Governments frequently collude in the commission of genocide, war crimes, and crimes against humanity. The crimes of the Nazis, the Khmer Rouge, the 1994 “interim government” of Rwanda, factions in the former Yugoslavia, and countless others were committed pursuant to official state policy and authority. This fact creates a fundamental problem for international criminal jurisdiction. It is unlikely that a government that is responsible for the crimes would be efficacious in their prosecution. For that reason, sole reliance on the usual mechanisms of municipal law enforcement would be misplaced in this context. The problem, then, is how to fashion a jurisdictional structure that circumvents obstruction by perpetrator regimes while still maintaining legitimate foundations for the exercise of judicial power.
One of the mechanisms commonly proposed for this purpose is universal jurisdiction. Under universal jurisdiction, the courts of any state may exercise jurisdiction without regard to the territory where the crime occurred or the nationality of perpetrators or victims. The rationale for universal jurisdiction is that crimes such as genocide, war crimes, and crimes against humanity are an affront to humanity and, therefore, are of concern to all states. The problem of governmental collusion in the crimes and resultant unwillingness to prosecute is ameliorated by vesting jurisdiction in all states.1
The attraction of universal jurisdiction is compelling. Law should intervene when innocent human beings are slaughtered, tortured, and subjected to other atrocities. If all states have jurisdiction over the relevant crimes, then at least some perpetrators may be prosecuted some of the time, thereby providing more deterrence, retribution, and condemnation of the crimes, and more incapacitation and perhaps even rehabilitation of perpetrators, than would otherwise exist.
The case for universal jurisdiction would be a strong one if we could assure that prosecutions would be brought and tried impartially and with due process; that the law applied would consist exclusively of the established content of international law; and that relevant national executive organs would hold a veto power over prosecutions, to be used (only) when a prosecution might bring dire international-relations consequences. The problem with universal jurisdiction is that we cannot ensure that these conditions are met. Rather, there is the real risk of prosecutions that are politically motivated; that are carried out without due process; that apply law that exceeds what is universally accepted as established international law; or that are undertaken without sufficient political control to avoid dire consequences on the international plane.
The question that must be asked about universal jurisdiction is whether the potential benefits are worth the risks. Sometimes universal jurisdiction will work well; perpetrators will be duly tried and punished, and the purposes of criminal justice will be served. Sometimes, universal jurisdiction will not work well; defendants will be tried without due process, or in politically motivated, biased proceedings that may themselves exacerbate interstate tensions.
Today, after identifying a number of important but often obscured features of the historical development of universal jurisdiction, I will consider, in the light of that history, the strengths and weaknesses of universal jurisdiction, in relation both to due process and to international relations. My intention is to pose – but, unfortunately, not to dispose of – serious questions to be confronted in fashioning mechanisms for the prevention of the most serious international crimes.
II. The Historical Development of Universal Jurisdiction
Universal jurisdiction has been created in haste over the centuries. Interspersed with long quiescent periods have been flurries of activity during which, in the press of events, facts have been overlooked or exaggerated and flawed analogies have been drawn. Viewed in the most positive light, a discontinuous series of less than careful steps have been taken. Viewed less generously, a vaulting ambition that misrepresents the conditions of existing law has prevailed. So often pressed by matters of compelling moral significance, proponents of universal jurisdiction have sometimes sacrificed circumspection and rigorous analysis. As we shall see, this manner of proceeding has not been without its casualties. The purpose of the present section is to understand better the substantive issues relating to the doctrine of universal jurisdiction by tracing its provenance and development.
The crime generally cited as the original subject of universal jurisdiction is piracy. The definition of “piracy” under customary international law was a matter of notorious ambiguity for centuries. The authors of the Harvard Research in International Law, writing in 1932, concluded that “[t]here is no authoritative definition.”2
There have been, however, some consistent themes in the legal definition of piracy. From its inception, the law of piracy distinguished “pirates,” who operated privately and for private gain, from “privateers” or others commissioned or authorized by states. As stated in the 1958 Convention on the High Seas, and restated virtually identically in the 1982 U.N. Convention on the Law of the Sea:
Piracy consists of any of the following acts:
1. Any illegal acts of violence, detention or any act of depredation, committed for private ends by the crew or the passengers of a private ship or a private aircraft . . . .3
A rather stable feature of the law of piracy has been, then, that the definition of piracy turns critically on its including private acts and excluding the official acts of states. This feature of the legal definition of piracy is significant. By excluding state acts from the definition of piracy, the law of piracy was designed to prevent universal jurisdiction over piracy from becoming a source of interstate conflict. This rationale has been elaborated by Professor Crockett:
[If universal jurisdiction] could be asserted vis-a-vis a State, potentially undesirable consequences are evident. The threat to international peace and stability could be of grave significance if a State whose interests have not been directly infringed sought to punish a State which authorized an act of piracy. . . .
[A]fter balancing the threat of State violence against the danger to international peace from allowing universal jurisdiction against the offending States, it is reasonable to opt for the rule that State acts will not be within the definition of piracy.4
In addition to reducing the potential for sparking interstate conflict, limiting the definition of piracy to private acts also has inhibited the use of universal jurisdiction over piracy as a tool of interstate conflict. The potential for such use was vividly reflected in the 1956 debate on the law of piracy, in the U.N. Sixth Committee, in which the 1958 Convention on the Law of the Sea was negotiated. As reported in the proceedings of a 1957 meeting of the Grotius Society, with Sir Gerald Fitzmaurice presiding:
Clearly, then, there was an element of controversy in the discussion on piracy [in the Sixth Committee debate]. In fact, it was this subject which produced the only heated incident in an otherwise orderly and constructive debate on the law of the sea as a whole. The reason for this becomes plain when it is realised that the representatives who were criticising the view that piracy is essentially a crime “committed for private ends by the crew or the passengers of a private ship or a private aircraft” came mostly from the Soviet Union and the countries associated with it, and that the representatives who took the other view came mostly from the Western powers. Nor was the controversy purely academic. It was alleged by the Soviet Union and its supporters that the activities in the China Sea of the Nationalist Chinese naval forces, aided and abetted by those of the United States, were “piratical” – a point of view which was of course vigorously denied by the spokesmen of the countries concerned.5
The result of those debates, as we know, was to retain the exclusion of all but private acts from the definition of piracy. The law of piracy continued specifically to exclude from the definition of piracy and, so, from the reach of universal jurisdiction, cases involving acts taken pursuant to the authority and policy of states. From its earliest days, the law of piracy has, in this way, minimized the extent to which universal jurisdiction over the crime could become a source or a tool of interstate conflict.6
B. World War II Following the development of universal jurisdiction over piracy and the possible extension of universal jurisdiction to slave trading (a crime, like piracy, pursued for private gain),7 the law of universal jurisdiction changed little until the aftermath of World War II (WWII). In the post-war years, numerous prosecutions for war crimes and crimes against humanity were conducted in both national and international tribunals. Although the jurisdictional bases for some of those trials are ambiguous and, in some cases, controversial, a major development in the doctrine of universal jurisdiction can be traced to this period. Unfortunately, this development was based in part upon faulty reasoning.
The courts that conducted the post-war trials at the national level based their jurisdiction on various combinations of territoriality, nationality, passive personality, protective principle, and universal jurisdiction, often listing several of those bases (and sometimes not specifying a jurisdictional basis at all). At least some of the national courts appear to have exercised universal jurisdiction, whether or not they explicitly so stated. According to the United Nations Law Reports on Trials of War Criminals, published in 1947, “there have been numerous . . . trials by the Courts of one ally of offenses committed against the nationals of another ally or persons treated as Allied nationals, and in many trials no victims were involved of the nationality of the state conducting the trial. . . . ”8
The Monte case is a particularly interesting example in this regard because the defendant was not a national of one of the Axis powers. In the Monte case, the defendant, a Spanish national, was tried before a French military tribunal in Paris for “murder and ill-treatment” of Belgian and Spanish as well as French inmates at a concentration camp in Germany.9
The best known of the post-war trials is probably that conducted by the International Military Tribunal at Nuremberg (“Nuremberg tribunal”). The jurisdictional basis for that tribunal is controversial.10 While the view that the Nuremberg tribunal exercised universal jurisdiction has gained considerable currency, the alternative hypothesis, that the Nuremberg tribunal’s jurisdiction was based on the Allies’ governmental authority within post-war Germany, comports more consistently with the historical evidence.
The four Allied States that established the Nuremberg tribunal had taken on supreme authority in Germany. As stated in the Berlin Declaration of June 5, 1945:
The Governments of the United States of America, the Union of Soviet Socialist Republics and the United Kingdom, and the Provisional Government of the French Republic, hereby assume supreme authority with respect to Germany, including all the powers possessed by the German Government, the High Command and any state, municipal, or local government or authority.11
In that position, the Allies exercised judicial and all other powers of sovereignty in Germany. At a minimum, the Allies, acting in their capacity as the effective German sovereign, consented to the prosecution of German nationals at the Nuremberg tribunal. A maximalist reading would be that the Nuremberg prosecutions were actually an exercise of national jurisdiction by the effective German sovereign, the Allies. As Professor Randall puts it, “the jurisdiction of the [Nuremberg tribunal] and the zonal tribunals arguably arose from the victorious Allies’ assumption of whatever jurisdiction Germany would have had over the specific offenses.”12 Indeed, this is the view reflected in the Judgment of the Nuremberg tribunal, which states: “[T]he making of the Charter [establishing the Nuremberg tribunal] was the exercise of the sovereign legislative power by the countries to which the German Reich unconditionally surrendered; and the undoubted right of these countries to legislate for the occupied territories has been recognized by the civilized world.”13 The jurisdictional basis of the Nuremberg tribunal was not delineated with greater precision than that in the tribunal’s Charter or Judgment.
While there is strong evidence that the Nuremberg tribunal based its jurisdiction on the consent of the Allies as effective German sovereign, the theory that the Nuremberg tribunal based its jurisdiction on universal jurisdiction has attained some credence over the years. The passage from the U.N. Secretary-General’s 1949 Report on the Nuremberg tribunal, from which this theory may have garnered some of its force, begins by quoting the same sentence from the Nuremberg Judgment quoted immediately above. It then goes on to say:
In this statement the Court refers to the particular legal situation arising out of the unconditional surrender of Germany in May 1945, and the declaration issued in Berlin on 5 June 1945, by the four Allied States, signatories of the London Agreement. By this declaration the said countries assumed supreme authority with respect to Germany, including all the powers possessed by the German Government, the High Command and any State, municipal or local government or authority. The Court apparently held that in virtue of these acts the sovereignty of Germany had passed into the hands of the four States and that these countries thereby were authorized under international law to establish the Tribunal and invest it with the power to try and punish the major German war criminals.
The Court, however, also indicated another basis for its jurisdiction, a basis of more general scope. “The Signatory Powers” [the Tribunal said], “created this Tribunal, defined the law it was to administer, and made regulations for the proper conduct of the trial. In doing so, they have done together what any one of them might have done singly; for it is not to be doubted that any nation has the right thus to set up special courts to administer law.” The statement is far from clear, but, with some hesitation, the following alternative interpretations may be offered. It is possible that the Court meant that the several signatory Powers had jurisdiction over the crimes defined in the Charter because these crimes threatened the security of each of them. The Court may, in other words, have intended to assimilate the said crimes, in regard to jurisdiction, to such offences as the counterfeiting of currency. On the other hand, it is also possible and perhaps more probable, that the Court considered the crimes under the Charter to be, as international crimes, subject to the jurisdiction of every State. The case of piracy would then be the appropriate parallel. This interpretation seems to be supported by the fact that the Court affirmed that the signatory Powers in creating the Tribunal had made use of a right belonging to any nation. But it must be conceded, at the same time, that the phrase “right thus to set up special courts to administer law” is too vague to admit of definite conclusions.14 The Secretary General was right to be wary of drawing, from that passage in the Nuremberg Judgment, the conclusion that the Nuremberg tribunal’s jurisdiction was based on either the protective principle (the reference to counterfeiting) or the universality principle (the reference to piracy). Rather, the assertion in the Nuremberg Judgment that, in establishing the Nuremberg tribunal, the Allies had “done together what any one of them might have done singly”15 is equally applicable to a sovereign-consent theory as to a universal-jurisdiction theory of that tribunal’s jurisdiction. Indeed, read together with the passage of the Judgment which states that “the making of the Charter was the exercise of the sovereign legislative power by the countries to which the German Reich unconditionally surrendered,”16 the meaning appears more consistent with the view that the jurisdiction of the Nuremberg tribunal rested on the effective sovereign powers of the Allies to prosecute or consent to the prosecution of German nationals. The Nuremburg tribunal, then, likely was not an instance of the exercise of universal jurisdiction in the post-war trials.
It is significant that, when and to the extent that universal jurisdiction actually was applied in the post-WWII trials, as it appears to have been in some of the trials in national courts, it was conceptualized, often explicitly, as analogous to universal jurisdiction over piracy. A British military court, for example, in its opinion convicting a German defendant for killing a Dutch civilian in the Netherlands stated that, “under the general doctrine called Universality of Jurisdiction over War Crimes, every independent state has in International Law jurisdiction to punish pirates and war criminals in its custody regardless of the nationality of the victim or the place where the offense was committed.”17
Because no specific precedent existed prior to WWII for subjecting war crimes and crimes against humanity to universal jurisdiction, it is unsurprising that the extension of universal jurisdiction to those crimes would have relied in part on analogies to the law of piracy. There was, however, an important flaw in that analogy. While the law of piracy limited the crime, by definition, to acts done for private gain, allegations of war crimes and crimes against humanity frequently concern conduct carried out under official state policy or authority. Universal jurisdiction over war crimes and crimes against humanity, therefore, can become a source and an instrument of interstate conflict, in a way that universal jurisdiction over piracy was designed to avoid.
This flaw in the analogy between universal jurisdiction over piracy and universal jurisdiction over war crimes and crimes against humanity was not acknowledged by those who extended universal jurisdiction to the latter category of crimes in the post-war trials. The significant implications of the flaw in the analogy went unaddressed. Nevertheless, as we shall see, the use of universal jurisdiction in the post-war trials – to whatever extent it was actually used (itself a matter of dispute), and whatever the shortcomings in the reasoning underpinning its use – became precedent for subsequent applications of universal jurisdiction to war crimes and crimes against humanity.
C. The Aftermath of the Aftermath of WWII
The jurisdictional bases of some of the trials following WWII, including particularly the Nuremberg tribunal, have remained ambiguous and, in some cases, controversial. The legal literature in the decades following the war reflected uncertainty concerning the jurisdictional bases of the post-war trials and, relatedly, the legal status of universal jurisdiction.
The Israeli Supreme Court, writing in 1962 in the Eichmann case, considered the range of conflicting views on universal jurisdiction:
One of the principles whereby States assume . . . the power to try and punish a person for an offence is the principle of universality. . . . This principle has wide currency and is universally acknowledged with respect to the offence of piracy jure gentium. But while general agreement exists as to this offence, the question of the scope of its application is in dispute. Thus, one school of thought holds that it cannot be applied to any offence other than the one mentioned above [piracy], lest it involve excessive interference with the competence of the State in which the offence was committed. . . .
A second school of thought . . . agrees, it is true, to the extension of the principle to all manner of extra-territorial offences committed by foreign nationals, but regards it as only an auxiliary principle to be employed in circumstances in which no resort can be had to the principle of territorial sovereignty or to the nationality principle, two principles on which all are agreed. . . .
A third school of thought holds that the rule of universal jurisdiction, which is valid in cases of piracy, is logically applicable also to all such criminal acts of commission or omission which constitute offences under the laws of nations (delicta juris gentium), without any reservation whatever or at most subject to a reservation of the kind mentioned above. This view has been opposed in the past because of the difficulty of securing general agreement as to the offences to be included in the above-mentioned class.
A fourth view is that expressed de lege ferenda by Lauterpacht in 1949 in the Cambridge Law Journal:
It would be in accordance with an enlightened principle of justice – a principle which has not yet become part of the law of nations – if, in the absence of effective extradition, the courts of a State were to assume jurisdiction over common crimes, by whomsoever and wherever committed, of a heinous character.18
The Eichmann court ultimately concluded that the application of universal jurisdiction was appropriate in the case before the court. In so deciding, it reasoned, based on the post-war trials, that “[t]he truth is – and this further supports our conclusion [that Israel is justified in applying the principle of universal jurisdiction in the Eichmann case] – that the application of this principle has for some time been moving beyond the international crime of piracy.”19
Notwithstanding whatever uncertainty may have existed after the war, and even into the 1960s, as to the legal status of universal jurisdiction, the Geneva Conventions of 1949 “codified” the use of universal jurisdiction over war crimes, treating the doctrine as an accepted feature of customary international law.
While not codified in any treaty, universal jurisdiction over crimes against humanity has also been treated as a feature of customary international law. Applying universal jurisdiction to prosecute Eichmann for crimes against humanity, the Supreme Court of Israel relied both upon the precedents created by the post-war trials and also, once again, on an analogy to piracy.20
Genocide, too, has been treated in recent decades as giving rise to universal jurisdiction under customary international law. The 1948 United Nations Convention on the Prevention and Punishment of the Crime of Genocide does not provide for universal jurisdiction. Universal jurisdiction over genocide was proposed but rejected during the negotiation of the Genocide Convention, in view of strong opposition by France, the Soviet Union, and the United States.21 The United States’ stated reasons for opposing provision for universal jurisdiction in the Genocide Convention were that “it would apparently seek to establish a rule of law applicable to nationals of States which have not consented to it, namely, such States as may not ratify the Convention,” and that universal jurisdiction over genocide could become a source of or be used as a tool in interstate conflicts.22 Quite apart from the terms of the convention on genocide, however, genocide has subsequently been treated as giving rise to universal jurisdiction as an accepted feature of customary international law. The Cvjetkivoc court in Austria in 1995, the Jorgic court in Germany in 1997, and the Pinochet court in Spain (in pretrial proceedings) in 1998, among others, have affirmed the use of universal jurisdiction in prosecutions for genocide. By 1987, the Restatement of the Foreign Relations Law of the United States, without qualification, listed genocide (as well as war crimes) as giving rise to universal jurisdiction.23