A review of international criminal jurisdictions


This contribution does not purport to examine the international criminal jurisdictions nor their case law from a technical standpoint. It is more a paper on legal policy. Nearly two decades after the first special international criminal tribunals were established and nearly a decade after the Rome Statute setting up the International Criminal Court came into force, various general questions are posed about them. These questions of principle are important enough for us to observe from the outset that international criminal jurisdictions have not yet secured their place among the galaxy of international organizations nor consolidated the way in which they are instituted.

In what context and why have these instances been set up? What are they all about? Why is there more than one jurisdiction and what characteristics and significant differences arise from comparing them — for it is common knowledge that there is no coherent whole nor any single international criminal law system but on the contrary separate, independent and somewhat disparate jurisdictions? What are the limits or weak points they all share? What use is criminal prosecution and what can be expected of it? Lastly what is to be made of these jurisdictions overall and what possible future is there for them? More broadly what alternatives can be contemplated so they can better fulfil the functions ascribed to them?


This contribution[1] does not purport to examine the international criminal jurisdictions nor their case law from a technical standpoint[2]. It is more a paper on legal policy. Nearly two decades after the first special international criminal tribunals were established and nearly a decade after the Rome Statute setting up the International Criminal Court came into force, various general questions are posed about them. These questions of principle are important enough for us to observe from the outset that international criminal jurisdictions have not yet secured their place among the galaxy of international organizations nor consolidated the way in which they are instituted.

In what context and why have these instances been set up? What are they all about? Why is there more than one jurisdiction and what characteristics and significant differences arise from comparing them—for it is common knowledge that there is no coherent whole nor any single international criminal law system but on the contrary separate, independent and somewhat disparate jurisdictions? What are the limits or weak points they all share? What use is criminal prosecution and what can be expected of it? Lastly what is to be made of these jurisdictions overall and what possible future is there for them? More broadly what alternatives can be contemplated so they can better fulfil the functions ascribed to them?

1. The sources of international criminal instances

We shall be returning to the underpinnings of criminal actions (II below). At this point it is sufficient to observe that organized criminal justice, whether internal or international, originates in the will to maintain or restore peace in the public sphere. The aim is to replace private vengeance by collective prosecution for both individual and collective ends: individual, since justice is done to people who have suffered harm to their persons or property; collective, since protection is afforded to common values that are wrongfully infringed. In the international arena this undertaking does not necessarily involve the law courts and indeed still only does so in exceptional circumstances. In this, international law still lags very far behind municipal law, that of states, or remains very different from it.

The judicial prosecution of international crimes by international criminal instances does not predate the twentieth century. And even in the twentieth century, it only became effective in the aftermath of the Second World War. Admittedly, one of the earliest manifestations of international criminal punishment was Napoleon’s internment on Saint Helena—but that was above all a police measure to detain a tyrant who had disturbed Europe’s quiet and to prevent him from doing any further harm[3]. A hundred years later, article 227 of the Treaty of Versailles provided for resort to an international legal system to judge Emperor Wilhelm II, who was considered responsible for the First World War and for violations of international law in the conduct of it—but having taken refuge in the Netherlands, he could not be arrested[4].

As for the Nuremberg and Tokyo trials[5] that followed the Second World War, while they took on a judicial form that was more marked for Germany than for Japan based on the rights of the occupying powers and on the unconditional surrenders, they remained very much the justice of the victors. This prosecution of crimes by the tribunals of the Second World War remained very incomplete, then. Moreover, punishment was sometimes swifter and more summary—immediate shootings of camp warders, discreet elimination of the leaders, expeditious purges in national settings, etc. Initially both Churchill and Roosevelt thought that a Nuremberg type trial was pointless and would readily have had a number of the leaders shot forthwith[6], given the less than convincing results of the end of the First World War.

It ought to be added that the criminal prosecution has continued over the many years down to the present day in national legal systems, particularly through the action of Nazi hunters[7]. As for the Allies’ war crimes, they never made it to the stage of international legal charges. Despite prosecution being piecemeal and partial, the fact is that the image of the Nuremberg and Tokyo tribunals—especially that of Nuremberg—is a very positive one, if only because of its symbolic scope. For one thing it laid down the principles of international criminal law that have been applied since then, and even enhanced; for another, it fuelled demand for international judicial prosecution of certain crimes that are considered to be international in conflict situations that were very different from the Second World War but also very different from each other—but with the common feature that they were a series of breakdowns of peace and failures of the mechanisms for maintaining or restoring it.

2. The breakdown of peace and the failure to maintain or restore it

The sudden return to favour of international criminal instances dates back, as is widely known, to the conflicts that marked the break-up of Yugoslavia after the disappearance of the USSR and the re-uniting of Germany. These conflicts, while formally international ones insofar as they involved what were recognized to be sovereign states, but also internationalized conflicts through the UN interventions, nonetheless retained typical features of civil wars, especially through the role of paramilitary forces and the direct violence against non-combatants. A comparable situation arose a little later with Rwanda. The cruelty of these conflicts and the involvement of public authorities in the massive infringements of humanitarian law that characterized them entailed the resort to international criminal prosecution of the crimes perpetrated. Many commentators saw this as an advancement of international law and the buttressing of humanitarian law[8].

  • That is an overly optimistic view. One should see this resort more as a triple failure of peace and security. A failure first of the domestic and international mechanisms devised to avert conflicts. Any system for maintaining peace and security—and especially the collective security system organized by the UN Charter—has several functions:[9] a preventive function, a dissuasive function and only finally a corrective function. If the preventive function fails, the dissuasive function kicks in: the coercive threat is supposed to ward off the danger. Both are designed to rule out violence as a means of settling differences, either because preventive diplomacy has worked or because the Security Council has taken the necessary measures to dissuade any use of armed force. In the cases in point, it is plain that these mechanisms did not work.

The failure is next, faced with the violence unleashed, that of the international corrective mechanisms available to the Security Council, which ought to have put an end to the conflicts as promptly as could be done. And finally the failure is one of the domestic legal mechanisms that could have effectively punished crimes committed in this part-international, part-civil context. It is true that in the conflicts in former Yugoslavia and in Rwanda, the powerlessness or even the complicity of the governments and local authorities made such justice illusory.

  • That being so, by setting up the International Criminal Tribunals for former Yugoslavia and then forRwanda[10], the Security Council looked to be asserting its authority and enhancing the range of resources available to it for safeguarding international peace and security. It seemed to be taking back control in matters where it had lacked effective coercive capacity for reasons that need not be developed here. In fact, the Security Council was thus taking note of its ineffectiveness in preventing or promptly halting acts about which something could only be done with hindsight—that is, too late in the day, for law is successful when it prevents offences from being committed, not when it prosecutes the perpetrators.
  • The creation of the International Criminal Court some years later[11] was an attempt to make good this shortcoming in part, since, independently of any existing or ongoing offence, the aim was to put in place a permanent mechanism the role of which ought to have been fundamentally dissuasive. In response to pressure from a very active coalition of humanitarian NGOs[12] supported by public opinion, states constituted by treaty a criminal court whose jurisdiction was not confined to a specifically designated conflict, as with the ICTs, nor defined by an international organ like the Security Council. Beyond this difference in legal method, the foundations of these two types of instance, ICT and ICC, seem very different. This leads to questions about the foundations of criminal law systems—foundations that are common to municipal and international legal systems alike.

3. The three foundations of criminal justice

The foundation of criminal prosecution is a composite thing. It invariably involves law but law that rests on a more basic legitimacy. It always includes several separate, even antagonistic, elements but that are associated in varying proportions. Let us go back to Greek vocabulary and the images or myths that it conveys to present them in a human and not just an abstract manner, with the mixture of sensitivity or even of passion that they contain and not just the rigour of logic. For in the sacred dimension that law always contains, and above all the law that allows punishment to be inflicted, reason must invariably struggle with or come to terms with vindictive and retributive impulses. While it is reason that convinces, only emotion can impel to action. We can distinguish Nemesis, the deepest but also the least rational foundation; Themis, a more rational foundation but with a reason that is the reason of state; and Dike, which allies reason and values, promoting both social and spiritual harmony. In other words, we can differentiate justice as impulse, justice as an institution, and justice as reason—or the sword without the scales, the sword and the scales, and the scales without the sword.


Nemesis is probably the deepest and oldest foundation but also the most rudimentary, irrational and impulsive. The blood of the victims cries out for vengeance, it calls down the ire of their kin and vindicates their direct violence against the guilty. Crime is also an attack on the order of the world as the gods have ordained it and it therefore defies them, and the impulse of vengeance seeks to restore that order by counterbalancing the crime by its punishment. It is even more than that, since in the vendetta an innocent person may be killed to avenge the death of another innocent person—a primitive principle of equivalence or of reciprocity without symmetry. That is the justification of private vengeance, the lex talionis that makes each person their own virtual avenger as much as the proxy of some sort of natural law. Purges, executions without trial, even collective guilt and punishment of an entire group are how it translates—at a push, the US intervention against Iraq: 9/11, a crime of lese-majesty against the heart of the only world power, requires some pay-back, someone must be killed.

We can gauge the drawbacks of such a conception of justice and its dangers where law and order is concerned since private vengeance leads to an uninterrupted cycle of private violence and of war of all against all. Accordingly Nemesis is the anger of the gods of which everyone is ultimately a victim. It has been the impetus behind many international conflicts; it is a permanent threat to law and order. It engenders diffuse insecurity; it destroys social bonds. Justice and peace risk being dissociated and this primitive form of justice invariably includes a seed of violence to come. How can one not see contemporary traces of it for example in some US behaviour after 9/11 with secret prisons, unlimited detention without trial at Guantanamo, recourse to torture, individuals deprived of any status, excluded from humankind as it were?[13]

How can one not see a trace of it, that was particularly unjustified in relation to 9/11, in the circumstances of Saddam Hussein’s trial?[14]


Themis is a qualitative leap corresponding to the strengthening of the public authorities, a strengthening whereof the function of trying the accused is an essential instrument. It looks to go beyond the disadvantages of Nemesis, by imparting to criminal prosecution a proper, impartial, rational and collective character. By substituting organized and abstract punishment for the feelings of individual vengeance, by focusing on the guilty to the exclusion of their kin, the public authority, be it city or state, performs a dual mission of protecting individuals and of upholding law and order. It is a matter of reconciling the personal redress to which the victims or their kin are entitled and maintaining law and order. Both the authority and the impartiality of Themis also maintain or restore the social bond. Naturally, when punishment is public and not private it will be more lenient both than the act that triggered it and than private vengeance, because it is based on general rules and not on the lex talionis. By placing law and a criminal trial between the crime and its prosecution, there is a break in the reciprocal action between crime and punishment.

That is how state justice works in an ideal world. That ideal, however, is not always achieved, far from it. One on side, compromise may lead to a denial of justice. This is what happens when a legal system applies the principle of discretionary prosecution, meaning that the authorities reserve the right not to prosecute for reasons of their own based on public policy. If they deem that prosecution would be more dangerous for law and order than forgetting a serious crime or offence, they are entitled to refrain, thus sacrificing private interest to the public interest. The same is true with amnesties, early releases, or reductions in sentence. On the other hand, judicial errors are not that rare, and they are sometimes spectacular, and the great trials which in the light of history underscore abuse by or errors of the authorities have deeply scarred the public conscience, from the trial of Socrates to the Dreyfus affair. In a more minor and more recent way, many death sentences in the US have been passed on individuals who were subsequently proved innocent[15]. It is even one of the most sound arguments against the death penalty, that a mistake can never be righted.

Even so this is within the framework of the rule of law based on a principle of legitimacy—and criminal jurisdiction is an essential attribute of sovereignty. States put their public power in the service of the legal system, which has advanced forms of investigation and judicial systems, defence rights and rights of appeal. What of the international arena? Is there a public authority that is legitimate enough, impartial enough, organized enough to mete out punishment that is both proper and effective? This is one of the major problems, with its two facets of principle and of technique. It is to be feared that institutions that cannot maintain the peace will have no less difficulty in upholding justice. It is the international criminal tribunals set up by the Security Council that correspond most closely to this attempt to put a public international authority in the service of international justice and to put international justice in the service of peace. The threesome of public authority, criminal prosecution and maintaining public peace is very much present. But is it effective? Is it impartial?


Dike, the most spiritual and abstract form is the search for harmony, for an order of the world that is at the same time an ideal order based on the observance of natural law. It is justice as for Antigone, calling for flawless norms as absolute and universal values, including against the city or state. This vision of justice plays a useful role for example when it denounces judicial errors and secures their redress. It allows criminal law and Themis to be perfected, when for example it leads to the prohibition of torture—being put to the ‘Question’—to the reinforcement of defence rights, and to the abolition of the death penalty. It may imply pardon and reconciliation but it also commands repentance and the duty to remember. It transcends vengeance in the name of higher values than the individual or the group—but it does not go so far as to forget; it sometimes implies the absence of statutes of limitation and leads for example to the permanent hunting down of former Nazis.

It may be argued that international criminal justice is from this standpoint preferable to state criminal justice, because it can judge states themselves or their representatives in the name of universal values; in a sense in the name of humankind. A precedent was provided by the Russell Tribunal[16], a private instance created in the context of the Vietnam War, on which Jean-Paul Sartre among others sat, and which defined itself as the Peoples’ Tribunal, but that was not based on any public authority nor had any basis in positive law[17]. This legal ideology was first embodied by the promotion of norms invested with special authority, norms that could not be transgressed, or norms of jus cogens[18]. It is the opposite feature for feature, in its near-metaphysical dimension, to Nemesis, which is a form of impulse. It joins up with it despite all this on another plane: in the Weberian sense of the term, it derives from the ethic of conviction, of an action without concern for its concrete consequences because it is based on a certain form of higher necessity, whereas Themis is entirely part of the ethic of responsibility, where one weighs the consequences of one’s acts.

With Dike we recognize the inspiration behind the International Criminal Court and the ideology of NGOs that militated for its creation. They thus expressed a degree of mistrust of state justice, of its loopholes and compromises. The immunity of state leaders is dismissed, as much insistence if not more is on the trial rather than on sentencing the guilty, and the death penalty is excluded. But its ideal character cannot mask the weakness of its institutional underpinnings, especially when it is claimed that Security Council intervention can be set aside or reduced[19]. This form of justice may therefore weaken the social bond, although in a very different way from Nemesis, by leaving open wounds that cannot heal. As abstract justice, it lacks public authority. In practice, deviating from reason, it can only work with the support of states whereas its ambition is to circumvent them, or even to steal a march on them. Dike therefore has its limits and contradictions, reflected in the low workload of the ICC and its structural difficulties. But the ICC is but one of the components of an incomplete and disparate galaxy of international criminal instances.

4. The international criminal instances as a work in progress

I speak of an unfinished job of work and a shambles because the constitution of an organized international criminal system is far from over and the components making it up are neither coordinated nor in the course of unification. Each of the instances remains independent, governed by its own statute, with varied jurisdictions. This job of work dates back some twenty years, but it is already a welter of different attempts and unequal achievements. The organizational empiricism behind their creation may be an asset insofar as each tribunal can be adjusted to a mission that is specific to it. It also has its drawbacks, as there is no overall vision to this construction and the creation of each new instance gives the impression that the previous ones were unsatisfactory, that the preventive or dissuasive functions of the prospect of criminal prosecution are not fulfilled, that we are still searching around for an effective system of justice. Five steps can be made out in that search.

  • A first, rather piecemeal, formula was that of the tribunal tasked with judging the perpetrators of the Lockerbie bombing[20]. After the Security Council had taken coercive measures against Libya, which was thought to be behind the attack and which refused to prosecute or deliver up its nationals who were suspected, it was agreed that an ad hoc court, sitting in The Hague and broadly applying the laws of Scotland, where the bombing had occurred, should be set up. It tried and sentenced Libyan nationals. In doing so, it confined itself to a minimal approach, stopping with individual agents without directly implicating the action of the state authorities[21]. The Security Council’s attitude marked with a flourish its condemnation of terrorism after the socialist camp had disappeared and therefore its intervention in the area. The Security Council’s attitude was also characterized by the power of its intervention, as it asked a state to surrender its nationals to foreign instances, by using coercive measures to achieve this. But the Security Council did not—not then—create an international criminal court.
  • The setting-up of the two special criminal tribunals for the former Yugoslavia and for Rwanda followed soon after[22]. This was a qualitative leap, both because of the assertion of the Security Council’s authority to create them on the basis of Chapter VII, that is, by a mandatory decision, and because statutes were established, fixing the basis for incrimination, criminal procedure and the scale of sentences. Both tribunals were designed along the same lines and are subsidiary organs of the Council within the meaning of the Charter[23]. Terrorism was not formally made a crime but the acts it involves can be covered since for terrorism to be amenable to criminal law presupposes that specific acts are identified and condemned. These tribunals have all the characteristics of courts of law, especially as regards the independence of the judges sitting. Their creation is fully part of Themis, a system of justice backed by a public authority with means of coercion and which pursues its ends for the general interest—to contribute to restoring and maintaining law and order, in the case in point international peace and security.
  • The International Criminal Court was then created under entirely different circumstances. It is not an authoritarian response to a concrete situation. It was set up not by a unilateral act of a universal organization but by a multilateral treaty that is formally outside of the United Nations, the Statute of Rome[24]. It is thus accepted by those states that will be subject to its jurisdiction—although this relates to individuals only—and is not imposed on them. It answers, as is known, a more general call from coalitions of NGOs for the establishment of a permanent criminal instance, more independent of states and of the Security Council alike. It is thus closer to Dike than to Themis—universal values of civil society and not of transient and perhaps impure interests of states. But it must come from state commitments and be binding on states. So it was ‘pilot states’[25], notably states steeped in the values of the European Union—of which Commissioner Emma Bonino[26] was the agent and the symbol—that were to join in the movement and see it through. It no more makes any mention of terrorism than do the statutes of the International Criminal Tribunals.
  • The Rome Statute, its entry into force in 2002 and the setting up of the International Criminal Court in The Hague were mostly commended as steps forward, or even as a qualitative leap in international law[27]. But the history of criminal instances does not stop there, however. After the creation of the ICC new special international tribunals were set up, mixed tribunals that one might call of the fourth kind, notably for Sierra Leone and Cambodia[28]. They borrow certain features from the earlier ICTs in that they depend closely on the UN and have a limited and specific subject matter. They also share features with the ICC in that they rely not on coercion but on the agreement of the states concerned. While they do not specifically aim at terrorism, they might even be reminiscent of the Lockerbie tribunal with the important difference that the states sitting in judgment are the national states of the accused and of the victims. They are original as they leave scope for local judges and systems of law so that this outside justice can be appropriated by the populations and the legal systems that are to be protected, maintained or restored.
  • Lastly we observe a certain return to the authoritarian formula of the ICTs, but in the specific framework of a particular terrorist act, the assassination of Rafic Hariri[29], with the Special Criminal Tribunal for Lebanon, which was tortuous to set up and for which there are no guarantees as to whether it will operate successfully. The tribunal illustrates the many difficulties encountered in designing and implementing international criminal prosecution. Attempts were made to set it up by agreement, but it ultimately took an authoritative decision by the Security Council[30]. The investigation into the killing has shown how necessary cooperation among states is and how difficult it can be to secure it. It can be asked whether this ICT truly contributes to international peace because it has rather aggravated internal tensions in Lebanon, made Syrian positions and then those of Hezbollah more radical as they came under suspicion. Is this the domain of Themis or of Nemesis?

5. ICTs, ICC: Opposing rationales, variable effectiveness

The brief and varied history of the creation of international criminal instances denotes a definite wavering, a somewhat erratic empiricism both as regards the modes of creation and the competence and operation of international criminal courts and tribunals. Can one see in this an ultimately rational method of trial and error? Not really since no perceptible advance can be observed whether in the method or in the results. Whether resort is had to a unilateral and authoritarian decision, a multilateral convention, or to a special agreement between one state especially involved and the United Nations, the difficulties remain. However, two opposing rationales transpire: that of the special tribunals, and that of the permanent court[31]. We have seen that they rest on very different perceptions and conceptions of international criminal prosecution. Must a choice be made? Short of this, comparison can be made. Each technique has its advantages and drawbacks—plainly the right formula has not been found—but is there a right formula?

The Special Criminal Tribunals

At first sight, the advantage seems to lie with the Special Criminal Tribunals, and above all with those founded upon a mandatory resolution of the Security Council. Their jurisdiction is incumbent upon states and individuals, municipal courts and non-state actors alike. They are in a position to benefit from the coercive support of the Security Council, which can compel states to cooperate. Their jurisdiction is retroactive, since at the time they are created it is to punish serious international crimes and offences that have already been committed. But this is not an exception to the non-retroactivity of criminal law, which is probably one of the ‘general principles of law recognized by civilized nations’ within the meaning of article 38(1)(c) of the Statue of the International Court of Justice.[32] The offences prosecuted were in fact already defined and sanctioned under international law, if only because, for most of them, they were violations of humanitarian law[33]. Finally, they are constituted for a specific situation, for a limited time, which means they can be adapted to the objectives of criminal prosecution and judicial policy that are to be pursued, but also that they can take account of the constraints of restoring civil and public peace.

The special criminal tribunals set up by the Security Council have functioned, tried and sentenced[34]. They have therefore proved their effectiveness. However, their outcome is far from being entirely positive, both because of what they do and of what they do not do. For what they do, the use of secret mandates[35], in the context of the ICT for former Yugoslavia especially calls into question the rights of the defence. Trials that last too long and come to no specific conclusions, like the Milosevic trial, do not leave the feeling that justice has been done[36]. For what they do not do, the impossibility of securing the handover of the persons to be prosecuted impedes their operation. Witness protection against outside pressure is sometimes patent. Above all, their highly political origin leads to twofold discrimination. For one thing, such tribunals have only been set up in certain specific cases, ignoring other situations that might have justified it—as with certain behaviour related to the conflict in Afghanistan after 2001 in Iraq after 2003, or the Israeli-Palestinian conflict. For another thing the existing tribunals sometimes seem to have concentrated on the vanquished, ignoring the exactions made of the victors—as in the matter of Kosovo, for example.

The International Criminal Court

As for the International Criminal Court, while scholarship has generally welcomed its establishment, it is at the same time the most controversial of the international criminal instances and the least effective of all those in operation. Its permanent status, its institutionalization, its independence of any specific conflict were supposed to be advantages, as were the conditions for bringing a criminal case by the Prosecutor, with a judicial filter[37] to prevent abuse of process. Being permanent and more institutionalized than the ICTs may hold out hope that in the long term the ICC will enjoy greater legitimacy and that its role will develop. It must be added that alongside the ICC as a court there is the Rome Statute as a treaty. This treaty provides that the ICC has only subsidiary jurisdiction, unlike the ICTs[38]. So states parties are invited to establish their own criminal jurisdiction, reinforcing the dissuasive role of the ICC while contributing to the harmonization of criminal laws of states and to the ubiquity of criminal prosecution. Thus, the absence of trials in the ICC is not necessarily a sign that it is ineffective.

  • The shortcomings and limits of the ICC are no less substantial. The first of these is its lack of universality. Neither China, India, Russia nor Israel and especially not the United States are parties to the Rome Statute[39]. Non participation by the US is not merely passive, since the US has undertaken to take up an international legal fight against the ICC, withdrawing its initial signature to the treaty, concluding with states parties that accept bilateral agreements by which they undertake not to surrender US nationals, and even passing domestic legislation authorizing the release by force of any of its citizens held at The Hague, the headquarters of the ICC[40]. Beyond motives clearly related to the presence of US troops on foreign battle fields, and who are therefore particularly exposed to prosecution[41], this must be read as an opposition to the principle that US nationals should be tried before an international criminal jurisdiction. The situation is different when the US supports the creation of the ICTs by the Security Council, since its nationals are then a priori not implicated[42].
  • The other shortcomings and limits specific to the ICC relate to its lack of authority. It is because of deliberate pressure from NGOs backed by small countries that the Rome Statute seeks to keep the Security Council from any involvement in its functioning. The Security Council may admittedly bring actions before the ICC and suspend its action if it deems necessary, but only for a renewable one-year period. The distrust thus expressed does not entail serious legal consequences since the Security Council acts under the UN Charter and is not subordinated to the Rome Statute[43]. It may thus, on the basis of Chapter VII, set aside the jurisdiction of the ICC in a specific situation if maintaining international peace and security so require. It may also pre-empt its jurisdiction by setting up a special ICT. If support from the Security Council is not forthcoming, the ICC counts on NGOs, whose role is officialized within it to trigger investigations, as ‘reliable’ sources[44]. It must above all count on the support of states, subject to many chance conditions and that cannot be obtained in practice other than on a voluntary basis. Under these circumstances—a point we shall return to—it is not surprising that the ICC has not yet in eight years completed any trial, even if several accused persons have been surrendered to it.

6. ICTs, ICC: Common weaknesses

Nearly twenty years on from the establishment of the first International Criminal Tribunals, one cannot help observing a degree of ‘international criminal justice fatigue’ within the international community. Its high operating costs raise questions about whether the money could not be better spent elsewhere[45]. Some commentators think the conditions of detention of the accused or convicted are too mild, and describe a luxury system of justice compared with state systems. The strongest support comes from NGOs and from judicial lobbies of law firms and judges. More broadly this extra-ordinary system of justice is invariably a political system of justice, with all the contradictions the formula may convey. Between political and security considerations, international criminal justice has trouble finding its place and its role. It relies essentially upon law, which is its supreme value, its legitimacy and its founding principle. However, political and security approaches only use international law as an instrument and set it aside where they deem necessary. In this context, the general limits of the international criminal jurisdictions are additional weak points.

  • The limits of jurisdiction of these courts and tribunals are striking. Admittedly war crimes are now an integral part of ICC incriminations, whereas initially they could be set aside by those states parties who so wished[46]. Admittedly a definition of aggression, lack of which meant this crime escaped the ICC, has recently been supplied by the Kampala Conference[47]. But it may be wondered whether it will be operational and if aggression is not above all behaviour that can be ascribed to states and not to individuals. Above all, there is no jurisdiction over criminal behaviour that is detrimental to international peace and security. It can be accepted that ecological questions are of a separate order and that cases should only be brought before criminal jurisdictions if they involve physical violence against people and infringement of humanitarian law. But international terrorism, an international crime by its very nature, is neither defined nor specifically prosecuted outside of specific instances within their framework. But piracy at sea, which is growing, still escapes such prosecution[48], as does hostage taking, which gives rise to negotiations rather than criminal prosecution.
  • Where there is jurisdiction, what of the investigations? In municipal legal systems, they are accorded particular importance since they are used to define charges, identify suspects and collect evidence. The police and the judicial system are called upon to cooperate in this venture, which is expedited and supervised by the judicial authorities, and the coercive resources of the state are made available to them. In this respect the international criminal judicial system is largely deprived of effective resources. The limits in the case in point are not so much judicial as material. Collecting testimony, checking its authenticity and that witnesses are bona fide, protecting witnesses where need be in a context of disturbances, all of this raises difficulty, particularly when the investigators do not necessarily have the cultural empathy for evaluating the various allegations. This human dimension therefore extends further with the qualifications of the investigators. They ought to have adequate training on the subject, which is far from being always the case, especially when the investigators are drafted in from NGOs.
  • Another general limit lies in the criminal procedure implemented in these jurisdictions, and which is mostly an accusatorial procedure of Anglo-American inspiration[49]. Accusation and defence are equals as it were, unlike in the inquisitorial procedure, which is more widespread in Romano-Germanic systems, where the investigation is in principle conducted in terms of both prosecution and defence by an independent judge. In any event, this procedure is highly specific, with the result that the accused are not necessarily familiar with its constraints and do not necessarily master it. It is more difficult for them under these circumstances to ensure their defence effectively, and the principle of being equal with the prosecution is often breached in practice. Added to this is the point that the accused are seldom tried in their own language and that the need for simultaneous translation makes proceedings that much longer and more complicated. All of this contributes to a the system of justice being perceived as independent, yes, but foreign to the facts, foreign to the accused and foreign to the victims. This perception does not make for appeasement, which is invariably the purpose of the trial. In this respect, hybrid or mixed jurisdictions are certainly the better formula.
  • A particularly regrettable consequence of the preponderance of Anglo-American type procedure lies in it being impossible to have judgments in absentia. This formula, which is known in French law for example, means that an accused can be prosecuted even if not present, whenever it has been impossible to lay hold of the person physically[50]. The purpose is to ensure that a crime does not go untried. However, the defence rights of the person prosecuted are not omitted. An attorney can always speak for the accused who may make an appearance before the jurisdiction at any time. If otherwise the person is apprehended later—and a conviction reinforces the application for the person to be surrendered—, the verdict is not enforceable and there must be a retrial, this time in the presence of the accused. It has been considered that there can be no trials in absentia in the international criminal jurisdictions. Yet the advantages would be spectacular: a public trial, substituted for a mere accusation, involving the exposition of evidence by both sides, leading to a judicial decision would give more substance to the applications for the transfer of the people thus convicted, without infringing their rights, since a new trial would be arranged.
  • International criminal jurisdictions have, more generally, clear difficulty in fitting in with an international judicial order. Admittedly there is not in the international order any organized, hierarchized and centralized jurisdictional system under a supreme court capable of harmonizing case law and so contributing to the unity of international law around commonly construed principles. But the International Court of Justice, sitting in The Hague and whose statute is an integral part of the United Nations Charter, appears to be the keystone of classical international law, and arbitration panels generally comply with its case law. International criminal jurisdictions are in another register. They do not settle legal disputes between states; they try and may convict individuals[51]. Their proliferation as independent jurisdictions, applying the same law as the others, carries a risk that international law will become fragmented, since it may be diversely perceived and applied from one court or tribunal to the next. This is all the more likely as these jurisdictions are sometimes attached to promoting conceptions or judicial ideologies that stem more from doctrinal considerations than from states, as for example with respect to the recognition of jus cogens.
  • Another difficulty results from the security constraints that weigh on the functioning of these jurisdictions. Insofar as the objective of criminal prosecution is to restore civil peace and international security, this may lead to waiving the prosecution of suspects because their trial might entail new disturbances and some deeds are best ignored or forgotten. This is the case, for example, if one intends to negotiate with people who might be prosecuted but who are necessary to any negotiation. Impunity may then become a condition for any return to peace. That is the point of the right that the Security Council is acknowledged to have to suspend an action before the ICC—although the Council needs no authorization to exercise an option that the Charter confers on it anyway. Again this is the case when the military support of states is required for peace operations and those states only intend to provide that support if they enjoy criminal immunity. For several years the US thus obtained from the Security Council that its armed forces could not be tried by the ICC in the context of such operations[52].

7. The different purposes of the criminal justice system

The fundamental purpose of any criminal justice system is to maintain or restore law and order, which in international society is called international peace and security. This ultimate objective may lead to emphasis on various aspects, or various moments of criminal prosecution. We return here in another way, from a different angle, to a basic question, that of the foundation of the international criminal justice system. The different aspects or moments it includes invariably co-exist in any criminal prosecution system, whether municipal or international. What are they? Observance of the law, and more generally the violation of norms; the guilty, who are to be prosecuted; the victims, who must have justice; the trial, which must be fair; the prosecution, which triggers the process. In an ideal system all of these aspects are present, fit together and are balanced. Is international criminal justice in a position to respond to this ideal? It has to be observed that while the elements it contains are as rich and complex as those of state legal systems, the ends it can achieve are often partial and so incomplete.

The Law

The first purpose of a criminal justice system is to recall and restore the intangibility and the sacred character of the law which has been violated by the act being prosecuted. In this, prosecution exceeds an individual and subjective context and takes on an objective character that is of public interest. For the international criminal justice system, it is humanitarian law, as mentioned in or referred to by the statutes of the jurisdictions, that is to be consecrated and sanctioned, sanctified almost. The norms that were disregarded then stand at the heart of the trial, in a sort of purifying ritual, a sort of expiatory ceremony. By compelling compliance with them, we are reminded that they are there, the infringement of them is wiped out, and their authority is strengthened. Such an object is common to Themis and to Dike, whether the law be that of the city or of some transcendent absolute. This is also the most general function, which may be ostensible or underlying depending on the case in question.

The Guilty Party

A second object is the guilty party, who is to be punished. The criminal is at the same time caught in the spotlight, exposed and excluded; the punishment is individual and exemplary, repressive and dissuasive all at once. The criminal may prompt a morbid fascination, a hazardous attraction, which the penalty inflicted is intended to erase by the loss of rights imposed on the criminal. It is then the punishment and no longer the guilty party that educates and fascinates, as in the metaphor of the hangman dear to Joseph de Maistre: the hangman is the connection and the horror of human association[53], the sacred terror he elicits maintains the social connection at the same time. The sword of the law has everything to do with Themis. The eye of God is certainly in Cain’s tomb, but the social body is appeased more by a worldly punishment. As the foundation of domestic security policies, the real and visible punishment must make the evil tremble and reassure the good citizens. This security-based conception is the one which as a whole dominates the current-day international criminal justice system.

The Victim

A third object is the victim or those the victim leaves behind. They need to be satisfied by suitable reparation, so as to erase not just the social disorder that any crime produces but the individual prejudices incurred. Criminal justice is then not just vindictive but compassionate. In truth, victims obtain limited compensation rather than any complete restoration of their rights, since the consequences of the crime cannot ever be completely erased. To that extent victims are themselves the source of a new disorder since their protest, their cry for justice cannot be fully appeased. While a necessary component of the trial, the victim is always also an element that interferes. Spontaneously, the victim tends towards Nemesis, which disturbs Themis as much as Dike. As for the guilty party, certain conceptions of justice make him too a sort of victim who is to be rehabilitated as much as punished. However, this approach is foreign for the time being at least to international criminal justice.

The Trial

With international criminal justice, the main object tends to be the trial itself. The important thing is that the crime is not to be left without any official recognition and that, on the contrary, a solemn act should enshrine and purge it after due process that respects the right of the defence. That implies a certain theatralization of the trial, a sort of sacrificial ceremony. The trial then finds an end in itself. Justice is not necessarily satisfied by it, in three different ways. First, the trial may end without a judgment, as with the Milosevic trial, which one tends to forget did not reach a verdict. Then when it is claimed to judge for History, in the name of a duty to remember, of a diffuse and general repentance. Lastly, when symbolic trials are undertaken where an accused is charged not just with his own acts but becomes the scapegoat for something that goes beyond him, to which he only contributed in part and did not instigate. Such changes in course are not rare in the international arena.

Lastly, and here the regression of criminal justice is even clearer here, the object may simply be the accusation. The tendency is then to consider that the important thing is that named, or even unnamed individuals can be accused, that criminal behaviour itself is condemned and held up to public condemnation. It matters little whether or not the trial is actually held in the end; the accusation establishes a true presumption of guilt. As with the previous object, the media dimension means more than the judicial process proper. For want of sufficient investigative resources, for want of means to arrest the accused, things stop at the accusation. It is these two objects rather, trial and failing that accusation, that international criminal justice cultivates. They are the mark of its incompleteness, as illustrated by the difficulties of the criminal tribunal for Lebanon or the impossibility of obtaining the appearance of Sudan’s President Bechir[54]. The possibility of judgments in absentia would provide an at least partial solution to this regression.

The Prosecution

Accusation, though, falls far short of satisfactorily answering all of the hypotheses of the demand for international criminal justice. It suffers from various points of bias and shortcomings that may give rise to new frustration. There are the objective difficulties just pointed out, the difficulty of conducting investigations, of obtaining reliable testimony, of protecting witnesses[55]. Added to this are other constraints that hold for all jurisdictions but are probably more sensitive in the context of the ICC. First, attempted manipulations by political leaders of some states anxious to eliminate rivals whom they accuse[56]. Then the selective character of prosecutions that may be confined to second fiddles so as not to incriminate leaders to whom it is not wished to give offence. Lastly, in the event of prosecution, the very strategy of the prosecution, which chooses the charges, thus defines the scope of the trial to come and may lead to it being restricted for various reasons, whether judicial or other. All of these elements may increase rather than diminish the victims’ frustration and resentment.

8. Mixed outcome, constant need

The overall picture of international criminal jurisdictions is a contrasted one. On the one hand, they exist and so must play some role. The very fact that they do exist means that a certain resistance has been overcome and the idea of criminal prosecution by such jurisdictions has been universally accepted—at least insofar as it is imposed by the Security Council with the ICTs, as acceptance is still a long way off for the ICC. On the other hand, while the need for such jurisdictions is widely felt, the existing ones seem to be one the wane. Perceptions of them are scarcely positive and states are recalcitrant about them. Now, as the French international law scholar Louis Renault emphasized more than a century ago, ‘il ne suffit pas que la justice soit juste ; encore faut-il qu’elle le paraisse’ (cf. the adage ‘justice must not only be done; it must also be seen to be done’). This raises questions about the conditions under which such instances might work effectively. Is it realistic to think that these conditions might be fulfilled in the context of today’s society and international law? Failing this, are there practical alternatives that might offset their weakness and respond to the demand for justice in the face of certain international acts of delinquency?

A Weakened Perception

Perceptions of international criminal jurisdictions are hardly positive. Scholarship, especially European scholarship, remains in favour of them but more as a hope for the future than a tangible reality. NGOs, many of which militated in favour of the ICC, have often become disillusioned. It may be asked whether the dynamic behind these jurisdictions did not correspond to a moment in time, to a privileged decade between the fall of the Berlin Wall and the collapse of the USSR and the 11 September 2001. The institutional and financial means deployed hardly seem in proportion to the results achieved. Added to that is the fact that these jurisdictions are perceived as being instrumentalized by states eager to dissimulate or offset their powerlessness or their actual inaction, as in the case of the former Yugoslavia, Rwanda or Kosovo. It would have been better to prevent or halt the exaction rather than endeavour to prosecute their perpetrators after the event, even if that is preferable to impunity. Lastly, the idea often prevails of de facto discrimination in criminal prosecutions that are aimed above all at African subjects. Is the North undertaking to judge the South in this way?

  • States are increasingly recalcitrant to this type of jurisdiction, and their growing resistance adds to the diffuse perception just recalled. Even those states that are favourable to them hesitate to finance them in a deteriorated economic context. The ICC in particular has run into many difficulties, not the least of which is the attitude of the United States[57]. Is it imaginable that, in the long run, an institution that purports to be universal can subsist without US involvement, or even against the US? Yet it seems unrealistic to expect that the US position of hostility to the ICC will change, particularly as the definition of aggression adopted by the Kampala Conference cannot possibly suit the US. Compounding this is the persistent absence of China, India and Russia. Without necessarily deserving to be compared with a Potemkin village, is the ICC not condemned to become a regional jurisdiction, with the European Union at its heart, in the face of growing distrust of African states, as strikingly exemplified by their reluctance to see the Sudanese President Bechier tried?
  • As for the existing ICTs, they are in principle of limited durations and some are programmed to end[58]. It can be argued that they have performed their missions in the cases of the former Yugoslavia and of Rwanda, but their role is not exhausted and the image of a piecemeal and partial justice remains. Their actual contribution to the return to law and order is not easy to evaluate. The case of the Tribunal for Lebanon stands apart and is rather negative for the time being. While the idea of international judgment of those responsible for the assassination of Rafic Hariri was initially well received, actually achieving it has run into resistance from various sources and into the difficulty of identifying who perpetrated and who commissioned the killing, which underscores the limited means of these institutions. Henceforth it is law and order in Lebanon that is threatened by the prospect of Lebanese nationals being accused. The position of the Tribunal for Cambodia, tasked with trying those responsible for the genocide by the Khmer rouge, is scarcely less fragile and both public opinion and the authorities seem to wish the trials would end in the name of reconciliation and law and order.

A Constant Need of International Criminal Cooperation

And yet, the need for an international criminal justice system has increased in the context of the globalization and transnationalization of many activities. These phenomena generate new forms of delinquency the international dimension of which is plain to see. The regulation they call for should involve suitable criminal prosecution. This is limited for the time being to violations of humanitarian law. But what of ordinary international crime, organized crime, mafias, arms smuggling, drugs smuggling and trafficking in human beings, what of international terrorism and hostage taking, of piracy at sea, wilful pollution, economic and financial delinquency, patent and copyright infringement, and of fraudulent speculation, all of which play on the hermetic character of state systems to enjoy impunity? It is clear that the classical forms of interstate criminal cooperation, notably extradition conventions, are insufficient. Globalization should imply an internationally organized quest to prevent and prosecute such forms of delinquency. In practice, international law only asks the questions it has to resolve, so that the efforts made are piecemeal and lack any initial overall design.

In this way the Security Council has acted only in a partly mandatory way and with limited success in the case of international terrorism, and again by calling for better cooperation among states[59]. It did not choose to create a new specialized international jurisdiction. On what conditions might effective international criminal jurisdictions be set up? They are etched out in the limits and weaknesses of the existing jurisdictions. At bottom, it is not only necessary to provide for clear incriminations but also sufficient means to conduct independent and qualified international investigations, and to apprehend suspects so they can be brought before the courts. That presupposes strong and legitimate public institutions—a super state, ultimately, for criminal jurisdiction is intrinsically tied to a public power which, in the current state of international society, belongs to states alone. In terms of method, it seems futile to expect states to accept to be stripped of this criminal jurisdiction by means of a multilateral treaty. And even the ICTs created by the Security Council concretely need the states to cooperate.

9. Alternatives and possible developments

If we rule out therefore the political-judicial revolution that would be brought about by internationalizing criminal jurisdiction, what are the alternatives that might strengthen the ubiquity of prosecution where it is perceived to be necessary? Alternatives are to be found listed for example in a recent Security Council resolution 1918 of 27 April 2010. This relates to piracy off the coast of Somalia and in adjoining maritime areas. We shall come back to the specific approach to this situation (paras. 53-56). The resolution contains more general lessons above all. It is not based on Chapter VII of the Charter, and is therefore made by way of a recommendation. Paragraph 4 of the resolution ‘requests the Secretary-General to present to the Security Council within 3 months a report on possible options…’[60]. Hence the Council is wondering about the criminal law formulas for internationally regulated prosecution of piracy in the region. It mentions three formulas, marking the emergence of options somewhere between international criminal jurisdictions and more classical criminal cooperation between the states concerned. The Security Council also notes that allowance must be made for ‘existing practice in establishing international and mixed tribunals’: creating special chambers within domestic jurisdictions, a regional tribunal, or an international tribunal. The first two are original and deserve closer attention.

Delegation of Criminal Jurisdiction?

For the specific situation in the Indian Ocean, the immediate origin of the resolution, the Security Council welcomes another option that is not in the final list. It commends the efforts of Kenya and the Seychelles to prosecute people suspected of piracy. In fact, these two states act as it were by delegated criminal jurisdiction of the countries whose vessels have been victims of piracy, substituting their criminal jurisdiction, which resolves intricate questions of criminal procedure, so providing a local solution to an international problem. But this formula is not entirely satisfactory, because the two states in question do no necessarily have the resources to conduct investigations and trials, nor to hold the people convicted, nor any interest in doing so. So third states must put up the finance, despite the risks of corruption, of charges being dropped, or even of escape. The Council shows itself to be somewhat sceptical with respect to the formula which it seems to consider provisional pending more sustainable solutions, since it declares it acknowledges the ‘difficulties’ encountered especially by Kenya.

Special Chambers within Domestic Tribunals and Universal Jurisdiction

The ‘special domestic chambers’ formula looks like an attempt at rational implementation of universal jurisdiction. In respect of piracy at sea, states have an internationally recognized universal jurisdiction, which is doubly different from unilaterally universal jurisdiction laid down by a state’s criminal code. In the latter case, the incrimination is immediately established by a municipal law but is not necessarily accepted by other states, who may fear their nationals will be prosecuted whereas their home states do not prosecute them. Universal jurisdiction recognized by international law cannot be contested by third states, but it is not automatically applicable in domestic law since it must be introduced by appropriate legislation. And again a distinction must be drawn between universal jurisdiction based on custom, which a priori is binding on all, and similar jurisdiction based on treaty, which is binding on the parties to the treaty alone and may include a more specific requirement as to its embodiment in domestic law. This the case, for example of the 1984 Convention on torture[61]. Universal jurisdiction is an attractive formula, but one that is difficult to implement.

In resolution 1918, the Council ‘Calls on all States, […] to criminalize piracy under their domestic law […] consistent with applicable international human rights law’. By adding the idea of special chambers, it opens the way to a legal assistance that could be offered to applicant states to facilitate the creation of such instances. This is an implicit acknowledgment of the concern to rectify the exercise of universal jurisdiction and so to make it acceptable and feasible. From the moment the mechanism exists, one hurdle to its implementation has been removed. Universal jurisdiction prompts much debate and rather hostile debate. Nothing legally prohibits a state from establishing its criminal jurisdiction as it sees fit, but it may seem to be an intrusion into the jurisdictions of other states that may be more concerned by the crimes or offences prosecuted[62]. Conversely, what is the point in prosecuting for a state whose nationals are neither accused nor victims?[63] Thus universal jurisdiction is disturbing whether it is implemented or whether it is not used. If besides it can be triggered by direct complaints from individuals to foreign legal systems, the recipient states may become involved in matters they might prefer to know nothing of.

Regional Tribunals

Another formula suggested by resolution 1918 is that of a regional tribunal. Its creation presupposes agreement among the states in question and therefore an international treaty among them[64]. A priori, the solution has a number of advantages. It does away with the perception of a discriminatory jurisdiction since it relies on the states directly concerned and may be based upon a synthesis of local systems of law. As with mixed tribunals, this formula makes it easier for the societies in question to make the criminal prosecutions their own. It implies shared costs and responsibilities. The drawbacks and difficulties with it are no fewer for all that. Such tribunals will be cumbersome to set up[65], will presuppose a balanced agreement among their sponsor states, may call for outside technical assistance and probably international financial support to run them[66]. Africa seems to be a particularly propitious continent for this type of regional tribunal, which might be placed under the aegis of the African Union. It would then be desirable to extend its jurisdiction to other international crimes. This might prove an elegant response to the continent’s sometimes negative perceptions of existing international criminal jurisdictions[67].


By way of conclusion, it should be recalled that justice must ultimately remain in the service of peace. It must maintain peace objectively, and it is then called security. Its task is also to restore peace subjectively, to repair damage, salve moral injuries and limit resentment. In the name of peace itself, solutions other than criminal prosecution must not be neglected, as alternatives or supplements—reconciliation commissions, symbolic recognition and reparation, but also amnesties, that are always effective when the time is right. The former French monarchy used them successfully after the wars of religion, and then after the confrontations of the French Revolution[68]. There comes a time when historical reason must substitute for legal and even judicial reason. In this spirit, refusing to put a time limit on the prosecution of crime is too stringent a measure and verges on being dangerous. It is better to accept statutes of limitation, so that wounds can be closed, even in the long run—say, fifty or sixty years on. The past cannot hold the future imprisoned indefinitely. The day always comes when one must let the dead bury the dead and we must observe with Louis Aragon that les temps sont déraisonnables où l’on met les morts à table (Bereft of reason are the times when we sit the dead down to dine).

[1] My thanks to Emmanuel Bourdoncle, doctoral student at Panthéon-Assas University, for helping to collect the data and references for these notes.

[2] On international criminal law in general, see Hervé Ascensio, Emmanuel Decaux and Alain Pellet (eds), Droit international pénal, (CEDIN Paris X, Pedone, 2000); Antonio Cassese, Damien Scalia and Vanessa Thalman, Les grands arrêts du droit international pénal, (Dalloz, 2010). See also the case law comments in Annuaire Français de Droit International (AFDI) by Hervé Ascensio and Rafaëlle Maison, passim; the papers by Pierre Hassner, Rafaëlle Maison and Nicolas Michel in Droit international et relations internationales – divergences et convergences, Journée d’étude de la Société Française pour le Droit international (Pedone, 2010), p. 80-94; a series of articles on ‘Un bilan des juridictions pénales internationales’, in Annuaire Français de Relations Internationales (AFRI) 2011, vol. XII. This study concentrates above all on international jurisdictions and not on the applicable criminal law, which raises issues of a different order.

[3] The question of trying Napoleon had arisen, but the plan was clearly a matter of political justice. ‘The idea, suggested above all in England, to have him tried by representatives of all the sovereigns of Europe has something attractive about it; it would be the greatest and most imposing judgment the world has ever seen; the finest principles of the law of nations could be developed by it ... and however things turned out, it would be a great monument in history’, wrote Joseph de Maistre in a letter to the omte de Front on 27 July 1815.

[4] Treaty of Versailles, article 227:

The Allied and Associated Powers publicly arraign William II of Hohenzollern, formerly German Emperor, for a supreme offence against international morality and the sanctity of treaties. A special tribunal will be constituted to try the accused, thereby assuring him the guarantees essential to the right of defence. It will be composed of five judges, one appointed by each of the following Powers: namely, the United States of America, Great Britain, France, Italy and Japan. In its decision the tribunal will be guided by the highest motives of international policy, with a view to vindicating the solemn obligations of international undertakings and the validity of international morality. It will be its duty to fix the punishment which it considers should be imposed. The Allied and Associated Powers will address a request to the Government of the Netherlands for the surrender to them of the ex-Emperor in order that he may be put on trial.

[5] The decision to set up the tribunals was taken at the Potsdam Conference (2 August 1945) among the three main Allies. Notice that their representatives were different in part from the Yalta Conference (Churchill, Roosevelt and Stalin) since Harry Truman represented the US and, at the end, Clement Attlee the UK. The international military tribunal set up in Nuremberg was established by an international agreement among the US, the USSR, Great Britain and France, signed on 8 August 1945. Mandated with trying the principle leaders of the Third Reich, the trials lasted from 20 November 1945 to 1 October 1946. Japanese war criminals were tried by an international military tribunal that sat in Tokyo from 3 May 1946 to 6 April 1948. It was set up by a Charter of the allied forces occupying Japan, introduced on 19 January 1946.

[6] See for example Telford Taylor, The Anatomy of the Nuremberg Trials – A personal Memoir (New York: Alfred A. Knopf, 1992).

[7] These are private individuals, often Jewish (such as Simon Wiesenthal, Serge Klarsfeld, etc.) who from the 1950s, set about tracking down Nazi leaders who had escaped any trial so as to have them arrested and tried. The best known case remains that of Adolf Eichmann, who had been in charge of organizing the extermination of Jews, who was abducted in Buenos Aires on 11 May 1960, and tried in Jerusalem, sentenced to death and executed.

[8] See especially Philippe Weckel, ‘L’institution d’un tribunal international pour la répression des crimes de droit humanitaire en Yougoslavie’ (1993) Annuaire français de droit international, 232-61; Mubiala Mutoy, ‘Le Tribunal international pénal pour le Rwanda: vraie ou fausse copie du Tribunal international pénal pour l’ex-Yougoslavie ?’ (1995) 4 Revue Générale de Droit International Public, 929-54 and Carsten Hollweg, ‘Le nouveau tribunal international de l’ONU et le conflit en ex-Yougoslavie’ (1994) 5 Revue de Droit Public, 1357-97.

[9] Serge Sur, ‘Sécurité collective’ in Thierry de Montbrial and Jean Klein, Dictionnaire de stratégie, (PUF, 2000), p. 305-09.

[10] The International Criminal Tribunal for the Former Yugoslavia was set up on 25 May 1993 by UN Security Council resolution 827. Based in The Hague, it is mandated with prosecuting and trying the persons responsible for serious violations of international humanitarian law in the former Yugoslavia since 1 January 1991.

UN Security Council 955, adopted on 8 November 1994, set up an international criminal tribunal for Rwanda to try persons responsible for serious violations of international humanitarian law committed in Rwanda or by Rwandan citizens in the territory of neighbouring states between 1 January 1994 and 31 December 1994.

[11] The Rome Statute of 17 July 1998, a treaty signed by 120 states, provides for the establishment of the International Criminal Court. The Statute came into force on 1 July 2002. The ICC is a permanent jurisdiction tasked with trying persons accused of genocide, crimes against humanity, war crimes or crimes of aggression.

[12] Jean-Marie Chasles, Le mythe de la démocratie internationale, PhD thesis (Panthéon-Assas University, 2009).

[13] Notice that this behaviour does not comply with Security Council resolution 1368 of 12 September 2001. That resolution observes that the United States is acting in self-defence because of the attacks of 11 September, which is the legal basis for the armed intervention in Afghanistan. But the Council calls on all states to bring to justice the perpetrators, organizers and sponsors of the attacks and that request was not followed up. The supposed perpetrators or organizers, when they could be arrested, were detained without trial, and even subjected to treatment that was not consistent with humanitarian law. Resolution 1368 (3): The Council ‘Calls on all States to work together urgently to bring to justice the perpetrators, organizers and sponsors of these terrorists attacks and stresses that those responsible for aiding, supporting or harbouring the perpetrators, organizers and sponsors of these acts will be held accountable’.

[14] The former Iraqi President was judged by the Special Iraqi Tribunal between 19 October 2005 and 5 November 2006. The trial was punctuated by the resignation and dismissal of several of Saddam Hussein’s lawyers and by the resignation in the course of proceedings of the tribunal’s president. Tried on a single count, S. Hussein was sentenced to death and was executed on 30 December 2006. Cf. Nada Mourtada Sabbah, ‘Quelle justice pour Saddam Hussein ? Aspects de politique judiciaire d’un procès futur’ (2005) 6 Annuaire français de relations internationales, 197-217.

[15] US abolitionists base their combat on criticism of the US court system rather than on moral grounds. The fact is that 128 people (1.75% of convicts on death row) have been released after review of their trials. Many other cases have featured in the media because of serious doubts about the guilt of the accused and the fairness of the trial (an experts’ report in 2009 concluded that Todd Willingham, executed in Texas in 2004, was innocent; three stays of execution of Troy Davis who was sentenced to death in 1991 despite the absence of any material evidence, 7 out of 10 prosecution witnesses retracting their statements and official doubts of some jurors; or Claude Jones, who was convicted in 1990 and executed in 2000, but a DNA test in 2010 invalidated the main prosecution evidence).

[16] A public opinion tribunal established in 1966 by British philosopher Bertrand Russell to ‘try’ the US intervention in Vietnam. It was to give rise to a new ‘tribunal’ on Latin America in the 1970s and then a ‘Permanent Peoples’ Tribunal’ established on 23 June 1979. The formula met with fresh success recently with the creation of the ‘World Tribunal on Iraq’ in 2005 and then the ‘Russell Tribunal for Palestine’ in 2009.

[17] President De Gaulle refused to authorize the Russell Tribunal to sit in France, as Jean-Paul Sartre had asked, arguing that the state had a monopoly over the system of justice.

[18] As part of the protracted debate within the International Law Commission (ILC) on the international responsibility of states, some members wanted to make the violation of these ‘peremptory’ norms ‘an international state crime’, which was not included in the final draft. Fundamentally, the establishment of international criminal jurisdictions to try individuals replaced this incrimination of the state as a juristic person. The question might arise again, though, over the crime of aggression, which, more than individuals, might implicate a state apparatus. See Pierre-Marie Dupuy, ‘Action publique et crime international de l’État – A propos de l’art. 19 du projet de la CDI sur la responsabilité des États’ (1979) AFDI 539-54.

[19] The intention to oppose the principles of the ICC to those of the ICTs set up by the Security Council was patent at the Rome Conference. As the policy instrument of the great powers, the Security Council was thought unsuitable or even dangerous by the partisans of justice with clean hands, especially the NGOs.

[20] On 21 December 1988 a bomb destroyed a Pan Am airliner in flight, which came down on the Scottish town of Lockerbie, killing 259. On 21 January 1992 the Security Council asked Libya to surrender two of its citizens to the UK courts (resolution 731); on 31 March that same year (resolution 748) it suspended air traffic to and from Libya because of its refusal to cooperate, and on 18 November 1993 it adopted resolution 883 taking coercive measures. The accused were finally surrendered to a special tribunal sitting in the Netherlands in 1999, tried in 2000, one of the two being sentenced to life imprisonment before being released in 2009. In the meantime, the evidence presented to the tribunal was seriously called into question.

[21] Although the facts are not of the same order, this matter can be compared with that of the Rainbow Warrior, in which French agents were found responsible for sinking that vessel belonging to the NGO Greenpeace in a New Zealand port. They were prosecuted and convicted individually in that country in the ordinary criminal courts. Alongside this, through an arbitration instance, France recognized its international responsibility and compensated the victims. Libya also accepted to pay compensation for the victims of the Lockerbie bombing—and of a previous bombing of a French-registered UTA aircraft—by means of an agreement with the relevant states. These interstate proceedings were not of a criminal character.

[22] Resolutions 808 of 22 February 1993 and 827 of 25 May 1993, respectively.

[23] The possibility for the Security Council to establish subsidiary organs is laid down in articles 7(2) and 22 of the UN Charter. It has enabled it to put in place peacekeeping operations but also commissions mandated with important missions such as UNSCOM for disarming Iraq and monitoring its disarmament after it invaded and annexed Kuwait (resolution 687 of 3 April 1991). These subsidiary organs are therefore not secondary organs. In the case of the special criminal tribunals, their jurisdictional independence, for example, is fully enshrined.

[24] The Rome Convention was adopted on 17 July 1998. Seven states voted against (Bahrain, China, India, Israel, Qatar, United States, Vietnam). The US and Israel then signed it, and then the US withdrew its signature under the George W. Bush Administration. Jean-François Dobelle, ‘La Convention de Rome portant statut de la CPI’ (1998) AFDI 356-69.

[25] Within the ‘like-minded group’, some sixty states formed the founding core of the International Criminal Court, at the heart of the negotiations that led to the Rome Statute. Led by Canada, the group of ‘pilot states’ rallied almost all of the European Union countries (except for France), countries from Eastern Europe, Latin America, West Africa and Southern Africa.

[26] This Italian politician was EU commissioner for consumer policy and consumer health protection and head of the humanitarian office ECHO from 1994 to 1999 and so deeply involved in the talks leading to the Rome Statute.

[27] Marie-Claude Robergé, ‘La nouvelle Cour pénale internationale : évaluation préliminaire’ (1998) 832 Revue Internationale de la Croix-Rouge, 725-39. The most positive assessments are from NGOs, which were particularly active in creating the ICC. It is helpful to consult the Human Rights Watch July 2008 report Courting History. The Landmark International Criminal Court’s First Years (available on the website) or the regular publications of the Coalition for the International Criminal Court (CICC), which brings together 2500 organizations worldwide. The latest evaluations of NGOs, however, show some reservations, calling for continued efforts to make justice more universal and to make the Court’s more independent of states.

[28] Security resolution 1315, adopted on 14 August 2000, established the Special Court for Sierra Leone, tasked with trying the persons responsible for crimes against international humanitarian law committed in Sierra Leone since 30 November 2006. By a bilateral agreement between Cambodia and the UN of 6 June 2003, after the UN General Assembly had adopted resolution 57/228, Extraordinary Chambers were established, within the Cambodian judicial system, and tasked with trying the persons responsible for the Khmer rouge regime between 1975 and 1979. To this list may be added the specialized sections for international crimes established within the UN administrations in Kosovo (UNMIK) and Timor-Leste (UNTAET).

[29] On 14 February 2005, a suicide bomb attack on Rafic Hariri’s car killed 20 people in Beirut, including the former Lebanese Prime Minister.

[30] The tribunal was established from 2006 as part of an agreement between Lebanon and the UN as called for by Security Council resolution 1664, adopted on 29 March 2006. However, once the agreement was signed, arguing they faced financial and legal difficulties, the Lebanese authorities delayed its performance. The establishment of the tribunal provided for by the agreement was finally imposed on Lebanon by Security Council resolution 1757 of 30 May 2007 on the basis of Chapter VII of the Charter.

[31] Serge Sur, ‘Le droit international pénal entre l’État et la société internationale’, in Le droit pénal à l’épreuve de l’internationalisation (Geneva: LGDJ-Georg-Bruylant, 2002), p. 49-68.

[32] The Court applies ‘general principles of law recognized by civilized nations’. The controversial term ‘civilized’ nations refers generally to organized legal systems, without establishing any hierarchy among civilizations.

[33] Generally, it can be considered that international criminal jurisdictions try serious infringements of humanitarian law, which includes in particular war crimes, crimes against humanity including genocide, and the crime of aggression, which is a crime against peace. But these general principles may contain many nuances and uncertainties, especially in respect of controversies over the customary or non customary character of some of these rules. See for example Ascensio, Decaux and Pellet, Le droit international pénal (note 2 above).

[34] The ICT for the former Yugoslavia has indicted 116 persons. The 113 proceedings completed have led to the conviction of 55 persons, 9 acquittals, 13 referrals to tribunals of former Yugoslavia and 36 cases closed because the charges have been withdrawn or the accused is deceased. With 26 cases being heard and 11 being appealed, the ICT for Rwanda has 34 convictions, 8 acquittals, 2 referrals to national courts and 3 cases ended because charges have been dropped or the accused is deceased. The Special Court for Sierra Leone has indicted 13 persons while the Extraordinary Chambers in the Courts of Cambodia have, for 10 indictments, arrested 5 persons and convicted one.

[35] As provided by article 53 of the rules of procedure and evidence of the ICT for the former Yugoslavia, an indictment may not be made public until served so as to avoid the accused fleeing.

[36] Indicted in 1999 by the ICT for the former Yugoslavia and arrested on 1 April 2001, Slobodan Milosevic was surrendered to the Tribunal in June 2001. His trial began on 12 February 2002, but the former president died before it could be completed on 11 March 2006.

[37] The prosecutor may initiate an investigation of his or her own motion but it is examined by a pre-trial chamber composed of 3 judges (art. 15 of the Rome Statute).

[38] Article 1 of the Rome Statute recalls that the Court is ‘complementary to national criminal jurisdictions’.

[39] 78 UN member states, i.e. about 40%, have not signed or ratified the Rome Statute. This includes notably almost all of the Arab states (except Jordan) and those of southern and South-East Asia (except for Cambodia and Bangladesh).

[40] The American Service Members Protection Act (ASPA), passed by the US Congress in 2002, apart from making US support for UN peacekeeping operations conditional upon impunity being granted to US troops in respect of the ICC, authorizes the President to use ‘all means necessary and appropriate’ to bring about the release of US and allied citizens held by the ICC.

[41] France thus used a clause of the Rome Convention allowing a state exemption from the ICC jurisdiction for war crimes for seven years. The US subordinated its involvement in UN peacekeeping actions to exclusion of ICC jurisdiction and signed various agreements for stationing troops abroad providing for the immunity of private security firms or private paramilitary firms against local criminal prosecutions.

[42] Julian Fernandez, La politique juridique extérieure des États-Unis à l’égard de la Cour pénale internationale (Pedone, 2010).

[43] Charter article 103: ‘In the event of a conflict between the obligations of the Members of the United Nations under the present Charter and their obligations under any other international agreement, their obligations under the present Charter shall prevail’. So one cannot see how a convention outside of the Charter could modify it, or impede or guide decisions of the Security Council, which acts only by virtue of the Charter and pursuant to its provisions. Serge Sur, ‘Vers une Cour pénale internationale : la convention de Rome entre les ONG et le Conseil de sécurité’, (1999) Revue générale de droit international public (RGDIP), 29-45.

[44] Article 15 of the Statute, allowing an investigation to be initiated on the prosecutor’s own motion authorizes the prosecutor to seek further information from states, UN organs, intergovernmental organizations, NGOs or ‘other reliable sources’. By giving a general label of respectability to NGOs, which are of very various origins and inspirations, the Statute exposes the Court to claims and demands the main purpose of which may not be to see that justice is done, to use measured language. This wording is indicative of the weight of NGOs in negotiating the Rome Statute.

[45] The operating budget of the ICT for the former Yugoslavia rose from $39 million for 1994 and 1995 to $376 million for 2008 and 2009 (the forecast for 2010 and 2011 is $290 million). The ICT for Rwanda saw its annual budget rise from $40 million in 1996 to $135 million in 2009 (6.4% of the UN budget). The Special Court for Sierra Leone has an annual budget of $19 million (2003) and $28 million (2009) whereas the special tribunal for Cambodia cost only $5 million in 2009. For its first year of activity, in 2009, the Special Tribunal for Lebanon had a budget of $51 million. The ICC budget for its first year of operation in 2003 stood at $30 million; it rose to $53 million in 2004 and its budget for 2010 is $103 million.

[46] Article 124 of the Statute provides that, upon becoming a party to the Statute, a state may for seven years declare it does not accept the Court’s jurisdiction for war crimes, as France for example has done (cf. note 39 above).

[47] This Review Conference was provided for by the Rome Statute. It was held at Kampala in June 2010. Resolution 6 (11 June 2010) defines the crime of aggression as ‘the planning, preparation, initiation or execution, by a person in a position effectively to exercise control over or to direct the political or military action of a State, of an act of aggression’. The definition of act of aggression takes up that in UN General Assembly resolution 3314 of 14 December 1974. Generally formulated, it is ‘use of force by a State against the sovereignty, territorial integrity or political independence of another State, or in any other manner inconsistent with the Charter of the United Nations’, which reiterates the actual terms of Charter article 2(4), with the clarification of ‘armed’ force. It may be doubted whether it is applicable by a criminal jurisdiction. Michael Glennon ‘Le crime d’agression : une définition sans rime ni raison’ (2010) AFRI 269-303.

[48] See IX below.

[49] This preference for a procedure along Anglo-American lines, even if it is not total, is part of a fight over systems of law that relates to globalization. It is not enough to negotiate international rules that are consistent with one’s interests. It is necessary, and this may even be an alternative solution, to foster imitation of one’s domestic law, to universalize it even, not just by projecting it outside its home country, but via a mimetic effect in foreign domestic systems of law. The US tries to do this by various means. In the case in point, it is successful even though it is not party to the Rome Convention and is guarding against the criminal jurisdiction of the ICC—except when initiated by a Security Council requests as in the Bechir case (see note 54 below).

[50] Even if it is somewhat exceptional, France applies the in absentia procedure. After the disappearance of several of its citizens in Argentina, because the local jurisdictions failed to act, in 2000 it tried and convicted in absentia one Alfredo Astiz, a military officer, for torture and sentenced him to life imprisonment. His trial in Argentina finally began in 2009. In another field from humanitarian law, France has more recently secured the extradition from the United States of Manuel Noriega, former head of state of Panama, who had served his sentence for drug smuggling. Re-tried after his surrender, he was convicted in France in 2010 for money laundering.

[51] The ICJ is an interstate jurisdiction and not a criminal jurisdiction. But it has been called upon on several occasions to adjudicate questions relating to humanitarian law and more generally the exercise of criminal jurisdiction by states. The PCIJ had the opportunity to do so especially in the Lotus case (Judgment of 7 September 1927) between France and Turkey in which it observed the existence of competing criminal jurisdiction between plaintiff and defendant. But applications and judgments on the subject have multiplied in recent years. They concern, for example, alleged violations of the 1948 Convention on Genocide (Judgment of 26 February 2007, Application of the Convention on the Prevention and Punishment of the Crime of Genocide, Bosnia and Herzegovina v. Serbia and Montenegro), of the 1984 Convention on the Prohibition of Torture (Request by Belgium of 19 February 2009 versus Senegal, affair pending of Questions relating to the Obligation to Prosecute or Extradite), questions of the immunity of public authorities (Judgment of 14 February 2002, Arrest Warrant of 11 April 2000, DRC v. Belgium; Judgment of 4 June 2008, Certain Questions on Mutual Assistance in Criminal Matters, Djibouti v. France). These are either violations of basic rules or problems of cooperation in criminal matters between states. However, interference with the case law of international criminal jurisdictions may occur.

[52] This protection was put in place by UN Security Council resolution 1422, adopted on 12 July 2002, modified and renewed by Security Council resolution 1487, adopted in June 2003, which guaranteed immunity for 12 months. It has not been renewed since 2004.

[53] Joseph de Maistre, Les soirées de Saint Pétersbourg – Entretiens sur le gouvernement temporel de la Providence, 1821; (Lyon: Louis Lesne Editeur, 1842).

[54] On 14 July 2008, the prosecutor of the ICC asked for an arrest warrant to be issued against President Omar El Bechir. Two warrants were issued on 4 March 2009, for war crimes and crimes against humanity and then on 12 July 2010, for genocide. That has not prevented the Sudanese President from being re-elected to office, in April 2010, and from travelling in Africa. The African Union considers the warrants are dangerous for the situation in Darfour and has asked its member states not to make any arrest.

[55] These difficulties are patent notably in the case of Thomas Lubanga, the leader of a Congolese militia, whose trial, currently underway in the ICC, has been challenged and postponed several times for doubts about the evidence communicated and pressure on the witnesses. International Herald Tribune, 23 November 2010.

[56] Rightly or wrongly this is so in the case of Jean-Pierre Bemba, whose trial for war crimes and crimes against humanity began in the ICC in November 2010. Former Vice-President of the DRC defeated in the 2006 elections, he has thus been removed from the local political scene. International Herald Tribune, 23 November 2010.

[57] This attitude is part of a general challenge from the US to the interest of multilateralism and of a reserved or even hostile attitude to a number of recent negotiations and instruments in various domains.

[58] By its resolution 1932 (29 June 2010), the Security Council extended for the third time, until the end of 2010, the functioning of the ICT on Rwanda, which should have ended in late 2008. By resolution 1931 of the same day it hoped for the prompt completion of the work of the ICT for the former Yugoslavia.

[59] Especially with resolutions 1373 (28 September 2001) and 1540 (28 April 2004). These two resolutions are based on Chapter VII of the UN Charter; the first contains a general, civil programme for the prevention and prosecution of terrorism; the second gives preferential attention to the risk of proliferation of weapons of mass destruction or their precursors for the benefit of non-state actors liable to perpetrate terrorist attacks. Serge Sur, Terrorisme et droit international, Centre for Studies and Research of the Hague Academy of International Law, (Martinus Nijhoff, Leiden-Boston, 2008), especially p. 44-51.

[60] The Secretary-General filed his report on 26 July 2010 (Doc. S/2010/394). It details seven possible options, which are developments of formulas indicated in resolution 1918. They range from the establishment of a new ICT by a Security Council resolution, to regional intergovernmental cooperation, via a regional tribunal set up by agreement among interested states.

[61] The Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment was adopted on 10 December 1984 by UN General Assembly resolution 39/46. Senegal’s delay in drawing the consequences of the Convention for its domestic law and establishing its criminal jurisdiction on this basis is the source of a dispute with Belgium pending before the ICJ over the trial of Hissène Habré, former head of state of Chad, living in Senegal and facing prosecution in Belgium (note 51 above). Senegal argues that it cannot afford the expense of a trial and requests the support of states and institutions. After various controversies over the costs of trial and its organization, a donor conference was held on 24 November 2010 in Dakar, with announcements of contributions from the African Union, Belgium and the EU in particular.

[62] By a 1993 statute law Belgium attributed universal jurisdiction to its courts to prosecute serious violations of international humanitarian law. After being extended to cover genocide in 1999, the statute was reduced in scope in 2003 further to various international protests and US pressure in particular.

[63] It is generally considered that while a state can establish the scope of its criminal jurisdiction at its own discretion, save its international obligations by custom or by treaty, there must be some connection between it and the offence prosecuted. See the separate opinion of judge Gilbert Guillaume in the Case of the Arrest Warrant of 11 April 2000, ICJ, Judgment of 14 February 2002 (note 50 above). Some of these international rules establish universal jurisdiction as in piracy or slave trading.

[64] Mauritius has taken unilateral initiatives to organize regional cooperation facilitating prosecution, commended by resolution 1918.

[65] Resolution 1918 in its paragraph 4 takes into account ‘the time and the resources necessary to achieve and sustain substantive results’. The same resolutions calls on the states of the region to criminalize piracy under their domestic law (para. 2) and acknowledges the decision of the Seychelles to engage in prosecution and to consider ‘hosting a regional prosecution centre’ (preamble). The resolution also approves the creation of a trust fund administered by the UN Office on Drugs and Crime to ‘defray the expenses associated with prosecution of suspected pirates’ (preamble).

[66] On the basis of resolution 1918, the Secretary-General mandated Mr Jack Lang to prepare a report on the possible options for effective prosecution in this domain. The choice of a Frenchman was justified by the presence of a European naval force, Atalante, in the zone and by France’s involvement in the search for solutions acceptable to the coastal states. After visiting the various countries in the region, he must collect expert opinion and submit his report by the end of 2010.

[67] It may be wondered whether a regional criminal court might not be set up in the context of the European Union, with jurisdiction for international crimes committed against citizens of its member states. That would be an extension of the spirit of the European arrest warrant, and perhaps a model for other continents. Good use could thus be made of the confirmed experience of the EU as regards regional jurisdictions.

[68] This capacity for reconciliation encouraged and sanctioned by public authorities is one of the finest of European traditions. As a root of international peace or civil peace, it is a world away from Nemesis, and may be a form of Dike. Franco-German reconciliation is a recent international example of it. Going further back, in the context of French civil wars, wars of religion and then the French Revolution, one might meditate on the examples of the Edict of Nantes, promulgated by Henri IV in 1598: ‘May the memory of all things that have happened since March 1585 as well as all the foregoing disturbances remain extinguished and lulled as a thing that had not been’ and may the French ‘contain themselves and live together like brothers, friends and fellow citizens’; or the 1814 Charter granted by Louis XVIII: ‘We have erased from our remembrance, as we would wish they might be erased from history, all of the ills that have afflicted the homeland during our absence’. Cited in Roseline Letteron, ‘Le droit à l’oubli’ (1996) 2 Revue de Droit public et de science politique, 385-409.