Superior Orders and Duress as Defenses in International Law and the International Criminal Tribunal for the Former Yugoslavia


The purpose of this paper is to explore the changes in our understanding of the defenses of duress and “superior orders” in international humanitarian law over the past several centuries. International humanitarian law is an attempt to restrict the sovereignty of leaders and to hold accountable those in power for violations of human rights. This paper draws on many renowned discussions of these topics in the literature and in earlier tribunals, while focusing primarily on the case law of the International Criminal Tribunal for the Former Yugoslavia (ICTY) to support conclusions made by the author and to provide a multi-disciplinary perspective on these topics. The author concludes that our understanding of these defenses has swung from one extreme to the other and back again, and will no doubt continue to change as new situations arise. This paper is submitted in memory of the late Honorable Richard May.

Table of Contents: 


    Since the dawn of time, civilizations have sought to create order and enforce various codes and principles for their people. When these cultures clashed, there were few, if any, rules regarding conflict, and there were virtually no efforts by external powers to influence the domestic policies or actions of another region. In the last 100 years, this has begun to change. As nation-states and the idea of liberalism emerged as powerfully guiding forces in the 17th, 18th, and 19th centuries, governments began to give more concern to concepts like due process rights as well as to their judicial systems. After two world wars in the first half of the 20th century, governments tried to enforce some seemingly universal precepts upon the instigators of these horrific and costly campaigns. While many agree that these trials represented little more than a puppet show, it became clear that a body of international law was developing that could actually punish the actions of individuals within governments. Though these initial efforts were not perfect, a consensus was undoubtedly forming: there are certain rules of governance and war, and those who violate these rules need to be punished. The rise of the Iron Curtain and the Cold War stymied the movement toward international law and justice, but it did not kill it. Several notable treaties were signed in the 1960s and 1970s that indicated even the two superpowers, the United States and the Soviet Union (USSR), could agree with each other and work toward a better and more peaceful world.

    The collapse of the Soviet Union in 1991 produced new opportunities as well as new problems for many of the former republics. One very troubled location was Yugoslavia. During the Cold War, President Jospeh Broz Tito was able to quiet the ethnic tensions that had plagued the Serbs, Croats, Muslims, and others in the area for hundreds of years, but after his death in 1980 and a decade of half-hearted and misguided reforms, a power vacuum enabled an intelligent and ambitious psychiatrist named Slobodan Milosevic to seize power and exploit those ethnic tensions for his own gain. The subsequent civil war, lasting nearly a decade in some areas, devastated a great deal of the infrastructure and cultural relics of this historically important and diverse area. Reports of massive human rights violations surfaced and, after several inquiries, the international community agreed to hold those perpetrators liable for their actions by creating the International Criminal Tribunal for the Former Yugoslavia (ICTY). While several infamous actors remain at large, this institution has created a large body of law through its decisions that will influence policy and actions for years to come. Even though some questioned the merit or wisdom of creating such a body while the conflict was in progress, it seems clear that the world wanted to do something to try to ensure these kinds of acts were not allowed to take place again.

    This organ of international law seeks to apply the principles of fairness, due process, and justice to acts committed in the course of war, whether civil or international. But how effective is it and other similar bodies in delivering justice? Are they simply a means of “victors justice?” Are the accused treated fairly? These are all legitimate questions. While certainly these institutions are a substantial boost for human rights around the globe, there are still several problems with them. One of the greatest concerns deals with the applicable body of law. In the Statutes of the ICTY and the International Criminal Tribunal for Rwanda (ICTR), a considerable amount of attention is paid to the prosecution, the judges, and the registry, as well as the enumeration of the crimes within the courts’ jurisdiction. However, the obligations of the accused and legally sanctioned defenses are not included. In spite of the fact that nearly every penal code in the world allows for certain defenses to serve as excuses for crimes, the Statute for each of these courts does not. As one author writes:

    There can be no question that the inclusion of defenses in the sphere of international criminal law constitutes an important stage in the recognition of the principle of fairness in international litigation and therefore the application of human rights standards in international criminal proceedings. (Knoops, 2001, p. 8)

    The question, then, must be asked: what defenses are acceptable in the ICTY and the ICTR and how did they develop that way? What do these defenses mean, and are they supported either by previous decisions in international law or national jurisdictions? This paper will focus exclusively on the defenses of “superior orders” and duress. To adequately answer the above questions in this context, it is imperative that several terms be defined. Following that, the historical trajectory of each of these defenses should be examined to determine how each was originally understood and how that understanding has changed over time, reflecting new opinions and situations. Finally, the defenses as understood at the ICTY will be examined. Some brief conclusions, recommendations, and a discussion of some of the problems that lay ahead will follow, hopefully shedding some light on the future of international criminal and humanitarian law. This paper will explain how these defenses came to be, as well as how they are currently understood.


    First, a meaningful discussion about the laws and rules of a court must describe the composition of that court. In common law systems, there are two parties competing against each other in an effort to come to a final decision. In criminal cases, the prosecution represents the state, not the victim, and brings charges against the accused. After each side presents its arguments, a judge, jury, or both, decide whether the accused is guilty of breaking the law in question. This process is often described as adversarial and uses the notion of precedence to decide many cases. The United States, the United Kingdom, and many other countries use a common law system. On the other hand, civil law systems are closer to a truth seeking organization. This is not to imply that these courts do not levy punishments, but simply that their goal is not retributive. Judges take a more active role and all sides tend to work together. France, Switzerland, and many other European nations employ this type of a system. The ICTY, however, falls into neither of the above categories. Because it was created in a political environment, compromises needed to be made to ensure that all parties would support the creation and adoption of this court. The judges at the ICTY are more engaged than those deciding cases in a common law system, but the process is more adversarial than most civil law systems. Judges and attorneys alike have had to adapt to this unique situation. Like every other liberal justice system in the world, the accused are granted certain rights in Article 21 of the statute (ICTY Statute). Rules of procedure and evidence are laid out in Article 15, and the crimes and conditions of liability are stated in Articles 2 – 7 (ICTY Statute).

    Yet, defenses are not defined nor provided for in the Statute of the Tribunal itself. A defense to a crime “refers to a claim for an acquittal on a matter which the accused must show is in issue” (Scaliotti, 2001, p. 111). There are generally two kinds of defenses: procedural and substantive. Procedural defenses deal with the rules of the court and whether they were adhered to in the case at hand (Scaliotti, 2001, p. 111). Substantive defenses, on the other hand, can be further broken down into standard and affirmative defenses. A standard substantive defense involves a dispute over some facts, like whether the accused had an alibi or whether the prosecution proved its case to the necessary standard (Scaliotti, 2001, p.111). When presenting an affirmative defense, “defendants have the duty to present evidence that would disprove the presumption of criminal responsibility” (Wrightsman, 1998, p. 295). This is distinct from a standard substantive defense, in that it obligates the defense to prove something, rather than simply calling into question some aspect of the Prosecutor’s case. The specific standards for these defenses vary, but most jurisdictions use something close to “the preponderance of the evidence,” or “more reasonable than not.” The ICTY has adopted the standard of “on the balance of probabilities,” a similar notion to those just mentioned. This standard is not as stringent as the level the prosecution must meet (usually, “beyond a reasonable doubt”) but it does create a burden for the defense, though these defenses are not required to be presented. Interestingly, Article 67(1)(i) states that “The accused shall be entitled…not to have imposed on him or her any reversal of the burden of proof or any onus of rebuttal” (ICTY, 2002). While it is possible to interpret this provision to mean that the defense never has a burden of proof, it is more accurate to say that this clause prohibits the inversion of the burden of proof in general terms; it does not state that there could never be a situation where the defense would need to prove a claim. It simply states that the reversal of the burden (from the prosecution to the defense) will not occur carte blanche. Essentially, there is a distribution of the burden of proof to both sides, and the prosecution cannot simply shift their burden on to the defense.

    Specifically, there are several important defenses that have been brought up both domestically and internationally at various times. The first, and probably the most well-known, of these defenses in international and military law is the defense of superior orders. This defense argues that an individual is not responsible for any act he or she commits if he or she was ordered to perform that act by a higher-ranking officer. According to one author:

    Military discipline is founded on complete obedience to superior orders, and it was considered impractical to expect a member of the armed services in conditions of war “to weigh scrupulously the legal merits of the orders received.” (Mitchell, 2000, p. 5)

    Essentially, since militaries are built on a hierarchical structure that is to be rigidly adhered to, lower-ranking members of the armed services do not have the necessary intent or motive required for most crimes. A very closely related concept, though not a defense, is the notion of command responsibility. Command responsibility assumes that a superior officer is able to control the actions of his subordinates and is therefore liable for their actions provided three conditions are met: a superior/subordinate relationship exists, the superior either knows or has reason to know that a crime has occurred or will occur, and the superior must not take any steps to prevent or punish those said acts. Command responsibility is raised as an issue here because of its close association with the defense of superior orders, and an understanding of both of these concepts is necessary for a richer discussion of both topics.

    Another defense commonly associated with superior orders is duress or necessity. Duress requires that a person be put into a situation where he/she, or someone they know, is being threatened with imminent and severe bodily harm if the person does not comply. Again, a person under duress lacks the necessary mens rea or intent to be found guilty of committing a crime because they were simply trying to avoid a greater harm by committing their crime.

    A pair of defenses rest upon a reasonable mistake of some kind being made by the accused – the first is a mistake of fact, and the second is a mistake of law. A mistake of fact occurs when an individual performs a certain action believing one thing when in fact another is true. For example, if a person suspects a robbery is taking place and they attack the “robbers,” who were in fact actors portraying a scene, that person would not be found liable for his/her actions because he/she mistakenly believed that a crime was taking place and sought to prevent it from happening. A mistake of law can occur in three situations: a person does not know that a rule exists prohibiting his action; a person, while knowing that a rule exists, does not think it applies; or a person misunderstands a rule he knows exists, and acts (Scaliotti, 2002, p. B4. This is not normally considered a full defense, since almost every legal jurisdiction assumes that every citizen knows the law of his/her land. Essentially, ignorance is no excuse. Consequently, mistakes of fact and law will not be examined in the bulk of this paper, but will become an issue again in the conclusion section.

    It is important to note before proceeding into the historical development of each of these defenses that not all of these qualify as full defenses; some are simply understood to be mitigating factors in punishment. This distinction will prove to be important when analyzing the position of the ICTY on many of these provisions.

    Historical Development

    Superior Orders 

    While some defenses are relatively new, others have existed for several centuries. “Superior orders” is probably the oldest defense used in military trials. Dating back to the 15th century, Peter von Hagenbach was tried and convicted for mistreating, and permitting those under his command to mistreat, the people of Breisach (Levie, 1991, p.187). His defense was that he was adhering to the orders of his superior, the Duke of Burgundy. Henry Wirz, a captain for the Confederate Army in the American Civil War, made a similar defense against allegations of his mistreatment of prisoners of war (Levie, 186). Wirz stated:

    I think I may also claim as a self-evident proposition that if I, a subaltern officer, merely obeyed the legal orders of my superiors in the discharge of my official duties, I cannot be held responsible for the motives that dictated such orders. (Levie, 1991, p. 186)

    Wirz’s claim was not given much credence and he was sentenced to death.

    Lassa Oppenheim, a well-known authority on international law in the first half of the 20th century, produced nine editions of his comprehensive works. His views, however, changed considerably over time. In his first edition, published in 1906, Oppenheim wrote that “If members of the armed forces commit violations by order of their Government, they are no war criminals and cannot be punished by the enemy….” (Levie, 1991, p. 186) However, by the sixth edition written in 1935, Oppenheim writes:

    The fact that a rule of warfare has been violated in pursuance of an order of the belligerent Government or of an individual belligerent commander does not deprive the act in question of its character as a war crime…[M]embers of the armed forces are bound to obey lawful orders only…. (Levie, 1991, p. 187)

    It is certainly probable that Oppenheim and his editors revised his views after witnessing World War I and the trials following it.

    The Dover Castle case following World War I provided one of the first cases involving the defense of superior orders in the period of modern warfare (Scaliotti. 2001, p. 133). The Dover Castle was a hospital ship carrying the sick and wounded from Malta to Gibraltar. In spite of the fact that this was known as a peaceful ship, it was torpedoed, resulting in the death of everyone on board. The National German court found the sailor who actually launched the torpedo not guilty because, according to German law, subordinates were required to follow all orders from their superiors (Scaliotti, 2001, p. 133). The Llandovery Castle case involved a similar situation where, subsequent to the boat sinking, most of the survivors in the water were shot by the accused. In this case as well, the accused were found not guilty because they acted pursuant to a superior’s orders (Scaliotti, 2001, p. 133).

    The International Military Tribunals in both Nuremberg and the Far East following World War II created the largest discussion of command responsibility and superior orders ever assembled in one place at that time. While most of these trials involved liability stemming from a command position, the development of this body of law directly relates to the defense of superior orders in that it established the guidelines and a rubric for understanding superior/subordinate relationships. And, most importantly, these decisions set a standard that liability can exist on both sides of that relationship. Several notable cases merit attention and discussion.

    The most well-known case from the World War II Tribunals regarding command responsibility was the trial of General Tomoyuki Yamashita at the International Military Tribunal for the Far East (IMTFE). After assuming command of the Philippines in October 1944, he was charged, convicted, and sentenced to death for the murders and rapes committed by troops under his command (Lippman, 2000, p. 142). The tribunal determined that because the crimes were so widespread and systematic, Yamashita either knew and ordered these acts to occur or that he at least condoned them (Lippman, 2000, p. 142). In an oft-quoted statement, the Tribunal declared:

    where murder and rape and vicious, revengeful actions are widespread…and there is no…attempt by a commander to discover and control the criminal acts, such a commander may be held responsible, even criminally liable, for the lawless acts of his troops, depending upon their nature and the circumstances surrounding them. (Lippman, 2000, p. 143)

    This decision was not universally supported. Justice Frank Murphy dissented, arguing that there was no precedent for such a decision in international law and that the court had erroneously ignored the principle of individual responsibility, the very foundation of all justice systems (Lippman, 2000, p. 143). This view was passed over in favor of placing an affirmative duty on commanders to be responsible for and informed of their subordinates’ actions.

    Another significant case was the High Command case before the Nuremberg Tribunal. This case limited command liability to those situations where the commander had either actual or constructive knowledge of wrongdoing by his/her subordinates. In this case, the Court stated that a commander had the right to assume his subordinates were acting lawfully and that criminality is not ascribed based upon a position in the chain of command (Lippman, 2000, p. 149).

    Another important and famous case for the development of the superior orders defense was the trial of Adolf Eichmann in 1961. This case codified the “manifest illegality” principle, which states that a subordinate should disobey all orders that are clearly illegal (Scaliotti, 2001, p. 131). Another particularly relevant case arising from crimes committed during World War II was the trial of Klaus Barbie. In this case, the French Court of Cassation stated that the defense of superior orders is not an excuse, and may not qualify as a mitigating factor for punishment (Scaliotti, 2001, p. 132).

    Essentially, these standards of responsibility allowed defendants to make certain defenses against accusations. A superior could argue that he/she did not know, nor could he/she have known, that these acts were taking place. Similarly, as long as the crimes committed were not so widespread and notorious that they could not be ignored, a commander could be found innocent for his inaction. However, there was still considerable debate as to whether superior orders constituted a defense to charges of war crimes. Drawing from some of the conclusions of the World War II cases, subordinates had an obligation to disobey orders that were manifestly illegal and that following orders may not result in a decrease in punishment.


    Duress as a defense is accepted in nearly every jurisdiction around the world in some form or another. This defense has existed throughout time, yet has a relatively short history in international law. In the cases following World War II, duress was raised several times. First, in the Einsatzgruppen case, the Court found:

    There is no law which requires that an innocent man must forfeit his life or suffer serious bodily harm in order to avoid committing a crime which he condemns. The threat, however, must be imminent, real, and inevitable. No court will punish a man who, with a loaded pistol at his head, is compelled to pull a lethal lever. (U.S. Military Tribunal, 1950, p. 480).

    Other cases, such as Stalag III, Holzer, and Feurstein, took a different approach, understanding that duress is not a defense when an innocent life is taken but not denying that duress is a defense in general terms. This approach has been adopted by many civil law systems, while many common law systems have not required that condition.

    The Masetti case, decided before the French Court of Cassation on appeal, ruled that duress qualified as a defense, and that the defendant should have been acquitted (Scaliotti, 2001, p. 147). The Eichmann case did not raise duress as an issue specifically, but the Supreme Court of Israel did not exclude duress as a defense either (Scaliotti, 2001, p. 147). The Jepsen case, decided by a British court, ruled that duress could qualify as a defense if “the evil threatened the accused was on balance greater than the evil which he was called upon to perpetrate” (Scaliotti, 2001, p. 149). This decision was the first to incorporate a standard of proportionality between the threat levied and the act committed into the consideration of the defense. In the Hostage case before the Nuremberg Tribunal, duress was ruled as a defense, even to murder, when “the threat is imminent, real, and inevitable” (Scaliotti, 2001, p. 149).

    ICTY Interpretations

    The ICTY, not surprisingly, relies upon precedence a great deal, though not exclusively. Because of this use of stare decisis, it was necessary to explain as much of the history behind these defenses as possible in order to better explain the position of the Court. While the Statute states certain possibilities while prohibiting others, these issues will be addressed in the relevant subsections.

    Superior Orders/Command Responsibility

    Article 7 of the Statute of the ICTY (2002) discusses the conditions for individual responsibility. Articles 7.1 and 7.3 deal specifically with command responsibility (as does 7.2, in that it precludes “Head of State” immunity), while 7.4 directly addresses superior orders. Based upon the clear language of the Statute, superior orders are then clearly not a defense at the ICTY, though the fact that the defendant committed his/her acts in response to an order can serve as a mitigating factor for punishment according to Section 4. It is worthwhile to note that with the provisions laid out as they are in 7.1 and 7.3, a superior could theoretically be found guilty under two different statutes for the same act (Jia, 2000, p. 143). If, for example, a superior orders his subordinates to commit a crime against humanity, and then does nothing to prevent or punish it, he would be guilty under 7.1 (planning and ordering) as well as 7.3 (failure to punish or prevent). Something very similar to this in fact happened in the Blaskic case (Nybondas, 2003, p. 69). However, according to personnel at the ICTY, trial chambers usually only convict on one level of liability rather than under both possible frameworks (Harmon, personal interview, May 28, 2003).

    Command responsibility has received a great deal of attention by the court and has produced a considerable amount of literature. The court has been forced to decide issues concerning de facto versus de jure control, civilian versus military superior responsibility, and many others. De facto control exists when a person is able to order others to do things, whether they have an official position of superiority (Lippman, 2000, p. 159). This interpretation comes from the Delalic decision, which represented an important step in assigning blame accurately under command responsibility. While previous wars and tribunals had clear hierarchical structures with very regimented positions of authority that were never in dispute, the chaotic nature of fighting in Yugoslavia, as well as the introduction of paramilitary organizations into armed conflict, presented a troublesome dilemma for the Tribunal. The standard created in Delalic allowed the court to hold people accountable based upon their ability to control the situation, regardless of their rank. Consequently, a sergeant who can make soldiers do what he wants better than a captain would be more responsible because of his heightened level of control despite the former’s lower rank.

    On appeal, an interesting and perplexing point was brought out with regard to the distinction between de facto and de jure control. In Paragraph 197, the court states:

    [T]he possession of de jure power in itself may not suffice for the finding of command responsibility if it does not manifest in effective control, although a court may presume that possession of such power prima facie results in effective control unless proof to the contrary is produced. (ITCY, 1996, Paragraph 197).

    While some argue that this does not reverse the burden of proof, it certainly seems to indicate that if the Prosecution is able to prove that the accused had a title or a position of authority, then there is sufficient evidence for the court to likewise assume he/she had effective control, unless the Defense is able to prove that the accused had no effective (de facto) control (May 2002, p. 370). Hopefully, this issue will be clarified by subsequent decisions, though on further appeal nothing more was said on this matter.

    The court also has ruled that command responsibility is not a standard of strict liability. Again in Delalic, the court delineates between actual knowledge and constructive knowledge. The opinion states that when actual knowledge cannot be established, it can be inferred from circumstantial evidence. Twelve different standards are listed that are presented as kinds of evidence necessary to  prove such an aspect to a crime, ranging from the proximity of the commander to the crimes themselves to the scope and logistics involved (ICTY, 1996b, Paragraph 386). While the standard of “should have known” can be quite problematic, depending upon the interpretation, the Chamber attempted to make the standard as clear as possible by arguing that information of a nature that would cause a superior to open an investigation would be sufficient (ICTY, 1996b, Paragraph 393). Essentially, if the superior, having evaluated the available reports, should reasonably suspect that a crime was committed, is being committed, or will be committed, (s)he has a duty to investigate.

    A further distinction the Tribunal has made is between civilian and military command responsibility. Several cases have dealt with the development of this understanding. In the Celebici trial decision, the court concluded that superior responsibility extends to civilian commanders as well (ICTY, 1996b, Paragraph 393). On appeal, the Chamber clarified its position by explaining that “command” usually refers to people in the military, while “control” can have a broader meaning, including civilian authorities (Suominen, 2001, p. 771). The Aleskovski judgement, meanwhile, states that civilians must have de jure or de facto control and the ability to sanction, probably in the form of cutting off supplies or materials (ICTY, 1995a, Paragraph 78). The court seemingly acknowledges here that a civilian authority, during times of war, will not have the same degree of influence and power over members of the armed forces, but, if he/she is able to punish those forces in some manner for noncompliance, then he/she could be found guilty under command responsibility, specifically the “failure to punish” provision in 7(3). The Kordic case was the first time a political leader had been found guilty of war crimes and crimes against humanity based upon command responsibility (Nybondas, 2003, p. 61). The Chamber wrote in Paragraph 840 of the Kordic decision:

    [T]he Chamber holds that great care must be taken in assessing the evidence to determine command responsibility in respect of civilians, lest an injustice is done. In the first place, it is established that substantial influence (such as Kordic had), by itself, is not indicative of a sufficient degree of control for liability under Article 7(3). (ICTY, 1995b, Paragraph 840)

    This bifurcation is an important one – while it is necessary to use this new avenue of culpability to prevent the commission of future crimes, it is not sufficient to simply prove that a person had substantial influence, but that he/she had “material control.” Material control seems to mean the actual ability to issue orders and have them carried out without unreasonable delay or dispute. The Court actually found Kordic did not have material control, and was not liable under 7(3), but the decision reaffirmed that civilians could be liable.

    Following a superior’s orders has resulted in the mitigation of punishment for a number of people. This precedent was established in the Erdemovic case, along with several other particularly relevant decisions. In the Sikirica case, the Defense argued that Kolundizja would have been imprisoned for fleeing and that he had to follow his superior’s orders (ICTY, 1995c, p. 215), though the Court found his situation to be insufficient to mitigate his punishment. In the Tadic Sentencing Judgement, the court ruled that the leadership roles of individuals can be taken into account as either aggravating or mitigating factors (ICTY, 1996d, p. 1229). Similarly, in the “Mitigating Factors” section of the Vasiljevic case, the court acknowledged that the “relative significance of the Accused in the broader context” could be a mitigating factor, but that was not to mean that lower-ranking people would necessarily receive lighter sentences than those with more control (ICTY, 1998).

    Simply because orders were issued does not mean that the defendant will receive a mitigation of punishment. If the accused delighted in performing his acts and did so without hesitation, the court will show no leniency for this factor ( ICTY, 1996b, p. 1281). In essence, the defense must demonstrate that the defendant would not have committed the said acts if not but for the order of a commander, limiting the accused’s liability and mens rea.


    The defense of duress has become one of the most controversial issues the ICTY has decided upon. Some members of the Tribunal have even expressed reservations and concern over the finding of the Court regarding this issue (G. Mettreux, personal interview, May 20, 2003). Without a doubt, the most applicable case regarding duress at the ICTY is Erdemovic (ICTY, 1995a). A great deal has been written discussing not only how the decision was made, but also what implications it will have on subsequent cases and proceedings. While there is technically a difference between necessity and duress, in that necessity can involve natural occurrences that force a person to commit a certain act while duress implies a situation created by another that forces a person to commit a certain act, the two shall be used equivocally, as is the case in the literature (ICTY, 1996c). Judge McDonald and Judge Vorhah filed a joint, concurrent opinion, Judge Li filing a separate, partially concurring and partially dissenting opinion, while both Judge Cassese and Judge Stephen filed separate dissenting opinions. While each of these opinions warrants discussion, only the joint opinion filed by Judge McDonald and Judge Vorhah (ICTY, 1996b), as well as the dissenting opinion filed by Judge Cassese (ICTY, 1996b) will be analyzed here for the sake of brevity. A short summary of the case will be presented first to allow a better understanding of the arguments and the rulings made by the judges.

    Drazen Erdemovic was a young man who joined the Yugoslavian army to provide for his wife and infant son. He had tried to stay out of the conflict for as long as possible, but was in desperate need of money. Erdemovic allegedly joined a specific unit because it was more ethnically heterogeneous in the hope that this group would be less likely to commit war crimes. However, after a short time, the command and the composition of the group became predominantly Serbian. In July 1995, Erdemovic, along with the rest of his unit, was sent to a farm outside of Pilica, where he was ordered to execute Muslim men from 17 to 60 years in age who were unarmed and had recently surrendered to the Bosnian Serb army. When Erdemovic refused to participate, his commanding officer told him he would be shot along with the others if he did not comply. Reluctantly, Erdemovic took part in the killings, personally murdering between 10 and 100 people that day. Following his leaving the army, Erdemovic turned himself in to the ICTY, pled guilty, and cooperated fully with the Tribunal.

    The sentencing chamber found that duress, while potentially a legitimate defense, was not applicable in this situation, because there was no corroboration for Erdemovic’s claim that he acted under duress (ICTY, 1996a, Paragraph 91). Interestingly, the Court needed no additional evidence to convict him of the crimes he told the Court he committed. The standards for duress, as established in the Krupp case, were that the act threatened must be immediate and serious, that there was no means of escape, and that the act committed was not disproportionate to the act threatened ( ITCY, 1996a, Paragraph 17). Complicating the “disproportionate” standard was the understanding of the Trial Chamber that the life of the accused and the victim are not equivalent when dealing with a crime against humanity, where, tautologically speaking, humanity is the victim (ICTY, 1996a, Paragraph 19). The Chamber concluded that:

    On the basis of the case-by-case approach and in light of all the elements before it, the Trial Chamber is of the view that proof of the specific circumstances which would fully exonerate the accused of his responsibility has not been provided. Thus, the defense of duress accompanying the superior order will…be taken into account at the same time as other factors in the consideration of mitigating circumstances. (ICTY, 1996a, Paragraph 20)

    On Appeal, the Chamber split 3-2 against accepting duress as a defense, instead acknowledging it solely as a mitigating factor in punishment. The majority opinion, presented by Judge McDonald and Judge Vohrah, expressed the view that law “must serve broader normative purposes,” and that the potential for abuses would undermine one of the prime objectives of international law, namely, the “protection of the weak and vulnerable” (ICTY, 1996b, Paragraph 75). McDonald and Vohrah go on to argue that their decision cannot be examined in a purely legal framework, but must be understood as having serious and actual policy implications (ICTY, 1996b, Paragraph 78). They rejected the utilitarian framework adopted by Cassese and the Masetti case, and instead argue that “international humanitarian law should guide the conduct of combatants and their commanders” (ICTY, 1996b, Paragraph 80). While they concede their standard would not be met by an ordinary citizen (ICTY, 1996b, Paragraph 83), McDonald and Vorhah go on to assert that a soldier should be prepared to die given his/her job and that he/she consequently has a stronger resolve than other people (ICTY, 1996b, Paragraph 84).

    Cassese’s dissenting opinion is truly different in many respects. Cassese accepts the notion of duress as a complete defense, offering four standards for its application. Three of these are the same as those described above (immediacy, no recourse, and proportionality), but the fourth criterion states that “the situation leading to duress must not have been voluntarily brought about by the person coerced” (ICTY, 1996c, Paragraph 16). Essentially, this standard was subsumed in earlier understandings of duress in international law – it is not a very compelling argument to say that a person was forced to do something after he/she willingly put himself in a position to be forced. After analyzing the cases set forth as examples by the prosecution for the invalidity of duress as a defense, Judge Cassese moves on to analyze the Llandovery Castle case, the Eichmann case, and other previously discussed cases of precedent for duress. Interestingly, and undoubtedly because of his background and hence familiarity with them, Cassese discusses the relevance of three cases decided in Italian military trials reaffirming the legal basis for duress (ICTY, 1996c, Paragraph 35).

    Cassese spends a great deal of time analyzing the standard of proportionality, as well as its difficult applicability in cases of this nature. He seems to bear a similar reluctance demonstrated by Judge McDonald and Judge Vorhah to wholeheartedly endorse a purely utilitarian standard for duress, but because the present case involved a situation where the victims were going to die whether the accused participated in their execution, he seemed more willing to admit the need for it. Cassese even writes specifically about the difficulty a court would face in trying to weigh the life of the accused against that of the victim. He states:

    The third criterion – proportionality will, in practice, be the hardest to satisfy where the underlying offence involves the killing of innocents. Perhaps…it will never be satisfied where the accused is saving his own life at the expense of  his victim, since there are enormous, perhaps insurmountable, philosophical, moral and legal difficulties in putting one life in the balance against that of others in this way: how can a judge satisfy himself that the death of one person is a lesser evil than the death of another? (emphasis in original) (ICTY, 1996b, Paragraph 42)

    While concluding just as the majority of the Appeals Chamber had that there was no specific rule applied to duress as a defense to a crime involving the murder of innocent civilians, Cassese instead decided to apply the general rule that duress was a defense to this case in particular (ICTY, 1996b, Paragraph 41). Judge Cassese seemed quite willing to allow future courts to apply the rule of duress where necessary, in a case-by-case approach, much like the Trial Chamber had decided. As he wrote in Paragraph 42:

    Where it is not a case of a direct choice between the life of the person acting under duress and the life of the victim – in situations…where there is a high probability that the person under duress will not be able to save the lives of the victims whatever he does - then duress may succeed as a defense. Again, this will be a matter for the judge or court hearing the case to decide in light of the evidence available in this regard.…The important point, however…is that this question should be for the Trial Chamber to decide with all the facts before it. The defense should not be cut off absolutely and a priori from invoking the excuse of duress by a ruling of this International Tribunal….(ICTY, 1996b, Paragraph 42)

    Cassese, like McDonald and Vorhah, seemed to be very concerned with possible abuses of this defense, but he deemed its use necessary to set the bar very high to prevent misapplication of the rule (ICTY, 1996b, Paragraph 43). A point that seemed especially relevant to Cassese was the fact that the crime was going to occur whether the accused committed the act or not; it is unclear whether this defense would have received support from him had this not been the case (ICTY, 1996b, Paragraph 43b). Cassese also acknowledged the relevance of the rank of the officer and the actions of the accused after the disputed act (ICTY, 1996b, Paragraphs 45, 46). Cassese summarizes his point quite clearly when he writes in his conclusion:

    Law is based on what society can reasonably expect of its members. It should not set intractable standards of behavior which require mankind to perform acts of martyrdom, and brand as criminal any behavior falling below those standards. (emphasis in original) (ICTY, 1996b, Paragraph 46)

    The similarity to the Einsatzgruppen case’s rhetoric is surely not accidental.

    Various commentaries have made similar analyses of this case and the decision-making that took place in it. One alleges that it seems the judges in the majority decision already had a ruling in mind, and simply looked for case law that would support that view while eschewing precedence to the contrary (Newman, 2000, p. 163). While this assertion may be true, it seems to be more of an ad hominum attack rather than a legitimate indictment, since many judges probably have opinions on various issues, particularly when the case law is as divided as it was in the present situation. The same author goes on to later argue that Judge Li, the deciding vote, “took the easy way out,” instead of thoroughly analyzing the case law and the arguments presented (Newman, 2000, p. 169). Another author took a more nihilistic approach when arguing that neither ruling would prove to be acceptable or easy (Rowe, 1998, p. 213).


    Obviously, there are many more applicable defenses in international law than superior orders and duress. The ICC Statute actually includes several others that were not discussed in the present effort, and there are many procedural defenses that warrant equal attention. A great deal more could easily be written, even on just these two selected defenses, than has been. Several issues and problems with the current statutes and decisions exist that need to be either clarified or rectified. First, despite the amount of discussion it has received and even the enumeration of many different possible sources, there remains some question as to what the “should have known” standard actually means for those officers in positions of command. Epistemologically speaking, if someone has information that should lead them to know a given fact, what does he/she actually know? If a person knows that something should allow them to know something else, he/she would have to know the final conclusion in order to correctly make that inference. If this is the case, there is potentially no difference between “know” and “should have known.” Additionally, because the full picture can be seen in hindsight, it is very easy to say that someone should have known something based upon several pieces of available evidence, but there is a distinct possibility that people who are in command positions and who do not know specifically the actions of their subordinates could be found guilty for war crimes for simply being unable to infer what the court did from the given information.

    Second, while the “manifestly illegal” standard is not technically within the Statute of the Tribunal, if superior orders along with a combination of the defense of mistake of law were presented, it seems likely that the court would resort to that standard. However, this standard has substantial problems with it, as demonstrated by Hannah Arendt’s observations of Adolf Eichmann and her conclusions about the moral character of soldiers and even civilian bureaucrats in a situation of “administrative massacre,” not unlike the situation in Yugoslavia or Rwanda (Osiel, 2001). The fact that seemingly innocuous and otherwise virtuous people can willfully participate in the calculated and systematic murder of hundreds, sometimes thousands, of people can be quite shocking to many people. Sociological research in the 1960s and 1970s conducted by Stanley Milgram and Philip Zimbardo indicate people’s willingness to adhere to the orders of an apparent authority with little regard to their own moral codes. Whether this represents a defect of character is unclear, but it seems from the case law as well as the research that anyone could potentially do even the most horrific things and rationalize his/her actions by stating that he/she was just doing their job. In this situation, the order would not be manifestly illegal to the individual, nor would it be deemed manifestly illegal to others if these findings are valid for the majority of the population. It is very easy for judges and attorneys who have never had to confront these decisions or situations to pass judgment on others, particularly after the fact and with the backing of some of the world’s most powerful governments and individuals, but it is quite a different matter to try to place oneself into the position of the accused and decide honestly whether he/she would have acted differently (Osiel, 2001, pp. 35, 36). Instead, what is possible, maybe even probable, is that the breakdown of the customary sociopolitical and moral order that dictates the proper course for interactions between people is indeed the cause of these seemingly inexplicable crimes. In such a chaotic environment, the human mind, seeking to find order in a world without any, accepts whatever propositions are put forth without internal or reflective consideration because by doing so it can find a pattern within which to operate, as well as goals, norms, and expectations to follow. If this is in fact the case, it seems likely that prosecuting these kinds of people would serve little use and in fact may detract from the productive or efficient functioning of a society. These people were not “evildoers” before their war crimes, nor would they likely become criminals after order was restored. Truly, they were probably model citizens before the breakdown of their society and would be very likely to do the same after the reestablishment of order. If these propositions have merit, international as well as domestic law has a considerable task before it in figuring out what to do in these situations. Certainly, there are sociopaths and sadists who revel in torturing and killing others, but the recognition of another class of offender could well prove quite problematic for those in charge of administering justice.

    Equally, the Court should consider the stigmatizing effect that a conviction before an international criminal tribunal can have on an individual if and when that person tries to reintegrate into that area. No one would dispute that the guilty should be punished in some form, but labeling someone like Erdemovic as a war criminal or as a perpetrator of a crime against humanity seems excessive. One of the stated goals of the Tribunal is reconciliation and rebuilding, but it is very possible that the current legal categories are anathema to these goals. Perhaps a tiered approach would be better, not unlike the distinction in American jurisdictions between felonies and misdemeanors.

    Other problems exist as well, though they are probably of a less philosophical nature and are hence easier to resolve. One such problem involves the lack of a standard dictating the necessary level of punishment commanders must mete out if they find out about a criminal act committed by their subordinates after the fact. Would dismissal be sufficient? Would a suspension be necessary? A court martial? These issues have yet to be resolved before the ICTY, but if this Tribunal does not hear such a case, undoubtedly a future one will. There are other structural concerns that should be addressed as well, in order to aid in the fair administration of justice. For example, the guilt and punishment phases of a trial should be separated to allow a full and vigorous defense to the charges brought against the defendant. While the judges acknowledge and understand the strategies of defense attorneys at the Tribunal, it creates undue stress and an unnecessary burden on them. According to one official at the Tribunal, it is “awkward for the Prosecution and detrimental to the Defense” (M. Harmon, personal interview, May 28, 2003). Without addressing this issue, the problem will only worsen, particularly since the ICTY is seen as a court establishing a great deal of precedent and as setting an example for future international courts. It should not be overlooked that the difficulties for defense counsel are limited purely to statutory elements and judicial rulings. Concerns about the quality of defense counsel, as well as the resources and rights of the accused also are matters of great concern. No one thought the Tribunal would get everything correct in its first effort, but unless these problems are at least given attention, there is little hope for actual justice in future international criminal tribunals.


    • International Criminal Tribunal for the Former Yugoslavia. (1995a). Prosecutor v. Aleksovski. Judgment. IT-95-14-1. Retrieved August 16, 2004 from
    • International Criminal Tribunal for the Former Yugoslavia. (1996a). Prosecutor v. Erdemovic (Pilica Farm). Sentencing Judgment. IT-96-22. Retrieved August 16, 2004 from
    • International Criminal Tribunal for the Former Yugoslavia. (1996b). Prosecutor v. Erdemovic. Joint Separate Opinion of Judge McDonald and Judge Vorhah. IT-96-22. Retrieved August 16, 2004 from
    • International Criminal Tribunal for the Former Yugoslavia. (1996c). Prosecutor v. Erdemovic. Separate and Dissenting Opinion of Judge Cassese. IT-96-22.
    • International Criminal Tribunal for the Former Yugoslavia. (1995b). Prosecutor v. Kordic and Cerkez. Judgment. IT-95-14/2. Retrieved August 16, 2004 from
    • International Criminal Tribunal for the Former Yugoslavia. (1969). Prosecutor v. Delalic et al (The Celebici Case). Judgment. IT-96-21. Retrieved August 16, 2004.
    • International Criminal Tribunal for the Former Yugoslavia. (1995c). Prosecutor v. Sikirica et al. Sentencing Judgment. IT-95-8. Retrieved August 16, 2004 from
    • International Criminal Tribunal for the Former Yugoslavia (1996). Prosecutor v. Delalic. Judgement. IT-96-21-A. Retrieved August 24, 2004 from
    • International Criminal Tribunal for the Former Yugoslavia. (1998). Prosecutor v. Vasiljevic. Judgment. IT-98-32. Retrieved August 16, 2004 from
    • Jia, B. B. (2000). The doctrine of command responsibility: Current problems. Yearbook of International Humanitarian Law. 3, 131 – 165.
    • Knoops, G. (2001). Defenses in Contemporary International Criminal Law. Ardsley, New York: Transnational Publishers.
    • Levie, H. (1991). The rise and fall of an internationally codified denial of the defense of superior orders. Revue de Droit Militaire et de Droit de la Guerre (The Military Law and Law of War Review). 30, 183 – 209.
    • Lippman, M. (2000). The evolution and scope of command responsibility. Leiden Journal of International Law. 13, 139 – 170.
    • May, R. (2002, May). Command responsibility – A new basis of criminal liability in English law? The Criminal Law Review. 38, 363 – 378.
    • Mitchell, A. (2000, September). Failure to halt, prevent, or punish: The doctrine of command responsibility for war crimes. The Sydney Law Review. 22, 381 – 409.
    • Newman, S. (2000). Duress as a defense to war crimes and crimes against humanity – Prosecutor for Drazen Erdemovic. Military Law Review. 166, 158 – 171.
    • Nybondas, M. (2003). "Civilian superior responsibility in the Kordic case." Netherlands International Law Review. 1, 59 – 82.
    • Osiel, M. (2001). Mass atrocity, ordinary evil, and Hannah Arendt: Criminal consciousness in Argentina’s dirty war. New Haven, Conn.: Yale University Press.
    • Rowe, P. (1998). Duress as a defense to war crimes after Erdemovic: A laboratory for a permanent court? Yearbook of International Humanitarian Law. 1, 210 – 228.
    • Scaliotti, M. (2001). Defences before the International Criminal Court: Substantive grounds for excluding criminal responsibility – Part 1. International Criminal Law Review. 1, 117 – 172.
    • Scaliotti, M. (2002). Defences before the International Criminal Court: Substantive grounds for excluding criminal responsibility – Part 2. International Criminal Law Review. 2, 1 – 46.
    • Suominen, S.B. (2001). Prosecuting superiors for crimes committed by subordinates: A discussion of the first significant case law since the Second World War. Virginia Journal of International Law Association. 41, 747 – 785.
    • Wrightsman, Lawerence, Michael Nietzel, and William Fortune. 1998. Psychology and the Legal System. Pacific Grove, California: Brooks/Cole Publishing.
    • United Nations. (2002, May 17). Updated Statute of the International Criminal Tribunal for the Former Yugoslavia. Retrieved August 16, 2004 from
    • U.S. Military Tribunal. (1950). Trial of Otto Ohlendorf and others (Einsatzgruppen case). Law Reports of Trials of War Criminals. Volume 8. Nuremberg.
    PDF icon HensonC.pdf124.26 KB