security Council in particular, would necessarily have to be bound by the amnesty provisions of such accords: they might be able to decide that former war criminals who stepped outside former Yugoslavia could still be vulnerable to prosecution.

Questions concerning implementation of the laws of war by UN peace-keeping forces have arisen in several recent conflicts, including in Somalia and Rwanda, as well as in the former Yugoslavia. The presence of the United Nations Protection Force (UNPROFOR) in the former Yugoslavia from February 1992 onwards contributed to the world's awareness of war crimes there, and to the sense that something should be done about them. These forces often helped journalists to reach war zones, and their very presence made UN inaction in face of war crimes unacceptable.

One curious accusation concerning the role of UNPROFOR can simply be dismissed. The legal affairs department of Mcins Sans Frontis has accused UNPROFOR of violating the rules of war. This accusation seems to be based on a serious misunderstanding both as to the content of international law relating to armed conflict, and as to the fact that UN forces have up to now been clearly viewed as bound by such law, not least because the countries providing contingents are so bound.

A major question raised sharply by events in Yugoslavia is: Should UN peace-keeping forces gather information about war crimes, and/or arrest suspects? A similar question has arisen for the personnel of UNHCR and other agencies, for UN Human Rights Action Teams, and for European Union monitors. In the case of UNPROFOR, inasmuch as a clear answer has emerged, it appears to be that information on violations may be recorded and passed on, including by some national contingents through their own national authorities. However, it has not yet been made any part of the mandate of UN peacekeepers to arrest suspected war criminals and hand them over for possible trial. Sometimes UN peacekeepers have been passive onlookers at atrocities. There were accusations that UNPROFOR in general, and Dutch forces who were in place at the time, knew of atrocities committed against Muslim men in Srebrenica at the time of its capture by Bosnian Serb forces in July 1995 and did little. There have also been press suggestions that the governments of Britain, the USA and the Netherlands sought to play down the massacre.

Another question, no less difficult, is: Are UN peace-keeping personnel entitled to a special legal status under which, for example, any attack on them would be classed an offence against international law? The Security Council has passed numerous resolutions requiring belligerents to respect the special status of UN peace-keeping forces. In December 1994, the UN General Assembly approved the text of a Draft Convention on the Safety of United Nations and Associated Personnel. There are strong reasons for sympathising with this proposed convention. It responds to serious problems of attacks and hostage-taking. It has precedents in the protection given to impartial workers in the four 1949 Geneva Conventions and in the 1977 Protocols. Indeed, 1977 Geneva Protocol I already gives specific protection to the UN emblem. Like the 1949 Geneva Conventions, the 1994 convention will require states parties to follow an "extradite or prosecute" rule regarding alleged offenders - that is, those suspected of attacks on UN personnel. There may be difficulties in the application of this convention. Like so many treaties, it relies heavily on states being willing and able to take action against their own nationals. If this does not work, and UN forces then decide to take action themselves against alleged offenders, there is a risk of UN forces finding themselves parties to an armed conflict, in which case the convention would probably cease to apply, being replaced by the law of international armed conflict.This implies an important transition of the status of a peace-keeping operation, with broad policy ramifications.

The most important means of dealing with violations of the laws of war in the former Yugoslavia has probably been the threat and use of force by NATO in conjunction with the UN. It was violations of the UN-declared "safe areas", repeated obstruction of humanitarian relief, and then the atrocities and bragging accompanying the Serb capture of Srebrenica in July 1995, that led to a change of Western policy. The greater willingness of Western powers to use force was most evident in the "Operation Deliberate Force" bombing campaign of 30 August to 14 September 1995. This campaign involved moral ambiguities typical of warfare: however reluctantly, the NATO and UN forces operating in Bosnia were in a form of "co-belligerence" with Croat forces which were themselves, both in the Krajina and in Bosnia-Herzegovina, engaging in practices hardly distinguishable from so-called "ethnic cleansing". The NATO operation could not be expected to stop all atrocities in a peculiarly vicious war. However, it did perhaps lead to a degree of realization on the part of the Bosnian Serbs that verbal condemnations by outside bodies could actually lead to serious military action, which was contributing to the disastrous loss of Serb-held territory. Yugoslavia exemplifies the lesson that if outside powers seek seriously to change the practices of belligerents and to stop atrocities, they may have to be prepared to intervene militarily. In so doing they may find themselves working with dubious belligerents, and they may have to be prepared to exert serious pressure on co-belligerents as well as on more open adversaries.

4. Civil War and Humanitarian Intervention in Somalia 1992-95

Somalia was the first case in the UN era of a forceful armed intervention specifically authorized by the UN Security Council, and partly justified in terms of international humanitarian law, including the need to protect international relief and peace-keeping personnel. Security Council Resolution 794 of 3 December 1992 made several references to international humanitarian law, deploring widespread violations, and stating that the Council

"5. Strongly condemns all violations of international humanitarian law occurring in Somalia, including in particular the deliberate impeding of the delivery of food and medical supplies essential for the survival of the civilian population, and affirms that those who commit or order the commission of such acts will be held individually responsible in respect of such acts[...]"

The rest is history. It has involved several elements: first, the absence of any consistent effort to deal with the persons responsible for such acts; second, the pursuit of General Aideed, eventually abandoned; and third, the gradual emergence of a doctrine which seemed to privilege forces acting with a UN mandate. Following an attack on UN forces earlier, the UN got involved in the slaughter of civilians in the terrible incident in Mogadishu on 9 September 1993. The following day the UN military spokesman in Mogadishu, Major David Stockwell, was quoted as saying: "Everyone on the ground in that vicinity was a combatant, because they meant to do us harm. In an ambush there are no sidelines and no spectators". This appeared to be a broader justification than earlier ones based on earlier cases of Somali use of women and children as shields for gunmen. The whole story is a warning against using international humanitarian law as a basis for military intervention without thinking through exactly what is to be achieved and how.

5. International Conference, Geneva, August-September 1993

One effort to call for more effective implementation was the International Conference on the Protection of War Victims, held in Geneva from 30 August to 1 September 1993. The most important impetus for this conference was the need to do something about the widespread and flagrant violation of the laws of war in recent conflicts; another was the need for a partial substitute for two International Conferences of the Red Cross and Red Crescent that had collapsed due to problems over the representation of South Africa and of Palestine. Representatives of 160 states attended the Geneva conference. Like the Paris Conference on the Prohibition of Chemical Weapons four years earlier, the Geneva Conference tried to restore the sanctity of battered norms. Virtually all the recommendations in the conference declaration were on the subject of increasing the number of formal adherents to the existing rules of international humanitarian law, and improving the dissemination and practical implementation of those rules. The conference showed that the international community was largely bereft of convincing ideas about the application of international humanitarian law in civil wars. The declaration did require expert level follow-up meetings on implementation, the results of which were to be reported to the next Red Cross Conference and to the UN.

6. Rwanda 1994

The acts of genocide in Rwanda in the first half of 1994 required a response from the UN Security Council. It failed to secure the cooperation of states to take effective action to stop the killings, but in November 1994 it took steps to establish the International Tribunal for Rwanda. This was the first time that an international criminal tribunal had been established in respect of an essentially non-international situation. Successive articles in the tribunal's Statute encompass the crime of genocide; crimes against humanity; and rules of instruments governing non-international armed conflicts - in particular common Article 3 of the four 1949 Geneva Conventions, and 1977 Geneva Protocol II. In addition, many offences are also violations of the law of Rwanda. The establishment of this tribunal constitutes further evidence of a desire to bring massive human rights violations inside states, including in internal armed conflicts, within the ambit of international rules. The fact that the tribunal was established in pursuance of a request from the Government of Rwanda may increase the possibility of its getting sufficient local co-operation to be able to achieve significant results. However, there are grounds for nervousness. When on 8 November 1994 the Security Council debated the resolution establishing the Tribunal, the representative of Rwanda voted against, and expressed his country's dissatisfaction with the arrangements for the tribunal on grounds roughly summarized as follows:

1. Its competence is limited to events of 1994, whereas in fact there had been massacres in Rwanda since 1959, and the genocide there had been long prepared.

2. Its composition and structure is inappropriate and ineffective. It "would only appease the conscience of the international community rather than respond to the expectations of the Rwandese people and the victims of genocide in particular". The sharing of the office of Prosecutor, and the Appeals Chamber, with the International Tribunal for the Former Yugoslavia is unsatisfactory.

3. It risks dispersing its energy in trying not only genocide but also crimes that come under the jurisdiction of internal tribunals.

4. Certain countries which "took a very active part in the civil war in Rwanda " are able to propose candidates for judges and participate in their election.

5. Some of those condemned may be imprisoned outside Rwanda, and the countries concerned may have too much say in reaching decisions about the detainees.

6. Capital punishment is ruled out, whereas it is provided for in the Rwanda penal code.

7. The seat of the tribunal should have been in Rwanda rather than Tanzania.

Many of these objections resulted from fears that the international community, having failed to take decisive action during the time of mass killings, was only taking token action now. The very slow progress of the Rwanda tribunal to date risks confirming these fears, and undermining confidence in the tribunal both in Rwanda and internationally. Kenya's refusal to extradite those wanted by the tribunal, announced on 5 October 1995, added to the doubts.

Internal atrocities, including those in Rwanda, can also be addressed by national legal procedures, even in states not directly involved. This is part of a broader process by which internal atrocities are increasingly seen as international crimes. Thus the 1993 Belgian law on Crimes de Droit International includes provision for Belgian criminal jurisdiction over certain breaches of 1977 Geneva Protocol II, regardless of the nationality of the victim or perpetrator, or of where the offence was committed. In accord with this, on 29 May 1995 the Brussels prosecutors' office issued several international arrest warrants against persons involved in the atrocities in Rwanda.

1. Woodrow Wilson's Dilemma in 1914

The question of whether non-belligerent states should take an active role in seeing to the implementation of treaty-based humanitarian rules is not a new one. Countless such issues arose in two world wars. Such a case arose in August 1914, when the Counselor for the US State Department raised with the Secretary of State the question of whether, in view of the bombardment of Antwerp by a German military balloon, the United States should make a formal protest. He prepared two possible drafts, the first on the limited ground of endangering the lives of American citizens, and the second on the general ground of a violation of 1907 Hague Convention IV:

"The other draft based on the general ground of violation of the usages of civilized warfare would undoubtedly accord with the almost universal indignation expressed by the press of this country, which I believe in this case represents general public opinion. However strong may be the inclination to express abhorrence of such deeds, if we begin to make protests general in nature as to violations of civilized and humane methods of slaughter where are we going to stop"?

One week later President Wilson made a clear decision on this issue:

"My Dear Mr. Secretary:

I have thought a great deal about the matter of protest with regard to the dropping of the bombs and my present judgment is that we do not know in sufficient detail the actual facts and that we ought to be very slow to make formal protests, chiefly because we shall no doubt be called upon by every one of the belligerents before the fighting is over to do something of this kind and would be in danger of becoming chronic critics of what was going forward. I think the time for clearing up all these matters will come when the war is over and the nations gather in sober counsel again.

Cordially and faithfully yours, Woodrow Wilson"

Such an approach, largely influenced by the US anxiety to remain neutral, was hard to justify then, and would be hard to justify today. Yet the sense that making general protests about violations of the laws of war could be a difficult and complex business, full of pitfalls, and easier to start than stop, still rings true today.

A possible weakness of Woodrow Wilson's response is his assumption that when a war is over, matters relating to the laws of war can be cleared up: frequently they are quietly dropped as part of a settlement. Amnesties and waiving of legal claims are still very much part of statecraft. Thus in 1990, when the UK and Argentina restored full diplomatic relations, they dropped all claims between the two arising out of the 1982 Falklands War. In the Russian Federation in 1995, the negotiation of military and political agreements over Chechnya necessitated the Russians dropping their previous call for the arrest of the Chechen leader Dzhokar Dudayev.

2. Successors' Responses to illegal Acts of Previous Regimes

The ways in which violations of fundamental norms in many internal situations are handled, including by successor regimes, illustrates the complexity of the whole subject of enforcement. Sometimes such violations lead to judicial processes, but often they do not.

Terrible crimes occurred in South Africa in the apartheid years, and in many South American states in the 1970s, and Ethiopia in the 1980s. In these countries, successor regimes have taken very different approaches to the question of whether to prosecute. Decisions not to do so are naturally contested, but they reflect some serious considerations: that it is hard to pinpoint individual responsibility and invidious to select out a few for trial; that there were some mitigating circumstances at the time; that the quiescence or positive co-operation of those with overall political responsibility for the offences may be needed if the successor regime is to survive; that those who have lost power, privileges or jobs are perceived by the public to have suffered enough already; or that trials might reopen old wounds.

Similar considerations frequently apply as regards violations of the laws of war, m conflicts between as well as within countries. Decisions to prosecute, or not to prosecute, are frequently the outcome of complex political processes and calculations. We are very far from anything approaching a system where it can be taken for granted that offences will be the subject of trials.

3. One Alternative Vision

There are some grounds for scepticism about the idea that international bodies can apply a criminal law analogy to major violations of international norms. The idea that acts deemed to be crimes can and should be tried and punished in courts is a feature of all national legal systems, but its application on the international plane, as a means of enforcing the laws of war, is problematic. This is not to say that such efforts should not be made, but rather than they should not be seen as the sole or even principal means of implementation. Events in some major conflicts of the past two decades confirm the difficulties of the criminal law approach.

One might well therefore ask: What is the point in international norms, including the laws of war, if there cannot be a vigorous and consistent effort at international enforcement? Alfred Rubin has expressed one alternative vision of what the law is and how it may be implemented:

"International law is not a criminal law system; it is more akin to constitutional law, where enforcement rests on political counterpressures and foreseeable middle- and long-term reactions. A militarily organized movement that commits atrocities is likely to lose allies, unify its enemies, waste its energy in daring strikes of dubious military or political value, and ultimately turn on itself".

While this view of law may itself be tinged with too much optimism, it does highlight the truth that the means by which international norms are upheld are far more complex and wide-ranging than what is provided for in the conventions.

(introduction...)

The demand for effective action to implement the laws of war is likely to remain very strong, and to be very difficult to translate into effective policies. The difficulties of ensuring compliance need to be understood better than they have been if effective action is to be taken. The following conclusions are offered in this spirit.

1. Realist and Idealist Images of the Laws of War

A fundamental question to be addressed is: Why have there been so many cases of massive, cynical violations of the laws of war in recent years, whether by Iraq in its occupation of Kuwait, or in many of the civil wars which have followed the break-up of states and empires?

An explanation may be that the laws of war have come to be seen as too idealistic - neither conforming to the facts of power, nor reflecting the interests of belligerents. Unfortunately, many of the advocates of international humanitarian law may have unintentionally contributed to such negative perceptions. There are dangers in a picture of the law as coming out of Geneva, as a gospel which needs merely to be disseminated and applied in the rest of the world. There may also be dangers in the excessive complexity of some contemporary law, and of the writing about it. It is desirable to see more emphasis on the idea that the law is intensely practical - that it represents, at least in part, a set of deals between professional soldiers, and bargains among states, and that its implementation can have consequences which are for the most part compatible with the interests of belligerents.

2. Still a World of States

For better or for worse, we live in a world of states, and in most cases the laws of war, like other parts of international law, must be implemented through national mechanisms of various kinds: deliberations in government departments, national laws, manuals of military law, rules of engagement, government-established commissions of inquiry, and courts and courts-martial. Often when these mechanisms are employed, it is not immediately obvious that what is at issue is the implementation of the laws of war. This is because an issue may well be formally expressed in terms of a violation of a national law, or of internal military discipline, or of an ethical code which is seen as in some respects national in character.

3. Humanitarianism as a Substitute for Policy

When states and international organizations not directly involved in a particular conflict are moved to demand better application of humanitarian rules in that conflict, they need to be very careful about the manner in which they do so. In particular, if they do not at the same time show some understanding of the positions and aspirations of the belligerents, and fairness in relating principles to complex situations, their efforts may backfire. In respect of the former Yugoslavia, for example, there have been several statements by leading Europeans demonstrating a remarkable lack of comprehension of the depth and seriousness of the conflict. There has been a lack both of serious analyses of the problem and of convincing policies on it. In these circumstances, the emphasis on humanitarian issues can easily seem, or even be, a substitute for policy.

4. Application to Non-International Conflicts

Since 1945, the main form of conflict in the world has been civil war. In some cases, these wars have involved outside armed forces as well. Many developments suggest that we are witnessing a move of opinion in favour of applying a range of rules, including some of the norms of conduct designed for international wars, to civil wars. Since 1945, the development of the idea of "crimes against humanity, the growth of human rights law, and the enactment of 1949 common Article 3 and 1977 Protocol II, have all pointed in this direction. So, more importantly, has much practice of such bodies as the European Court of Human Rights. With the establishment of the International Tribunal for Rwanda in 1994 the UN Security Council has sought to clarify the application of a wide range of international rules to internal conflicts. This process is bound to meet many rebuffs: it will mean little if the belligerents themselves do not get the message; it will be especially hard to apply m those civil wars which involve conflict of neighbour against neighbour, and where the important distinction between soldier and civilian breaks down. Yet the situation whereby civil wars were hardly subject, even in theory, to a body of international rules was an anachronism, the ending of which should not be mourned.

5. Mines

The issue of land mines is one on which the public outcry against inhuman and illegal means of conducting warfare is currently very strong and likely to remain so. In many uses, land mines violate the most basic principles of the laws of war because they kill innocent individuals during wars, and also kill and maim thousands even after wars have ended. There is room for honest disagreement as to whether or not their manufacture and trade can be completely prohibited However, the requirements under the 1981 Weapons Convention; Protocol II, to record the location of minefields, and to assist in their removal, need to be taken more seriously than they have been both during and after wars. The provisions of the Protocol could be strengthened in various ways. The Convention's review conference in Vienna in September-October 1995 failed dismally to produce further commitments on land mines. However, work to add specificity to the Protocol's provisions (especially as regards detectability and deactivation of mines) will certainly continue. So will efforts to improve the Protocol's implementation and to extend its scope of application. As a bare minimum its rules should be made formally applicable, and should be applied, in civil as well as international wars.

6. Limits of Compliance Provisions

The formal provisions for ensuring compliance with the laws of war, which are by no means necessarily the same as the actual processes which induce compliance, include requirements that states should: (1) instruct their armed forces in the relevant conventions; (2) enact any necessary legislation to punish grave breaches of the conventions, and in general to take necessary measures for the suppression of breaches; (3) hand over persons suspected of having committed grave breaches for trial in another state; (4) co-operate in designating states not parties to a given conflict as "Protecting Powers"; and (5) assist the activities of impartial humanitarian organizations such as the ICRC, and, if no Protecting Power system is operating, assist other impartial organizations or neutral states or, in certain cases, the International Fact-Finding Commission.

While aspects of the system crudely outlined above have been of considerable use, much of the compliance system thus outlined has scarcely worked at all, and shows few signs of doing so. It is therefore necessary, even while trying to make states honour their existing obligations, to be receptive to the possibility that much implementation may take a different form from that prescribed in the conventions. This has in fact been the case. A wide range of bodies and mechanisms, domestic and international (including, at times, the European Community/Union and the CSCE/OSCE) has assumed an unanticipated role in the implementation process.

7. Trials

In the pre-1945 history of the laws of war the issue of trials for major violations was not greatly emphasized, and indeed in some early agreements reparations are mentioned more. There was an assumption, overtaken by events of this century, that for the most part the governments of civilized states would be responsible for the implementation of the laws of war. The 1949 Geneva Conventions did make extensive provision for penal sanctions in respect of grave breaches, but these have been little used by states.

These arrangements leave unsolved the acute problem of the state or non-state entity which will not put its criminals on trial, or which indeed is headed by its criminals. This phenomenon, which aggravated the barbarity of some recent wars, especially those with a strong element of communal conflict, has necessitated a new approach to the issue in the form of the International Criminal Tribunal for the Former Yugoslavia. This approach is not likely to yield quick results on anything like the scale of the atrocities to which it is a response, but may in the long run have some small part in the restoration of battered norms. The consequences for the laws of war of raising public hopes and then seeing them dashed could be serious. The tribunal merits support, but at the same time there is a need for understanding of the inherent difficulties of the tasks with which it is entrusted.

The European Union has assisted in the funding of both the Yugoslavia and Rwanda tribunals. It has done so without much public declaration of moral and political commitment, or fully-elaborated rationale as regards the possible contribution of the tribunals to eventual political reconciliation. Both tribunals face extreme funding crises in view of the freezing of expenditures in the UN system. The European Union and its member states should urgently consider increasing their contributions and presenting a modest rationale for so doing. It would do no harm to recognize publicly that, for the kinds of reasons advanced in this study, the tribunals are only likely to have a minor impact on vast problems, and are not necessarily the most important mechanisms even for the limited objective of securing implementation of the laws of war. In particular, the European Union must neither let the International Criminal Tribunal for the Former Yugoslavia starve for lack of funds, nor let it suffocate under the weight of exaggerated expectations.

8. International Criminal Court

The proposal for an International Criminal Court, discussed at the UN for many decades, has attracted new interest in the wake of the emergence of the international tribunals for Yugoslavia and Rwanda. It is likely that the UN General Assembly will in due course approve the text of a treaty establishing such a court. Following such a decision, the process of ratification by states will take time.

Complex questions, including the scope of competence of such a court, have yet to be answered definitively. There is a strong case for such a court being set up initially with a primary purpose of dealing with issues relating to the laws of war. It could, in effect, extend the geographical scope of the two existing tribunals, on the former Yugoslavia and Rwanda. Within the area of the laws of war, its focus must inevitably be on exceptional and grave breaches: such a court should not undermine national responsibility for punishing breaches, nor should it get into the situation of being a "back-seat driver" engaged in continuous ad)udication on all uses of force.

Whatever institutional forms may develop, the important aim should be to create a situation in which political leaders and military personnel pursuing policies that constitute grave violations of the laws of war would always fear the prospect of court proceedings, be it through national, European or international courts. In the system of states as it exists today, many would be able to evade such proceedings by staying in a country sympathetic to their cause. The fear of trial may thus be significantly reduced. The idea of an International Criminal Court is thus not a complete substitute for other approaches to the question of implementation.

9. Reparations

The whole question of post-war reparations as a form of sanction requires careful reconsideration. Reparations, as in the case of those demanded from Iraq since 1991, are usually demanded from a state which has lost a war on the grounds of its responsibility for the outbreak of the war. They thus relate more to ius ad bellum than ius in bello, but may encompass an element of the latter. They have merits, including the fact that they involve a clear concept of state responsibility, and can sometimes be fixed relatively quickly in negotiations between the states concerned.

Is it wise to demand reparations, especially in the extreme form of payment for the entire costs of all damage caused by the war, and in cases where the repayment process would last for decades? The historical precedents of reparations are mixed. If reparations are seen as unfair, or are seen as part of a more general policy of economic strangulation and the cause of complete economic collapse, they may backfire. The case for reparations may be stronger in cases where the sanctions are applied at the same time as a credible assurance is given that they will be lifted as soon certain reasonable conditions are met.

10. The United Nations

Different parts of the UN system have long had an active role in shaping and interpreting laws of war agreements and relating them to particular issues. The General Assembly has in the past been particularly active in this sphere, sometimes producing a particularly heady mix of law and political opinion. Since the mid-1980s, starting in the Iran-Iraq War, the UN Security Council has acquired a role in the implementation of the laws of war which was only dimly foreseen in the conventions, and hardly at all in the UN Charter. It has investigated violations (Iran-Iraq War), imposed reparations (against Iraq over Kuwait), authorized an intervention one of whose purposes was to restore respect for humanitarian law (Somalia), authorized a major use of force to stop attacks on a "safe area" (Bosnia-Herzegovina), and set up international tribunals (former Yugoslavia and Rwanda). Most of these roles have not, so far, been conspicuously effective. They have involved the UN in upholding standards in circumstances in which it is exceptionally difficult to ensure their application, in which its involvement could be counter-productive, or in which the Security Council itself may be the subject of accusations of violations. While the UN's role has certainly contributed much to international awareness of the laws of war, there is a need to temper the optimistic arguments about implementation which were made, for example, at the time of the establishment of the Yugoslav War Crimes Tribunal and the more or less simultaneous proclamation of the "safe areas" in Bosnia-Herzegovina.

11. Barbarians?

Is there a case for reviving old and deplorably ethnocentric distinctions between "civilized" and "barbarian" countries? Not in this form. The problem is not that certain countries per se are barbarian, but rather that they have brutal leaders; or that they are involved in conflicts over state formation, including ethnic conflicts, that by their nature challenge the laws of war. What the old distinction between "civilized" and "barbarian" countries may valuably highlight is the sense that some problems concerning universal implementation of the laws of war are by their nature extremely hard to solve. There are strong and legitimate concerns, particularly in some post-colonial states, that the increased diplomatic attention to international humanitarian standards could have the unintended effect of providing the basis for external intervention, and even a new form of colonialism.

12. A Set of Professional Military Standards?

Despite the many ongoing attempts to strengthen the means of formal international legal redress against major war crimes committed by a state, there remains a strong case for viewing the laws of war as having thus far consisted principally of a set of internationally-approved national professional military standards, backed up by national military and civil legal systems, rather than as a system of international criminal justice. As in the 1990-91 Gulf Conflict, there can be powerful reasons for a state or coalition to apply the laws of war even in the absence of reciprocity by the adversary. Such an approach may contribute to the public acceptance of a military action, to the maintenance of internal discipline within the armed forces, to concentration on action against useful military targets, and to the unity of a coalition. While such a view of the laws of war can never completely displace other views which place more emphasis on reciprocity and on punishment of states violating the rules, it does have the merit of concentrating attention on what is practically achievable within our own societies and their armed forces.

13. Need to Keep Our Own Houses in Order

An important priority may have to be keeping our own houses in order both at the level of our own individual countries, and at the level of regional alliances, including NATO. In recent years the armed forces in many NATO member states have in fact devoted unprecedented attention to laws of war questions. One issue which presents a potential problem concerns the incomplete adherence to 1977 Geneva Protocol I. Now that the UK government is in process of ratifying the 1977 Geneva Protocols, only three NATO states France, Turkey, and USA) remain definitely outside the Protocol. The United States administration, having effectively applied parts of it in the 1991 Gulf War, is now engaged in a careful reconsideration of Protocol I. This is unlikely to lead to rapid US ratification of the Protocol, but it could contribute to the process of effective harmonization in NATO to secure improved common understanding and implementation of the laws of war generally and the Protocol in particular.

14. The Relation between Ius in Bello and Ius ad Bellum

The idea that the laws of war are applicable in armed conflicts irrespective of the righteousness of each side's cause is well grounded in tradition and logic. However, this does not mean that there is no connection between ius in bello and ius ad bellum. Recent practice confirms that observance of the laws of war impinges on the question of justification for war in many and complex ways. For example, the distinction between the US-led coalition and Iraq in the matter of observance of basic norms of conduct contributed strongly to the sense of the reasonableness of the coalition cause. In 1992 the UN Security Council explicitly included violations of humanitarian norms as part of its justification for the intervention in Somalia. In 1995 it was violations of safe areas, including the commission of atrocities by Bosnian Serb forces in Srebrenica and the shelling of civilians in Sarajevo, that led to the UN and NATO joint authorization of "Operation Deliberate Force ". Ultimately it is extremely hard to separate laws of war matters from the broader political issues from which they spring; and the use of military force against persistent violators may be one of the most important forms of implementation.

15. Taking Implementation Seriously

The many failures to find effective means of implementation in respect of violations of the laws of war in the past twenty years, coupled with a high level of rhetoric on the subject, have had deeply damaging effects. They have contributed to a view, quite widespread today, that the laws of war are virtually a dead letter, and can be ignored with impunity. Serious violations in one conflict, publicized but not checked by international reaction, have lowered international standards, making such violations more probable in subsequent conflicts.

A critical intellectual weakness which has seriously affected understanding and implementation of the laws of war is the almost complete divorce between two important schools of thought about security matters in the post-1945 period. On the one hand, theorists of deterrence (a concept not limited to its most extreme form, nuclear deterrence) have shown little interest in the laws of war; while on the other hand proponents of international humanitarian law have had little to say about deterrence of any kind, nuclear or conventional. It is not surprising that France, a country committed to an unusual degree to the idea of deterrence, has indicated that it is not acceding to 1977 Geneva Protocol I because of "the lack of consensus among the signatory states of Protocol I as to the exact meaning of the obligations they have undertaken so far as deterrence is concerned ". The questions of how to deter adversary states from initiating war, and how, during a war, to deter them from violations of the laws of war, are central to international security debates, and indeed to the achievement of humanitarian objectives. Such questions should not be neglected by proponents of international humanitarian law. Some developments in the past twenty years, including the severe attenuation of the right of reprisal in 1977 Geneva Protocol I, risk not only weakening deterrence, but also undermining what may still be one important means of implementation of the laws of war. Although there is bound to be a degree of tension between the idea of deterrence and the laws of war, there are also many areas of intersection between the two approaches. For proponents of the laws of war to neglect such areas is to risk consigning themselves to a position of doctrinal purity and practical irrelevance.

There is an urgent need for a much fuller and more reflective international study and debate on the whole question of implementation. Such a process should involve representatives of states, alliances and armed forces, as well as of the UN, the ICRC and other interested bodies. It should employ other methodologies beside those of law. It should take account of neglected and uncomfortable realities, including the continued role of reprisals, and of the use of force, in preventing or opposing major violations. It should recognize the importance, and also the pitfalls, of the UN's new roles regarding implementation of the laws of war. It should look critically at how well or otherwise particular rules emerge from the test of war, as well as judging the acts of warriors by a legal yardstick. Above all, it should be based on a frank recognition of the enduringly complex realities, the unremarked successes and the conspicuous failures of the implementation process.