Hostname: page-component-745bb68f8f-mzp66 Total loading time: 0 Render date: 2025-02-06T10:16:54.308Z Has data issue: false hasContentIssue false
  • English
  • Français

Duress: From Nuremberg to the International Criminal Court, Finding the Balance Between Justification and Excuse

Published online by Cambridge University Press:  30 July 2015

MARCUS JOYCE
Rights & Permissions [Opens in a new window]

Abstract

Article 31(1)(d)1 of the Rome Statute of the International Criminal Court (ICC Statute) presents an important opportunity to reconsider the defence of duress in cases of unlawful killing. While the case of Erdemović has done much to substantiate the existence of the defence of duress at international law it appears to have curtailed the doctrine by interpreting it with reference to a strict form of proportionality characteristic of duress as justification. On the other hand, duress as excuse requires some measure of proportionality. This article will contend that the hybrid approach of Nuremberg Military Tribunals (NMTs), defined duress and the moral choice test primarily by reference to culpa in causa, not resorting to duress, and a ‘softer’ proportionality and in doing so, provided a more flexible and workable model for duress. Article 31(1)(d) of the ICC Statute, although an interesting attempt to find the balance between duress as excuse and justification, is a missed opportunity to redefine the defence in international criminal law. An alternative test for duress, with reference to the principles that emerged from the jurisprudence of the NMTs, is required in order to find the correct balance between duress as excuse and justification.

Keywords

duressexcusejustificationmoral choiceproportionality
Type
INTERNATIONAL CRIMINAL COURTS AND TRIBUNALS
Information
Leiden Journal of International Law , Volume 28 , Issue 3 , September 2015 , pp. 623 - 642
Copyright
Copyright © Foundation of the Leiden Journal of International Law 2015 

1. Introduction

The debate about whether duress should be a complete defence in cases of unlawful killing, or merely a factor to be considered in mitigation of sentence has long been a feature of international criminal jurisprudence. The majority in Erdemović, the first international criminal law case in recent years to consider the defence followed the current position in English law, and concluded that duress in cases of murder could only be raised in mitigation of sentence.Footnote 2

This article will suggest, by conducting a comparative analysis between post- Second World War jurisprudence and Erdemović, that it may be possible to find a satisfactory balance between duress as excuse and justification but that in order to do so a softer approach to proportionality must be adopted, combined with more emphasis on culpa in causa and not ‘resorting to’ duress.Footnote 3 The article will suggest that Article 31(1)(d) of the ICC Statute is essentially a failed attempt to find the correct balance between justification and excuse, as it still defines duress with reference to justification. The Article will conclude by proposing, again relying on the principles emanating from Second World War jurisprudence, an alternative test which attempts to find the balance between duress as justification and excuse.Footnote 4

2. Nuremberg: Articulating the moral choice test

The moral choice test first emerged at Nuremberg in the context of the defence of superior orders. Article 8 of the Charter of the International Military Tribunal (IMT) constituted a blanket ban on raising superior orders as a complete defence, choosing instead to permit the IMT to exercise its own discretion as to whether to consider superior orders as a mitigating factor.Footnote 5 In its judgment, the IMT introduced a moral choice test when considering superior orders as a defence in relation to Article 8 of the IMT charter:

The provisions of this Article are in conformity with the law of all nations. That a soldier was ordered to kill or torture in violation of the international law of war has never been recognised as a defence to such acts of brutality, though, as the Charter here provides, the order may be urged in mitigation of the punishment. The true test, which is found in varying degrees in the criminal law of most nations, is not the existence of the order, but whether moral choice was in fact possible.Footnote 6

Principle IV of the Nuremberg Principles, drafted to correspond to the case law of the IMT, provides that: ‘The fact that a person acted pursuant to order of his Government or of a superior does not relieve him from responsibility under international law, provided a moral choice was in fact possible to him’.Footnote 7 Legal commentators remain divided as to the significance of the inclusion of the moral choice test in both the IMT judgment, and in Principle IV.Footnote 8 Bassiouni argues that the inclusion of a moral choice test does not affect the ban on raising superior orders as a defence as set out in Article 8 of the IMT Charter but that it may have an important function to play in situations of duress:

If for example a defendant who was in a position where if he did not comply with the illegal order, he would be killed (i.e. no ability to make a moral choice), then the defendant may be acquitted once all relevant circumstances were examined pursuant to general principles of law (e.g. the traditional criminal law defense of coercion without regard to the defense of superior orders).Footnote 9

In raising a defence of superior orders, the primary focus is on the order itself, as opposed to the choice facing the recipient of the order. It is suggested that this is correct: An illegal order does not remove the choice of the recipient of that order to weigh its legality and, if illegal, choose not to follow it. It may be an uncomfortable choice, but in the absence of the threats to life and limb arising in cases of duress, it is a meaningful choice.Footnote 10

As such it is arguable that it is the legality of the order and the actor's knowledge of its legality, not the issue of choice, are central to the defence of superior orders. Conversely, in considering duress, the focus is on the threat occasioned and, assuming the requisite level of threat has been made, whether the reasonable man would succumb to that threat. This necessarily includes considering the impact of the threat on the choice open to him. Thus, when a threat to life or limb is made, the defence of superior orders no longer applies and one considers whether the accused was acting under duress.Footnote 11 Having completed even this brief analysis of both defences, it is contended that Bassiouni's view is preferable and the use of the moral choice test in cases of duress is more logical than in cases of superior orders.

Having concluded that the moral choice test is more logically applied to duress rather than superior orders, one significant difficulty remains: Duress is often viewed, as it was in Erdemović, as a defence of justification which requires the accused to do the ‘right thing’ in the circumstances. Central to justification is the concept that the act committed, although criminal, is the lesser of two evilsFootnote 12 and that the act committed be proportionate to the evil threatened.Footnote 13 Applying this test to duress, a balancing act is undertaken between the severity of the harm threatened to the coerced, and the harm he has to cause in order to avoid that threat in repelling the attacker, all within limits of proportionality.Footnote 14 So, if A is threatened with a punch unless he stabs B, his action in stabbing B could not be said, in general terms, to be justified on the basis that (i) it was not the lesser of the two evils, and (ii) it was not proportionate to the harm threatened, in this case a punch.

Few would claim that there is anything controversial in the above. The harm caused and that meted out are clearly disproportionate. We can fairly expect A to suffer a punch, rather than stab B. There is a moral or meaningful choice to be made between these two very different alternatives. This equation becomes much more complicated where the two harms under consideration are similar in nature, such as in cases of unlawful killing. Can it be said that A has a moral or meaningful choice when his own life is under threat? Can he fairly be expected to act in a strictly proportional manner in such a situation?

Thus, the difficulty with duress as justification is that it imposes overly demanding requirements on the availability of duress as a defence. In the case of an excuse, the righteousness of the accused's behavior is irrelevant – the issue is that the accused should not be held criminally liable for his actions.Footnote 15 Excuse concedes that the act is wrongful but seeks to avoid the attribution of liability for that act to the accused.Footnote 16 As will be contended below, this categorization of duress as excuse would be more consistent with the nature of the defence and how it has been interpreted in the jurisprudence arising out of the Second World War.

3. The Nuremberg Military Tribunals: A hybrid approach

After the IMT, in a series of trials conducted under Control Council Number 10 by what shall be referred to collectively as the Nuremberg Military Tribunals (NMTs), the issue of duress in the context of superior orders was considered.Footnote 17 The NMTs did not make any distinction between necessity and duress, nor between justification and excuse. Their interpretation of the defence of duress can best be described as a hybrid which incorporates elements of the lesser evils test, duress as excuse and duress as justification.Footnote 18 As noted by Ambos, the jurisprudence that developed did not require a strict balancing of interests, rather it focused primarily on whether or not the accused's free will had been limited to the extent that the attachment of criminal liability to his acts appeared unjust.Footnote 19

Einsatzgruppen specifically dealt with charges of unlawful killing. The Tribunal, in reference to the moral choice test of the IMT, held:

Let it be said at once that there is no law which requires that an innocent man must forfeit his life or suffer serious 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. . . . The test to be applied is whether the subordinate acted under coercion or whether he himself approved of the principle involved in the orderFootnote 20

For the Tribunal, the principal issue in determining the existence of a moral choice was whether the coercion directly led to the commission of the act, or whether the actor went along with the principle in the order. In considering proportionality, the Tribunal provided an example of what it considered would be disproportionate in the context of a case of unlawful killing:

If one claims duress in the execution of an illegal order it must be shown that the harm caused by obeying the illegal order is not disproportionally greater than the harm which would result from not obeying the illegal order. It would not be an adequate excuse, for example, if a subordinate, under orders, killed a person known to be innocent, because by not obeying it he himself would risk a few days of confinement.Footnote 21

On the basis of both extracts above, it is reasonable to conclude that the Tribunal viewed proportionality in broader terms by considering whether the two harms were comparable in terms of their severity, and not as a strict balancing act between the harm threatened and the harm meted out. It is also noteworthy that the defence was rejected by the Tribunal essentially on the basis of culpa in causa: the Einsatzgruppen were set up for the precise purpose of carrying out killings on a massive scale.Footnote 22 Thus, the principle that an accused who was threatened with death or serious harm may plead duress for taking another life is not undermined by the Tribunal's rejection of the defence on the particular facts of the case. Equally, the judgment left open the question of whether an accused who ‘passed’ the test of culpa in causa, (because when he joined the group it did not have a nefarious purpose, but who killed on a large scale as a result of duress), would be able to successfully plead the defence.Footnote 23

A number of additional NMT authorities are worthy of mention in this context, although, as further discussed below, they did not specifically deal with charges of unlawful killing. The Flick trial was the first of a series of three ‘industrialist trials’ which tried those deemed responsible for aiding the war effort through the enforced labour of civilians in their industrial enterprises. The Tribunal acquitted four of the accused of Count 1, (crimes against humanity and war crimes on the basis of enslavement and deportation), on the basis of necessity, finding that the defendants were subject to a ‘clear and present danger’:

The defendants lived within the Reich. The Reich through its hordes of enforcement officials and secret police, was always “present”, ready to go into instant action and to mete out savage and immediate punishment against anyone doing anything that could be construed as obstructing or hindering the carrying out of governmental regulations or decrees. In considering the application of rules to the defense of necessity, attention may well be called to the following statement:

The law of cases of necessity is not likely to be well furnished with precise rules; necessity creates the law, it supersedes rules, and whatever is reasonable and just in such cases is likewise legal. It is not to be considered as a matter of surprise, therefore, if much instituted rule is not to be found on such subject.Footnote 24

While the Tribunal required that the remedy should not be disproportionate to the evil it sought to avoid, it is evident that the Tribunal did not enter into a strict balancing act.Footnote 25 If it had, the defendants would most likely have been convicted – few could seriously argue that the potential deaths of four defendants outweighed the incarceration, forced slave labour, and resulting deaths of at least some of thousands of civilians.

The Tribunal contrasted the factual situation of the four defendants whom it acquitted with that of the remaining two defendants, namely Flick and Weiss. The Tribunal held that they could not avail themselves of the defence of necessity because they had taken ‘active steps’ to procure an increased production quota of freight cars and in so doing had not acted out of ‘compulsion or fear’ but instead, ‘to keep the plant as near capacity production as possible’.Footnote 26 In other words, they went beyond what was strictly required to offset the threat to them. This indicated that, contrary to acting out of fear, they approved of the principle behind the order. In Farben, the second of the industrialist trials, the Tribunal implied that coercive conditions, as existed in Nazi Germany for example, may give rise to a situation where the actor had no moral choice.

nevertheless, such an order is a complete defense where it is given under such circumstances as to afford the one receiving it no other moral choice to comply therewith. As applied to the facts here, we do not think there can be much uncertainty as to what the words “moral choice” mean. The quoted passages from the IMT judgement as to the conditions that prevailed in Germany during the Nazi era would seem to suggest a sufficient answer insofar as this case is concerned.

The Tribunal discussed the Flick trial, distinguishing the facts giving rise to the acquittals in that case from the facts as they related to the defendants in Farben on the basis of culpa in causa:

It is plain, therefore, that Hermann Roechling, von Gemmingen, and Rodenhauser, like Weiss and Flick, were not moved by a lack of moral choice, but, on the contrary, embraced the opportunity to take full advantage of the slave-labour programme. Indeed, it might be said that they were, to a very substantial degree, responsible for broadening the scope of that reprehensible system.

It follows that the defense of necessity is not available where the party seeking to invoke it was, himself, responsible for the existence or execution of such order or decree, or where his participation went beyond the requirements thereof, or was the result of his own initiative.Footnote 27

Thus, the hallmarks of the Tribunal's interpretation of the moral choice test in both Flick and Farben, as in Einstazgruppen, were whether (i) the circumstances as they existed were coercive and (ii) assuming circumstances were indeed coercive, whether the actor actually acted by reason of fear or compulsion arising from those coercive circumstances.

In Krupp, the third trial of industrialists, the Tribunal rejected the defendants' pleas of necessity. The Tribunal pointed in particular to culpa in causa, and/or an exclusion of resorting to duress, in evaluating whether the will of the accused was overpowered, or instead whether it coincided with the will of those from whom the threats emanated.Footnote 28 The Tribunal also referred to the harm that the accused would have suffered, (the potential loss of their industrial plants), as being insufficient to satisfy the proportionality requirement. The Tribunal further noted that the possibility of the accused being sent to a concentration camp was theoretical. Their reasoning contains elements of the lesser evils test:

If we may assume that as a result of opposition to Reich policies, Krupp would have lost control of his plant and the officials their positions, it is difficult to conclude that the law of necessity justified a choice favourable to themselves and against the unfortunate victims who had no choice at all in the matter. Or, in the language of the rule, that the remedy was not disproportioned to the evil.Footnote 29

It is also notable that the Tribunal in Krupp adopted a subjective standard in evaluating whether the accused acted as a result of the threat of harm or for some other purpose, the principal issue being not just whether the danger existed but whether the accused had a ‘bona fide’ belief in the danger which he claims forced him to act.Footnote 30 In High Command, the Tribunal drew a distinction between the defendants in that case ‘who received obviously criminal orders [and] were placed in a difficult position’ from a situation of coercion where they concluded there must be ‘a showing of circumstances such that a reasonable man would apprehend that he was in such imminent physical peril as to deprive him of freedom to choose’.Footnote 31 As such, the Tribunal in High Command was even more explicit than the tribunals in Flick and Farben, in characterizing a situation of duress as one where the actor is deprived of the actual freedom to choose whether to commit the impugned act or not.

Considering this jurisprudence, an attempt at summarizing the NMTs interpretation of moral choice and duress may look as follows: (i) assuming cogent evidence of coercive circumstances existed, one considers whether the actor acted as a result of that coercion, or whether he in fact approved of the principle involved; (ii) in assessing (i) above, one considers whether he only did what was necessary to save his life, or whether he went beyond what was strictly required (first limb of proportionality test); (iii) when assessing whether what he did was proportional to the harm he caused, one is not entering into a strict balancing of harms, but a broader consideration of whether the threat of harm to the actor and the harm to be meted out were comparable in terms of their severity (second limb of proportionality test).

Therefore, where there is coercion, the actor acts due to that coercion, he only does what is strictly required to meet the threat, and where the act committed is not, in broad terms, disproportionate in terms of severity with the threat, then the actor may be said to have no moral or meaningful choice. As such, the NMTs hybrid approach of merging justification, excuse, necessity and duress in one provision, did not lead to the adoption a strict lesser evils test, but rather a soft proportionality used in conjunction with more emphasis on culpa in causaFootnote 32 and the exclusion of ‘resorting to’ duress.

How then did the Appeals Chamber in Erdemović evaluate moral choice, proportionality, and the jurisprudence of the NMTs, over 50 years later?

4. Moral choice and proportionality in Erdemović

Although not included in the statute of the International Criminal Tribunal for the former Yugoslavia (ICTY), the issue of moral choice as reflected in the judgment of the IMT and Principle IV was reconsidered in the case of Erdemović. Erdemović was a foot soldier at Srebrenica who personally shot and killed 70 civilians.Footnote 33 He was indicted for murder as a crime against humanity, contrary to Article 5 of the ICTY Statute, and murder in violation of the laws or customs of war, contrary to Article 3 of the ICTY Statute. He pleaded guilty to crimes against humanity and was sentenced to 20 years imprisonment. He appealed the sentencing judgement on the basis that he had acted under duress – Erdemović had maintained from the beginning that he had been ordered by his superiors to shoot his victims and that if he did not do so he would be killed along with the other victims.Footnote 34 This was not disputed by the prosecution. Amongst the issues that arose for determination by the Appeals Chamber was whether or not the defence of duress could serve as a complete defence or merely as mitigation for reduction of sentence. In its opinion, the Appeals Chamber held by majority, that there was no rule under customary international law establishing duress as a defence in cases of unlawful killing, and that it could act as a factor in mitigation of sentence only.Footnote 35

4.1. The majority opinion: Moral and social policy considerations

In considering whether duress held the status of a defence to unlawful killing under customary international law the Tribunal said as follows regarding the IMT's moral choice test:

This unelaborated statement, in our view, makes no significant contribution to the jurisprudence on this issue. It does little to support the contention that the decisions of post-World War Two international military tribunals established a clear rule recognising duress as a defence to the killing of innocent persons which would then by now have become customary international law.Footnote 36

Whatever may be said as regards the status of duress in customary international law, it is perhaps inaccurate of the majority to dismiss the moral choice test as an ‘unelaborated statement’ that has made ‘no significant contribution’ to the jurisprudence on duress, particularly when one considers the jurisprudence of the NMTs. However, the majority dismisses that jurisprudence on the dubious ground, as refuted by Cassese, of the ‘questionable international character’ of the NMTs.Footnote 37

The majority then proceeded to consider afresh whether duress should be a defence in cases of murder, deciding it should not, on what they perceived to be the ‘social, political and economic role’ of the law. Their position, based on numerous citations from English case law regarding the ‘sanctity of human life’ appears less than convincing, as is their contention that duress would confer impunity on subordinates.Footnote 38

The majority thus concluded that their position as a matter of policy, was preferable to entering into what they called a ‘tortuous investigation into the relationship between law and morality’.Footnote 39 But is this a fair criticism if one applies it to the moral choice test? The moral choice test, as it emerged from the NMTs, is a determination of what a reasonable person might be expected to do in a situation where their own life is at risk. Of course the situations under consideration often raise difficult ethical issues, but that does not mean that in order to apply the moral choice test objectively one must enter into a ‘tortuous investigation’ of law and morals.

It is contended that in describing duress in these terms and in dismissing the moral choice test and the jurisprudence of the NMTs, the majority missed the point of moral choice and with it, an important opportunity to build on the existing jurisprudence relating to it.Footnote 40

4.2. Cassese's dissent: The right to a subordinate's life depends on what his death achieves

Cassese puts forward in a persuasive manner, through an extensive analysis of case law, his contention that in the absence of a rule in customary international law permitting duress as a defence to murder, the general rule on permitting duress as a defence should be followed. He firmly rejects the moral and policy considerations which influenced the majority into concluding that duress should not be permitted as a defence in cases of murder:

An international court must apply lex lata, that is to say, the existing rules of international law as they are created through the sources of the international legal system. If it has instead recourse to policy considerations or moral principles, it acts ultra vires.Footnote 41

Having denounced the moral and policy arguments of the majority, Cassese embarks on an examination of how the defence would apply in practice. In relation to the killing of innocents where it is a situation of one life or another, he suggests that while he would allow the defence of duress to be raised in such situations, it would rarely if ever succeed. Footnote 42 Cassese's principal basis for his position rests on the moral difficulty in choosing between one life and another:

Perhaps – although that will be a matter for a Trial Chamber or a Judge to decide – 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?Footnote 43

The final question is obviously rhetorical in nature and Cassese's conclusion is that a judge cannot so satisfy himself: In situations requiring the balancing of one's own life against that of an innocent other the accused should elect to die or, if he does not, he may at best achieve some mitigation of sentence. There is an apparent contradiction: Cassese rejects the moral arguments of the majority as a proper basis for making decisions regarding the defence of duress but then employs moral difficulties in denying duress in cases involving the death of innocents where an accused must choose between his own life and that of an innocent other.Footnote 44

It is contended that Cassese's analysis lacks the high-handed moral tones that dominate the majority opinion, which rejects duress outright as a matter of principle.Footnote 45 However, it still involves a rejection of duress in cases of unlawful killing on a basis which bears a similarity to the reasoning of the ‘sanctity of life’ expounded by the majority. Thus there is, in defining the matter as one of justification, too close of a link to the moral reasoning on which the majority relies for rejecting the defence outright.Footnote 46 An additional ground for Cassese's reluctance in permitting the defence to be successfully raised in cases of one life or another is explained in his analysis of post-Second World War jurisprudence:

But for the Italian and German cases mentioned above (paragraphs 35–39, supra), which stand out as exceptional, the only cases where national courts have upheld the plea of duress in relation to violations of international humanitarian law relate to offences other than killing. In this connection mention can be made of the well-known cases brought before United States Military Tribunals sitting at Nuremberg, Flick and Farben, as well as a few German cases. To my mind, this bears out the strong reluctance of national courts to make duress available in case of offences involving killing.Footnote 47

Cassese is correct in pointing out that the NMTs, apart from Einsatzgruppen, only acquitted on the basis of duress in cases which did not involve underlying offences of unlawful killing. However, is it right that their jurisprudence is dismissed as having nothing useful to say about duress in cases of unlawful killing?

On the one hand it may be said that the application of a proportionality test in situations other than unlawful killing is less problematic. Krupp, for instance, is an example of one of those ‘easier’ cases of duress where the harm to be suffered (in that case loss of property) is clearly disproportionate to the suffering of the defendant's victims. This, does not of course mean by extension, that the NMTs considered that an actor would be justified or excused in the much more contentious situation of killing to save his own life. One may therefore argue that the hybrid approach of the NMTs, in merging the differing concepts of necessity, duress justification, and excuse, only worked because they addressed cases which did not involve unlawful killing and where, consequently, the issue of proportionality was not so problematic.

On the other hand, while the jurisprudence of the NMTs may not amount to establishing a rule of customary international law in favour of duress in cases of unlawful killing, broad principles emerge from that jurisprudence which should not automatically be considered as confined to cases other than unlawful killing.

A number of principles which emerged from the jurisprudence of the NMTs were handed down in general terms without drawing a distinction between cases of unlawful killing and others. For example, it is contended that the Tribunals' approach of focusing primarily on culpa in causa and the exclusion of resorting to duress, rather than strict proportionality, is unlikely to have been intended to be confined solely to cases other than unlawful killing.

Further, in High Command, although the Tribunal did not have unlawful killing charges before it, its finding that duress removes the freedom to choose fairly would be arguably more, not less, applicable to cases of unlawful killing where the actor's life is at stake and the fairness of his choice is all the more doubtful.

Moreover, disregarding the case law of the NMTs on the basis that they did not address cases of unlawful killing may not entirely bear scrutiny. In Flick, for example, the Tribunal dealt with defendants whose actions were responsible for sending thousands of victims to labour camps, at least some of whom perished. It therefore seems somewhat artificial to conclude that the accused in that case were only permitted to successfully plead duress because, although their actions led to the death and suffering of thousands, they were not charged with murder.

Finally, although much is made of the fact that Einsatzgruppen was the only case to specifically address charges of unlawful killing, it is perhaps overly formulaic to treat the remaining jurisprudence of the NMTs as entirely separate. It is contended, given that the remaining jurisprudence did not disapprove of Einsatzgruppen either expressly or impliedly, that it is equally plausible that the remaining jurisprudence, in the same series of trials and under the same Control Council order, can be viewed as complimentary to Einsatzgruppen.

4.3. The exception to the general prohibition on duress in unlawful killing

Cassese points to one exception to his general reluctance to allow duress in relation to the killing of innocents in referring to those who would die along with their victims if they did not do as ordered. In such circumstances he considers that the proportionality test may be satisfied:

the case-law seems to make an exception for those instances where – on the facts – it is highly probable, if not certain, that if the person acting under duress had refused to commit the crime, the crime would in any event have been carried out by persons other than the accused. . . . In this case the evil threatened (the menace to his life and his subsequent death) would be greater than the remedy (his refraining from committing the crime, i.e., from participating in the execution).Footnote 48

Judge Stephen essentially follows Cassese's reasoning in distinguishing those cases where an accused has to choose between himself or another, from those cases where the accused would have joined his victims in being killed had he refused to succumb to the threat.Footnote 49 In doing so he relies on the moral choice test:

Such a moral choice was [referring to the moral choice of whether to kill an innocent or be killed in his place], according to the statements of the Appellant, not open to the Appellant to make. However he chose, the lives of the innocent would be lost and he had no power to avert that consequence. It is in this sense that it can be said that the Appellant had no moral choice. Of course he did have a choice, whether or not to lay down his life for the sake of the highest of ethical principles. But that is not the sort of choice the making of which criminal laws should enforce with penal sanctions.

However, in adopting an exception based on strict proportionality test of numbers, Cassese and Stephen are vulnerable to the criticism of having adopted an entirely utilitarian approach, as levelled at them by the majority.Footnote 50 More importantly, is it just that a man facing his own death, is imputed a moral choice on the basis of whether or not his victims will die regardless of his actions: Does this reasoning not encourage him to tell himself, or worse, a Tribunal, that they would have died regardless of his actions? On the other hand, few would disagree with the outcome of this scenario. Therefore, perhaps the problem is not the exception, rather it is justification and the lesser evils test which require such extraordinary circumstances to enable an accused to successfully plead duress.

5. Conclusions on duress as justification and moral choice

Justification as a strict balancing act of lesser evils is less problematic for those less controversial cases of duress where the harm threatened and that sought to be avoided are clearly disproportionate, as in our previous example of stabbing to avoid an assault. However, where the two forms of harm are comparable in terms of their severity, such as in cases of murder, the test has the effect that the actor is expected to die.Footnote 51 It is therefore contended that when the lesser evils test is applied, moral choice is determined on the basis of what his death might achieve rather than whether or not he actually had a fair or meaningful choice to make.Footnote 52 Thus, it has been suggested by Dressler that the lesser evils test is only truly compatible with duress when it is not allowed for cases of unlawful killing.Footnote 53

Is there any satisfactory interpretation of duress as justification in cases of unlawful killing? Using the example of Erdemović, Weigend provides an alternative to the exception advocated by Cassese. He suggests that the critical issue is not whether the actor could be replaced by another executioner, but rather, it is the short amount of time the victims have left to live until another gunman kills them along with the actor that is determinative of the actor's right to save himself. He suggests an appropriate test in determining liability would be ‘whether a reasonable victim would tend to insist on staying alive even for the short time still available’.

However, his proposal does not appear to apply to cases of one life or another, where the actor is still expected to die. Furthermore, a test based on the time the victims would have left to live is not without its own significant difficulties. What length of time would be considered reasonable? Would it not take any account of subjective factors? To take an extreme example, to a victim who already has a terminal illness another day might make no difference or, on the contrary, make all the difference. Footnote 54

6. Duress as excuse: Excusing the unjustifiable

By classifying duress as an excuse, the focus is not on whether the act is inherently ‘right’. On the contrary, it is accepted that the act is ‘wrong’.Footnote 55 Consequently, there is no attempt to seek to justify it by determining whether it is the lesser evil.Footnote 56 The issue is the attribution of liability which may be excused on two principle grounds: (i) a lack of mens rea on the part of the accused Footnote 57 or because (ii) although the accused's conduct cannot be condoned, due to special circumstances, society deems it unfair to criminalize it.Footnote 58

Eser contends that the lack of distinction between justification and excuse plays a detrimental role in attempting to solve what he describes as the ‘problem of duress’: When categorized as a defence of justification, one automatically sees duress as an attempt to condone what cannot be condoned.Footnote 59 While as an excuse, one can accept more readily that the accused is seeking forgiveness for having committed an act, which, although it is wrong, is not one which he would, in circumstances absent a threat, have committed. This approach has the benefit, because it does not seek to justify, of countering to some extent the moralistic and policy-driven reasons for denying the defence of duress in cases of murder.Footnote 60 Furthermore, as noted above, duress as justification is subjected to a strict balancing act which determines the moral choice of the actor on the basis of what his death might achieve.Footnote 61

Duress as excuse is also distinct from the utilitarian approach rejected by the majority in Erdemović: It does not focus on the numbers of individuals killed. Duress as excuse acknowledges that due to the unfairness of the choice open to the accused, the wrong-doing is not voluntary and therefore it is not in the broader interests of society to punish.Footnote 62

Duress as excuse may also be viewed as the negation of mens rea, although opinions are divided as to whether mens rea is actually negated. Cassese states that duress negatives the subjective element of the person under coercion: “the criminal intent of the person causing duress in a way substitutes for his [the actor's] mens rea”. In contrast, he suggests that in a case of necessity the ‘agent intends to cause an unlawful harmful effect.Footnote 63

It is also worthy of note that the view of the NMTs appears to have been that necessity and/or duress removed the mens rea of the offence. The following passage from Stratton's case as cited in Wharton's Criminal Law is often quoted in their judgements:

Necessity forcing a man to do an act justifies him, because no man can be guilty of a crime without the will and intent in his mind. When a man is absolutely by necessity forced, his will does not go along with the act.Footnote 64

In any event, whichever view is preferable, as we shall see Articles 30 and 31(1) (d) of the ICC Statute requirements as regards intent all but remove the possibility of duress by negation of mens rea.

7. A need for limits?

It must be conceded that duress is an atypical excuse. The accused acting under duress of threats is distinguishable from the accused suffering from insanity or another defect of the mind, as he has the capacity, both physical and mental, to make a choice.Footnote 65 Neither does the accused suffer from a physical defect which renders his act physically involuntary.Footnote 66

This has led one commentator in particular to suggest that limits must be placed on the extent to which acts committed under duress can be excused. Fletcher proposes what appears to be akin to a type of justificatory determination within a framework of excuse, which effectively serves to blur the line between both duress and excuse. In determining whether an act is voluntary or not, he advocates the use of a competing interests test which may be summarized as follows: if the benefit of the conduct is far outweighed by the harm meted out then this points to the voluntariness of the act, and therefore the conduct should not be excused.Footnote 67 The test is objective, with reference to what a person of ‘reasonable firmness’ would be expected to resist.Footnote 68

Perhaps the primary difference between Fletcher's objective test and the NMTs' approach can be summarized as follows; in considering whether the actor acted out of fear, or voluntarily, the NMTs add an element of culpa in causa or exclusion of resorting to duress: in addition to considering whether the two harms were broadly speaking proportionate they also analysed whether the actor only did enough to meet the specific threat occasioned to him or went beyond that strictly required to offset that threat, which would suggest voluntariness of the actor's actions.Footnote 69 While it remains difficult to pin down the NMTs' approach given their mixing of concepts, it is suggested that this is a form of ‘softer proportionality’ more characteristic of duress as excuse than justification. It is an approach which places limits on duress as excuse by focusing more on the actor's motivation but does not define duress in terms of the strict balancing characteristic of justification in order to determine whether or not he had a moral choice.

Furthermore, the NMTs approach, based as it is on mixing of concepts, offers some support for the possibility of adopting a hybrid approach to necessity and duress, without ‘justifying duress’. How does Article 31(1)(d) fare in its attempt to find a balance between duress as justification and excuse?

8. The ICC: Article 31(1) (d): A new approach

As the first international court or tribunal to provide a written definition of the defence of duress in its statute, the ICC has adopted a previously unseen approach to duress. Necessity and duress are combined in one provision in the ICC Statute. Duress is contained in Article 31(1)(d)(i) of the Statute which refers to threats being made by ‘other persons’, while subsection (ii) of the same provision refers to necessity in providing that the threat may come from ‘other circumstances beyond that person's [the accused's] control’.

The body of the text of Article 31(1)(d) provides that in order to rely on the defence, the accused must have acted reasonably and necessarily to avoid the threat to him and, in relation to his actions, must not have intended to cause greater harm than the one sought to be avoided by the threat to him.Footnote 70 This evaluation of the intention of the accused as to the level of harm caused amounts to a subjective element which has not been previously employed in cases of necessity or duress. It is arguably an attempt to soften the effect of the lesser evils test of necessity, on duress.Footnote 71 As with the ICTY Statute, there is no reference to the moral choice test.

This formulation, as included in the final draft of the ICC Statute is the result of compromise on the part of the member states.Footnote 72 It is materially different to all pre-Rome conference drafts of the provision, which, unlike the final draft, distinguished between necessity and duress.Footnote 73 Despite their combination in the same provision, it remains to be seen whether necessity and duress will be treated in exactly the same terms, or whether a distinction will be drawn between them.

8.1. Article 31(2): The promise of flexibility?

The objective requirement in the ICC Statute that the act to avoid the threat should be necessary and reasonable implies that it must also be proportionate, a characteristic of justificatory defences.Footnote 74 Ambos contends that the objective requirement that the reaction be ‘necessary and reasonable’ would apply only to the necessity limb of Article 31(1)(d) and not duress. In other words, acts committed under duress would not be required to be necessary and reasonable.

Upon initial consideration, it may seem difficult to share Ambos' interpretation of Article 31(1)(d) as capable of treating duress in a manner different to necessity given that the defences are combined under the same provision.Footnote 75 A potential solution however, may lie in Article 31(2) which provides that the Court ‘shall determine the applicability of the grounds for excluding criminal responsibility provided in this statute to the case before it’.Footnote 76

The author of this article agrees with Schabas' rejection of Eser's extremely broad interpretation of Article 31(2) that the Court can interpret each of the Statute's codified grounds as it so wishes, depending on the case before it, resulting as it might in numerous different versions of the same defence(s).Footnote 77 However, it is submitted that Article 31(2) is of greater significance than Schabas suggests. It may be used for more limited purposes, providing the Court with sufficient latitude to draw the traditional distinction between duress by threats and necessity as justification despite their inclusion as one provision in Article 31(1)(d).

Thus, Ambos' interpretation of the ‘necessary and reasonable’ provision would be possible within the framework of the statutory provisions. But is the objective limb the biggest obstacle to a ‘softer’ interpretation of duress? What of the new subjective element?

8.2. The subjective element

The actor must not intend to cause a harm greater than the one to be avoided. As suggested by Weigend, this element of intent may cover situations of mistake, where the actor acts in the mistaken belief that his actions will not lead to unlawful killing, but they in fact do.Footnote 78 However, this effectively removes the possibility of duress negating mens rea, unless he is deemed not to have the necessary intent by reason of mistake.Footnote 79

The intention not to cause ‘greater harm’ (as opposed to an objective requirement of committing a lesser harm) suggests that in situations where the harms are equal it may now be possible to successfully plead duress. This may therefore cover situations where the actor only intends to kill one victim in order to save his own life and thus suggests a softer form of proportionality. The exception outlined by Cassese where the actor would die along with his victims for not following the order would also still fall within the potential situations permitted by the provision.Footnote 80

However, should the victim intend to kill more than one victim, he will not be successful in his plea of duress. In other words, applying the moral choice test, in such situations he will be deemed to have a moral choice and will thus fail the test. This remains highly problematic. Therefore, and despite the apparent concession in cases concerning one life or another, Article 31(1)(d) still appears to retain a strict balancing element characteristic of justification. Furthermore, in contrast to the ‘necessary and reasonable provision’ where a common law distinction already existed between necessity and duress, it is difficult to envisage how a basis could be found to contend that Article 31(2) could serve to limit this subjective element so that it would only apply to necessity.

It is submitted that a preferable model for duress would have placed, returning to the principles of the NMT jurisprudence, a stronger emphasis on the following principles:

  1. 1. Culpa in causa: If one is voluntarily a member of a group which is reasonably likely to carry out criminal acts, duress cannot apply.

  2. 2. No resorting to duress: There must be (i) cogent evidence of a threat [Here one could add, in the alternative, a subjective element as was included in Krupp: Did the accused have a bona fide belief that he was in danger even if such threat did not objectively exist] and (ii) a causative link between that threat and the harm meted out – i.e. did the actor only act as a result of the coercion or did he approve of the principle involved.

  3. 3. Proportionality: (i) (first limb) was the harm meted out only what was strictly necessary to meet the threat occasioned and nothing more; and (ii) (second limb) were the two acts broadly speaking, proportionate: for example, unlawful killing could not be excused due to the threat of a broken arm.

Thus, in the NMT's proportionality test, the accused's fate swings not on whether the two acts were capable of being strictly balanced, but on whether what he did was only what was required to save himself and nothing more.

The proposed text of a new Article 31(1)(d) might look like this;

In addition to other grounds for excluding criminal responsibility provided for in this Statute, a person shall not be criminally responsible if, at the time of that person's conduct: the conduct which is alleged to constitute a crime within the jurisdiction of the Court has been caused by duress resulting from a threat of imminent death or of continuing or imminent serious bodily harm against that person or another person, and the person acts necessarily and reasonably to avoid this threat, provided that:

  1. (i) the person is not a member of a group which he might reasonably be expected to know will commit unlawful acts;

  2. (ii) the threat is reasonably capable of being considered sufficient to place the person in fear of imminent death or serious bodily harm or the person had a bona fide belief that death or serious bodily harm would occur;

  3. (iii) there is a direct or causative link between the threat occasioned and the person's acts;

  4. (iv) the acts carried out were comparable in nature and substance to the acts threatened;

  5. (v) the person only carried out such acts as might be reasonably considered strictly necessary in order to avert the threat occasioned.

By defining duress in these terms, as the NMTs did, one focuses on the actions of the accused with the aim of discovering his motivation, rather than the reverse: defining his motivation by reference to a mathematical balancing of harms. It is contended that this is not only a more just and workable definition of duress, but that it is also a stringent test which may just strike the right balance between the proportionality requirement, demanded by justification, and the more normative approach of duress as excuse.

9. Concluding remarks

Duress challenges fundamental concepts of innocence and guilt. We identify with the actor's dilemma, but we are repulsed by his actions. Yet it seems very difficult in principle to criminally condemn someone who has been placed as per the words of James Stephen ‘between two fires’ for choosing the fire which burns them least. Opposite our desire not to condemn the victim lies justification and proportionality: how many victims are we prepared to allow to be consumed by the flames resulting from the actor's actions?

Finding the right balance between these competing concepts is enormously difficult. But we can condemn the acts of violence committed by any individual, regardless of why they were committed, without condemning the actor who carried them out. On the one hand, duress as justification fails to draw this distinction by assuming a choice on the part of the actor with reference to the outcome of his actions rather than his motivation for committing them. On the other hand, while duress as excuse, in not seeking to condone, is successful in distinguishing between actions and actor, it requires some limits to be placed on those actions.

In adopting an approach similar to that of the NMTs, by emphasizing culpa in causa and ‘no resorting to duress’, combined with the proposed model for proportionality, one retains limits on the defence, but determines the fairness of the actor's moral choice primarily with reference to what motivated him in his actions. In weeding out the false claims of those who seek to abuse the defence, and permitting it for those who truly acted as a result of fear of harm, such approach is surely more holistic and just than strict proportionality. While it may be that the ICC's attempt is a failed one at least for those who have to choose between their own life and that of another, it is unlikely to be the final chapter in what is an undeniably controversial defence.

References

1 Art. 31(1)(d) ICC Statute. In addition to other grounds for excluding criminal responsibility provided for in this statute, a person shall not be criminally responsible if, at the time of that person's conduct; the conduct which is alleged to constitute a crime within the jurisdiction of the Court has been caused by duress resulting from a threat of imminent death or of continuing or imminent serious bodily harm against that person or another person, and the person acts necessarily and reasonably to avoid this threat, provided that the person does not intend to cause a greater harm than the one sought to be avoided. Such a threat may either be: (i) made by other persons; or (ii) constituted by other circumstances beyond that person's control.

2 DPP v. Lynch [1975] AC 653 (Lynch) permitted the defence to be raised in cases of murder, but was subsequently overturned in R v. Howe and Bannister [1987] 2 WLR 568.

3 As acknowledged by McGoldrick, Rowe, and Donnelly, it is thanks to the dissents of Cassesse and Stephen that duress was acknowledged as a defence in the ICC statute. See McGoldrick, D., Rowe, P., and Donnelly, E., The Permanent International Criminal Court (2004), at 275Google Scholar.

4 Ambos, K., ‘Grounds for Excluding Criminal Responsibility’, in Cassese, A., Gaeta, P., and Jones, J. R. W. D. (eds.), The Rome Statute of the International Criminal Court: A Commentary, Vol. 1 (2002), at 1044Google Scholar.

5 In practice it was not considered in mitigation either. Wilhelm Keitel, e.g., openly admitted his guilt and pleaded superior orders, but was hanged following conviction. The Tribunal held: ‘In the face of these documents Keitel does not deny his connection with these acts. Rather, his defence relies on the fact that he is a soldier, and on the doctrine of “superior orders”, prohibited by Article 8 of the Charter as a defence. There is nothing in mitigation. Superior orders, even to a soldier, cannot be considered in mitigation where crimes as shocking and extensive have been committed consciously, ruthlessly, and without military excuse or justification’. Judgment, International Military Tribunal, 1 October 1946, (‘IMT Judgment’), § 493.

6 Ibid., para. 447.

7 Principles of International Law Recognised in the Charter of the Nuremberg Tribunal and in the Judgment of the Tribunal (‘Nuremberg Principles’), Principle IV, Yearbook of the International Law Commission, Vol II (1950).

8 Greenspan, e.g., contends that the moral choice test does undermine Art. 8. Greenspan, M., The Modern Law of Land Warfare (1959), 493Google Scholar, at note 343.

9 Bassiouni, C., Crimes against Humanity in International Criminal Law (1992), at 427Google Scholar.

10 For a succinct summary of scholarly views on superior orders, see Bassiouni, C., Crimes Against Humanity in International Criminal Law (1999), at 457–63Google Scholar.

11 Casesse, A.et al., International Criminal Law Cases and Commentary (2011), 464, at 472Google Scholar.

12 Cassesse, A., ‘Justifications and Excuses in International Criminal Law’, in Cassese, A., Gaeta, P., and Jones, John R. W. D. (eds.), The Rome Statute of the International Criminal Court: A Commentary, Vol. 1 (2002), at 951CrossRefGoogle Scholar.

13 Writing in relation to self-defence: ‘The conduct in self-defence is proportionate to the offence to which the person reacts’, Cassese et al., supra note 12, at 461.

14 Defined in English law as ‘reasonableness of force’. See Smith and Hogan, Criminal Law (2008), at 361.

15 Hall, J., ‘Comment on Justification and Excuse’, (1976) 24 AJIL 639Google Scholar, at 639. It is contended that the usage of ‘right’ in this context is problematic. Take for example the classic justificatory defence of self-defence: an individual faced with an attacker ‘armed’ with his fist and in fear of being assaulted is entitled as a matter of law to defend himself. Few would disagree that if the attackee, in order to save himself, were to hit the attacker with his fist that it is right that he should be allowed to do so without being criminally liable. However, the author questions whether it can properly be said in a case of necessity, another defence of justification, that the underlying act is necessarily right. Consider the case of Dudley and Stevens (1884) 14 QBD 273 (Mignonette Judgment): it is perhaps difficult to go so far as to claim that the act of killing the cabin boy was ‘right’ even in the circumstances as they existed.

16 Fletcher, G. P., Rethinking Criminal Law (2000) at 759Google Scholar.

17 Control Council Law No. 10, Punishment of Persons Guilty of War Crimes, Crimes Against Peace and Against Humanity was a law enacted in 1945 by the Allied Control Council which created a framework for the prosecution of cases of a similar nature to those tried by the IMT.

18 van Sliedregt, E., Individual Criminal Responsibility in International Law (2012), at 251Google Scholar.

19 See Ambos, supra note 4, at 1005; Judgment, US v. Von Leeb et al. (case 12), in TWC XI (‘High Command judgment’); Judgment, US v. Krauch et al. (case 6), in TWC VIII (Farben Judgment), 1081–210, at 1174 et seq;, Judgment, US v. Ohlendorf et al. (case 9) (‘Einsatzgruppen Judgement’) § 480.

20 Einsatzgruppen Judgment, Ibid., para. 480.

21 Ibid., para. 470

22 Ibid., para. 479–82.

23 Dinstein, highly critical of the defence and in particular of Einsatzgruppen, contends that the correct approach is that ‘no degree of duress or necessity may justify murder’, Dinstein, Y., ‘International Criminal Law’, (1985) 20 Israel Law Review 206Google Scholar, at 235

24 Flick et al. US Military Tribunal Nuremberg, Judgment of 22 December 1947 Trials of War Criminals Before the Nuremberg Military Tribunals, Vol. VI (‘Flick Trial’) § 1200; Wharton's Criminal Law, Vol. I, Ch. III, subdivision VII, para. 126.

25 Flick trial, ibid., para. 1200.

26 Ibid., paras. 1199–201

27 See Farben Judgment, supra note 19, para. 1178.

28 Judgment, Krupp, US Military Tribunal Nuremberg, judgment of 31 July 1948, in Trials of War Criminals Before the Nuremberg Military Tribunals, Vol. IX (‘Krupp Trial’) at 1436; van Sliedregt, supra note 18, at 251.

29 Krupp Trial, supra note 28, paras. 1142–4.

30 Ibid., para. 1438.

31 High Command Judgment, supra note 19, 462 at 509.

32 Fitchelberg supports the view that an assessment of culpa in causa is a valid exercise in considering the legitimacy of a claim of duress and, in particular, an accused's claim that (s)he had no moral choice at the time of the impugned act. Fitchelberg, A., ‘Liberal Values in International Criminal Law’, (2008) 6 Journal of International Criminal Justice 3, at 15CrossRefGoogle Scholar.

33 Art. 7(4) ICTY Statute: ‘The fact that an accused person acted pursuant to an order of a Government or of a superior shall not relieve him of criminal responsibility, but may be considered in mitigation of punishment if the International Tribunal determines that justice so requires’.

34 Judgement, Erdemović, (IT-96-22-A), Appeals Chamber, 7 October 1997, (Erdemović Judgement), paras. 8 and 11.

35 Ibid., para. 19.

36 Joint Separate Opinion of Judges Vohrah and McDonald, Judgement, Erdemović (IT-96-22-A), Appeals Chamber, 7 October 1991 (‘Joint Opinion Vohrah and McDonald’), para. 45.

37 Majority opinion, supra note 34, at 52–55; Separate and Dissenting Opinion of Judge Cassese, Judgement, Erdemović, (IT-96-22-A), 7 October 1997, (‘Cassese Dissent’), paras. 21, 27; see also Paphiti, ‘Duress as a Defence to War Crimes Charges’, (1999) 38 Mil. L & L. War Rev 247.

38 The majority relies on excerpts from case law including references to the ‘sanctity the law attaches to human life’, citing R v. Howe and Others [1987] 1 All ER 771, at 785. and that ‘the law regards the sanctity of human life and the protection thereof as of paramount importance’ citing R v. Gotts [1992] WLR 284, at 292–3. See Joint Opinion Vohrah and McDonald, supra note 36, paras. 23, 71–8; See also Van der Wilt, Justifications and Excuses in International Criminal Law: An Assessment of the Case-law of the ICTY, The Legacy of the International Criminal Tribunal for the former Yugoslavia (2011), 277, at 290–1; Weigend also rightly rejects the majority argument that duress in unlawful killing would have the effect that superiors can confer impunity on subordinates, Weigend, T., ‘Kill or be Killed, Another Look at Erdemović’, (2012) 10 (5)The Journal of International Criminal Justice 1119Google Scholar, at 1227.

39 Majority opinion, supra note 34, para. 77.

40 As noted by Van Verseveld, the majority failed to draw the vital distinction between justification and excuse and had it done so, it may well have reached an entirely different conclusion. Van Verseveld, A., Mistake of Law, Excusing Perpetrators of International Crimes (2012), at 65.Google Scholar

41 Cassese Dissent, supra note 37, para. 49 (emphasis added).

42 Conversely, in a situation such as that presented in the case of Erdemović where, regardless of the accused's actions, he would die along with those he had been threatened to kill, then like Stephens, Cassesse considers that there the defence may be raised. See Cassese Dissent, supra note 37, para. 43.

43 See Cassese Dissent, supra note 37, para. 42 (emphasis added).

44 Also known as the doctrine of ‘inexcusable choice’. G. J. Knoops, Defences in Contemporary International Criminal Law, (2001), at 94, citing the Mignonette Judgment.

45 See also the comments of Lord Coleridge: in referring to the taking of an innocent life he labels necessity as ‘a temptation to murder’ which if allowed as an absolute defence to murder would be ‘the absolute divorce of law from morality’, Mignonette Judgment, supra note 15, para. 287.

46 See also, e.g., La Fave and Scott who state that for reasons of ‘social policy it is better that the defendant faced with the choice of evils do the lesser evil in order to avoid the greater evil threatened by the other person’, La Fave, W. and Scott, A., Criminal Law (1986), at 433Google Scholar.

47 Cassese dissent, supra note 37, para. 43; The Italian cases to which Cassese refers include those of Massetti, Bernardi and Randozzo and Sra et al. all of which concerned executions carried out by fascists in Italy during the Second World War. The German cases were essentially an extension of Control Council Law No. 10, although as noted by Cassese that law had been repealed in 1956.

48 Cassese Dissent, supra note 37, para. 43; see also Van Verseveld, supra note 40, at 66 and Van der Wilt, supra note 38, at 292–3.

49 Separate and dissenting opinion of His Honour Judge Stephen, Judgement, Erdemović, (IT-96-22-A), 7 October 1997 (‘Stephen Dissent)’, para. 52.

50 Majority opinion, supra note 34, paras. 80–1.

51 If, e.g., the accused referred to above was told to cut off someone else's arm or his own arm would suffer the same fate then presumably he would be convicted because the harm with which he was threatened was not greater than that which he caused. It is submitted that this is an obviously unsatisfactory result. Dressler, J., ‘The Exegesis of the Law of Duress: Justifying the Excuse and Searching for its Proper Limits’, (1989) 62 (5)Southern California Law Review 1331Google Scholar, at 1366, note 194.

52 Dressler states that what we mean when we say moral choice is actually whether the actor had a meaningful choice to make, ibid., at 1366, note 194.

53 Dressler provides numerous cogent reasons as to why the categorization of duress as a defence of justification is unsatisfactory. He underlines that it is wrong in principle to base the availability of duress on the lesser evils test and indeed, this is contrary to the common law definition of duress. In illustrating the point, he uses the example of a threat to cut off the accused's arm if he does not commit rape. It is likely, he submits persuasively, that the accused would be acquitted of rape, but not on basis of an analysis of whether the cutting off of an arm is a lesser harm than rape. Ibid., at 1352.

54 See Weigend, supra note 38, 1219, at 1228–31.

55 See Fletcher, supra note 16, at 798.

56 See Ambos, supra note 4, at 1037.

57 The common law position, as highlighted by the majority, is that duress does not negate mens rea. See Majority Opinion, supra note 34, at 70–2.

58 See Cassese, supra note 4, at 952.

59 Eser underlines that if duress were properly categorized as a defence of excuse and not one of justification, it may be seen in a more favourable light as it would not be associated with condoning the acts of the accused, but rather excusing those acts. See Eser, A., ‘Defences in War Crimes Trials’, (1994) 24 Israel Yearbook on Human Rights 201Google Scholar, at 214; See also Ambos, ‘Remarks on the General Part of International Criminal Law’, (2006) 4 Journal of International Criminal Justice 660, at 666.

60 See Eser, supra note 59, at 209, 214; Van Sliedregt, supra note 18, at 247.

61 See Van Sliedregt, supra note 18 at 246; the conflation of justification and excuse, necessity, and duress of which Eser complains is arguably more of an Anglo-American concept than a continental one. The Mignonette case was one of necessity and was considered with reference to justification. The fault of subsequent courts, particularly in the case of Lynch, lay in their defining duress in terms of justification, as it had been in Mignonette, and not in terms of excuse. On the other hand, German law, e.g. draws such distinction, permitting duress as an excuse to be pleaded in cases of murder, and necessity is reserved for all other cases. See Lynch, supra note 2.

62 Fletcher distinguishes this involuntariness, which he terms normative involuntariness, from physical involuntariness. In doing so, he states, in reliance on Hart, that the ‘distribution of punishment should reserved for those voluntarily break the law’ on the basis that we should ‘live in society where we have the maximum opportunity to choose whether we become subject of criminal liability’. See Fletcher, supra note 16, at 802–4; Hart, H. L. A., Punishment and Responsibility, (1968), at 22–24Google Scholar; See also Paphiti who suggests that the distinction between regarding the act as voluntary and involuntary is at the ‘root of the difference between common and civil law systems’, supra note 37, at 274.

63 Cassese, A., International Criminal Law (2008), at 280–1Google Scholar.

64 See Flick trial, paras. 1199–200; Krupp trial, para. 1438.

65 See Dressler, supra note 51, at 1357–9. It needs emphasizing that ‘capacity’ in this sense means ‘possessing the mental capability to’ and is not a concession by the author that an accused under duress has any type of meaningful choice open to him.

66 See Weigend, supra note 38, at 1232–33. Weigend draws a distinction between duress as ‘mere excuse’ contrasting it with insanity and unavoidable mistake which he contends are grounds of exculpation. However, it is more correct to draw a distinction on the basis that duress, due to its normative nature, is an atypical excuse. Furthermore, the author does not make the important distinction between the capacity to make a choice and the free will to give effect to that capacity. In a case of excessive self-defence, as highlighted by the author, the actor has both the capacity and freewill to use whatever force he sees as necessary. This is not the case in duress where the only choice is to comply or be killed, a choice which the author concedes amounts to the use of ‘extraordinary self-control or restraint’.

67 See Fletcher, supra note 16, at 803. In illustrating the gap between harm done and benefit accrued Fletcher uses the example of blowing up a city to prevent a broken finger. He is correct in stating that in such a case it is highly unlikely that the accused would be excused, but perhaps this is not because of the gap between the harm avoided and the harm suffered, rather there would be no defence of duress because the harm avoided is not sufficiently grave to meet the threshold of threat of serious injury or death required to plead duress.

68 See Fletcher, supra note 16, at 804.

69 See discussion of the Flick trial above: Flick and Weiss had increased the production quota beyond what was required which was determinative of the rejection of their plea of duress.

70 Art. 31, ICC Statute.

71 As Eser states ‘Thus, this defence requires less than “justifying necessity” would afford, and . . . more than excusing “duress” would be satisfied with’. See Eser, A., ‘Article 31, Grounds for Excluding Criminal Responsibility’, in Triffterer, O. (ed.), Commentary on the Rome Statute of the International Criminal Court (1999), 537Google Scholar, at 552.

72 Legal commentators have raised doubts about the provision, Eser referring to it as a ‘failed attempt’, ibid., at 550, and Ambos on the basis that it should have maintained separation between necessity and duress. See Ambos, supra note 4, at 1047.

73 See Eser, supra note 71, at 550; McGoldrick, Rowe, and Donnelly, supra note 3, at 274.

74 Ambos points to what he terms as only a ‘terminological’ difference between the ‘necessary and reasonable’ language of Art. 31(1)(d) and ‘proportionate’ as contained in Art. 31(1)(c). See supra note 4, at 1040.

75 See also Van Verseveld, supra note 40, at 66.

76 Art. 31(2) ICC Statute. Furthermore, the case law of the international criminal tribunals offers little assistance on this matter: as noted by Schabas, in Kupreskić, the Appeals Chamber of the ICTY considered the defence of necessity, but held that it was unnecessary to decide whether duress and necessity are the same defence under international law. See W. A. Schabas, The International Criminal Court: A Commentary on the Rome Statute (2010), at 491.

77 Ibid., at 491.

78 See Weigend, supra note 38, at 1224.

79 See also the mens rea requirements of Art. 30 ICC statute.

80 This subjectivity therefore differs from that in Krupp, which avoided any balancing: The test in Krupp required only that the accused hold a bona fide belief in the danger to him, not that he must balance that danger against the harm he is coerced into causing. See Krupp, supra note 28.